Documente Academic
Documente Profesional
Documente Cultură
SRI LANKA
PROCEEDINGS OF THE
PERADENIYA UNIVERSITY RESEARCH SESSIONS
PURSE 2007
Volume 12
Part II - Engineering, Physical Sciences, Humanities and Social Sciences
Editorial Board
Prof. C. Sivayoganathan
Faculty of Agriculture
Prof. V. Thevanesam
Faculty of Allied Health Sciences
Prof. R. Sivakanesan
Faculty of Medicine
PROCEEDINGS OF THE
PERADENIYA UNIVERSITY RESEARCH SESSIONS
PURSE – 2007
ISSN-1391-4111
ISBN-978-955-589-101-1
PUBLISHED BY
THE UNIVERSITY OF PERADENIYA
PERADENIYA, SRI LANKA
Effect of Curry Leaf (Murraya koenigii) in Beef Sausage on Blood Cholesterol Level 9
of Mice
N.S.P. Malmuthuge, J.K. Vidanarachchi, S.M.C. Himali, R. Sivakanesan and H.W. Cyril
iii
Organizational Design Features and Its Influence on Purchasing Performance: The 27
Case of Garment Manufacturing Firms
D.N. Tirimanne and A. Ariyawardana
iv
Antibiotic Activity of Tephrosia purpurea (Fabaceae) and Mimusops elengi 55
(Sapotaceae) against Some Clinical Bacterial Isolates
B.N.L.D. Rangama, C.L. Abayasekara and G.J. Panagoda
Ecto and Endo Parasites of Rodents from Two Selected Sites in Kandy District 86
K. Sumangali, R.S. Rajakaruna and R.P.V.J. Rajapakse
v
Screening for Antibiotic Activity of Tea Catechins Against Pseudomonas aeruginosa 89
and Enterococci
A.A.K. Karunathilake, W.W. Kumbukgolla, E.W.M.A. Ekanayake, V. Thevanesam, N.S. Kumar
and B.M.R. Bandara
Prognostic Significance of the Over Expression of Type I Growth Factor Receptor 113
Family Proteins on the Survival of Patients with Oral Squamous Cell Carcinomas
U.B. Dissanayake
vi
Changes in Periodontal Tissues and Gingival Crevicular Fluid with Friction and Non- 120
friction Methods of Orthodontic Treatment
N.C. Wellappuli, J.K.C. Amarasena, A. Chandrasekera, N. Amarasena, K.A. Kalyanaratne and
S.M.P.M. Ruwanpura
Development of an Antenatal Growth Chart using Maternal BMI in the Second Half 122
of Pregnancy for Sri-Lankan Mothers
S.B.A. Dissanayake, U.B. Dissanayake and W.I Amarasinghe
IL-1α, IL-1ra are Differentially Induced by Candida in Experimental Human Oral 124
Candidiasis
J.A.M.S. Jayatilake, L.P. Samaranayake and L.J. Jin
Influence of Personality Traits on the Attitudes Towards the Teaching Program and 126
the Academic Performance of Dental Undergraduates
R.W. Pallegama, S. Ariyasinghe and H.A.E.D. Perera
The Prevalence of Known Risk Factors for Road Traffic Accidents (RTA) in Kandy 129
Police Administrative Area
I.L. Wickramanayake, G.G.A. Gunasena, H.B.P.W. Wickramanayake and C.D.A. Goonasekera
The Effect of Training on Some Physiological Parameters Related to Physical Fitness 139
in a Population of Sri Lankan School Boys
K.C. Ratnatunga, K.I. Karunaratne, B.K. Dassanayake, I. Silva, A.A.J. Rajaratne and
A.S.P. Mallawarachchi
Vital Functions During and Outcome Following Filtration Based Continuous Plasma 141
Exchange in Guillain Barre Syndrome
S.P. Mudalige, Y.A.A. Jayasena, , G.S.M. Manchanayake, P.K.A.P. Kanankearachchi,
P.V.R. Kumarasiri, H.L.P.S. Dharmapala, S.P.Ekanayaka and C.D.A. Goonasekara
Knowledge on Basic Trauma Care Among the Final Year Medical Students Gained 144
from Clinical Work
M.D. Lamawansa and R.N.R. Arachchi
vii
The Effect of Occupational Exposure to Paddy Husk Dust on the Respiratory Health 146
of Paddy Mill Workers in Sri Lanka
B.K. Dassanayake, K.C. Ratnatunga, K.I. Karunaratne, D. Nandadeva, S.D.I. Nanayakkara
and A.A.J. Rajaratne
Brain Scoop: A Simple Post-Mortem Instrument to Collect Brain Samples for 155
Diagnosis of Rabies
G.S.P. de S. Gunawardena
A Comparison of Clove Oil and Lidocain as Short-term Anaesthetics in Black Ruby 157
Barb (Puntius nigrofaciatus)
S.K.N. Priyangika, S.R.D. Fernandopulle, B.G.S.J. Abeyrathne, M.D. Munasinghe and
H.B.S. Ariyaratne
Poultry Offal Meal as a Substitute for Fish Meal in Feeding Tilapia Cultured in 160
Cages
A.P. Jayasooriya, B. Samarasinghe, S.P. Gunaratne, E.R.R. Wimalasinghe,
N.K. Weerasundera, N. Kudagama and G.S.P. de S. Gunawardena
The Effect of Different Doses of Cobra Venom on Mice as a Model for Investigation 162
of the Tissue Damage
S. Halpage, K.B.J. Kumara, D.P. Nanayakkara, A. Dangolla and J.G.S. Ranasinghe
A Preliminary Study on the Proteins found in the Epididymal Fluid of the Goat 170
J.K.A. Deepanee, R. Subramanium, V.K. Gunawardana, H.B.S. Ariyaratne,
R.P.V.J. Rajapakse, N.A.N.D. Perera, D.C.A. Gunawardana and D.M.S. Munasinghe
viii
Seasonal Variation in the Occurrence of Acanthamoebae Species in the Eyes of a 172
Group of Captive Elephants in Sri Lanka
P.N.K. Wijesekara, K.A.P.A. Bandara, A. Dangolla, J.S. Edirisinghe and I.D. Silva
Intoxication of a Police Working Dog following Ingestion of Explosives (Tnt, C-4 175
and Plastic Explosives)
S.J. Wijesinghe, W.P.T.T. Weerasinghe and I.D. Silva
Entrance Zone Effect on the Sediment Trapping Efficiency in Desilting Tanks of 186
Run-of- River Type Mini-Hydropower Plants
U. Rathnayake, M.R.T.S. Harishchandra and S.B. Weerakoon
Issues Arising from Water Encroachments along Hurulu Wewa Feeder Canal 190
A. Abeynayaka, D.R.I.B. Werellagama and G.N. Paranavitana
Study of Cassava Starch Hydrolysis by α-Amylase and an Airborne Mould Culture 193
W.A.L.D. Wickramasinghe, S. Sotheeswaran, U.S.K. Weliwegamage, S. Rajapakse and
C.S. Kalpage
ix
Design of a Paddy Husk Charcoal Adsorption Unit for the Treatment of Parboiled 199
Rice Processing Wastewater
S.M.W.T.P.K. Ariyarathna, C.S. Kalpage and R. Shanthini
Use of Light Absorption in the Fabrication of a Device for Measuring Ambient Black 205
Carbon Concentration
N.M.N.V. Bandara, P.S. Palugaswawa and M.A. Elangasinghe
Strength and Ductility Evaluation of Rigid-Frames Using Finite Element Analysis 211
K. A. S. Susantha and T. Aoki
Equivalent Electric Network Model for Ionic Polymer Metal Composite Dynamics 224
B.G.L.T. Samaranayake and A.M.U.S.K. Alahakoon
Full Instruction Encoding for Heterogeneous Multi Pipeline Application Specific 230
Instruction-Set Processors
S. Radhakrishnan and R.G. Ragel
Design of an All-In-One Billing Meter for Three Wheelers in Sri Lanka 238
B.G.L.T. Samaranayake
x
Memory Efficient Algorithm and Closed Form Formulas for Computing Odd Magic 241
Squares of Higher Order
P.C. Perera
Controller Design and Implementation for Cost Effective Mobile Robots 249
I.B. Wijayasinghe and B.G.L.T. Samaranayake
Error Correction Model for Predicting the Carbon Dioxide Emissions of a Coal-based 255
High-income Economy
R. Shanthini and K. Perera
Limitation on Connecting Mini Hydro Power Plants to the Sri Lankan Power System 263
Network: A Case Study at Balangoda Grid Substation
A.I. Weerasekera, A. Arulampalam and J.B. Ekanayake
Modelling Sri Lankan Power System to Study the Effect on Dynamic Stability with 272
Large Scale Wind Power Integration
R.H.B. Ekanayake, A. Arulampalam and J.B. Ekanayake
Structural Studies of Plant Pigments Used in Dye Sensitized Solar Cells 275
F.M. Hussain, V. Sivakumar and K. Tennakone
xi
Deriving Shrunken Estimators for the Variance in One-Parameter Natural 281
Exponential Families
A. Laheetharan and P. Wijekoon
Major and Trace Element Composition of Bottom Sediments of the Malagane Tank: 284
Relationship with Sources
K. Mahatantila, R. Chandrajith and H.A.H. Jayasena
Integer Solutions of Non - Linear Diophantine Equations Using Continued Fractions 288
A.G.K.N. Alupotha and A.A.I. Perera
Chemical Reactivity of Mica, K-Feldspar and Apatite in Organic Acids and Water 298
T.H.N.G. Amaraweera, A. Pitawala and G.W.A.R. Fernando
Synthesis of a Novel Copper (II) Complex Capable of Trapping Hazardous Cations 304
Present in Water
A.M.K.S.P. Adhikari and M.Y. Udugala-Ganehenege
xii
Preliminary Investigation of the Liquid Crystal Behaviour of β-sitosteryl-β-D- 318
glucopyranoside
G.L.A.D. Liyanage, A.D.L.C. Perera and D.N. Karunaratne
Equilibrium Studies on Cadmium (II) Interactions with Aspirin and Vitamin C in 320
Aqueous Medium
H.H.M.C.U. Herath and C.V. Hettiarachchi
Preparation of a Zoning Plan for the Madu Ganga Special Management Area 327
E. Vidanapathirana, P. Wickramagamage and S.P. Nishshanka
Farmers` Perceptions Towards Climate Change and Farm Management Practices 332
K.W.G.R. Nianthi
The Frankfurt School of Social Research and Critical Theory, with Special Reference 342
to Max Horkheimer
M. Rajaratnam
Children’s Posttraumatic Emotional Distress and Coping After a Natural Disaster: 344
The 2004 Tsunami
G.R.K. Dissanayake and V. Ganepola
xiii
Philosophical and Systematic Analysis of Diotima's Definition of Love and Its 346
Relation to Plato's Theory of Forms in the Symposium
A. Samaratunge
Euripides’ Alkestis “A Drama of a Devoted Wife?” Translation from the Original 348
Greek into Sinhala with an Introduction and Commentary
C.S.M. Wickramasinghe
Issues in Translating 'Sexuality' into Sinhala: Inadequacy of the Terms Used in the 350
Sinhala Translation of Lady Chatterley's Lover By D.H. Lawrence
D.M. Kumarasinghe
Foreign Direct Investment and Economic Growth in Sri Lanka: A Causality 355
Approach
N. Balamurali
The Causes of Domestic Violence and Its Impact on Rural Women Victims 357
J.A.P.S. Perera
External Intervention and Conflict Outcome: Analyzing the Effects of Intervention 359
Strategies in Sri Lanka
M.M. Ihjas
A Study on Gender Differences in Food Habits and Sport Performance among 363
Secondary School Boys and Girls in Sri Lanka
W. Senevirathne and P. Sethunge
A Study of Western Musical Techniques Used in Sinhala Light Songs During the 365
20th Century
M. Alawathukotuwa
xiv
Message from the Vice Chancellor
Peradeniya University Research Sessions 2007
Peradeniya University Research Sessions (PURSE) is an annually held event and this year it is
scheduled to be held on the 30th November, 2007. This has become an important academic event of
the University calendar.
Generation of new knowledge through research into important problem areas by both university
teachers and students is part of the higher education process. In order to facilitate the teachers and
students to take up this challenge, the University of Peradeniya conducts over two hundred
internationally and locally funded research projects, annually. The PURSE provides an ideal
forum for staff and student researchers to present their research findings to a wider audience
enabling their work to be reviewed evaluated by a learned panel of judges. The experience gained
during PURSE generates confidence amongst the researchers and help them to perform better in
their future endeavours. Therefore, continuation of PURSE in a sustainable manner is of utmost
importance to the University of Peradeniya.
I wish to express my deepest gratitude on behalf of the University of Peradeniya for Honourable
Minister Dr. Sarath Amunugama for accepting our invitation to be the chief guest of the PURSE,
2007. Organizing the PURSE is not an easy task and the members of the Organizing Committee
have taken up the responsibilities assigned to them. The request of papers from all faculties and the
review and publication procedures take a substantial time of all members, referees and editors.
This year’s Chairperson, Dr. S.D. Dewasurendra, took up the challenge of coordinating the
activities of the research sessions with dedication to complete the task assigned to him. I take this
opportunity to compliment him on this occasion. I am also very grateful to the members of the
organizing committee comprising of academic and non- academic staff and students of all faculties
for their invaluable support and assistance in organizing the annual research Sessions 2007. I take
this opportunity to wish all the paper presenters and participants a rewarding day.
xv
xvi
Message from the Chairperson, Organizing Committee
Peradeniya University Research Sessions 2007
Peradeniya University Annual Research Sessions are being held for the 12th time when the
university completes its 65th year. Being the most comprehensive university with the oldest
tradition in the country, these sessions must certainly mean something special. The line-up of
academics and their credentials are impressive by regional standards and, one would expect to find
a hive of research and development activity in the midst of an enlightened group as ours. Certainly
such activity has to reflect amply on the research outputs as presented in its annual research
sessions.
We have some 160 peer-reviewed papers being presented, covering most of the disciplines in the
eight academic faculties. However, we must not hesitate to confront the reality that a large
proportion of what is being presented report work carried out in basic undergraduate
laboratories/facilities.
Research is an inspirational activity which explores the limits of possibility, exercising the creative
faculties to the limit. When economies operate at below subsistence levels, like in the case of ours,
there are overwhelming needs of a very basic nature in every sphere of human activity. This
coupled with the fact that there is a continuum of levels of development visible in different
countries, which have suddenly come to our very door-step through hyper-quick communication
media, open up almost limitless possibilities to those who seek some activity to keep occupied.
Therein, the craftsman, very justifiably, finds a fertile activity field and the versatile student too,
simple, real-life applications to sharpen newly acquired knowledge and skills. However, it is very
unlikely that such activity would reach the threshold of breakthroughs that form the back-bone of
any nation in the vanguard.
It is not possible to force creativity out of people, and all those who have the grasp of fundamentals
and the tools to manipulate them cannot necessarily be expected to use such knowledge to generate
inspirational levels of research activity. Most seem to content themselves in rejoicing exploits,
worthy though, of the versatile student in his act of sharpening teeth. No wonder then, that we are
where we are and we see the horizons that we see. Let us not fool ourselves. How many really
come looking for advanced learning to our shores? How many times has the country been able to
take advantage of its own research findings to create a difference to its own lot?
Then how, one would ask, does it happen where it really does? Arguably, when those who can and
are willing get what they need to exercise their inspired activity. And, what exactly is this that they
need? They need their domain, the domain that contains the facilities for their activity and the
freedom to explore and create. And how would they get this domain? In nations at advanced
degrees of development in the respective spheres, the domains are being created, either through the
needs of private enterprise or through the planned research & development programmes of the
state. The necessary funds come through regeneration. In a country where resources are scarce in
both these forms, the only possibility remains the tapping of resources of outside domains:
basically, bringing in work that can be carried out locally and delivered to the overseas domain,
securing precious funds for local capacity building. This could continue till local capacity builds up
to sufficient levels. Otherwise it would be inevitable to stay within the vicious circle of ignorant
bliss.
The task is not easy by any means, because gaining visibility in overseas domains and developing
the requisite level of confidence would be an immense barrier to break through. However, there are
xvii
things that we could do to improve our chances in this venture. To start with, let us not despise the
enterprising academic for building his own domain. Let us facilitate the networking of such
enterprising academics into functional research groups. Let us identify the difference between
research and resulting creation of new knowledge and the activities of teaching and forming young
graduates and undergraduate project work which is meant to sharpen their skills in the use of newly
acquired knowledge. Let us understand clearly that at any given moment of time, there are these
two focuses and that correspondingly there are two distinct groups of adepts. Let us also understand
that the former group could potentially perform better than the latter in its own task. Let us not
permit the majority power of mediocrity stifle the efforts of the researcher to kindle the little flame
of inspiration that could light up the lot of the majority too.
Having said that, we must place on record our deep appreciation of all those who toiled hard to
make PURSE-2007 another significant landmark in our pursuit of excellence. The researchers
presenting papers, the editor and the editorial committee are to be commended first. The unfailing
support and guidance of the Vice Chancellor have been a source of strength. The commitment
shown by the different committees and their chairpersons, personnel from the Council & Academic
Unit and University administration in general once again demonstrated the hidden strengths we can
draw upon. We also note with gratitude, the donations by the National Science Foundation and the
Postgraduate Institute of Science towards the printing of conference proceedings. We are grateful
to Honourable Minister Dr. Sarath Amunugama for accepting our invitation to be the Chief Guest
and our Chancellor, Dr. Premadasa Udagama for accepting to grace the occasion as Guest of
Honour. We record our gratitude to Professor G. P. Karunaratne for accepting our invitation to
deliver the Guest Lecture.
xviii
PERADENIYA UNIVERSITY RESEARCH SESSIONS PURSE – 2007
ORGANIZING COMMITTEE
Faculty Representatives
Faculty of Agriculture Dr. D.N. De Costa
Mr. B.M.L.D.B. Suriyagoda
xix
xx
PANEL OF CO-EDITORS
xxi
xxii
Guest Lecture
Is Research Needed?
G.P. Karunaratna
International Consultant to Southern Transport Development Project, Sri Lanka
Adjunct Professor, National University of Singapore
In a research community such as a university, the answer to the above is generally clear. Many
university professors undertake post-graduate training and research as part of their profession. But
should this be limited only to the beginning of the career?
If the meaning of research is to ‘search’ for knowledge then there is always something different to
look for. Should the ‘knowledge’ be limited to theory? Is the acquired knowledge useful only for
the researcher who raised the level of knowledge? Should he be complacent that his knowledge is
expanded? Will his results lead to knowledge being shared among the community? Should the
shared knowledge lead to welfare of and to uplift the society? Taking a more humanistic approach
it would be ideal to focus one's energy for searching and advancing knowledge for the benefit of all
species, far and near.
Following the perception expressed above the research finding should advance the intangible
knowledge and/or tangible applications stemming there from. The former eventually may lead to
‘tangible’ applications such as spiritual emancipation. The latter provides a seeding ground for
development work in all spheres.
Much of the effort devoted for research is based on the human mind. Certainly sharper the mind
better would be the analytical assessment and the grasping power of the work. Those who need
equipment and associated material for research have to depend on funding which may not be
available readily. Funding agencies are sought after by the researchers at this stage. These may
constitute public or private organizations and institutions, based on the need for a solution
(knowledge that is necessary) for a given problem. Private organizations thrive in the West as
funding agencies through reasonable assessment of the potential commercialization of research
finding.
Public organizations, such as government institutions, provide little to large-scale support in the
West, whereas only a handful of governments in the East encourage research. Such applications on
a large scale include US space program and defense. Private sector dominates in pharmaceutical,
biological and electronic industries among thousands of different research areas. Thus it would not
be easy to imagine how this world would be without the research conducted by all the humans
since the inception of the world, be they ancestors or not. Therefore, whether the finding is
stoneware in the prehistoric time or Edison’s electric bulb or nanotechnology or space structures, to
take a minuscule of examples, research has yielded knowledge in a practical sense for the benefit of
the mankind. Occasional use of such findings towards destruction is disheartening for many.
Choice of a viable topic is researcher’s option, and sometimes the funding agency that needs results
for an unresolved issue. Funded research should produce achievable targets. They should
preferably be completed within the stipulated time frame and budget. However it would be more
enjoyable to witness the application of one’s research findings if the opportunity arises.
xxiii
At the time the speaker accepted an offer to become a professor at the University of Singapore
funding for research was an attractive thought. No sooner he arrived with his young family he was
requested to provide advisory services to the panel of engineers for the construction of the Changi
International Airport in Singapore replacing the much congested smaller airport in the island of
525 km2. The natural sand available in the sea bed was dredged for reclaiming a 750 ha of new land
upon which the new airport facilities were to be built. Sand was placed on 40 m deep marine clay
in the sea bed which was to compress about 3 m under 10 m thick new land. Since it takes more
than 30 years to accomplish this settlement if left untreated, it became obvious that the soft clay
must be pretreated so as to preclude future settlement. The only available band drain to reduce time
of treatment at that time was made with a plastic core and a paper filter. During installation it was
observed that the paper filter tended to tear off in the air when gusty wind blew at 40 km/hour. The
speaker was able to assemble a completely natural band drain composed of coconut coir and jute
fibre that surpassed the tensile strength of all band drains.
Being a third world country, in mid seventies, Singapore University had to seek the services of a
London Solicitor for patenting the product which became the first patent of the Singapore
University. Subsequently the product was tested in the same marine clay with identical
performance of the other band drain. The patent has expired after 20 years and stands to benefit the
community of civil engineers. It has been used in Singapore, Malaysia, Japan and Sri Lanka on
commercial projects for soft ground improvement.
Enjoyment and monetary return from such patented innovation are benefits of the research. When
the patent expired the development becomes people’s inheritance. Benefits accrued by the
community gives an additional enjoyment which is in the realm of sharing knowledge. Hence
research is to advance knowledge in all fields, be they intellectual or practical. The community
should be able to share the benefits as part of the research effort.
xxiv
Part II
Where, C = U + gy and R is hydraulic radius. Where, yb is the water depth in the base flow,
yp is the water depth at peak, u*b is the shear
When passing a hydrograph in the channel, it velocity in the base flow and DT is the duration
has been shown that the rate of change of water of hydrograph. The HYDP of test cases varied
level is more significant than that of mean between 3.17×10-4 (most unsteady) and
velocity and, as a result, the contribution of the 1.52×10-5 (least unsteady). As a typical result
179
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
m a x ( Δ u * /u *4 )
base flow
Figure 2 compares the u* values based on the
20%
10%
8.0 6.0
HYDP= 1.25E-04
y 0%
7.0 u*steady (u*s) 5.0 0.0E+00 5.0E-05 1.0E-04 1.5E-04 2.0E-04 2.5E-04 3.0E-04 3.5E-04
u* (cm/s)
u* unsteady HYDP
y (cm)
6.0 4.0
References
Figure 2. Comparison of shear velocities
computed by steady and unsteady Henderson, F.M. (1996) Flood waves in
prismatic channels, Journal of Hydraulic
state formula Engineering, ASCE, 89(4), 39–67.
Tu, H. and Graf, W.H. (1993) Friction in
It can be seen from Figure 3 that the maximum unsteady open-channel flow over gravel
difference in u* computed by the two methods beds, Journal of Hydraulic Research, 31(1),
is generally larger in the rising limb than that in 99–110.
the falling limb and that it always remains less
180
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
was designed and fabricated using perforated velocity of rain drops (V), soil characteristics
PVC pipes and a nozzle as shown in Figure 1. (K), support factor to represent any erosion
To generate raindrop distribution similar to
181
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Figure 4. Variation of X7 Vs X6
− 1 . 46
⎡ d g ⎤ ⎡ Iρ d s ⎤
q = 4 . 4 E 9 ⎢ 2s 2 ⎥ ⎢ ⎥
⎣S I ⎦ ⎣ drc ⎦
250
200
q experimental
182
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Main V3
V2 Supply CT – Collection tank
DG – Depth Gauge
FGB – Fixed gravel bed
V1 LA SF MGB – Movable gravel bed
DG LC LA – Lifting arrangement
LC – Load cell
P2
P1 SF – Sediment feeder
ST
ST – Sediment trap
UT UT – Upstream tank
MGB FGB
CT P1, P2 –Pumps
V1, V2, V3 –Valves
Figure 1: Experimental set-up
183
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
⎡ δy ⎤
yp
( DT)g 2
1
Quns ⎢ yb y p y ρ δt ⎥ (1)
= F⎢ , , , , S, , ⎥
ρs gd ( 3
) 1
2
⎢ d50 d50 yb ρs g(DT) yb 1
d50 2 ⎥
⎣⎢ ⎦⎥
50
K a b ir ( 1 9 9 3 ) K a b ir ( 1 9 9 3 )
150 N e w F o r m u la 150 N e w F o r m u la
100 100
50 50
0 0
0 20 40 60 80 100 0 20 40 60 80 100
T im e / (s ) T im e / (s )
Run No. M B 56 R un N o . M B 70
250 200
Bed load transport / (g/m/s)
K a b ir ( 1 9 9 3 ) K a b ir ( 1 9 9 3 )
200
N e w F o r m u la 150 N e w F o r m u la
150
100
100
50
50
0 0
0 20 40 60 80 100 120 0 20 40 60 80 100
T i m e / (s ) T i m e / (s )
184
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
185
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
186
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
90
Q=15 l/s
85 Q=20 l/s
75
70
65
60
55
50
0 10 20 30 40 50 60 70 80 90
Expanding Angle (Deg)
However, as the flow entrance angle is covering different sediment sizes are
decreased, the construction cost of the de- recommended.
silting tank increases and therefore the
optimum entrance angle can be decided only References
after a cost analysis. When the discharge
through the tank is below the design value the Janssen, R.H.A. (2004) Analysis and design of
sediment removal efficiency increases.tank sediment basins, 8th National Conf. in
increases and therefore the optimum entrance Hydraulics and Water Engineering,
angle has to be decided only after a cost Institution of Engineers, Australia.
analysis. When the discharges through the tank Ranga Raju, K.G., Kothyari, U.C, Srivastav, S.
is below the design discharge, the sediment and Saxena, M. (1999) Sediment removal
removal efficiency increases. efficiency of settling basins, J. Irrigation
and Drainage Engineering, ASCE, 125(5).
Conclusions Ranga Raju, K.G. and Kothyari, U.C. (2004)
Sediment management in hydroelectric
The entrance zone of the de-silting tanks has a projects, Impact of silt on hydro turbines,
considerable impact on the trapping efficiency Himalayan Small Hydropower Summit,
of the sediment in the de-silting tanks. The Dehradun, India.
trapping efficiency of the tank increases with Singhal, S.K. and Ratnendra, S. (2006) Impact
the reduction of the expansion angle of the of silt on hydro turbines, Himalayan Small
entrance zone in the de-silting tanks. The Hydropower Summit, Dehradun, India.
optimum expansion angle is found to be about
10 degrees. Nevertheless, further investigations
187
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
188
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
45.00
30.00
25.00
20.00
15.00
10.00
5.00
0.00
ril
y
er
ch
ry
er
ar
Ap
mb
mb
ua
ar
nu
M
br
ce
ve
Ja
Fe
De
No
Month
Figure 1. Simulated and observed stream flow of the Maha Oya for the calibration period
25.00
Observed discharge
20.00 Simulated discharge
Discharge (ft /sec)
15.00
3
10.00
5.00
0.00
ne
ay
Ju
M
Month
Figure 2. Simulated and observed stream flow of the Maha Oya for the verification period
189
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Issues Arising from Water Encroachments along Hurulu Wewa Feeder Canal
190
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
pumps). Once a siphon is set up, water HFC, ID, and MASL discussed and agreed on a
generally is allowed to flow out continuously water management plan, providing more water
to sustain the siphon function. This consumes a to HW. HW farmers now provide labour for the
considerably larger amount of water than a well maintenance of the upstream canals and the
designed outlet irrigation canal, which can also labourers also serve to police unauthorized
be used as a discharge control device. tapping. Yet the informal agreement did not
consider improving water controlling structures
or land use planning of the area. When MASL
officers removed the siphons, local politicians
forced them to put the siphons back. This is a
case of the state acting against the promises
made to the donors, to provide water to HW.
Measures to prevent recurrence of such things
are important in future schemes.
Conclusions
Improper management including poor
catchment management and land use planning
reduces the productivity of water resources and
land. Future land developments must take these
Figure 2. Water encroachments facts into account at project planning stage
itself. Water management policy must be
The design flow rate of 4.3 m3/s gives a water coupled with land management and land
flow of 44.6 MCM (million cubic metres) for a development.
four month cultivation season (“Yala”) and
55.7 MCM for a cultivation season of five References
months (“Maha”). Yet the HW water budget Abeynayaka, A., Werellagama, D.R.I.B.,
shows a very low inflow to the HW (Figure 3) Yatawara, S.C.B. and Paranavitana, G.N.
starting from 1985. On many occasions the (2007) Lessons on water management; Past
HFC flow is far below the designed flows due to the Future, International forum on water
to water pilferage. environmental governance in Asia,
Thailand, 26-32.
Owing to water scarcity in the HW command
area, the farmer organizations (FO) of HW
have campaigned to adopt a water management
plan for the HFC. In late 1990s’ FO of HW and
45
40
35
HFC inflow /(MCM)
30
25
20
15
10
5
0
19 5
19 6
19 7
88
19 9
19 0
91
19 2
19 3
94
19 5
96
19 7
19 8
20 9
00
20 1
20 2
03
20 4
05
8
8
8
8
9
9
9
9
9
9
0
0
0
19
19
19
19
19
20
20
Season
191
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
192
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
193
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
4 0C in the refrigerator and were sub-cultured was monitored after liquefaction and
twice per month. saccharification. Results are given in Table 1.
Samples are denoted by an English letter and a
Results and discussion number. The letter refers to starch
concentration and the number indicates the
When starchy slurry is heated above 60 0C, the thermal treatment process. Conversion of
starch containing granules get swelled and starch to glucose in each test sample is shown
rupture and are gelatinized. The gelatinized in Table 2.
cassava starch gets partially hydrolyzed rapidly
by heat stable α-amylase in liquefaction. The According to Table 2, the highest glucose
partially degraded starch chains are called conversions were found among samples ‘C’
dextrins, which are suitable as starting (above 54%) in which the initial starch
materials for the later steps in ethanol concentrations are only 5%, while samples with
production. highest initial starch concentrations (A) have
reported the minimum Conversion Yields (less
For this study cassava powder was prepared than 21%). The best result; 57.6% was obtained
from cassava root by a wet milling process. with sample C3 which was prepared by the
The average cassava powder yield was method explained before.
375 g/kg of fresh cassava. It was estimated that
the starch and glucose contents in dry cassava Decreasing glucose conversion yield with
powder are 19.55 and 3.48 g/100g, increasing initial starch concentration is
respectively. attributed to a possible inhibition activity
taking place in starch solutions. Vasquez et al.,
Hydrolysis of cassava starch was studied under (2004) reported that plants contain anti
none different conditions using α-amylase as bacterial and anti fungal substances (such as
explained before; three different starch polyphenolics, cyanogenic compounds and
concentrations under three thermal processing peroxidases) in order to inhibit the bacterial
methods. The selected sample size was 30 ml. and fungal attacks. Therefore, it seems that low
After the liquefaction stage, the air borne concentrated starch solutions are more suitable
fungal culture which was identified as for the
Aspergillus niger was introduced to each of the
sample. The change in glucose concentration
Saccharification
After 2 days 0.22 0.17 0.20 0.17 0.14 0.17 0.11 0.13 0.10
After 5 days 0.22 0.22 0.23 0.17 0.18 0.16 0.15 0.15 0.16
Sample A1 A2 A3 B1 B2 B3 C1 C2 C3
Saccharification
After 2 days 19.8 15.3 18.0 30.6 25.2 30.6 39.6 46.8 36.0
After 5 days 19.8 19.8 20.7 30.6 32.4 28.8 54.0 54.0 57.6
194
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
195
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
Methodology
Chlorpyrifos (O,O-diethyl O-3,5,6-trichloro-2-
pyridyl phosphorothionate) is a highly active, Five random water samples were collected
non-systemic pesticide applied to control a from the Kiwullinda Oya in May 2007, which
wide range of soil pests and arthropods on was during off-peak pesticide application
various crops including vegetables, citrus, period. On June 16th, 2007, which fell within
cereals, maize, potato and tobacco. the peak pesticide application period, 14 water
Chlorpyrifos containing pesticides are samples were collected over a 3 km stretch of
commercially available as Durban, Lorsban, Kiwullinda Oya. The sampling locations are
Syren, Commando, Judo, Bullet, Chlofos, shown in the schematic diagram of Figure 1.
Unifos, Loseban, Patas, Lidoband, Dhanwan, From each sampling point, a 600 ml sample of
Dorson and Omexan in the local market. This water was collected in a 1-litre amber glass
chemical is moderately toxic following acute bottle with a glass lid, and the samples were
oral, dermal, and inhalation exposures (Zhao et transported to the Analytical Laboratory of the
al., 2004). The embankment of Kiwullinda Oya department within 4 hours of the sample
(a small stream, 9.5 m wide and 1 m average collection.
o
Samples were refrigerated below
depth), running through the Marassana area in 5 C until the time of analyses. Water samples
Kandy District and feeding the Victoria were analyzed for the presence of chlorpyrifos
reservoir, is densely cultivated with vegetables using a High Performance Liquid
and paddy. Chlorpyrifos
o
, with 40 days half-life Chromatograph (HPLC) method (CIPAC 1 C,
in buffers at 30 C (WHO, 2002), is one of the 1985). In this method, the target compound,
popular pesticides in use in Marassana. chlorpyrifos, was extracted using
Therefore, the probable presence of dichloromethane. The extract was then
chlorpyrifos in surface and ground water at concentrated and dried by means of blowing
Marassana, especially in the Kiwullinda Oya, ambient air, which is at 35 0C or below, over
has long been suspected. The objective of this the extract. The residue was then dissolved in a
research was to quantify chlorpyrifos mobile phase (acetonitrile). The solution so
concentrations (if present) in the water of prepared was injected to the HPLC and the
Kiwullinda Oya stream basin. It is important peak area readings were obtained. These
because, according to the discharge water readings were then converted to concentrations
quality standards (CEA, 1990), no pesticide by use of calibration charts, which were
should be detected in surface water even in generated in this study itself.
trace amounts.
inlet
inlet inlet inlet inlet inlets
inlet
1 2 3 7 10
4 5 6 8 9 11 12
13 14
Sample Anicut across
points the river Reservoir
inlet of stream
Figure 1. Schematic diagram of the 14 sampling points and the relative positions of inlets feeding
significant amount of water to the investigated stream
196
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
2.5
30 Concentration in extract, μg/l
(b)
25 (a) 2
Peak area, mAU
20 1.5
15 1
y = 0.7293x
10
y = 0.0316x 0.5 R2 = 0.9913
5 R2 = 0.9386
0
0 0 1 2 3 4
0 500 1000 Concentration in wastewater, μg/l
Chlorpyrifos Concentration, μg/l
Figure 2. Calibration curves for (a) HPLC peak area versus chlorpyrifos concentration, (b)
chlorpyrifos concentration in extract versus chlorpyrifos concentration in wastewater
Chlorpyrifos concentration in
3
2
2.5
the stream , μg/l
2 11
1
1.5
13
1 5 14
4 7 9
0.5 8
3 6 10 12
0
0 50 250 500 551 750 795 1045 1138 1222 1472 1722 2500 3000
Distance along the stream, m
Figure 3. Variation of chlorpyrifos concentration along the stream on a day during peak pesticide
application period
197
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
The chlorpyrifos concentration in the water pesticide application periods. The cross-
sample collected from one of the community sectional assessment of the chlorpyrifos
wells in this area was also as high as reported in this paper is being repeated at
0.79 µg/l. The highest chlorpyrifos present to study the variations in the
concentration was detected at the sampling concentration of chlorpyrifos in the stream
point 2 where several irrigated channels and with time.
field surface runoffs enter the stream.
Comparison of concentration data (Figure 3) Acknowledgements
and the sampling locations (Figure 1)
provides a telling relationship. Local peaks RG/2006/33/E, University of Peradeniya, is
in concentrations are seen at points 2, 5, 7, 9 acknowledged for funding. Special thanks
and 13, where main irrigated canals enter the are due to the Pesticide Registration Office
Kiwullinda Oya. In the span of sampling of the Department of Agriculture and to the
points 13 to 14, surface runoffs from technical staff at the Department of
cultivated lands located at high elevations Chemical and Process Engineering for the
enter the stream water at no specific location, technical support provided.
which keeps the chlorpyrifos concentration
approximately a constant. Stream water at Reference
points 10 and 12 are diluted with large CEA (1990) General standards for discharge
quantities of fresh water entering via of effluents in to inland surface waters,
channels, which are not running through Act no 47 & 56, Part 1, section 1,
cultivated land. It was observed that the both Gazette extraordinary of the Democratic
sides of stream banks between point 10 and Socialist Republic of Sri Lanka -
11 were cultivated with vegetables. The high 1990/02/02.
peak at point 11 is therefore, attributed to CIPAC handbook 1 C (1985) Collaborative
accumulation of pesticide contaminated International Pesticide Analytical
water seeping through the stream banks. council, 2028-2031.
WHO (2002) Geneva WHO Specifications
Conclusions And Evaluations For Public Health
Though, according to Sri Lankan Pesticides, Chlorpyrifos O,O-diethyl O-
regulations, not even trace amounts of 3,5,6-trichloro-2-pyridyl
pesticides should be detected in discharge phosphorothioate, World Health
water from industrial and agricultural Organization publication, retrieved on
applications high concentrations of 08/03/2006 from
chlorpyrifos (up to 2.48 µg/l) have been http://www.who.int/whopes/ quality/en/.
observed along a 3 km stretch of Kiwullinda Zhao, Q., Gadagbui, B. and Dourson, M.
Oya, a small stream running through the (2004) Lower birth weight as a critical
cultivated lands in Marassana. It is apparent eVect of chlorpyrifos: A comparison of
that chlorpyrifos entered the stream with the human and animal data, Regulatory
water from irrigation channels and surface Toxicology and Pharmacology, 42, 55–
runoffs. It is found that the presence of 63.
chlorpyrifos in surface waters is not limited
only to the peak, but also to the off-peak
198
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Design of a Paddy Husk Charcoal Adsorption Unit for the Treatment of Parboiled
Rice Processing Wastewater
Introduction
Materials and methods
Soaking paddy in water until it attains a
moisture content of 30 to 35% by weight is The wastewater sample was collected from
the key unit operation in parboiled rice the soaking tanks of a commercial rice mill
processing. Wastewater exiting the soaking situated in Kalawewa, Mahaweli H zone, Sri
tanks of the rice mills generally registers Lanka. The initial COD of the sample was
chemical oxygen demand (COD) at a very measured to be 4500 mg/l and the pH was
high value, for example, 4500 mg/l, which is 4.5. PHC was also collected from the same
well above the permitted COD (< 250 mg/l) mill, and sieved to separate particles of size
for wastewater discharged to surface waters 0.2 to 0.5 mm, which were used as the
in Sri Lanka (CEA, 1990). However, owing adsorbent in this study without subjecting to
to the location of a majority of rice mills in any pre-treatment. A column of 150 cm
the relatively environmentally-healthy rural height and 2.7 cm diameter was fabricated
areas of Sri Lanka, no stringent effluent with PVC conduit pipe. Of this column, a
control is imposed on these rice mills. As a height of 120 cm was packed with 75 g of
result, a good number of the rice mills in Sri PHC, which was the adsorption media.
Lanka discharge their untreated rice mill Wastewater was passed through the bed at the
effluent (RME) into surface waters, and experimentally determined optimum
thereby polluting them. superficial velocity of 5 mm/min. Samples of
the column effluent were collected at the
Since some of the pollutants in the RME are bottom of the bed at selected time intervals,
adsorptive, adsorbents such as activated and their respective COD were measured. All
carbon and paddy husk charcoal (PHC) could experiments were at least duplicated. The
be used to selectively remove these data so obtained were used to construct the
substances from the wastewater. PHC, breakthrough curve, showing the profile of
containing about 65% carbon is readily the column effluent COD against time; and
available at rice mills. It is a solid waste the principles of adsorption column design
generated from the boilers which use the were then employed for scale up for the
paddy husk waste originating from the industrial unit (Christie, 1993).
milling section as fuel. In boilers, about one
fourth of the paddy husk mass gets converted Results and discussion
into PHC during combustion. It has been
shown that PHC could successfully remove Figure 1 shows the breakthrough curve
COD, pH, odour, colour and TDS in RME obtained with a 5 mm/min superficial flow
(Ariyarathna et al., 2004). velocity with the laboratory-scale PHC
adsorption column. The ratio of COD of the
In this study, a detailed investigation was wastewater leaving the column (c) to the
carried out to design PHC adsorption units COD of the wastewater entering the column
for commercial-scale parboiled rice (co = 4500 mg/l) was plotted against time.
processing mills of capacity up to 10,000 The long time taken for bed saturation (about
kg/day of paddy. A design curve, constructed 3600 min) shows that the rate of adsorption is
with the help of experimental data obtained relatively slow.
from laboratory-scale PHC adsorption unit,
was used to scale up for an industrial-scale The levelling off of the breakthrough curve
fixed bed PHC adsorption column which shown in the Figure 1 close to c/co =1 implies
could reduce COD from 4500 mg/l to 500 that the entire bed of PHC has come to
mg/l. equilibrium with the wastewater fed to it. In
which case, according to theory, the total
capacity of the PHC packed bed is
proportional to the area between the
199
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
breakthrough curve and the vertical axis Table 1 summarises the results obtained for
(denoted by A1 = 32). A value of COD is the laboratory-scale column and the
chosen for the column effluent, at which the corresponding scaled up values for the
column operation would cease. This value of industrial-scale column for an operating time
the COD is known as the breakthrough COD. of three days using the procedure outlined in
It was chosen be 225 mg/l which is below the Christie (1993). Both the columns must have
permissible COD level of discharge water the same input and output CODs, as well as
quality into surface water. Since the initial flow velocity during scale up.
COD of the wastewater was 4500 mg/l, the
ratio at the breakthrough point was 0.05, The industrial columns were designed for the
which was reached 240 min (i.e. 4 hours) treatment of 33,000 l of wastewater, which is
from the commencement of the experiment, the maximum volume of wastewater
as shown in the figure. The area up to this produced per day by the rice-mill (chosen for
point gives the usable capacity of the bed this study). Therefore, three identical
(denoted by A2 = 4). The ratio of the two adsorption columns (throughput of each is
areas, A2/A1, which is the fraction of the total 11,000 l/day) are needed to be operated
bed capacity or length utilized up to the break simultaneously to treat the above wastewater.
point, becomes 0.125. Hence, usable length Table 1 shows that 677 kg of PHC is required
for selected bed was calculated to be 15 cm to treat 33,000 l of wastewater with the bed
(= 0.125 x the PHC bed height). The being operated for 3 consecutive days.
unusable bed height which remains a constant
as long as the superficial flow velocity
remains a constant, was then 105 cm.
C/Co
0.05
Figure 1. Breakthrough curve obtained with the laboratory-scale PHC adsorption column.
200
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
201
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
Table 1. Selected process parameters and
Tea is the main agricultural export of Sri Lanka
contributing around 15% to the total exports of levels
the country. Sri Lankan tea is renowned as one
of the best blends of tea available in the global Level
market. The number of direct and indirect Parameter Lower Upper
beneficiaries from the tea industry is (-1) (+1)
considerable. Extraction time /(min) 10 20
Extraction temperature 75 85
At present, Sri Lanka exports several varieties
of tea including bulk tea, tea packets, tea bags, /(oC)
green tea, and instant tea. Among these, the Evaporator temperature 85 95
demand for instant tea is continuously /(oC)
increasing. There are two main types of instant
tea: hot water soluble instant tea and cold water
soluble instant tea. (Abeysekera, 2005). Production of instant tea
The principal operations carried out in Five liters of distilled water was pre-heated to
manufacturing hot water soluble instant tea are the required temperature of the extractor as
extraction, filtration, evaporation, aroma given in Table 1. Then 500 g of tea (tea grade
stripping and aromatization, spray drying and of Broken Orange Pekoe Fannings) was added
packing. The main indicators of the quality of to pre-heated water, and the temperature was
instant tea include particle size distribution, the kept constant at the required extractor
total polyphenol content, the thearubigins to temperature for the required period of time
theaflavin ratio, and the brightness of the tea while stirring. Then the tea brew was strained,
(Roberts and Smith, 2001). and was evaporated in the climbing film
evaporator at the Department of Chemical and
In this work, the effect of the important process Process Engineering, University of Peradeniya.
parameters such as extraction time, extraction The first 100 ml of vapour leaving the
temperature and evaporator temperature, on the evaporator was condensed, collected and added
quality of instant tea was examined. back to the concentrated brew to simulate the
aroma stripping operation. The evaporation
Materials and methods was then carried out for 90 minutes. The
concentrated tea brew was then spray dried at a
Experiment planning constant temperature of 95 ºC in the spray
dryer at the Food research institute,
As there were several variables under study and Gannoruwa. The resulting tea powder was then
since the production of instant tea and analysis aseptically packed and sealed in triple layered
of the quality were expensive and time aluminium foil bags to prevent possible
consuming, a two level factorial experiment moisture absorption.
design approach was employed to screen out
the influential process parameters. The selected Quality measurements
process parameters and the relevant levels are
shown in Table 1. The upper and lower levels The standard procedure for determining total
of each parameter were decided based on the Polyphenol level in instant tea was performed
range of process parameters used in the using a Folin-Ciocalteu Colorimetric reagent,
industry. with optical density measurements at 765 nm.
(ISO working draft, 2005). The standard
procedure for determining thearubigin and
theaflavin in instant tea was implemented using
202
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
isobutyl-methyl-ketone (IBMK) and sodium best-fit line in the plot of the normal
hydrogen carbonate by measuring optical probability against factor effect. (Montgomery,
density at 380 nm and 460 nm. (Roberts and 1976)
Smith, 2001).
By analyzing the four normal probability plots
Calculation of quality measurement related to total polyphenol, theaflaving,
thearubigin and brightness, the only outlier
Theaflavin and Thearubigin content were which deviates from the best-fit line is found in
calculated from the optical density the plot related to total polyphenol, and it
measurements at 380 nm while the brightness corresponds to the interaction of evaporator
was calculated from the optical density temperature. (Figure 1).
measurements at 460 nm (Roberts and Smith,
2001). The total Polyphenol content was Interaction plots were drawn to identify
calculated from the optical density significant interactions (Montgomery, 1976).
measurements at 765 nm. (ISO working draft, The only significant interaction found was
2005). between extractor time and evaporator time.
This is indicated in Figure 2. This interaction
Data analysis indicates that the Evaporator temperature has
little effect at low Extractor temperature, but a
The calculated values for thearubigin content, large negative effect at high level of Extractor
theaflavin content, total polyphenol content and temperature. This may be due to the sensitivity
the brightness values were analysed using the of Thearubigin to high temperature. When both
factorial experimental method. extractor and evaporator temperatures are high
the exposure time of tea for high temperate
Results and discussion increase, which leads to the destruction of
Thearubigin.
According to the factorial method, the most
influential parameters stay as outliers of the
1 99
2 98
5 95
10 90
20 80
30
60
Normal probability
50
40
70
80 20
90 10
C
95 5
98 2
99 1
99.8 0.2
99.9 0.1
99.99 0.01
-2 -1 0 1
Effect estimate
203
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
References
Conclusions
Abeysekera, S.A. (2005) Technical Feasibility
Among the three process parameters, extractor Study on the Manufacture of Hot Water
time, extractor temperature and evaporator Soluble Instant Tea, Tea Research Institute,
temperature, analysed in this research, Thalawakele.
evaporator temperature was identified as the ISO working draft (2005) Tea and Instant tea-
parameter which has the highest influence on Determination of total Polyphenols in tea-
the quality of instant tea. Part 1: Method using Folin – Ciocalteu
colorimetric reagent.
40.00 Montgomery, D.C. (1976) Design and Analysis
35.00 of Experiments, John Wiley and Sons.
30.00 Roberts, E.A.H. and Smith, R.F. (2001) The
Thearubigin /%
204
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
205
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Table 1. Mass concentration of PM10 and resistance obtained from the experimental set up for
different filter samples
Resistance from the
Concentration of
experimental set up
No Air pollution measurement Site PM10 (μg/m3)
( Ω)
1 Blank Filter paper (Reference) 0 4.5
2 In front of Kandy Girl’s High School 171 52.5
3 In front of Art’s theater (Peradeniya) 28 9.6
4 Infront of Kandy Post Office 176 14.6
5 Good’s shed Bus Stand 167 21.9
6 Gannoruwa 13 5.6
7 Idamegama 17 5.4
8 Inside Peradeniya Botanical Gardens 4 5.9
9 In front of Peradeniya Botanical Gardens 113 12.4
10 Near Katugastota Bridge 346 64.0
Filter and
filter Convex Light
Convex Dependant
holder Lens Amplifier
Lamp Lens Resistor and multi
meter
70
Resistance Measured (Ω)
60
50
40
30 R = 0.8759
20
10
0
0 100 200 300 400
3
PM10 concentration (in μg/m )
Figure 2. Resistance measured from the device plotted against the PM10 concentration
206
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
207
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
Experimental method
Vehicle service stations generate considerable
amount of wastewater and existing treatment Wastewater for the study was obtained from a
methods in most of these facilities in Sri Lanka service station, which has a reasonably good
are not up to the required standards. Most of treatment facility, and the samples collected
the methods employed, such as the use of fat were effluents from the treatment plant. In
traps, sand filters and charcoal filters, are electro-coagulation, the treatment of
limited to the physical treatment of wastewater. wastewater was examined by varying the
By physical treatment only free oil and grease current and the reaction time for each type of
are removed but the dispersed oil, which has a anode plate; aluminium and mild steel.
smaller particle size than the free oil, is not Aluminium cathode was used in both cases. In
removed effectively. Therefore, a proper chemical coagulation, effect of alum dosage on
chemical treatment method should be removal was studied at different reaction times.
implemented to reduce the oil content in the
discharged wastewater from the vehicle service Measurements
stations.
After a given reaction time, the turbidity was
The intention of this study is to determine a measured in 10 minutes intervals up to a total
suitable chemical treatment method and to of 120 minutes of sedimentation time. The oil
optimise it to treat both free and dissolved oil and grease test was carried out for samples
and grease in the wastewater from vehicle which had been treated for a given reaction
service stations effectively. The chemical time and had been allowed to settle for a time
treatment methods tested in this study were period of 90 minutes.
electro-coagulation using aluminium and mild
steel electrodes and the use of alum as a The oil and the grease content of the
coagulant. wastewater sample were measured using the
standard gravimetric method (Eaton and
Materials and methods Awwa, 1995) and the Turbidity was measured
using the turbidity meter in NTU.
Experiment planning
Results and discussion
For electro-coagulation, the effect of reaction
time, current density and the sedimentation The turbidity and oil removal for different
time on oil removal was investigated. The other combinations of current density values and
factors such as pH, the distance between the reaction times were measured for the three
electrodes, and the conductivity of the mixture types of treatment methods tested. Figure 1
were kept constant at the optimum values indicates the variation of turbidity removal
reported in the literature. Accordingly, for with sediment time and reaction time when Al
every trial, the initial pH value was kept at 6, anode was used. Oil removal efficiency for
the conductivity of the wastewater sample was these conditions is shown in Figure 2.
adjusted using 0.6 g/l of NaCl, and the distance
between the plates was kept at 10 mm (Xu and From the results obtained, it is evident that an
Zhu, 2004). In addition, the contact area of the increase in reaction time and current density
electrode plates was also kept constant. will increase the removal of turbidity and oil
for both types of anodes. For the chemical
coagulation process, the removal of turbidity,
oil and grease increased with the alum dosage.
208
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
100
90
80
Turbidity removal /%
70
60
50
40
30
Reaction times
20
60min 50min 40min
10
30min 20min 10min
0
0 10 20 30 40 50 60 70 80 90 100 110 120
Sedimentation time /min
Figure 1. Turbidity removal vs sedimentation time for Al anode at different reaction times
It should be noted that the wastewater used in maximum oil and grease removal was archived
this study had already gone through the with mild steel anode at a current supply of 4 A
physical treatment process and most of the oil and a reaction time of 40 minutes.
remaining is dispersed oil, which is difficult to
remove. This is the reason for low oil removal A comparison of the three treatment methods is
in this study. as shown below in Figures 3 and 4. Hence, is
not that suitable to use in daily applications.
Oil removal efficiency
100.00
30
90.00
25
80.00
20
Turbidity Removal /%
70.00
15
60.00
10
50.00
5
40.00
0
10 20 30 40 50 60 30.00
209
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Conclusions
Maximum Oil Removal
Considering both oil removal and turbidity
removal, it can be concluded that, for the range
35.00 of dosage used in the experiment, the Electro
30.00 coagulation process is more efficient than the
25.00 use of Alum for the treatment of wastewater
O il re m o v a l %
210
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
There are eight diaphragm plates in one column
Improvement of highway network of Sri Lanka and six in the beam. Yield strength of steel was
includes construction of motorways, flyover 298 MPa. The finite element mesh of the model
crossings, various bridge structures, etc. These is shown in Figure 2. The model consists of
structures can have sub-structure components four-nodes doubly curved shell element (S4R)
such as cantilever type columns, inverted L- at column bases and adjacent panels of beam
shaped piers, and rigid-frames, which are column joints, four-node linear beam-column
primarily constructed using concrete. However, element (B31OSH) at mid segments of
steel is also a very useful construction material columns and the beam, and rigid beam
since it offers a great flexibility in space elements (R2D2) at the beam column joints and
utilization, has speedy construction time and the interface between shell and beam-column
high earthquake resisting characteristics. In elements. The load bearing capacity beyond the
particular, steel rigid-frames supporting peak load occurs mainly due to the local
highway bridge systems are popular in many buckling deformation at column bases and
countries due to their effectiveness in probably at beam-column joints. Thus it is
overcoming space restrictions. Introduction of essential to use shell element at those locations
rigid-frames in large-scale motorway in order to capture the effects of the local
improvement projects in Sri Lanka would be buckling deformations. The lengths of the
very helpful when densely constructed areas segments where the beam-column elements
are concerned. Meanwhile, there is a were assigned were decided so that the
considerable interest among professionals elements undergo elastic or very little inelastic
involving infrastructure projects to ensure deformation. Therefore the use of beam-
adequate earthquake resistance in view of column elements would not lessen the accuracy
many recent earthquake events reported in the but increases the computational efficiency. In
region. Therefore, an analytical investigation order to simulate the proper boundary
that involves a large displacement finite condition at the interface of beam-column and
element analysis procedure is carried out to shell elements and column and beam joints
examine the performance of rigid-frames rigid beam elements were used. Only a half of
subjected to earthquake-induced forces. the frame was modeled considering the
symmetry in loading and the geometry. Mesh
Analytical model size was decided based on a trial and error
Finite element analysis method (FEM) is method. The length between the base and the
known to be accurate in analyzing steel first diaphragm is divided into 18 segments
structures provided that realistic material model while subsequent lengths are divided into 6
and proper elements with optimum element segments. Each sub-panel between longitudinal
mesh are employed. In this study, numerical stiffeners consists of 4 columns of shell
model of a steel rigid-frame pier was made elements. Three columns of elements were
using general purpose finite element analysis assigned in longitudinal stiffeners. The material
program Abaqus (Abqus, 2003). The model behavior of shell elements was simulated using
consisted of a one-story, rigid-frame pier with a the modified two surface plasticity model
beam length of 5.0m, a column height of 5.8m, (2SM) (Shen et al., 1995) externally linked to
and a 600 × 600 × 6 mm rectangular cross the Abaqus program. The beam-column
section that has 12 longitudinal stiffeners elements were simulated using a bilinear
(sizes: 50 × 6 (column), 80 × 8 (beam flange), elasto-plastic material model available in the
60 × 8 (beam web)), as shown in Figure 1. program. The 2SM has been extensively
More details of the model can be found in checked and found to be very accurate in
Nishikawa et al., (1998). inelastic cyclic analysis of steel structures. A
subroutine has been developed for 2SM and is
connected to the ABAQUS through the user
defined material option. Since the beam-
211
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
column elements undergo elastic or very little load carrying capacity was observed. Normally,
inelastic deformations the use of bilinear model the strength is given in terms of the ratio of
would not affect the accuracy. When loading maximum lateral load Hmax to Hy (i.e.,
procedure is concerned, first the vertical load P Hmax/Hy). The ductility (μ) is given by either
given by 0.15 times the yield load of the the ratio of lateral displacements at Hmax to δy
column was applied at columns top and then (i.e., μm=δm/δy) or lateral displacement at
the incremental cyclic lateral loads were 0.95Hmax beyond the peak to δy (i.e.,
applied in terms of multiples of yield μ95= δ95/δy) (Aoki and Susantha 2005). Above
displacement δy (i.e., ± δy, ± 2δy, etc.) at the two definitions are graphically shown in Figure
top left corner, as indicated in Figure 2. The 4. The values of Hmax/Hy, δm/δy, and δ95/δy of
displacement corresponding to the yield load the frame are 1.30, 2.0, and 2.70, respectively.
was taken to be the yield displacement.
The failure modes such as local buckling at
Result and discussion column bases and in adjacent panes of beam-
column connections could be observed from
The strength and ductility of the frame can be this analysis as shown in Figure 5. The local
evaluated using the envelope curve of the buckling patterns at the column bases were
lateral load-lateral displacement hysteretic similar to those normally observed in cantilever
curve shown in Figure 3. It is seen here that the type steel columns subjected to cyclic lateral
maximum load occurred at displacement level loads.
of 2δy and beyond that a gradual decrease in
1000
μm=δm/δy
Lat. Load (kN)
0 Hy
-500
-1000 δy δm δ95
-200 -100 0 100 200
Disp. (mm) Displacement, δ
Figure 3. Lateral load-lateral displacement curve Figure 4. Ductility indices μm and μ95
212
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
213
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
214
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
for the purpose of finite element analysis with a face, the highest value of major principal stress
maximum temperature of 77 0C at the core and was observed as 6.2 MPa at the location C in
the outer surfaces at the ambient temperature of the direction of 890 to X-axis.
27 0C as shown in Figure 2. Since the weir was
cast as a series of longitudinal segments, each For the plane strain analysis the highest major
nearly 5 m long, the stress analyses were done principal stress occurred at A', shown in Figure
for two extreme cases of plane stress and plane 4, and identical to A in the earlier figure. On
strain to understand the distribution of thermal the inclined surface of the upstream end of the
stresses in the weir. weir, the highest major principal stress was
observed at B' again identical to B in the earlier
Results figure. The principal stresses at locations o
A'
and B' were 8.6 MPa in the odirection -38 and
The contours of major principal stress for the 7.5 MPa in the direction -45 , respectively. On
case of plane stress analysis are shown in the upstream face, the highest value of major
Figure 3. principal stress was observed as 4.4 MPa at
Sign Convention:
location C', identical location
o
to C in the earlier
figure, in the direction of 89 to X-axis.
C Tensile stress - Positive
Compressive stress - Negative The major principal stress variations were also
studied at locations A, B and C for different
B core temperatures and are reported in Figures 5
A and 6, respectively, for plane stress and plane
strain conditions.
14
Location A
12
Major. Prin. Stress / MPa
Location B
10
Location C
8
6
Figure 3. Contours of major principal stresses
4
in MPa for plane stress analysis
2
C′ 0
27 37 47 57 67 77 87
0
Core Temperature / C
B′
A′ Figure 5. Maximum principal stress
variations (plane stress) at critical
locations
12
Location A
10 Location B
Major. Prin. Stress / MPa
8 Location C
Figure 4. Contours of major principal stresses 6
in MPa for plane strain analysis
4
Discussion of results and conclusions will get healed towards the end of the hydration
period and with the maturity of concrete.
From the results, the locations, where high
principal stresses occured were identified as A, The results obtained in this study mainly
B and C (also identified as A', B' and C’). Since depend on the mesh size used, aspect ratios of
the location A is an intersection point where elements and the boundary conditions. The
inclined and horizontal surfaces meet together, results would be slightly different slightly if a
a significant amount of reinforcements were finer mesh having better aspect ratios for
provided at this location even without having elements is used or different boundary
any knowledge about thermal effects and that conditions are used for the analysis.
might be enough to resist the tensile stresses
developed by the hydration heat. Therefore, no This analysis has provided a reasonable
cracks may appear near location A during the understanding of the process of initiation of
initial hydration period of the weir. surface cracks and their location in mass
concrete during the initial hydration period.
Locations B and C were also identified as
possible critical locations for initiation of References
cracks and actually there were two surface
cracks initiated from those locations and SAP2000 Non-linear, Version 10.0.1 (1998)
propagated in inward normal directions of the Analysis Reference User’s manuals,
surface during the initial hydration period of Computers and Structures Incorporation
the concrete. The values of principal stresses (CSI), Berkeley, California, USA.
and their directions were not significantly Shyamalee, M.M.G.V., Pathirana, C.K. and
different at locations B and C for both plane Herath, K.R.B.(2007a) Analysis of a weir
stress and strain analyses. for earthquake loading, Proceedings of the
International Conference on Mitigation of
For plane strain condition, the stress intensity the Risk of Natural Hazards, March 27-28,
factors calculated for 1 mm surface flaws of the University of Peradeniya, Sri Lanka, ISBN
concrete at locations B and C are 0.5 MPa m1/2 955-589-099-1, 50.
and 0.3 MPa m1/2 respectively. These values Shyamalee, M.M.G.V., Pathirana, C.K. and
exceed the fracture toughness of concrete Herath, K.R.B. (2007b) Earthquake effect
(0.2 MPa m1/2) and initiations of cracks due to on a weir, A paper submitted to the annual
small surface flaws at these locations are hence congress of the Sri Lanka Association for
justified. Advancement of Science (SLAAS) to be held
in December 2007.
Even though the visible cracks appeared in the
concrete surfaces during initial hydration they
216
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
217
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
100
Secondary stress evaluation
0
The critical member without 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 1.E+08 1.E+09 1.E+10
100
S tre sh isto ryw h e n 2 ivr e sat re a c ivt e
more realistic residual fatigue life of riveted
bridge connections where the detailed stress
50
histories are known.
0
-50
0 1 2 3 4 5 6
References
T i me (s e c ) Alampalli, S. and Lund, R. (2006) Estimating
fatigue life of bridge components using
Figure 4. The von Mises stress histories at measured strains, Journal of Bridge
critical locations due to heaviest rail Engineering, 11(06), 725-736.
traffic Imam, B.M., Righiniotis, T.D. and
Chryssanthopoulos, M.K. (2007) Numerical
Determination of fatigue curve modeling of riveted railway bridge
connections for fatigue evaluation,
The obtained mean S-N curve obtained from Engineering Structure, Article in press, doi:
the assessment code (Network Rail, 2001) was 10.1016/j.engstruct.2007.02.011.
transformed into a fully known Wöhler curve Mesmaque, G., Garcia, S., Amrouche, A. and
by using Kohout and Vechet Wöhler curve Rubio, G.C. (2005) Sequential law in
modeling technique as in Figure 5. multiaxial fatigue, a new damage indicator,
International Journal of Fatigue, 27(4),
Fatigue life estimation 461-467.
Network Rail (2001) RT/CE/C/025, Rail track
The new damage indicator based sequential line code of practice: The structural
law in multiaxial fatigue was utilized to obtain assessment of under-bridges, Rail track.
a more realistic service life for the riveted
connection..
218
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
219
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Table 1. Assumed failure probabilities of shown in the numerical example, different path
bridges failure probabilities can be found and using this
process, path performance can be evaluated.
Bridge System System failure Depending on the path failure probabilities,
reliability probability resources can be allocated in a comprehensive
index way.
B1 4.94 3.5×10-5 This is the first publication of the authors
B2 5.35 4.5×10-6 regarding the concept of resources expenditure
B3 4.87 5.5×10-5 using reliability index. They expect to do more
B4 4.81 7.5×10-5 research in this area for greater understanding.
B5 4.81 7.5×10-5
B6 4.80 8.0×10-5 Acknowledgements
B7 5.03 2.5×10-5 National Science Foundation of Sri Lanka
B8 5.25 1.0×10-5 (NSF/SCH/2005/02 and RG/2002/E/01).
Initially, from the values of system reliability References
indices of bridges, system failure probabilities
can be found as shown in the third column of Atashi, M., Lachemi M. and Kianoush (2007)
Table 1 and using those, path failure Numerical modeling of the behavior of
probabilities can be found as outlined in overlaid slab panels for reinforced concrete
Table 2. bridge decks, Journal of Engineering
Structures, 29(2), 271-281.
Table 2. Path failure probabilities Christensen, P.T. and Baker, J. (1982)
Structural reliability theory and its
Path Lower bound Upper bound applications, Springer-Verlag, Berlin,
path path Germany.
Failure Failure Christensen, P.T. and Murotsu, J. (1986)
Application of structural systems reliability
probability probability theory, Springer-Verlag, Berlin, Germany.
AEB 8×10-5 19×10-5 Kong, J.S. and Frangopol, D.M. (2004) Cost–
ACDB 7.5×10-5 11×10-5 Reliability interaction in life–cycle cost
ADB 7.5×10-5 13.5×10-5 optimization of deteriorating structures,
Journal of. Structural Engineering,
Discussion 130(11), 1704-1712.
This paper proposes a way to evaluate bridge
paths performances using reliability theory. As
220
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
221
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
222
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Maximum displacement of the highway bridge The total weight indicates vulnerability. If the
at Peradeniya (middle of span) is 465 mm. total score is low, the vulnerability is high and
if total score is high the vulnerability is low. To
Maximum displacement of the Akbar bridge get a marginal score a survey has to be carried
(middle of span) is 360 mm. out and then detailed analysis should be done
for bridges in a particular network.
Visual Screening by SRS (Weight)
According to the method used in this research,
the most seismic vulnerable bridge is the
highway bridge at Peradeniya and it has a score
Evaluation Priority of 34.7 and the Akbar bridge has a score of
W>50 Low 46.5. Considering their displacement, it can be
50<W<70 Medium said that the Peradeniya bridge has higher
70<W<100 High vulnerability than the Akbar bridge. The result
obtained from the SRS also indicated similar
results. Therefore, it can be said that the SRS
Low has enough accuracy. Using this SRS, one can
evaluate the condition of potential earthquake
Detailed Analysis risk on existing bridges.
References
Bandara A.C.R., Priyadarshana G.S. and
Tennakoon N.C. (2006) Condition
Upgrading Upgrade evaluation of bridges for potential
needed? earthquake risk, Undergraduate Research
Yes No Report, Department of Civil Engineering,
University of Peradeniya, Sri Lanka.
Discussion
By applying the aforementioned procedure to a
seismically less resistant bridge, this can be
validated to have a reasonable accuracy.
223
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Equivalent Electric Network Model for Ionic Polymer Metal Composite Dynamics
224
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Rout
I(t)
I (k ) = dI (k − 1) + KV (k ) − cKV (k − 1) (6)
C
R1
This parametric estimation method works well
R
Vs V(t)
for noise free measurements, which is hard to
R2
C achieve in practice. Therefore the measurement
noise component has to be incorporated in the
model. Different parametric system
(a) (b) identification models such as Auto Regressive
with Extra Input (ARX), Auto Regressive
Figure 2. Initially suggested passive electric Moving Average with Extra Input (ARMAX)
network model for the IPMC and Box and Jenkin (BJ), which can handle
actuator. (b) Since the initial current measurement noise are available as derived in
Ljung et al., (1999) With the results of the
is finite, a series resistor R1 is preliminary identification experiments, the BJ
introduced. The output resistance model structure, whose block diagram is shown
Rout of the voltage source Vs is in Figure 3, is used here for the model
also included. The voltage measured verification as that has the capability of
across the IPMC is V(t) estimating independent dynamical models
between input-output and noise-output.
Passive electric network model derivation
The strategy used in this paper is to obtain a
Referring to Figure 2(b), the transfer function basic input output model for voltage vs. current
from the input voltage V to the total network dynamics using non-parametric identification
current I is: methods and verify its validity using
⎛s + 1
⎜
⎞
⎟
(1) parametric identification experiments.
I ( s) ⎝ R2 C ⎠
=
V (s) ⎛ + ⎞⎞
R1 ⎜ s + ⎛⎜ R1 R2 e(s)
⎝ ⎝ R1 R2 C ⎟⎠ ⎟⎠
C (s )
D (s )
where ‘s’ is the Laplace operator. This can
also be presented as V(s) B (s ) I(s)
I ( s ) K (s + b ) B(s ) (2) A(s )
Σ
= =
V (s) (s + a ) A(s )
where K = 1 , 1 R1 + R2 Figure 3. BJ model structure to incorporate
b= , a=
R1 R2C R1 R2C measurement noise in the parameter
estimation. More accurate noise
For a step voltage input, it becomes (3) models may be obtained by suitably
I( s ) K Kb selecting the orders of C(s) and
= +
V ( s ) (s + a ) s (s + a ) D(s). A(s) and B(s) are same as in
equation (2)
In the discrete time domain, with the z-
transform operator and h being the sampling Results and discussion
period, it becomes
The model verification is done using both
I ( z) K (z − 1) (1 − e − ah )Kb (4) parametric and non-parametric approaches.
= +
V ( z) (z − e ) (z − e )
− ah − ah
Non-parametric identification
which can be simplified as
(5) Referring to step response of IPMC shown in
I ( z ) K (z − c ) Figure 1, K, b and a of the transfer function
=
V ( z ) (z − d ) model described in equation (2) were estimated
using initial value, final value and the time
where c = a(1 − e − ah ) − 1, d = e − ah . constant of the step response of current. The
values are given in Table 1. Figure 4 shows the
applied step, measured current and simulated
Introducing the pulse transfer operator q, it current using the above model.
becomes
225
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Parametric identification
A BJ model was fitted to a recorded voltage
and current data set. The input voltage was
varied as a pulse sequence of random
amplitude and duration. The strategy used in
trying to fit the BJ model was to keep the
voltage vs. current dynamics fixed according to
equation (6) and fine tune noise model to
achieve the best fit. The Figure 5 shows the
step sequence applied and the corresponding Figure 5. Comparison (lower) of BJ model
current variation and simulated BJ model and and Non-Parametric model with the
non parametric model outputs for comparison. experimentally obtained output data
corresponding to the input voltage
As can be seen in Figure 4, most of the time,
the non-parametric model output closely pulse sequence (upper)
follows the basic shape of the experimental
output. But the model does not take into However, especially from D to E and some
account the measurement noise, which is why other places in the figure, none of the models
matching in Figure 4 (d) is poor. can match the actual output. This is believed to
be due to un-modeled non-linearities.
The Figure 5 compares the models obtained by
BJ method and non-parametric method with the Table 1. Model parameters of the non-
experimental data, for the series of pulse input parametric identification
voltages. At A, C, and F, where the current rise
is shape corresponding to high frequency Estimated Value
excitations in the frequency domain, the non- Parameter
parametric model output is more closer to the K 0.01168
actual output. But when the changes are slower b 0.0915
in a place like B, the BJ model is better as it a 5.0813
incorporates a high order noise model.
Conclusions and future work
An equivalent electric network based dynamic
model for IPMCs has been introduced derived
and verified experimentally. It will be
improved in the future to incorporate the un-
modeled non-linearities.
References
Colozza, A. (2007) Fly like a bird – Flapping
wings could revolutionize aircraft design
Figure 4. Verification of the identified non- IEEE Spectrum, 32–37.
parametric model Ljung, L. (1999) System Identification: Theory
for User. 2nd Ed., Prentice Hall.
226
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
There are quite a number of mobile
applications such as, Guitar Tuner (Kohn,
An electronic tuner is a device used by 1997-2007), Tiny Tuner (GetJar.com, 2004-
musicians to tune instruments. There, a 2007), and Chromatic Guitar Tuner
musician plays a note and the display of the (4pockets.com) are available. The Guitar Tuner
tuner tells the musician how much the played and Tiny Tuner are simple applications which
note has deviated from the desired note in are only capable of playing a note so that
terms of frequency. Most of these devices, such musicians can hear the note and adjust their
as KORG Chromatic Tuner CA-30, are instruments. However Chromatic Guitar Tuner
standalone equipments which are solely used to is a sophisticated application which is capable
tune musical instruments. It would be of analyzing the frequencies and estimating the
advantageous if this functionality could be closest notes. The limitation of this application
integrated into existing electronic appliances is that it can only be deployed on high end
like mobile phones as these devices are mobile devices such as Pocket PCs.
equipped with all the necessary hardware
components to be used as an electronic tuner. Therefore the aim of this project is to develop a
In addition, mobile software applications are mobile application which could be deployed in
becoming increasingly popular and authors of most of mobile phones. It was a challenging
this paper uses this opportunity to develop a task to implement frequency analysis in low
tuner application to be deployed in mobile end mobile devices such as Java enabled
phones. mobile phones due to their limited
functionality. The authors of this paper have
Section two of this paper focuses on related succeeded in this by finding a mechanism to
researches on this area and section three implement Fast Fourier Transform (FFT) in
describes the methodology used during the Java 2 Micro Edition (J2ME).
development process. Section four shows the
results obtained and section five discusses the Development and implementation
issues related to development. Finally, section
six concludes the paper. The major part of the product is the frequency
calculation. The FFT has to be applied to
Related work obtain the frequency of the sound wave from
which the notes are identified. In order to
In the first part of this section we will brief two perform FFT the sound wave needed to be
academic researches and in the latter part we sampled. But the chosen platform, J2ME, does
will focus on few similar industrial not accompany an Application Programming
applications. Zaykovskiy and Schmitt (2007) Interface (API) to deal with sampling the
presented front end implementation of speech sound. Even though this limitation exists,
recognition systems for mobile devices. The J2ME was chosen as the development platform
signal analysis of these systems is being carried due to the higher number of mobile phone
out at the server side where the mobile client models supported by this platform. Then the
does not interfere with signal processing. The authors used the only available option, which is
shortcoming of this sort of application is the to sample the sound wave within the
communication overhead between the client application. The sound wave is sampled and
and the server. The Java signal analysis stored as a wave file, which comprises of
application by Clausen et al. (1998) can only header (chunk descriptor), “fmt” sub chunk and
be deployed on desktop computers and any “data” sub chunk as shown in Figure 1. As
other similar or advanced electronic appliances depicted in Figure 1, the “fmt” chunk has eight
due to the functions used in this particular fields which are represented in little endian
program. format except the field chunk ID. The fields
which are important from this chunk for the
227
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
FFT are SampleRate and BitsPerSample. The frequencies are known, it is mapped into the
value represented in BitsPerSample field is closest notes and displayed to the user. The
being used in deciding the number of bytes to first phase of our application samples the sound
read from data chunk per reading. The value in wave by recoding it as an audio clip in wave
the sample rate field is being used in file format as shown in Figure 1.
calculating the frequency. When the
Main:
While there is data in data chunk
Call Subroutine1 with 256 samples
End while
For (i = 0 to 256)
FindMax (|totalCos[i] + totalCos[i]|)
location = i
End
Frequency = SampleRate * location / (2*255)
End Main
228
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
229
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
230
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
number of pipelines until a design that meets Row 3 in Table 1 gives an example of such an
the given design criteria is obtained. encoding. Given a register array for a register
operand field, we determine the encoding
Instruction encoding values by using REE algorithm. The full REE
algorithm is omitted from this presentation for
Instruction encoding is the act of using binary brevity. Applying this algorithm to each of the
bits with certain format to represent operand fields, we can obtain encoding for all
instructions. The format of the instructions can operand fields in an instruction type with a
be generally divided into two parts: one, minimal number of bits.
operation field or the operation code (op-code),
for encoding operation; and the other, multiple Operation encoding
operand fields for encoding operands. Our
encoding approach uses a fixed instruction We group the instructions according to the
width for each individual pipeline, with the number of bits needed for the operand fields.
width varying from one pipe to another. The When the instruction type is unique no bits are
encoding starts with the existing instruction used for operand fields. The operation field
sequence produced by the scheduling encoding is summarized in Algorithm 1. For
algorithm. Encoding for each pipe is carried each instruction type, the instruction operation
out separately. The instructions in a sequence encoding progresses from inner most level to
are grouped based on the type of the the outmost level; and the code size grows
instructions. The pattern of operand values is accordingly.
analyzed and operands are encoded with
minimal possible number of bits. Based on the
operand encoding, the operations of the whole
instruction set used by the application program
is then encoded. Details for both encoding
tasks are given below.
Operand encoding
In the instruction set architecture (ISA) used in
our design, the operands can generally be
classified into two types: registers and
immediate values. The encoding strategy is
demonstrated in the following with register
operands (or registers for simplicity).
Conventionally, the size of a register field (or
Algorithm 1. Operation Encoding
operand field), is determined by the size of the
register file. However, with a given program
sequence, a certain type of instructions may not Results and discussion
use all of those registers. Therefore, the field We designed ASIPs for two applications from
size can be reduced. Rather than randomly Mibench embedded systems benchmark suite.
assigning code-words to each of the registers, The designs were later synthesized using
we used an encoding approach such that the Synopsys Design Compiler based on the
decoding will be simpler, thus reducing Taiwan Semiconductor Manufacturing
hardware complexity, and hence improving Company’s (TSMC) 90nm core library, and
performance, area and power. We call this simulated with the Modelsim simulator. Figure
technique reduced-bit efficient encoding or 1 depicts the area, clock cycle and leakage
REE. Table 1 compares a typical full bit power comparisons of the non-customized
encoding (FBE) against REE for a given set of (typically encoded) and customized (fully
register usage. encoded) designs. The average savings are
26.9% on area, 27.8% on leakage power and at
Table 1. FBE Vs. REE the same time there is a reduction in clock
Registers R2 R4 R13 R11 cycle by 1.4%. In addition, the average
FBE (b3b2b1b0) 0010 0100 1101 1011 instruction memory size saving is about 69%.
REE (c1c0) 10 00 01 11
For instruction encoding, many approaches
have been proposed. In (Lee et al., 2002),
authors presented a technique that encodes all
231
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
instructions, required by an ASIP, with a given implemented in the future work for
instruction size. A hierarchical instruction customization of processor.
encoding for VLIW-based architecture
application is presented in (Liu, 2005). These References
approaches are not tailored to our target
processors proposed in (Radhakrishnan et al., Lee, J.S., Choi, K. and Dutt, N. (2002)
2005). The techniques presented in this paper Efficient instruction encoding for automatic
exploit the unique architectural feature of our instruction-set design of configurable
target heterogeneous multi-pipeline ASIP. ASIPs, ICCAD, 649-654.
Liu, C.H. (2005) Hierarchical instruction
Conclusions encoding for VLIW digital signal
processors, ISCAS, 3053-3056.
We presented techniques to fully customize PEAS Team. (2002) ASIP Meister Toolset.
instruction encoding for a multiple pipe Retrieved 4 2007, from ASIP Meister:
processor. This approach best trades off the http://vlsilab.ics.es.osaka-u.ac.jp/dac2003/.
simplicity of fixed size encoding approaches Radhakrishnan, S., Hui, G. and Parameswaran,
and high density of varied size encoding S. (2005) n-pipe: Application specific
techniques. The encoding is customized, with heterogeneous multi-pipeline processor
each pipeline having its own fixed instruction design, Workshop for Application Specific
width and the instruction width varying from Processors.
one pipe to another, and hence achieving a Radhakrishnan, S., Hui, G., Parameswaran, S.
better trade-off between the design simplicity and Aleksandar, I. (2006) Application
of the fixed-width encoding and code reduction specific forwarding network and instruction
efficiency of the varied-width encoding. Due to encoding for multipipe ASIPs,
the resource and time limitation only few CODES_ISSS: 06, Proceedings of the
applications were implemented for the current fourth International Conference on
work. Many versatile applications will be CODES.
232
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
In CSP approach, instead of selecting
In a typical radio station, DJs play songs using individual music titles, one by one, like above,
pre-generated playlists. A playlist is a set of the idea is to produce a collection of music
music titles which should be played in a given titles, in an order, restricted to a set of global
order. Composing high quality playlists for constraints (Pachet et al., 1999). One of the
radio stations is very demanding because of the better CSP approaches to APG, with regard to
large number of songs available for selection. scalability is proposed by Aucouturier and
Therefore the person responsible for Prachet (Pachet, 2002), which starts by having
composing playlists have to either memorize or a random set of songs from the whole set, and
listen to the excerpts of the songs in order to iteratively tries to improve the playlist by
choose the right song for the right situation. making small changes to it, using adaptive
search (an adaptation of the local search)
All the radio stations in Sri Lanka build their algorithm (Codognet and Diaz, 2001).
playlists manually. In today’s competitive
world, even though maintaining a high Both these approaches so far have looked at
standard largely defines the popularity of the APG for playlist generation mainly for a set of
station, the above situation makes it difficult very few listeners or a large set of listeners
for a human to achieve this consistently. This which can provide immediate feedback. This
is the motivation for this research which paper, for the first time, presents an APG
examined the feasibility of using a computer to technique for radio stations (a large global set
generate music playlists for radio stations using of listeners from which continuous immediate
a newly developed Automatic Playlist feedback cannot be obtained), PlayGen, which
Generation (APG) algorithm. is a hybrid of the aforementioned techniques
and enjoys the benefits of both the schemes.
Related work
APG, which is formally defined by Pauws et Methodology
al., (2006) has enjoyed lots of research recently The APG algorithm presented in this paper,
and the researchers have come up with many PlayGen, uses the algorithm presented in
different approaches depending on different (Prachet, 2002) as a starting point due to its
applications, such as similarity based scalability. The main idea of that the algorithm
techniques for Personal music players to presented is that the qualities/characteristics
Constraint Satisfaction Programming (CSP) which a playlist should posses are described as
approach for Electronic Music Distribution constraints.
(EMD) systems. Similarity based techniques
are specially used for recommender systems, in Problem statement
which the listener selects one or more of her
favorite songs to be used as seed songs and the The playlist generation problem is defined as
generator will pick the rest of the songs one by follows in (Prachet, 2002). A playlist S is a
one for the unordered playlist. Collaborative sequence of variables v1, v2…, vn whose values
filtering (French and Hauver, 2001; Pestoni et xi can be taken from a catalogue of music titles
al., 2001), timbre similarity measures (Logan D. The problem is to assign values to variables,
and Salomon, 2001; Aucouturier and Pachet, so that the resulting sequence satisfies
2002) and machine learning using existing constraints c1, c2,…,cm.
playlists (Platt et al., 2002) are the popular
similarity based techniques. The major Constraints
advantage of using these techniques is that
their sensitivity in capturing the changes in the Constraints are seen as simple cost functions.
choices of a single listener. The cost represents how ‘bad’ the constraint is
satisfied, for the given assignment of variables.
Constraints may hold on attributes of music
233
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
titles and these are typically metadata of songs. moment of time considered. Playability
Therefore, a playlist will have a total cost function P is defined as follows.
associated with it. The algorithm starts with a
certain initial playlist (having a certain total P=P(release year, number. of times requested/
cost) and iteratively refines the solution until a recommended, last time played, time of the
playlist satisfying all the given constraints day).
(having a total cost less than a given threshold)
is found. Authors have developed the following Further, the algorithm addresses a set of global
three compulsory constraints for radio stations constraints that would cover various
along with several other optional ones. requirements of any radio station as detailed
under constraints. Therefore, this is the first
• C-ALL-DIFFERENT: Two songs of the time that these two (CSP and similarity based)
same album/artist should not be included. techniques are combined to develop a hybrid
• C-DISTRIBUTION: Genre (e.g. pop), APG algorithm for radio stations.
type of artist (e.g. female), release year
and tempos (in beats per minute) of two Let S ≡ playlist, P ≡ set of playable songs and dj ∈ D
consecutive songs should be different.
• C-VARIETY: Number of different
genres of songs should exceed a certain Assign random songs from P∩RNS to S
threshold. Until cost(S
S) is less than threshold
For each vi
For example, the cost associated with C-
Calculate cost vi
VARIETY is quantified as follows.
Find vw whose cost is the worst
Cost(C-VARIETY): For each song dj ∉ S and dj ∈ P
vw Å dj
Return the length of the playlist – number of Calculate cost(vw)
genres of music titles divided by length of the
Locate dL with the lowest- cost(vwÅdj)
playlist
vw Å dL
PlayGen algorithm S)
Find cost(S
234
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
station. Secondly, a set of frequent radio constraint satisfaction approach and the
listeners (L) of the above station rated them similarity based techniques, of playlist
considering their own satisfaction derived out generation, are combined to develop an
of listening to them, relating to qualities like automatic playlist generator, which would suit
variety etc. In a set of five playlists, the radio stations. The developers were able to test
playlists 1-4 were generated using all three and prove that it is feasible to generate quality
compulsory constraints (of which playlist 3 is music playlists using a computer.
depicted in Figure 2) while playlist 5 was
randomly generated using only C-ALL- References
DIFFRENT constraint. Figure 3 depicts the
playlist ratings from the test. All five playlists Aucouturier, J.J. and Pachet, F. (2002) Finding
were generated from a database containing songs that sound the same, Proceedings of
song attributes of 200 music titles. Most of the IEEE Benelux Workshop on Model based
attributes were metadata of songs like title, Processing and Coding of Audio,
artist, album, and tempo, which were extracted University of Leuven, Belgium.
from allmusic.com and id3v2 tags of the songs. Codognet, P. and Diaz, D. (2001) Yet another
local search method for constraint solving,
Discussion Proceedings of the AAAI Fall 2001
Symposium, Cape Cod, MA.
French, J.C. and Hauver, D.B. (2001)
Ratings for Playlists
Flycasting: On the fly broadcasting,
Proceedings of International Conference on
5.00 Web Delivering of Music (WedelMusic),
4.00 Firenze, Italy.
Average rating
235
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
Methodology
User dependant speech based lip
synchronization is an area that is being The first step in this research was to find a
researched thoroughly. Combination of the suitable recording method for the lip
auditory and visual speech recognition is more movements and corresponding sound capture.
accurate than only auditory or only visual. For this, several methods were tried out to see
Consequently, there has been a large amount of which was the most appropriate for the
research on incorporating bimodality of a required project. The recording was done using
speech into the human-computer interaction a Canon XL 1 digital video camcorder with 7.2
interfaces. A speech-driven face animation is V DC power supply. The recordings have a
one of the research topics in this area, since frame rate of 24 frames per second (fps) and a
using natural voice for the animation of 320x240 resolution. The first recording was in
synthetic faces remains a challenging area of Moving Picture Experts Group (MPEG) format
research in computer animation. but it was later discovered that Audio Video
Interleave (avi) was the better format for
Since the FAPs (Facial Animation Parameters) picture frame / sound editing.
are required to animate faces of different sizes
and proportions, the FAP values are defined in Most of the avi manipulation and specific
face animation parameter units (FAPU). The frame extraction was done in Adobe Premiere.
mouth is one of the most difficult face features The coordinates of the key points marked on
to analyze and track. It has a very versatile the lip were obtained using computer graphic
shape and almost every muscle of the lower methods of a Matlab program. For finer co-
face drives its motion, unlike some areas of the ordinate manipulations and storing of data in a
face such as hair line or eyebrows which more accessible form for those who do not
become mobile only when the user uses certain have Matlab (Ver 7.0), Microsoft Excel spread
tones and expressions. sheets were used. The wave files, in wave and
avi format were also processed using Matlab,
The aim of this research is to develop a speech Adobe Premiere and Goldwave. The neural
based lip synchronization technique which networks were constructed using Matlab and
operates in real time. The final system should the output files were also implemented in
be capable of analyzing a speech signal and re- Matlab graphics (which uses OpenGL which is
producing the coordinates of the critical points C based.).
of the lip. The first part of the research was
dedicated to investigating models for the Data collection and pre-processing
automatic lip synchronization by speech signal
analysis and to finding a method which is The work initially started aiming to develop a
suitable for recording speech and facial system that was capable of simulating a 3D
movements with the minimum of equipment. face movement with real time user independent
speech. However, the work was limited to 2D
The research dealt with a method by which the face movement capture and lip synchronization
key points (critical points) of the lip could be for user dependant speech signals due to
read as coordinates and stored in a suitable difficulties that arose during the project. These
format and also a method by which to process difficulties included extracting key coordinates
speech signals, so it can be fed into a neural from the recording, different levels of frame
network for quick results and easy training. clarity, the lack of proper lighting equipment
for the recordings, etc.
It is hoped that by further developing this such
a method could be used to in the animation The research mainly concentrated on capturing
industry and for web communication, such as mouth / lip movements and corresponding
webcasting and user friendly avatars. sounds. It also dealt with some attempts at
236
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
capturing facial movements using limited phonemes such as /k/ consisting of the
equipment and some techniques in sound unvoiced stop, are not that easily recognized.
processing which can be used for phoneme Sounds with the soft ‘sss’ sounds are extremely
recognition. difficult to accurately recognize. It is believed
that this is caused by the similarity in
Several methods were tried out to find a pronunciation between these phonemes.
suitable arrangement for the recording to be
made. These methods included using the two ANN
mirror approach (Lin, I-Chen, Jeng-Sheng,
Ouhyoung, Ming 2003 and Whit, Amelia and A technique based on Artificial Neural
Lees, 1999) for 3D motion capture, full frontal Networks (ANN) was tried to create a user
face capture and finally recording only the dependant system. In this technique, the speech
mouth and nose area. The marker coordinates analysis consisted of programming neural
on the lip contours were extracted from the networks by feeding direct speech made up of
recordings and stored in Ms Excel worksheets. 44 phonemes and using the critical points of
the lip movements as the desired out come.
Then, speech analysis was done using cross- Feeding direct sound to the ANN produced
correlation of the phonemes used to see if this large errors and was therefore deemed
process could be used for Phoneme unsuitable.
recognition. Also, a technique based on
Artificial Neural Networks (ANN) was tried to When the result from the first method proved
create a user dependant system. In this insufficient several other speech processing
technique, the speech analysis consisted of methods were used to extract the features from
programming neural networks by feeding direct the speech signal. The second attempt was to
speech made up of 44 phonemes and using the train an ANN using the wave pattern envelop
critical points of the lip movements as the and the corresponding key-point coordinates.
desired out come. This method gave better results than the earlier
method but was still not satisfactory.
When the result proved unsatisfactory several
other speech processing methods were used to The final method was to extract the energy
extract the features from the speech signal. levels of the wave using Discrete Fourier
These methods included reducing the waves Transform and training an ANN using this data
into its wave envelop pattern and extracting the with again, the key point co-ordinates as the
energy levels of the wave using Discrete desired outcome. This method gave for more
Fourier Transform. encouraging results as the error when training
ANN was very small.
Experimental results
Discussion
Cross-correlation The cross- correlation method can be used to
Speech analysis was done using cross- categorize phonemes into three different groups
correlation of the phonemes used to see if this and to identify them as being in one of these
process could be used for Phoneme groups but it is not possible to use that method
recognition. for actual identification of a specific phoneme
from an audio speech signal.
From this method The two original methods (feeding direct sound
:accurately recognizing a sound is 56.18%. to the ANN and training the ANN using sound
:25.00% phonemes gave ambiguous results. envelops) do not produce satisfactory results
:18.18 % of the phonemes were not but using the Fourier transform to process
recognizable at all. sound waves showed promise. When the ANN
where trained using this method gave more
It is believed that the recognition accuracy of satisfactory results than any of the previously
56.18% is due to the pronunciation distinctness tried methods.
between the phonemes such as /p/, /t/ and /th/,
as well as to the small class size. Certain of
237
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
B.G.L.T. Samaranayake
Department of Electrical and Electronic Engineering, Faculty of Engineering,
University of Peradeniya
238
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Calculate and display the fuel wasted due to Figure 2. Display screen of the ALL-IN-
traffic jams ONE-BILLING METER
This is done inside the micro-controller using
the engine feedback on idling.
239
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Figure 3. Effective radius of the road wheel influencing the distance calculation
Input Waveform
Output Waveform
on flat plane
Figure 4(a). When the two input coils are supplied with the same sinusoidal input signal, two
identical sinusoidal signals are induced in the two output coils, provided that the two
coils are identical. The black circles are metal balls
Input Waveform
Output Waveform
on flat plane
Output Waveform
on inclined plane
Figure 4(b). When the three wheeler is travelling on an inclined plane, the metal balls enter into
the coil core increasing the magnetic coupling (reducing the reluctance). Hence the
output coil with metal balls inside has a higher voltage induced compared to the coil
without the metal balls. This difference is used to detect the elevation and decide the
inclination.
240
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30th November 2007
Memory Efficient Algorithm and Closed Form Formulas for Computing Odd
Magic Squares of Higher Order
P.C. Perera
number.
Introduction
Even though the computational aspects of
Magic squares have been studied for at least lower order magic squares are fairly simple
three thousand years, the earliest recorded and straightforward, those of the higher or-
appearance dating to 2200 BC, in China. der counterparts demand memory and time
In the 9th century, Arab astrologers used efficient algorithms. The computational time
them in calculating horoscopes, and by 1300 as well as the memory requirement for the
AD, magic squares had spread to the West. computation of n × n magic square are both
An engraving by the German artist Albrecht of order n2 . The unavailability of a closed
Dürer included a magic square in which the form expression for the nth order magic squa-
artist embedded the date, 1514, in the form re results lengthy source codes. Moreover, as
of two consecutive numbers in the bottom n becomes very large, high memory require-
row. Because the concept of a magic square ment makes the computation of nth order
is so easily understood, magic squares have magic square impossible. In this endeavor,
been particularly attractive to puzzlers and it is attempted to devise a memory efficient
amateur mathematicians. In modern mathe- algorithm and formulate a closed-form ex-
matics, the notions pertaining to magic squa- pression to compute odd higher order magic
res are utilized to characterize magic graphs squares by formulating de la Loubere type
as well as magic cubes. (Jezny and Trenkler magic square.
et al. 1983 and Adler, Robert Li et al. 1978)
The objectives of this work are to develop
The nth order magic square is an n × n ar- memory efficient algorithms in the context
ray consisting of the first n natural numbers of nth (n mod 2 = 1) order magic square to
such that the sum along rows, along columns determine,
as well as along two diagonals are all equal (a) the cell (i, j) ∈ N × N for a given k
n(n2 + 1)
to . Of course, given any magic satisfying 1 ≤ k ≤ n2 , and
2
square, a rotation or reflection will produce (b) the entry k with 1 ≤ k ≤ n2 when the
another magic square. Not counting these cell (i, j) ∈ N × N is specified,
as distinct, it is known that there is only one
3rd order normal magic square, and there are where N = {1, 2, ..., n}.
880 normal 4th order magic squares. The
number of distinct normal magic squares in- Preliminaries
creases dramatically with its size. For in-
stance, there are over 13 million normal mag- Letting the entry of the ith row and the j th
ic squares of 5th order. The algorithm de- column of the magic square be xij , the prob-
vised in this endeavor produces only one such lem of magic square can be formulated as fol-
normal magic square for a given odd natural lows. For the sake of notational simplicity,
241
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30th November 2007
henceforth, unless specified otherwise, the algorithm may be very high as n becomes
entry of the cell corresponding to the ith row very large. The case is the same for the other
and the j th column of the nth order magic existing algorithms used for computation of
square is denoted by Mn (i, j). Then, magic squares. The algorithm devised in this
work requires much less computer memory
n
X n(n2 + 1) as well as computational time.
xij =
j=1
2
for i = 1, ..., n The main result
n
X n(n2 + 1)
xij = The following results are conjectured and pr-
i=1
2 oved in this work. For the sake of brevity,
for j = 1, ..., n their proofs are not presented here.
n
X n(n2 + 1)
xii =
i=1
2 Theorems
Xn 2
n(n + 1) Theorem 1. For a given n and k satisfying
xn+1−j,j = n mod 2 = 1 and 1 ≤ k ≤ n2 ,
j=1
2
Mn (a, b) = k,
where xij ∈ {1, 2, ..., n2 }.
where
It should be noted that the solution of the
(n + 2k − 3) k−1
problem is not unique. The optimum algo- a= − mod n
rithm available at present is used in MAT- 2 n
LAB. In MATLAB, two n × n arrays are de- +1
clared by the function MESHGRID in the
and
process of obtaining the nth order magic squ-
are. Then those two n × n arrays are manip- k−1
ulated to obtain the nth order magic square. b= n+k−2−2 mod n
n
Thus, it is obvious that the demand for the +1
memory is an issue as n becomes large. To
comprehend the demand for the memory and Example 1. If n = 7, using the above algo-
the associated complexity of other existing rithm, the MATLAB routine
algorithms, consider the following algorithm
used in MATLAB. n=7;
for k=1:n^2,
function M = magic(n) a=mod((n+2k-3)/2+
...floor((k-1)/n),n)+1;
n = floor b=mod((n-k-2+
...(real(double(n(1)))); ...2*floor((k-1)/n),n)+1
M(a,b)=k;
if mod(n,2) == 1 end;
[J,I] = meshgrid(1:n);
A = mod(I+J-(n+3)/2,n); produces 7th order magic square as
B = mod(I+2*J-2,n);
M = n*A + B + 1; 30 39 48 1 10 19 28
end 38 47 7 9 18 27 29
46 6 8 17 26 35 37
In the case of the above MATLAB routine, 5 14 16 25 34 36 45
A, B, I and J are of nth order square matri- 13 15 24 33 42 44 4
ces. Thus, the memory requirement and the 21 23 32 41 43 3 12
computational time pertaining to the above 22 31 40 49 2 11 20
242
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30th November 2007
243
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30th November 2007
1
Independent Researcher
2
Department of Engineering Mathematics, Faculty of Engineering
University of Peradeniya
244
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30th November 2007
Note that 2pi −2mi is directly obtained by In Phillips et al. (2001), the square of long
shifting mi−1 by one left shift while discard- integers was obtained by removing repeated
ing the most significant bit of mi−1 . The digit products from the accumulation tree
number mi is obtained by keeping the least which is called optimized squaring using pre-
significant digit of mi−1 as it is, while com- computed partial products. It can be shown
plementing the other bits. that the algorithm devised in this paper su-
Since, in this case, it yields persedes the method of optimized squaring
2
X using pre-computed partial products present-
N = m0 = m22 + 2pi (2pi − 2mi ) , (3) ed here. Shifting involved in the regular pro-
i=1 cess of squaring is not an issue in the con-
where m2 = 1, we have text of this new algorithm. Additions and
subtractions are also kept minimal in this
2p1 (2p1 − 2m1 ) = 1 1 0 0 case.
2p2 (2p2 − 2m2 ) = – – – –
m22 = – – – – References
N2 = 1 1 0 0 Phillips, B. (2001) Optimized squaring of
long integers using pre-computed par-
1 0 – – – – – –
tial products, Proceedings. 15th IEEE
– – 1 1 0 – – –
... Symposium on Computational Arithm-
– – – – – – – 1
etic
1 0 1 1 0 0 0 1
245
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30th November 2007
246
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30th November 2007
247
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30th November 2007
248
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30th November 2007
point, ground clearance, etc. Out of them, the Hence the electrical time constant
main criterions of the design are the speed and τe= La/ Ra can be calculated from the rise time
the maneuverability as the machines have to of the current waveform in Figure 4 and
travel about 50m in total in the game field in 3 assuming τe << τm, where τm is the mechanical
minutes, where the field is of a polygon shape. time constant.
The speed requirement is over 2 ms-1 for a
machine of weight of approximately 25 kg with In the current falling, τm =J Ra /K2 is dominant
all its accessories connected. and it can be used to evaluate the relevant
parameters. The armature current in steady
System modeling state is related to the torque T by T=K Ia. The
effect of the mass (M) of the mechanical
Permanent Magnet DC motors are used in the structure can be modeled by
drive platforms, in order to achieve the desired
2
dynamics, because the torque of the stepper T= ω, (3)
motors, which is the other alternative, Mr 2 s
drastically drops at higher speeds. The block
diagram model as shown in Figure 2 is used to where r is the radius of the road wheels. Hence
model the Permanent Magnet DC motor from the resultant inertia felt by the motor can be
the input voltage to the output angular velocity given by, J = J + J + Mr .
2
The identified
(Figure 3). s w
2
In the block diagram, La and Ra are the motor parameters are listed in Table 2, which
armature inductance and resistance respectively are used in the controller design.
and K is the torque constant, while J being the
rotor moment of inertia. Since the manufacturer Controller design
specified values for the above parameters were The controller mainly consists of speed control
not available for the motors, they were and position control. The purpose of the speed
identified by measuring armature current Ia and controller is to maintain the road wheel speed
the speed ω for a set of input voltages Va.
249
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30th November 2007
Heading 2
250
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30th November 2007
251
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30th November 2007
P.C. Perera
1
The binomial expansion of yields that
Introduction 1−x
∞ k−1
!j
In recent times, much research has been fo- 1 1 X X i−k
= k bi 2 .
cused on developing both time and mem- N 2 j=0 i=0
ory efficient algorithms for solving compli-
1
cated mathematical problems. Contrarily, Considering the binary expansion of N, we
very efficient algorithms for calculation of have
simple operations that are performed very ∞ k−1
!j
frequently, tremendously improve almost ev- 1 1 X X i−k
= k bi 2 =
ery complicated computation performed by a N 2 j=0 i=0
(1)
computer. The algorithm developed in this X∞
252
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30th November 2007
1 the outputs produced by the difference equa-
and = 0.142857142857... = 0.142857. If
7 tion converges to the reciprocal of our inter-
1 1
x= ∈ Q, then = 0.00110011...2 = est.
1012 1012 k−1
X
0.00112 . Theorem 2. Let N = bi 2i where bi ∈ A
i=0
Lemma 1. If N is a prime, N |2N −1 −1 and
∞
1 X
the upper bound for the length of the string ∀ i. Then = 2−i xi and yi for i = N
N i=1
repeated in the binary sequence of N1 is N −1.
are the sequence of solutions to the k th order
For a given prime N , there exists a ∈ N such k−1
!
that N |2(N −1)/a − 1 and the exact length of
X
difference equation yn+k = bi yn+i +
the string repeated in the binary sequence of i=1
1 N −1
N is a .
b0 yn with initial conditions yi = 0 for i =
1, ..., k − 1 and yk = 1 such that the se-
Example 2. Recall that N = 17 is a prime. quence {zj }j=1 has the property zr−k+1 =
∞
253
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30th November 2007
Conclusions
References
Bennett, W.S. (1973) Quotient generation
with conventional binary multipliers,
Proceedings of IEEE, 61(5), 664-665
254
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
The results obtained from regressing CO2 on
Oil shock of the 70s has been the major factor GDPpc, ln(CO2) on ln(GDPpc) and ln(CO2) on
that caused structural breaks in the ever- 1/ln(GDPpc) using the ordinary least square
increasing trend in carbon dioxide (CO2) (OLS) procedure are given in Table 1. In all
emissions experienced by high income cases, the R2 statistics are very high. The low
economies such as United States, Japan, France values of Durbin-Watson (dw) statistics
and Austria (see, Unruh and Moomaw, 1998; indicate the presence of serial correlation
Friedl and Getzner, 2003; Shanthini and Perera, among the residuals of the respective
2006). The ever-increasing trend in the CO2 regressions. The problem is, however, more
emissions of Australia, despite it being a high than just the presence of auto-correlated
income economy, remains relatively unaffected residuals. It is the fact the time series
owing to perhaps its dependence on coal as fuel considered are all nonstationary as revealed by
(BP, 2006; World Bank, 2007). This paper the results of Augmented Dickey-Fuller (ADF)
seeks to identify the effect the increasing test carried out to test the presence of unit
income has upon the CO2 emissions of roots. Regressing one nonstationary time series
Australia, particularly because it is one of the on the other is known to give significant
two countries that has defied the Kyoto statistical correlation, such as high R2, between
Protocol (UNFCCC, 1997) that assigned the variables, even though the two variables
mandatory greenhouse gas emissions may not be related at all.
limitations to the signatory nations.
In order to determine if there exists an
Since both income and emissions data of equilibrium relationship between the
Australia are trending non-stationary time nonstationary time series of our interest, i.e. if
series, an ordinary least square regression these series “move together” over time, we
model developed based on this data can be tested the residuals of the respective regression
spurious (Granger and Newbold, 1974). In this models shown in Table 1 for stationarity using
study, we show that there exists a cointegrating the ADF test. The results showed that the
relationship between the income and emissions residuals are stationary at the 5% level of
non-stationary time series, and thus the long significance for the first two regression models
term effect that income has on emissions can and at the 10% level of significance for the last
be captured using the error correction modeling regression model tabulated in Table 1.
approach proposed by Engle and Granger Therefore, we concluded that the respective
(1987). time series are cointegrated, which is indeed a
clear proof for the existence of an equilibrium,
Data analyses i.e. a long-term, relationship between the
The annual CO2 emissions (denoted by CO2) cointegrated time series (Engle and Granger,
and the gross domestic product per capita 1987; Stock and Watson, 1988).
(denoted by GDPpc) data for Australia used in
this study, obtained from World Development Results: Error correction model
Indicators (World Bank, 2007), are shown in The error (or equilibrium) correction model
Figure 1. The unit of CO2 is Mt (mega tonnes = (ECM) between two cointergrated first-
109 kg) of CO2 and that of GDPpc is thousands difference stationary variables, y and x, could
of constant 2000 US$. The time span used for be written (Engle and Granger, 1987) as
modeling is 1960 to 1996, which makes
available the emissions data from 1997 to 2003
for model validation.
255
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
256
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Figure 1. CO2 emissions and GDP per capita time series of Australia from 1960 to 2003
257
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Evaporator
Condenser
258
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Table 1. Comparison of results obtained from test rig with theoretical estimates
Refrigerant Convective heat transfer
Trial
flow rate coefficient (W/m2K)
3
kg/s x10 Experimental1 Theoretical2
1 4.225 267.7 304.5
2 4.630 301.3 327.1
3 8.712 510.0 521.0
1 - Obtained using the test rig given in Figure 1
2 - Obtained using correlation in Ref [1] and [2]
259
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
References
Dittus, F.W. and Boelter, K.L.M. (1985) 1930,
Heat transfer in automobile radiators of the
260
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
L.U. Bakmeedeniya
Department of Mechanical Engineering, Faculty of Engineering, University of Peradeniya
Introduction
Methodology
Provision of a congenial working environment
is an important matter that affects productivity Two factories were selected as case studies
of any process or service. Various measures are where the occupants in the production floor
in place in industrial environments to maintain experience thermal discomfort.
conditions that make the working surrounding
comfortable. In the relevant standards, The geographical location and the factory
specifications are stipulated on acceptable layout were studied and suitable measurement
levels of temperatures, humidity, noise, etc., as grid was constructed to ensure accessibility, in
applied to different working environments, order that the intended floor measurements and
which are to be strictly adhered to. However, other readings could be obtained within an
the situation in most Sri Lankan industries is hour. At each grid point, dry and wet bulb
far from this, owing to lapses ranging from temperatures were recorded for the different
those in planning, to the implementation and layers which were, respectively, 1 m and 2 m
commissioning of a process. Complaints about above the floor. The process was repeated
high temperatures and humid conditions are throughout the working day (8 a.m. to 5 p.m.).
common on production floors where the Air flow velocity was also noted. The
number of workers per unit area or the investigation procedure implemented in was
generation of heat/moisture is relatively high. repeated after one month.
For examples apparel industries, footwear
manufacturers, catering industry/hotels etc., are Other observations made included the currently
common instances where one experiences such implemented solutions, inside/outside factory
adverse working conditions. In these cases, the conditions and the environment, the number of
main problem appears to be insufficient occupants, sources of heat generation, location
ventilation and heat and effluent removal. of fans and their effective region, blowers,
These two aspects are important to maintain a diffuser fans, windows, doors and related
balanced heat and moisture exchange process measurements.
between the process or the production floor, the
workers and the surroundings. Analysis of observations
The objective of the reported study was to For both case studies, relative humidity was
identify prevailing problems that affect the found from psychometric chart using dry and
human comfort in the working environments wet bulb temperature measurements, and
and to propose remedial actions using tools of hourly plots of relative humidity and ambient
computational fluid dynamics in accordance temperature contours were individually
with the relevant standards. Investigations were obtained.
carried out to study the factors and analysis
done with the help of available thermal comfort The behaviour of temperature and RH variation
standards to draw conclusions from an were examined on an hourly basis and
engineering view point. compared to detect special patterns as well as
repetitive behaviours. In addition, the nature of
This study looks into the present state of the background, such as fans, occupants,
selected industries in the above context to functions etc., were considered to determine
understand the gravity of the issue using basic their contributions to the problem and the
heat transfer and fluid dynamic principles amount of heat generation was also calculated
applied to two-dimensional flow, as a part of as necessary. Considering the results the
the research. It is expected to extend the existing problems were determined.
analysis to three-dimensional flow situations
using tools of computational fluid dynamics.
261
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Comparisons and discussion relative humidity was also outside the comfort
zone. In this factory, the presence of fumes
Case 1 – An apparel industry situated in the from the chemicals used in the process
Central province contributes to the increase of temperature
levels. There was also a high rate of heat gain
It was observed that the temperature contours from several sources. The presence of a wide
followed a similar pattern in some regions monitor roof also led to the formation of eddies
throughout the day. Figure 1 shows such a at the high levels leading to the stagnation of
contour in the factory during the most critical warm air in the breathing zone.
period. All the pedestal fans were placed in a 4.00
way that air movement was forced towards the
rear wall. But the exit door and the window
areas were not enough to provide sufficient
space to expel the air. As a result, air was 3.00
trapped in the vicinity of the wall, which also
reaps the higher temperature. The
corresponding RH contours did not reflect the
2.00
pattern of the temperature contours. However,
the humidity range in the floor was not at the
desired level.
1.00
5.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00
4.50
Figure 2. Temperature contours at 2 pm
4.00 (Footwear manufacturing factory)
3.50
Conclusions
3.00
2.50
According to the results, observations and
comparisons it can be seen that the factory
2.00 floor environments did not have desired values
1.50
of ambient temperature, relative humidity, and
also the space per person. Thus, it can be
1.00
1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00 concluded that the major contributory factors
Rear wall for the thermal discomfort are insufficient
ventilation, insufficient heat and effluent
Figure 1. Temperature contours at 2 pm removal inside the floor owing to improper
(Apparel industry) implementation of the present solutions,
noxious building configuration and inadequate
Case 2 – Footwear manufacturing factory space per person on the floor.
situated in Western province
References
Here again, the analysis of the observations
(Figure 2) shows an example of a critical ASHRAE 55 (1981) Air quality and thermal
temperature contour in the factory, and other comfort in factory buildings, Construction
details show major factors, which affect the Technology Update, 64.
occupants’ thermal comfort. Temperature at the BOI Enterprises (2004) General Guidelines for
floor was above the accepted level, and the Factory Buildings
262
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Limitation on Connecting Mini Hydro Power Plants to the Sri Lankan Power System
Network: A Case Study at Balangoda Grid Substation
263
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Gomala Oya, Rakwana and Wijeriya bus bars exceed the limits.
Fault level of Balangoda grid is increased by 132 kV side has increased by 1.22 MW by
16% by connecting the mini hydro plants. connecting the mini hydro plants. It is 37%
Breakers at the grid at 33kV level have a short increase over the losses without mini hydro
circuit capacity of 1450 MVA. Hence at this plants.
moment it is not a problem. If any other plants
to be connected the switch gear of Balangoda Conclusions
grid sub has to be changed.
At Balangoda grid, mini hydro plants
The line losses due to the connection of the penetration is studied. This has shown that the
mini hydro plants have been increase by 3 voltages at the mini hydro plants terminals
MW. This is very significant as it was 0.14 have already exceeds the limits with above
MW with out mini hydros. The line losses of 60% injection at night peak loading conditions.
264
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Further over voltages occurred in several points Ceylon Electricity Board (2006b) Medium
at light loading condition. This has also Voltage System Planning Report 2006,
increased the fault level at the grid substation Sabaragamuwa Province.
by 16%. Therefore this study concludes that for Ceylon Electricity Board (2006c) Monthly
proper utilization of mini hydro plants, a Review Report - May 2006, System Control
voltage control mechanism must be adopted at Branch.
the grid.
References
Ceylon Electricity Board (2006a) Statistical
Digest 2006, Statistical Unit, Commercial
& Corporate Branch.
265
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
A simplified model of the system is shown in
Embedded generation technologies are proved Figures 2. The transformer and the series
by researchers as one of the solution to satisfy impedance of the line are lumped into one
the increasing electricity demand while series impedance of (0.354+j27.626) Ω. AVR
optimizing the existing network (Jenkins, and governor controls together with
2000). In Europe micro grid technology also synchronization control are shown in Figures 3
potentially studied to supply green electricity to and 4. The AVR consists of a PI regulator with
most important industries (Lasseter, 2002; an output derivative feedback function to
Arulampalam et al., 2004). However, represent the practical AVR response. The
challenges are being made on keeping the governor consists of a PI regulator with the
micro grid running when the main grid was
disconnected due to a fault and smooth M AIN GRID
PT2
LO AD
representing the non frequency sensitive loads Figure 2. Simulation model of the AVR
and other representing frequency sensitive
loads. Initially steady state stability of the
micro grid, connected to the main grid, was
studied. Automatic Voltage Regulator (AVR)
operation and governor controller operations
were checked in islanded operation. Finally
resynchronization and automatic grid
connection was checked, when grid was
resorted. Data of generator, transformer and
feeder is annexed. Figure 3. Simulation model of the governor
266
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Simulation results
0.60 0.60
0.40 0.40
1.000 1.000
pu
pu
0.950 0.950
line end voltage line end voltage
1.020 1.020
0.990 0.990
pu
pu
0.960 0.960
Speed Speed
1.040 1.050
1.040
1.020
pu
1.030
pu
1.000 1.020
turbine turbine
0.60
0.80
0.70 0.50
pu
pu
0.60 0.40
0 50
0.0 5.0 10.0 15.0 20.0 0.0 5.0 10.0 15.0 20.0
Figure 5a. Results at islanded operation Figure 5b. Results at shedding loads
267
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Conclusions
Table 1. Transient study results at islanded
This study confirms that synchronization of
and load shedding operations the Nilambe generator with the micro grid
Excessive load consisting of local loads after shedding the
Grid failure
shedding excessive load is possible. The proposed
Over Under Over Under AVR and governor control together with
shoot shoot shoot shoot automatic synchronizer has shown very good
Generator bus voltage/pu 1.025 0.983 1.026 1,001 performance within allowable responses
limits. Therefore this can be taken as a
Grid point voltage/pu 1.003 0.966 1.004 0.979
preliminary studied model in for developing
Generator Speed/pu 1.035 1.049 micro grid technology to the Sri Lankan
power system.
Guide vane opening /pu 0.556 0.378
A commercially available synchronizer can
be configured and used with transmitters and
receivers for this purpose of remote sensing
and synchronizing.
References
Arulampalam, A., Barnes, M., Engler, A.,
Goodwin, A. and Jenkins, N. (2004)
Control of power electronic interfaces in
distributed generation microgrids,
International Journal of Electronics,
91(9), 503-523.
Figure 6. Simulation results at synchronizing process Jenkins, N., Allan, R., Crossley, P.,
Table 2. Transient study results at Kirschen, D. and Strbac, G. (2000)
synchronizing process Embedded Generation, IEE power and
Pre
Over Under Steady
energy series 31, 2000, ISBN
synchroniza
shoot shoot state
0853967748
tion Lasseter, R.H. (2002) MicroGrids, IEEE
Generator bus voltage/pu 1.01 1.04 1.03 Power Engineering Society Winter
Grid point voltage/pu 0.991 1.021 1.02
Meeting, Volume 1, 305-308.
Generator Speed/pu 1.03 1.11 1.09 1.098 Appendixes
Guide vane opening /pu 0.41 0.851 0.346 0.438
Generator data: 3MVA, 6.9kV, 0.85 p.f.,
Xd=1.014 pu, Xd'=0.314 pu, Td0'=6.55 sec,
Discussion Xd"=0.28 pu, Tdo"=0.039 sec, Xq=0.77 pu,
The AVR and governor responses have Xq"=0.375 pu, Tqo"=0.071 sec, Inertia 2 sec.
proven that it can maintain the micro grid
stability after a grid failure with excessive Line data: 33kV tower line of "Links"
load rejection. The maximum generator bus conductor line length -2km, Positive
voltage overshoot is 2.6% and maximum sequence resistance - 0.177ohms/km,
grid point overshoot is 0.4% for the grid reactance - 0.313 ohms/km, Zero sequence
failure. The maximum frequency overshoot resistance - 0.326 ohms/km, reactance -
is 4.9%. The maximum bus voltage
overshoot during the synchronization is 3%. 1.588 ohms/km
The speed overshoot is 8%. Even though
these overshoots are little higher than the Transformer data: Capacity - 4MVA,
expected values, the system was stabilized Voltage ratio - 6.9kV/33kV, Impedance
within few seconds. voltage - 10%
268
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
269
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Next, a three phase capacitor bank was S2, Q2 = Apparent power and reactive power
connected in parallel to loads to implement after power factor correction
power factor correction, and waveforms were QN = Reactive power compensated by the
observed. capacitor
Vsupply ILoad*5
400
300
200
Volt, Amp 100
0
-100
-200
-300
-400
0.020 0.030 0.040 0.050 0.060
Vsupply Isupply*5
400
300
200
Volt, Amp
100
0
-100
-200
-300
-400
0.020 0.030 0.040 0.050 0.060
Vsupply Icapacitor*5
400
300
200
Volt, Amp
100
0
-100
-200
-300
-400
0.020 0.030 0.040 0.050 0.060
Figure 7. Waveforms of the capacitor current and harmonic load current with supply voltage
270
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
271
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Modelling Sri Lankan Power System to Study the Effect on Dynamic Stability with
Large Scale Wind Power Integration
Introduction
In the next step, the generator data required for
To meet the increasing demand for electrical transient studies was inserted in to the
energy, the estimated need of new generating database. In first attempt, no typical data was
capacity to the Sri Lankan power system is used in place of missing data. This allows the
over 4400 MW in next 15 years (Ceylon system to run on IPSA default values. Dynamic
Electricity Board, 2003; Ceylon Electricity stability of the system was then studied to
Board, 2005). Studies have revealed that wind demonstrate the stability model of the IPSA.
is the most promising option of the available Then in second attempt, the default damping
renewable sources for grid connected power factor (which was zero in IPSA) was slightly
generation in Sri Lanka other than the adjusted together with minimum required
conventional larger generating stations. Wind transient data using typical values. This was
energy resource assessment study carried out in done for machines, which were not had
2002 has confirmed the availability of 20,000 sufficient data. These values were stored in a
MW of wind resources in north-western coastal different database thus leaving the database
region from Kalpitiya Peninsular to Mannar with real data untouched.
Island, Jafna Peninsular and the Central
Highlands. The following Stability criterion is used. The
system shall remain stable during and after a
However not enough studies are available to system disturbance for three phase faults at any
assess the transmission grid accessibility of O/H line cleared by the protection with
these wind resources. And also no system successful and unsuccessful auto re-closing
studies are carried out on routine basis for including a possible loss of generation and load
transmission connection proposal for non- transformer. Few marginal dynamic stability
dispatchable power including the large scale cases were identified for this study through the
integration of wind power. However the studies experience on the recent faults events in the
associated with large scale integration of wind CEB system and by considering the three phase
power are less complication than that of the fault levels at all the bus bars.
distributed power generation. Yet dynamic
stability is the concerning issue due to the In present Sri Lankan power system stability
intermittency of the wind power availability issues are more vulnerable during the periods
and also due to the limitations of supporting the of more concentrated heavy hydro generation.
fault current by wind generators. These periods are generally occurred in July-
August and November-December, for normally
Methodology anticipated weather patterns. Therefore within
The Sri Lankan transmission system is initially this paper the stability analysis was limited to
modelled using IPSA software, with data system peak loading occurred at 19:30 hrs on
sufficient for steady state analysis. The best 7th August 2006. The considered peak had
option would be to build the model from the 50% concentrated hydro 30% concentrated
beginning using the database, which will thermal and 20% dispersed mixed generation
accommodate the easy maneuverability of the [Ceylon Electricity Board, 2006]. All loads
model for different applications and users. It were modelled as fixed load as it represent
also helps the gradual upgrading of the worst case. In other words the motor loads
database with real data when available. The generally supports towards greater stability
model thus developed was run for steady state than the fixed loads.
analysis and the computed values were
compared with actual measured values for Finally, the impact on the dynamic stability of
different loading conditions. the Sri Lankan power system on integrating
272
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
moderately large wind power to the system at done for successful three phase re-closure after
identified Grids, were studied for the selected 300 ms and also for unsuccessful re-closure.
marginal stability cases. Study within this The fault clearing time was assumed as 100 ms.
paper was limited to 100 MW wind integration Four cases were studied in simulation and their
at Puttlam Grid Substation for a night peak waveforms are shown in Figure 1, 2, 3 and 4.
loading condition.
Case 1: Fault at Kps_220 bus bar cleared
Results through bus bar protection
Following bus bars are more vulnerable for
dynamic instability when three phase fault
occurs at the line with heaviest current, closest
to the bus bar. Measured and calculated
voltages and load flow are listed in the Tables
1, 2 and 3 (Ceylon Electricity Board, 2006).
Table 1. Comparison of voltage results
Voltage Voltage
Bus Bar measured calculated
(kV) (kV) Figure 1(a) Bus voltages Figure 1(b) SM rotor angle
Kotmale_220 225.0 223.3
Victoria_220 230.0 226.2 In Figure 1(a), around 165 MW generation is
Kalnya_132 130.0 129.4 loss from the system when clearing the Kps
Fort_132 128.0 129.2 220 bus bar for a bus bar fault. In Figure 1(b),
SapGS_132 132.0 129.8 KGT1 is a machine closer to the affected bus
LaxNew_132 132.0 133.9
bar and Victoriya G1 is a machine far away to
the affected bus bar. Both machines are
remaining in stability after the bus bar fault.
Table 2. High fault level 220 kV bus bars & Case 2: Double Faults on Biyagama Kotmale
heavily loaded lines with load flow 220 kV Lines at Biyagama end
comparison
Symmetri. Heavily loaded Calcula. Measur.
Bus Bar fault level line / load load
(pu MVA) Generator (MW) (MW)
Biygma_220 46.601 Kotmale_220 141.83 140
Kps_220 44.228 Kps_CCG 104.99 105
Kotmale_220 40.948 Biygma_220 141.83 140
Panni_220 39.409 Biygma_220 58.63 60
Ktgoda_220 37.226 Biygma_220 81.33 80
Victoria_220 36.167 Kotmale_220 91.25 90 Figure 2(a) with re-closing Figure 2(b) without re-closing
273
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Case 3: ApuraNew line at Kotmale without re- that the wind is connected through a
closing 300 ms clearing delay representative synchronous generator. During a
disturbance wind generation will back off from
the system without supporting the fault current,
in worst case scenario.
Conclusions
Calculated load data through steady state load
flow analysis is tallied with measured data.
Instabilities shown in the system for
unsuccessful auto re-closing at Biyagama
Figure 3(a). No wind Figure 3(b). With wind Kotmale 220 kV double circuits for
simultaneous faults, is a known issue in the
system, which was instrumental on blackout of
Discussion
the system as well. Also few selected marginal
cases of instability were not worsened due to
The database approach is the most appropriate the introduction of 100 MW wind at Puttlam.
for power system model development in IPSA. This was evident from the case of three phase
The initial steady state and security analysis fault on ApuraNew line at Kotmale bus bar
was done through the partly populated without re-closing and 300 ms clearing delay.
database. Studying of the large power system Therefore the developed IPSA model of the Sri
was done in highly structured manner in the Lankan power system can be used for further
order of steady state load flow analysis, rate studies of the system.
exception handling and contingency analysis.
The dynamic stability of power system was References
studied only after solving and / or identifying
the unsolved issues in the order mentioned. In Ceylon Electricity Board (2003) Long Term
the process, the model is validated by Transmission Development Studies 2004 –
comparing to the measured load data and also 2013.
to the known instability issues. Finally the Ceylon Electricity Board (2005) Long Term
model was used to forecast the system stability Generation Expansion Plan 2006 – 2020.
for large scale wind integration at the most Ceylon Electricity Board (2006) System
promising site at Puttlam, where the base load Control & Operations – Monthly Review
coal power plant is also anticipated. In large Reports.
scale wind integration of wind, it is assumed
274
FACULTY OF SCIENCE
and
POSTGRADUATE
INSTITUTE OF SCIENCE
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
275
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Figure 2. Experimental DRIFT spectra for spinach (left) and calculated spectra for the cyanidin
models (right)
Figure 3. Experimental DRIFT spectra for purple yam (left) and calculated spectra for the
cyanidin models (right)
276
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Figure 4. Experimental DRIFT spectra for pomegranate (left) and calculated spectra for the
cyanidin models (right)
277
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
278
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
279
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
with four properties. We called the abstract set trigonometry, Journal of Mathematical
‘Layer’ and showed that the real and complex Analysis and Applications, 181(2), 431-457.
fields are trivial and simple cases of a Layer Fleury, N., Detrauberg, M.R. and Yamaleev
called harmonic Layer. We presented an R.M. (1995) Extended complex number
application of our abstract setting in the analysis and conformal-like
construction of harmonic functions. The transformations, Journal of Mathematical
definition is restricted to be primitive to meet Analysis and Applications, 191(1), 118-136.
our immediate needs, but could be refined with Nasir, H.M. (2007) Spherical harmonics
further properties. Further research is needed to transform in a non-polar coordinate system
find other properties to be satisfied by the and application to Fourier series in 2-
Layers. Sphere, Mathematical Methods in the
Applied Sciences, John-Wiley, 30(14),
Acknowledgements 1843-1854.
Nasir, H.M., Faham, M.A.A.M. and Waasim,
This research was partially supported by the M.S.M. (2007) Power representation of
Peradeniya University research grant spherical harmonics and fast discrete
RG/2006/48/S. spherical transform on cubed sphere,
International Workshop on Applied
References Mathematics and Computational Science
Fleury, N., Detrauberg, M.R. and Yamaleev, (IWAMCS07) Sept. 25-29, 2007, The
R.M. (1993) Commutative extended University of Electro-Communications,
complex numbers and connected Tokyo, Japan.
280
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Theorem 1
f \ (x;η )= h( x ) exp[ηT ( X ) − A( η )] , (2) Let X = ( X 1 , K , X n ) be a random sample
from a population with distribution f (x; θ ) and
g (θ ) be a real-valued function on Θ . Let T ( X )
281
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
[
Proof: Since E T * ( X ) = α kg (θ ) , and ] parameter p is T * ( X ) = ∑X i / r(n + τ 2 ) .
[ ]
Var T * ( X ) = α 2Var [T ( X )] , the mean squared n
i =1
∑ ∑
2
2 2 2
with ν = c( n) k τ , where c(n ) , is a known r Xi ⎜nr− Xi ⎟ /(nr−1) rτ +τ −1 + r(nr− 2)(nr−3) is
⎜ ⎟
i =1 ⎝ i =1 ⎠
constant depends on n , and τ is the coefficient
of variation of the probability distribution the optimal shrunken estimator of the variance
which is assumed to be known. V F (μ ) = rp (1 − p ). Similarly, the optimal
shrunken estimator of the variance functions of
Applications some basic types of NEF”s are given in the
The following applications illustrate the use of following Table 1.
the above results.
Conclusions
Binomial distribution: Let When the coefficient of variation τ of the
X i ~ Β (r , p ), i = 1L n , 0 < p < 1, and r be fixed. In distribution is known the above admissible
this case for an iid sample of B(r,p) one can shrunken estimators of the variance functions
identify have minimum mean squared errors in the
⎛ p ⎞ exp(η ) class CT (α ) = {αT ( X ) | 0 < α < ∞} .
η = log⎜⎜ ⎟,
⎟ p= , B ( p ) = −nr log (1 − p ) , A(η ) = nr log[1 + exp(η )].
⎝1 − p ⎠ 1 + exp(η )
282
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Table 1. Optimal shrunken estimators of the variance functions of several distributions, which
belong to NEFs
∑ X /(n + τ
i =1
i
2
)
3 Negative binomial ⎛ ⎞
( )
n n
∑ ∑
2
NB (r , p ) r2 Xi ⎜nr+ Xi ⎟ /(nr+1) rτ −τ −1 +r(nr+2)(nr+3)
⎜ ⎟
i =1 ⎝ i =1 ⎠
4 Gamma ⎛ n ⎞
2
Gam (r , λ ) r⎜
⎜ ∑ X i ⎟ /( nr + 2)( nr + 3)
⎟
⎝ i =1 ⎠
283
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Major and Trace Element Composition of Bottom Sediments of the Malagane Tank:
Relationship with Sources
284
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
The available metals considered are Na, K, Ca, including organic particles are flushed into the
Mg, Fe, Mn, Ni, Zn, Pb and Cr, and their mean tank body.
concentrations, in mg/kg, are 122, 170, 671,
2078, 1723, 240, 3.2, 5.2, 7.5 and 3.5, Conclusions
respectively. All the available metals show a
general distribution pattern which is an The analysis of sediments in the Malagane tank
association with clay sediments. The total emphasizes that the sediments are comparable
content of elements of bottom sediments was to the upper continental crust composition and
compared with that of the Upper Continental therefore mainly derived from the upper
Crust (UCC) values in order to determine the catchment of the tank. Apart from the
anthropogenic influence to the tank sediment. terrestrial materials, geochemical processes
Since the average elemental levels of the associated within the sediments also play a
terrain were not available, the bulk chemical major role. The total metal concentrations and
composition of the sediments of Malagane tank distribution indicates though extensive paddy
was compared with the UCC values proposed cultivations are practiced in the tank watershed,
by Rudnick and Gao (2003) (Figure 1). With the anthropogenic heavy metal pollution levels
few exceptions, the major and trace element are not significant in the study area.
compositions of Malagane tank sediments are
either similar or slightly depleted (alkali and Acknowledgements
alkaline earth elements, P, Mn and Cu) to that Authors of the paper like to express their
of the UCC composition. As and V are sincere acknowledgement to the National
enriched in sediments in the main tank body Science Foundation of Sri Lanka for the
and sediments close to the bund. The Zr and Ce financial support given throughout the research
contents are enriched in the Thaulla and in the project.
tank body, whereas Hf, Pb and Th are enriched
throughout the tank compared with that of the
UCC values. Since the Thaulla of the tanks References
acts as a sediment trap, heavy minerals Rudnick, R.L. and Gao, S. (2003) Composition
supplied from the watershed are deposited of the continental crust, Treatise on
mainly at Thaulla and only fine grains Geochemistry, 3, 1-64.
Figure 1. Average upper continental crust normalized elements from the Malagane tank
285
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
286
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
t-test
H0 : βi = 0 (Null Hypothesis)
H1 : βi ≠ 0 (Alternative Hypothesis)
ˆ
T = βi − βi ≈ t13 −10 ;(Figure 2) Figure 2. Probability distribution graph (t-test)
ˆ
Var ( β i )
Inf = -94.5658 + 1.7213 UeR - 0.7812 RER + 0.2782 GoE + 2.0309 EDS - 0.6988
GDPGR + 13.1975 Pop
287
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
(a) The positive integer solutions of
The origin of continued fractions is
traditionally placed at the time of the creation x 2 − dy 2 = 1 are (x, y ) = ( p is −1 , q is −1 ) for i ≥1
of Euclid's Algorithm. Euclid's Algorithm is when s is even and (x, y ) = ( p 2 is −1 , q 2 is −1 ) for
used to find the greatest common devisor (gcd) i ≥ 1 when s is odd.
of two numbers. However, by algebraically
manipulating the algorithm, one can derive the (b) The positive integer solutions of
simple continued fraction of a rational number x 2 − dy 2 = −1 are nonexistent when s is even
p q as opposed to the gcd of p and q. It is and (x , y ) = ( p(2i−1)s−1 ,q(2i −1)s−1 ) for i ≥ 1 when s is
well-known that integer solutions of a linear odd, where the two sequences {pn } and {q n } are
Diophantine equation ax + by = c, where given by,
a , b, c ∈ Z, can be obtained by expressing a b
as a simple finite continued fraction. This idea p0 = a0 , p1 = a 0 a1 + 1 , p k = a k p k −1 + p k − 2 ; k ≥ 2 and
can be generalized for the general linear
q0 = 1 , q1 = a1 , q k = a k q k −1 + q k −1 ; k ≥ 2.
Diophantine equation;
a1 x1 + a 2 x 2 + ... + a n x n = b , where (Anderson and Bell, 1997)
a1 , a2 ,..., an , b ∈ Z, can be solved by using
continued fractions. Here, when n is large we This can be extended to find the integer
need more conditions to solve the above solutions to x 2 − dy 2 = n by writing
Diophantine equation using continued
fractions.
[
n = 1.n = (x * ) − d ( y * ) (r 2 − ds 2 )
2 2
]
A Non-linear Diophantine equation of the form
x 2 − dy 2 = n , where d and n are integers, is
( )
2
(
= x * r ± dy * s − d x * s ± y * r , )2
known as the more general Pell’s equation, where (r, s ) is a particular solution to
which can be solved for integer solutions by
using continued fractions with additional two
( )
x 2 − dy 2 = n and x * , y * is an integer solution
to x 2 − dy 2 = 1 . Thus r * = x * r ± dy * s and
conditions. Further, this work has been s * = x * s ± y * r is an integer solution to
generalized to a Non-linear Diophantine
x 2 − dy 2 = n . Since we have infinitely many
equation of the form integer solutions for x * and y * , infinitely many
ax 2 + bxy + cy 2 + dx + ey + f = 0 , where integer solutions to our original Pell’s equation
a, b, c, d , e, f ∈Z. can be obtained.
Integer solutions to this equation can be Now consider a Non-linear Diophantine
obtained subject to three more conditions. equation of the form
ax 2 + bxy + cy 2 + dx + ey + f = 0 , where
Method a, b, c, d , e, f ∈ Z. Integer solutions to this
Use Pell’s equation; x 2 − dy 2 = n , where equation can be obtained with the following
d , n ∈ Z. Find integer solutions to Pell’s conditions:
equation when d >0 and d ≠ m for all m ∈ N.
(a) a, c ≠ 0
The following theorem can be used to find the
integer solutions to x 2 − dy 2 = 1 . (b) b 2 − 4ac >0
Theorem: Assume that d (> 0) is not a square (c) 4(b 2 − 4ac ) ≠ m for all m ∈ N.
of an integer and that pk /qk is the kth convergent
of d . Let s be the period of the simple Reduce this equation into a Pell’s equation of
continued fraction representation of d . the form z 2 − Mω 2 = N , and then obtain the
288
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
integer solution for it. Hence, infinitely many will be an integer solution to z 2 − 20ω 2 = 256.
integer solutions for x and y can be obtained.
Hence, y = − B ± z1 = − 26 ± 144 = −17 or 118 .
Result 2A 10 10
Therefore, y = y1 = −17 and hence,
− by1 − d ± ω1
Consider ax 2 + bxy + cy 2 + dx + ey + f = 0 . x= = 7 and 39.
2a
Solution − (by + d ) ± L , Then, (39,−17 ) and (7,−17 ) are integer
x=
2a solutions to x 2 + 3xy + y 2 + 5 x + y + 1 = 0.
where L = (by + d )2 − 4a(cy 2 + ey + f ) .
Take L = ω 2 = (by + d )2 − 4a (cy 2 + ey + f ) for Thus, we can find infinitely many integer
solutions for given Non-linear Diophantine
some integer ω . Then Ay 2 + By + C = 0, equation.
where A = b 2 − 4 ac , B = 2bd − 4 ae ,
Discussion
C = D − ω and D = d 2 − 4af .
2
289
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
Results
In the Northcentral and Northwestern regions
of Sri Lanka, groundwater is a limited resource The results show that the stream and canal
and is the main source of drinking water. The waters are rich in Ca and Mg (Figure 1)
area which is of flat terrain receives relatively although the other ions are considerably low. In
low rain fall with high seasonal and spatial contrast, lake water is characterized by many
variations, and has high evapotranspiration. dissolved ions except Ca and Mg (Figure 2).
The main groundwater bearing formations in The most interesting feature is that the highest
the area are the fractured crystalline bed rock values of Fe and Mn are in lake waters. The pH
and the weathered overburden. The alluvial is also more towards alkaline in lakes (4.68 –
deposits along the streams present in the area 9.40) compared to that of streams and canals.
also play a vital role in water supply. Almost
all the groundwater comes from precipitation, The dissolved ions in ground waters are
which soaks into the soil and passes down to comparably high. However, the deep
the aquifer. The regolith aquifer coincides with groundwater has more dissolved ions and the
the areas where there exists the small tank shallow ground water is markedly
cascade system since ancient periods. These characterized by higher K levels (Figures 3 and
aquifers are closely linked with the surface 4). In general, the concentrations of Fe and Mn
water in streams, canals and tanks. The geology are low in groundwater compared to surface
of the area is dominated by Precambrian waters. The pH is more alkaline in the
metamorphic rocks. The main metamorphic agricultural wells and dug wells while it is
rocks of the area are charnockites, charnokitic more acidic in the tube wells. The agricultural
gneiss, quartzite and calcgneiss. The objective wells are unusually high in Na.
of the study was to characterize the ground and
surface waters found in Northcentral and Discussion
Northwestern Sri Lanka.
During field investigations it was found that
Materials and methods the water table is shallow. In lake waters, the
high iron concentration gives rise to reducing
For the relevant study, the groundwater and the conditions and the other ions which show high
surface water was categorized as (a) Lakes (b) concentrations are due to high
Canals and streams (c) Shallow groundwater evapotranspiration (Guo and Wang, 2004).
(Dug wells, Agricultural wells) and (d) Deep However, Na concentration is dramatically
groundwater to study the variation of ions. The high due to mineral dissolution. The low Ca
pH, temperature and the conductivity of 296 and Mg values in lakes may be caused by the
samples were measured during the sample precipitation of calcite and dolomite (Guo and
collection in the field using the pH meter (PH Wang, 2004). The canal and stream water
320/Set 1) and the conductivity meter (Model chemistry is complex and has no relationship
CM-7B, cell type CG-201PL instrument), with natural conditions.
respectively. Dissolved cations such as Na, Ca,
Fe, Mn, K and Mg of the water were measured When one compares the dug wells and the
in triplicates using the Perkin Elmer Atomic agricultural wells, the latter has high water soil
Absorption Spectrophotometer (AAS). To interaction. Therefore, the dissolved ion
minimize analytical errors, appropriate concentrations may dramatically increase. In
standard techniques were followed. Statistical that case, due to the recycling of the same
software Minitab, Origin and Arc/GIS were water, the saline condition can be developed
used to carry out chemical data analysis. and increased with time.
290
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
In the case of tube wells, there is no clear Further, the data obtained will enable to
correlation with the geological formation. characterize the water in the area and be
However, in the areas that have quartzite, helpful to understand the water pollution due to
dissolved ion content is low. Thereby, it can be natural and anthropogenic activities.
suggested that the chemistry of groundwater
depends on the geological conditions and the References
flow direction.
Guo, H. and Wang, D. (2004)
Conclusions Hydrogeochemical processes in shallow
quaternary aquifers from the northern part
The water chemistry of the area is principally of the Datong Basin China, Applied
controlled by the redox conditions, pH, intense Geochemistry, 19, 19–27.
agricultural practices, hydrological conditions,
climatic conditions and underlying geology.
Figure 1. Cation variation (ppm) in canals/streams Figure 2. Cation variation (ppm) in lake water
Figure 3.Cation variation (ppm) in tube wells Figure 4. Cation variation (ppm) in dug wells/
agricultural wells
291
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
Extraction of proanthocyanidins
Monomeric flavan-3-ols (catechins) and
oligomeric proanthocyanidins are important An extract of PA’s in aqueous 70% acetone
constituents of tea (Camellia sinensis) flush. was prepared from 1 kg of fresh tea leaves
Proanthocyanidins (PA’s) are formed by the according to the procedure described
formation of inter-flavan links between previously (Tammer et al., 1994) and purified
monomeric catechins such as (+)-catechin and by solvent partition and chromatography on
(-)-epicatechin. PA’s are widely distributed Sephadex LH-20. The effluent from the column
plant defence compounds, and have a general was concentrated on a rotar vapor and freeze
toxicity towards fungi, yeast and bacteria dried to give the PA extract.
(Dixon et al., 2004). It has been postulated that
the composition and nature of PA’s may be
related to the increased resistance of some tea HSCCC - Solvent systems for HSCCC
cultivars to infection of tea flush by the fungus
Exabasidium vexans (Punyasiri et al., 2005), Four different two phase solvent systems
that causes Blister Blight leaf disease, the most containing different ratios of hexane, EtOAc,
serious leaf disease of tea in Sri Lanka. Tea is MeOH and water were prepared. The two
cultivated for its leaf, and tea flush is phase solvent system Hex : EtOAc : MeOH :
particularly susceptible to attack by E. vexans H2O (1:5:1:5) was selected for the separation of
which attacks young and succulent tea leaves the PA extract on the basis of the partition
and stems. coefficient and settling time.
292
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Time (hrs)
293
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
Isopropylation of benzene
Montmorillonite clay belongs to smectite
family and has an expandable layer structure of 2-Propanol (50 mmol), benzene (100 mmol)
aluminosilicate. The layered structure and Mn+-montmorillonite (0.5 g), 10-
facilitates the uptake of cations such as Zn2+, camphorsulfonic acid (0.5 g) were placed in a
Al3+ and Fe3+. Cation-exchanged clays show two necked flask equipped with a Dean-Stark
high Bronsted acidity due to protons in the apparatus. The reaction was maintained at
interlamellar zone formed by polarization of reflux temperature for 48 h. The mixture was
coordinated water molecules. Such clays can then allowed to reach room temperature and
function as solid acid catalysts in organic filtered to separate the clay catalyst. The filtrate
reactions (Richard, 1992). was distilled under reduced pressure to collect
the fraction containing cumene. The effect of
Cumene, the starting material for the industrial reaction time, molar ratio of reactants and
preparation of phenol, has been formed in 26% amount of catalyst on the yield of cumene was
yield by alkylating benzene with isopropanol in investigated. The product was characterized
the presence of Al3+-montmorillonite using an authentic sample of cumene (FT-IR,
1
(Wijeratne et al., 2006); isopropyl carbocation H-NMR, gas chromatography).
formed from isopropanol undergoes an
electrophilic aromatic substitution reaction. Results
With a view to improving the yield of cumene,
we envisaged facile generation of isopropyl Effect of different catalysts on isopropylation of
carbocation from an intermediate sulfonate benzene
ester, and examined the reaction of isopropanol
and benzene in the presence of metal-cation- As expected, camphorsulfonic acid alone did
exchanged montmorillonite and not produce cumene from the reaction of
camphorsulfonic acid. benzene with isopropanol. However, the
reaction mixture of camphorsulfonic acid,
Materials and methods benzene, isopropanol and Mn+-montmorillonite
(Mn+ = H+, Al3+ or Fe3+) furnished cumene in
Preparation of catalyst 60% yield; with Mn+-montmorillonite (Mn+ =
Zn2+, Cu2+ or Cd2+) the yield was 50%. The
Na+-Montmorillonite clay (5 g) was stirred reaction mixtures containing Na+ or Ce4+-
overnight with a 0.5 M solution of cation montmorillonite did not produce cumene.
chloride (pH 3.5, 200 ml) to obtain Mn+-
montmorillonite clay catalyst (Mn+ = H+, Na+, Effect of amount of clay catalyst
Zn2+, Cd2+, Cu2+, Al3+, Fe3+, Ce4+). The clay
was then centrifuged and washed with distilled Isopropylation of benzene in the presence of
water repeatedly until washings showed camphorsulfonic acid was carried out with
negative test for chloride ions. The clay sample varying amounts of Al3+-montmorillonite (0.1–
was then dried under ambient air for a week 1.0 g). With small amounts (0.1 or 0.2 g) of the
and ground to pass through a mesh of size 100. clay, the yield of cumene was 20%; with 0.3 g
The clay samples were characterized using the yield increased to 40%. Cumene was
their XRD patterns and FT-IR spectra. The isolated in 60% yield when 0.5 g or more (0.8,
catalyst was activated at 150 0C for 2 h prior to 1.0 g) clay catalyst was used.
the study of its catalytic activity.
294
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
295
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
Research design
Chromium is present in the environment in
both trivalent and hexavalent oxidation states. The amount of Cr(III) adsorbed by brick was
Cr(III) occurs naturally and is an essential determined by treatment of standard Cr(III)
micronutrient that is involved in metabolizing solutions of concentration varying from
sugar, protein and fat (Park and Jung, 2001). 1.00 mg dm-3 to 50.0 mg dm-3 with brick
Heavy metals such as chromium are not samples fired at each temperature. These
decomposed through biological means, and measurements were then used for adsorption
they are toxic to aquatic organisms as well as isotherm analysis. Comparison between the
humans if exposure levels are sufficiently high. percent Cr removal from aqueous solutions and
Chromium compounds are present in many the turbidity of aqueous brick suspensions was
types of industrial effluents, including also investigated.
electroplating, metal finishing, leather tanning,
pigments and photography. In order to comply Results and discussion
with the tolerance limits, it is essential that
industries treat their effluents to reduce the The amount of Cr(III) ions (in µg) adsorbed on
chromium to acceptable levels before discharge a unit mass (1.000 g) of laboratory fired brick
in order to maintain environmental quality and clay samples (d < 1.0 mm) varies linearly with
human health. Ion exchange, electrochemical the initial concentration of Cr(III) in solution
precipitation, chemical precipitation and for samples fired up to 400 °C. Samples fired at
adsorption are some acceptable treatment higher temperatures show less extent of
methods practiced worldwide (Kocaoba and adsorption at all concentrations investigated,
Akein, 2004). Most of these methods suffer and further they obey Type I adsorption
from drawbacks such as high capital and isotherm (amount adsorbed vs. concentration),
operational cost and/or problems associated reaching saturation at higher concentrations.
with the disposal of sludge. The objective of This observation supports monolayer coverage,
this research is to investigate the removal of suggesting the formation of inner sphere
Cr(III) species from aqueous solution, through complexes between Cr(III) and surfaces of
adsorption, by brick particles fired at different minerals containing Fe, Al and Si present in
temperatures, and to optimize the firing brick matrix. Change in the extent of
temperature for the most efficient removal. adsorption with temperature is probably due to
phase changes and chemical processes that
Materials and methodology would occur at high temperatures.
An interesting feature observed thorough these
Materials measurements is that the percent removal of
Standard solutions of Cr(III) were prepared Cr(III) by brick generally increases up to a
using Cr2(SO4)3 (98% assay). Unfired, dry certain intermediate temperature followed by a
brick clay samples obtained from Gelioya, decrease at high temperatures. However,
Kandy, were first sieved to obtain particles of turning points of percent removal –
diameter < 1.0 mm, and fired in the laboratory temperature curves, and the shape of the curves
at 100 °C, 200 °C, 300 °C, 400 °C, 500 °C, depend on the initial Cr(III) concentration
600 °C, 700 °C, 800 °C and 900 °C. (Figure 1).
296
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
297
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Chemical Reactivity of Mica, K-Feldspar and Apatite in Organic Acids and Water
Introduction
Mineralogical characteristic of minerals and
biogeochemical processes in nature are the key
factors controlling the chemical weathering
process that leads to the release of ions from
the mineral which provide the essential
elements for higher plants, with the exception
of nitrogen. Therefore, study of the short term
stability of mica, feldspar and apatite in
different geochemical conditions leads (1) to
identify the least stable mineral under natural
conditions, (2) to estimate the possible Figure 1(b) Biotite
potential for use as fertilizer and (3) to
understand the stability of mineral mixtures.
298
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
299
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
300
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
In December 2004, tsunami inundated and
thereby contaminated about 40,000 dug wells
and a considerable number of surface water
bodies. Therefore, potable water supply
systems that were based on groundwater
became unusable. Contamination of
groundwater became a massive problem to be
solved in the affected areas. In general, deep
sea sediments contain a large variety of organic
and inorganic chemical compounds including
heavy metals, some of which can be toxic
(Jacobs et al., 1987).
The present study was carried out in the west
coast of Sri Lanka, between Ambalangoda and
Rathgama towns in the Galle district
(Figure 1). This highly variable nature of
geology and environmental conditions within
short distance makes groundwater quality
highly variable. The purpose of the present
study is to understand the natural processes of
ground and surface water quality improvements
with time, under the influence of prevailing Figure 1. The study area showing the soil
environmental conditions. types and surface environmental
conditions
Materials and methods
Sixty five sampling points, including dug Results and discussion
wells, tube wells and surface water bodies were The concentrations of the heavy (trace) metals
selected, and sampling was done at four appear to be significantly low and are mostly
occasions during the period between below the detection limits. This indicates that
December, 2005 and December, 2006. Water the possible effect on water quality by the
samples were analyzed for heavy metals, (analyzed) heavy metals that were brought by
nutrients. pH, electrical conductivity (EC), tsunami water or the agitated sediments, is not
salinity, total dissolved solids (TDS), dissolved significant at most places. One out of five tube
oxygen (DO) and temperature. A portable wells sampled showed contamination by
multi-parameter meter and ELE portable nutrients. The dug wells in marshy
turbidity meter were used for determining environment where the water table was very
general chemical parameters. Heavy metals shallow were also showed contamination by
(Hg, As, Cd and Pb) were analyzed using nutrients.
Varian atomic absorption spectrophotometer
(AAS) while phosphate and nitrate levels were Based on the groundwater salinity levels
determined using standard methods (APHA, observed in December, 2005, the dug wells
20th edition). could be arranged in the order from low to high
salinity (Figure 2). When comparing the results
of subsequent salinity measurements in the
same wells, interesting behavioral patterns can
be seen (Figure 2). The wells with high salinity
301
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
as seen in Figure 2 have shown rapid decrease Figure 4 that there is no definite trend of
with time. Majority of these wells were situated temporal or spatial salinity variation other than
close to the coast and had inundated by the slight shrinking of the high salinity zones with
tsunami water. Wells with originally moderate time. This situation can be attributed to the
and low salinity on the other hand (Figure 2) previously mentioned diversified geological
do not show a fast or a significant change with and environmental conditions of the area.
time. These wells thus could be assumed to be Water quality also appears to be dependent on
saline even prior to tsunami. recharge conditions, sandy, peaty, or lateritic
soil compositions, local morphology and
presence of marshy conditions of the area. This
can be seen when compared to the geographic
distribution of salinity with the surface
geological and environmental conditions
(Figure 1) of the area. High evaporation
associated with shallow groundwater level in
the vicinity of marshes can retard the rate of
decrease of salinity created by the tsunami. In
addition, intrusion of saline water from
brackish water streams and other surface water
bodies can further complicate and mask the
natural salinity decreasing processes.
Conclusions
Figure 2. Salinity variation in dug wells Ground waters in the tsunami affected coastal
areas have been faced quality problems of
diversified nature. Increase in salinity has
become the most significant problem while
contamination with respect to nutrients has
been evident mostly in marshy areas with
shallow groundwater table. Heavy metal
contamination however, is not significant.
Groundwater quality has shown a slow
recovery with respect to salinity. Complex
geological and environmental conditions in the
area however, complicate and mask the natural
rates of recovery.
Acknowledgements
Financial support from NSF research grant
Figure 3. Salinity variation in tube wells RG/05/DMM/03 is acknowledged.
The salinity levels of the tube wells have References
shown an increasing trend with time (Figure 3).
Of these tube wells, TW100 is located near the Jacobs, L., Emberson, S. and Huested, S.S.
tsunami boundary, TW 101 and 103 within the (1987) Trace Metal Geochemistry in the
tsunami flood zone and TW102 and 104 Cariaco, Trench, Deep-sea Research,
outside the tsunami zone. Royaume University, Oxford: Pergamon.
265-284.
The area wise distribution of the salinity of
groundwater at different times is shown in
Figure 4. It is seen from the salinity maps of
302
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
303
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
304
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Table 1. Physical Properties of Complex-b b may be due to the loss of intensity as a result
of scattering of radiation between the pores
Colour and Shiny, black solids present in the particles. The particle sizes
Appearance calculated using XRD data and the Debye-
Hardness 2.5-3.0 (Mohs scale) Scherrer formula is 13.24 nm, which is 3 nm
Solubility Insoluble in almost all solvents smaller compared to that of the corresponding
except conc. HNO3 and HCl dithiooxamido complex. Interestingly, the
M.P. > 280 ºC reduction in the size of the new complex
together with the reduction in intensity of the
Streak colour Dark gray
XRD peaks compared to the copper(II)
dithiooxamido complex may also be indicative
of the packing of the molecule to form a
0.6 macrocyclic configuration. Analysis of the data
obtained in Tables 2 and 3 implies an
0.5
interesting affinity of the complex towards
0.4
Ag(I) ions. However, the test results reveal that
the ions such as Pb(II), Cd(II), Zn(II) and
Ni(II) neither trap nor release Cu(II) ions by
Intensity
0.3
180
160
140
120
Table 2. Testing for Ag(I) ions using AAS
Intensity
100
80
60
Concentration of Absorption Absorption
40 the Ag(I) solution for Ag(I) for Cu(II)
20
10 20 30 40 (3 cm3) passed
2 - th e ta v a lu e
5 ppm - -
10 ppm - -
Figure 3. XRD spectrum of Complex-b
20 ppm - 0.002
Discussion 20 ppm - 0.004
The hardness of the complexes is between 20 ppm - 0.005
finger nail and calcite according to the Mohs 100 ppm - 0.029
hardness scale. IR data show significant
differences in spectra of free dithiooxamide 100 ppm - 0.046
and metallo-derivatives in the region of N-H 100 ppm - 0.065
absorptions. The band at 3137 cm-1 attributed
0.001 M - 0.092
to N-H stretching frequency is sharper and
higher in frequency in metallo-derivatives 0.001 M - 0.122
compared to that of free dithiooxamide. The 0.001 M - 0.141
presence of additional strong bands at 872 cm-1
and 1518 cm-1, and the absence of a strong 0.002 M - 0.172
band at 1577 cm-1 corresponding to the C=O 0.002 M - 0.233
stretching frequency of Cu(acac)2 fragment
confirm the formation of a macrocyclic 0.002 M - 0.254
complex. The decrease in the intensity of the 0.010 M - 0.614
major band of the XRD spectrum for Complex-
305
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Table 3. Ion exchange ability (Complex-b) solution of Ag(I) ions through a column packed
by the copper(II) complex shows a substitution
Aqueous Precipi- Precipi Effluent with of Cu(II) ions of the complex only with Ag(I)
solution tating -tate conc. ions specifically and not with other ions such
passed agent with ammonia as Pb(II), Cd(II), Ni(II), Zn(II) at
effluent concentrations above 20 ppm. The unusual
0.01 M Ag (I) NaCl - Blue solution behavior of the new complex may be used for
the application of this type of systems for
0.01 M Pb (II) NaCl White No change
specific trapping of some other noble ions such
0.01 M Cd (II) NaOH White No change as Au(I).
0.01 M Ni (II) DMG Red No change
0.01 M Zn (II) K2CO3 White No change Acknowledgements
Financial support given by NSF (Grant No
Conclusions RG/2006/FR/02) is appreciated.
In summary, a blackish shiny Cu(II) complex
with the particle size of 13.2 nm and hardness References
of 2.5-3.0 has been synthesized by treatment of
Cu(II), dithiooxamide and acetylacetone. It is Martin, T.G. and Cumming, S.C. (1973)
difficult to study the solution chemistry of the Square-planar nickel(II) and copper(II)
compound because it is extremely insoluble in complexes containing 14- and 15-
almost all the solvents. IR and XRD data reveal membered macrocyclic ligands, Inorganic
that the structure of the new compound is Chemistry, 12, 1477-1482.
different from both copper(II) dithiooxamido Udugala-Ganehenege, M.Y., Jane, M.,
and copper(II) acetylacetonato complexes, and Hryhorczuk, M., Wenger, I.O.E. and
the N-end of dithiooxamide is primarily Endicott, J.F. (2001) Electron/atom transfer
bonded to the metal center showing a in halo-bridged homo-bimetallic
possibility of having pores in the complex. complexes, Inorganic Chemistry, 40, 1614-
AAS of the effluent collected by passing a 1625.
306
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
307
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
said, “I think it’s really important to give an difference at p=0.05 between the control group
analogy without explaining more and more and the experimental group.
when we get confused.” Pictorial analogies
seemed to clear up confusion about the Conclusions
analogue concept that teachers were using. The
students also thought that analogies would be Analogies had a motivational impact on the
useful to explain or describe objects or students. Students could be made interested to
processes that are too small that they cannot be learn abstract concepts in chemistry by using
seen with the naked eye. One student said, analogies in classroom teaching. Analogies not
“Whenever the concepts that the teacher only help to explain or introduce difficult
teaches at a level that we actually cannot see concepts in chemical equilibrium but also to
what’s going on, it’s good to use analogies.” make students attentive and actively involved
At the end of the lessons some students were in the lessons. In addition, use of analogies
able to make some analogies based on concepts enables to have high interaction between the
relating to their real world experiences. teacher and students, and among students. It is
recommended that teachers need to develop a
Analysis of the pre-test results showed that personal set of useful analogies in their
there was no difference between mean values teaching. It is necessary to select appropriate
of control (49.2) and experimental groups analogies that could be related to students own
(49.0). Standard deviations were 12.8 and 14.4, experience without misleading or confusing
respectively. Using the two sample t-test, it was them. Otherwise they may lead to
found that there was no significant difference at misconceptions. Therefore, teachers should pay
p=0.05 between the control group and the more attention in selecting analogies for their
experimental group of the pre-test results. explanations.
Analysis of the post-test results showed that
there was a difference between mean values of References
the control (32.29) and experimental groups Frieedal, A.W., Gabel, D.L. and Samuel, J.
(83.07). Standard deviations were 10.48 and (1990) Using analogies for chemistry
9.43, respectively. Using the two sample t-test, solving: Does it increase understanding?
it was found that there was a significant School Science and Mathematics, 90, 674-
682.
308
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction
Results
Simulation models are logical descriptions of
the interrelationships among elements of a
decision problem or the sequence of events that Regression analysis
occur in a system over time. The major The expected quarterly demands for each
advantage of a simulation model is its ability to quarter in 2007 given in Table 1 were
incorporate uncertainty in one or more input calculated with the use of the summary output
quantities. It captures probabilistic behavior obtained from the linear regression analysis
and allows one to estimate probability method (Utts and Heckard, 2004). Next, the
distributions and key statistics such as mean, error in this quantity was taken to be normally
variance, etc. that cannot be derived distributed with mean (μ) = 0 and standard
analytically. deviation (σ) = standard error obtained in the
summary output given in Table 2. As shown in
Due to the fluctuation of the sales according to Table 3, the actual demand for each quarter in
the market demand, effectiveness of the 2007 was taken to be the (expected demand) +
company promotional activities, seasonal (error), where error ~ N(0,σ).
variations and competition with other suppliers,
the quantity of sales in any given period is not Table 1. The expected quarterly demands
fixed. Therefore, the company has to make an
order under certain uncertainties. Quarter Exp demand
Quarter 1 336931.4 m
This paper uses a simulation model to
determine the optimal order quantity for a Quarter 2 296937.2 m
wholesale supplier for a certain period of time Quarter 3 294826.7 m
by analyzing the quantified risk involved in Quarter 4 459949.5 m
each demand quantity decision.
309
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Table 4(ii). 2007 demand distribution (assumed to be Normal with the following parameters)
μ σ
First quarter 336926.6 m 24050.88 m
Second quarter 296930.2 m 15626.32 m
Third quarter 294827.6 m 29435.96 m
Fourth quarter 459950.5 m 15188.3 m
Cash balance/Budget(A)
Rs 1000000
Table 4(v). Simulated quantities
Interest
Demand Revenue Cost Bank loan for the Profit
(m) (Rs) (Rs) (Rs) loan(Rs) (Rs)
First quarter 335626.1 5893750 5125000 4125000 185625 583125
Second quarter 296776.3 5422250 4715000 3715000 167175 540075
Third quarter 297992.9 4479250 3895000 2895000 130275 453975
Fourth quarter 458860.3 9430000 8200000 7200000 324000 906000
Net profit 2483175
310
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Mathematical model of the simulation model charts for the profit distributions, the risk
for the nth quarter : involved in each order quantity was quantified
and the order quantities given in Table 5 were
Let taken as the optimal order quantities which
Unit cost = c , Unit price = p , gave a relatively high profit for each quarter in
Demand = Dn , Order quantity = xn , 2007.
Revenue = Rn , Cost = Cn , Profit = Pn
Budget = A , Amount of loan = Ln , Discussion
Annual loan interest rate = r ,
Interest for the loan = In , As shown in Table 5, even with the best values
Depreciation rate of the remaining quantity = d for mean profit, there is a risk of getting
negative profit. But these preferred outputs
Decision variable : xn have a relatively a few number of influential
Input variables : Dn , c , p inputs compared to the other simulations for
each quarter. Also, these order quantities give a
Objective : Max Pn = Rn – Cn – (In + Ln) relatively high mean and a small standard
subjected to a minimum risk , deviation for the profit. Hence, the values
where 310,000 m, 270,000 m, 250,000 m and 440,000
Rn = c * min (Dn , xn) , Cn = c * xn , m can be considered as the optimal order
Ln = Cn – A , In = ( Ln * r) ÷ 4 quantities for the non-woven interlining in the
1st, 2nd, 3rd and 4th quarters for the year 2007,
For the year 2007 respectively.
Unit cost (c) = Current unit cost + cost
increment References
Unit price (p) = Unit cost * (1 +expected Albright, Winston, Zappe (2002) Data analysis
Profit percentage) and decision making with Microsoft
Current unit cost = Rs. 20, EXCEL, Duxdury Press, 891-977.
d = 2% , r = 18% Evans, J.R. (2000) Spreadsheets as a Tool for
Increment in cost ~Triang (0.30 , 0.50 , 0.80 ) Teaching Simulation, Volume 1, University
Expected profit percentage ~ Uniform (14.5%, of Cincinnati, Cincinnati,
15.5%) Utts, J.M. and Heckard, R.S. (2004) Minds on
Statistics, Second edition, Thomson, 492-
By analyzing the summary statistics report, 512.
sensitivity analysis report and the tornado
Table 5. The order quantities with the maximum mean for the profit
Quarter Simulation Order Minimum of Mean of the Maximum of
No. Quantity (m) the profit profit (Rs) the profit(Rs)
(Rs)
1 4 310,000 -568242 674806.7 755428.2
2 3 270,000 -77841.7 619493.9 663760.1
3 4 250,000 -541476 560140.5 613019.3
4 3 440,000 161871.1 975908.3 1053350
311
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
312
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
In te n sity
002
with exfoliated clay materials is suggested by
003
the absence of basal reflections associated with
the LDH, in the PXRD patterns. Introduction 001 Parent LDH
of exfoliated LDH into the amorphous sugar
matrixes improved the glass transition
temperature of the parent sugar. The LDH 10 20 30 40 50 60
nanolayers give the LDH filled nanocomposites
better thermal stability than the pristine sugar 2θ/degrees
due to differences in the chemical structure of Figure 1. The PXRD patterns of parent and
the two components and the restricted thermal exfoliated LDH
motion of the sugar molecules in the inorganic
matrix. On the other hand, the LDH/maltose
nanocomposites were stable for more than six
months. However, such thermal property
enhancement was not observed with
LDH/sucrose nanocomposites.
Conclusions
The nanocomposites prepared using maltose,
are amorphous and the glass transition
temperature of the composite materials, are
superior to the pristine maltose. The higher Figure 2. The SEM images of parent LDH
degree of hydrogen bonding and the restricted
thermal motions of the sugar molecules within
LDH nanolayers may improve the thermal
stability of the LDH/maltose nanocomposites
than pristine maltose, thus stabilize the 20% mal-nc
amorphous maltose by increasing the glass
transition temperature. Thus, we expect this 10% mal-nc
new finding may open up a new application of
LDH/sugar nanocomposites as drug excipients
Intensity
Amorphous maltose
313
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Acknowledgements
20% suc -nc The University Research Grant is
acknowledged for the financial support.
Intensity
314
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Introduction ⎛ ∂2 ∂2 ∂2
Δ s = s t ⎜⎜ (1 + t12 ) + 2 t1 t 2 + (1 + t 22 )
⎝ ∂ t12
∂ t1 ∂ t 2 ∂ t 22
Numerical computations in solving problems
defined on the sphere suffer from many + 2 t1
∂
+ 2 t2
∂ ⎞
⎟
difficulties near the poles, known as the ‘pole ∂ t1 ∂ t 2 ⎟⎠
problems’, when using spherical polar
coordinate system for the spherical surface. For and st = 1 + t12 + t22 , is given by pl(m,n) ,
yl(m,n) = l
example, in the computation of global weather st 2
prediction models, concentrated grid points m+n=l-1,l, where pl( m ,n ) are polynomials in t1
near the poles increase the amount of
computations in the pole region where and t2 satisfying the differential equation
quantities of interest are of less important than (1 + t 12 )
∂2p
+ 2 t1 t 2
∂2p
+ (1 + t 22 )
∂2p
the other parts of the globe. ∂ t 12 ∂ t1 ∂ t 2 ∂ t 22
⎛ ∂p ∂p ⎞
Avoiding pole problems has attracted some − ( l − 1 ) ⎜⎜ 2 t1 + 2 t2 − l p ⎟⎟ = 0
researchers in the recent past (Nasir and ⎝ ∂ t1 ∂ t2 ⎠
Faham, 2006). Among the recent developments
in this direction, one of the present authors has and are explicitly given by the non-zero real or
constructed weakly orthogonal and orthogonal imaginary parts of
spherical harmonics in a non-polar spherical m n
⎛ ( m − p + n − q) / 2 ⎞ ⎛ l ⎞ p q (l-p-q)
co-ordinate system based on the ‘cubed sphere’ ∑ ∑ 2 2 ⎜⎜ ⎟⎟ ⎜⎜ ⎟⎟ t1 t 2 i , i = −1
p = m1 q = n1 ⎝ (m − p ) / 2 ⎠ ⎝ p, q ⎠
defined from the surface of a unit cube (Nasir,
2007). with m1 = m mod 2, n1 = n mod 2 and the
subscript 2 in the summation indicates that the
Computational efficiency is also one of the index variables increases with step 2.
main targets in spherical transforms, which has
been considered for spherical harmonics in A set of continuous spherical harmonics can be
spherical polar coordinates by some constructed from the eigen-functions for one
researchers (MohlenKamp, 1999). In this work, face forming a six-tuple of functions
more properties and results for the orthogonal
Y ( t 1 , t 2 ) = ( y (t 1 , t 2 ), y ⎛⎜ −t 1 , t ⎞⎟ ,
−t 2
and weakly orthogonal spherical harmonics on ⎝ ⎠ 1(1) 1
315
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
316
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
317
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
318
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Conclusions
The glycolipid synthesized via Koenigs-Knorr
Figure 1. Texture of β-sitosteryl-2, 3, 4, 6- synthesis has thermotropic liquid crystalline
tetra-O-acetyl-β-D-glucopyranoside properties.
in the thermotropic hexagonal
columnar mesophase under crossed Acknowledgements
polarizer and analyzer
The authors thank the University of Peradeniya
The compound β-sitosteryl-2, 3, 4, 6-tetra-O- for providing funding for this work.
acetyl-β-D-glucopyranoside started to melt at (RG/2006/50/S).
131.7 °C. When cooling, it started to solidify at
127 °C forming the hexagonal columnar phase. References
This texture was clearly observed when the
temperature was around 28 °C. Jimmy, W. and Larry, S. Wikipedia: The free
encyclopedia [online], Available at;
Although the compound appears to be solid at http://en.wikipedia.org/wiki/Liquid_crystal
room temperature, it can exist in the liquid [Accessed 10 September 2006].
crystalline phase due to super cooling on the Smits, E. (1969) The sweet world of liquid
cooling cycle, and the transition from the crystals: the synthesis of non-amphiphilic
anisotropic liquid crystalline phase to carbohydrate-derived liquid crystals,
crystalline solid may occur below ambient [online]. Groningen, University of
temperature. Groningen, Available at:
http://dissertations.ub.rug.nl/faculties/scienc
This molecule consists of a rigid moiety which e/1998/e.smits/ [Accessed 21 July 2006]
is also called as mesogenic core (sugar group) Vill, V., Demus, D., Goodby, J. and Gray, G.,
to which several flexible alkyl chains are (1998) Handbook of Liquid Crystals. LCI
connected. Such kind of molecules have the [online]. 1(2A,2B), p.3. Available at:
ability to form discotic liquid crystals which http://liqcryst.chemie.uni-
can be stacked into columns (Figure 2). hamburg.de/lcionline/article/oldarticles/arti
According to the texture obtained under cle_0698/article_2.htm [Accessed 4
polarizing light microscopy, the compound November 2006].
synthesized in this study seems to arrange in a
hexagonal array forming hexagonal columnar
phase.
319
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Materials and methods When both ligands were present in the reaction
mixtures, either in 1:1 or 1:2 or 2:1 ratios with
Calvin type pH titrations were carried out using different metal concentrations (0.5 × 10-4 – 1.0
a Biochrom C18 pH meter. The mixtures × 10-2 mol dm-3), ternary complex systems of
320
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
MLX type were formed via either ML or MX remove toxic cadmium in the body by forming
binary complexes. In the statistical comparison more stable cadmium – aspirin – vitamin C
of the stabilities of ternary systems over binary complexes which are water soluble and hence
systems, it was found that the formation of they can be excreted from the body.
ternary complexes was much more stable than
that of binary complexes at all concentration References
conditions, as the values of Δ log Km were
greater than - 0.6, the reference value (Sigel, Irving, H and Rossotti, H. S. (1953) Journal of
1975). Chemical. Society, 3397-3405.
Kain, W. and Schwederski, B., (1994)
Conclusions Bioinorganic Chemistry: norganic
Elements in the Chemistry of Life,.
Both aspirin and vitamin C alone and together Chichester, UK: John Wiley and Sons.
can be chelated with trace amounts of cadmium Sigel, H. (1975) Angewante Chemie, 14, 394-
under physiological conditions. The results 402.
explain that there is a higher possibility of www.euro.who.int/document/aiq/6_3cadmium.
forming more stable ternary complexes in this pdf
environment. According to these results, it can
be suggested that vitamin C and aspirin
together in suitable concentrations be used to
321
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
322
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Conclusions
(a) (b)
The carbohydrate liquid crystal 3-O-decanoyl-
Figure 1. Texture of 3-O-decanoyl-1,2 : 5,6- 1,2 : 5,6-di-O-isopropylidene-α-D-
di-O-isopropylidene-α-D- glucofuranose was shown to form a lyotropic
glucofuranose in the (a) lyotropic hexagonal columnar phase with chloroform, as
hexagonal columnar mesophase (b) well as a thermotropic hexagonal columnar
thermotropic hexagonal columnar phase. According to the polarizing microscope
studies, the product may be an amphotropic
mesophase at room temperature liquid crystal.
under crossed polarizer and
analyzer. Acknowledgements
Thermotropic liquid crystal properties were The authors thank the University of Peradeniya
observed by heating a small amount of the for providing funding for this work
sample in a temperature controlling device. The (RG/2006/50/S).
solid melts immediately to an isotropic liquid
on heating (117-124 °C), yielding a colorful References
texture which might correspond to a hexagonal
columnar phase, on cooling (55.7 °C) (Figure Daoud, M. (1998) Matter diffusion in
1(b)). The texture remained unchanged even at hexagonal columnar phases. Brazilian
room temperature. Thus, although the Journal of Physics [online], 28. Available
compound appears to be solid at room at:
temperature, it may be super cooled and the http://www.scielo.br/scielo.php?pid=S0103
transition from anisotropic liquid crystalline -7331998000400005&script=sci_arttext
phase to crystalline solid phase may occur [Accessed 1 June 2007].
below ambient temperature. Smits, E., (1969) The sweet world of liquid
crystals: the synthesis of non-amphiphilic
carbohydrate-derived liquid crystals
[online]. Groningen, University of
Groningen. Available from:
http://dissertations.ub.rug.nl/faculties/scienc
e/1998/e.smits/ [Accessed 11 July 2006].
Vill, V. (1998) Carbohydrate Liquid Crystals.
LCI [online]. Available at:
http://liqcryst.chemie.uni-
Figure 2. Schematic diagram of a cross hamburg.de/lcionline/article/oldarticles/arti
section of the two-dimensional cle_0698/article_2.htm [Accessed 4
array of hexagonal columnar September 2006].
323
FACULTY OF ARTS
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
325
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
fact that conceptualizing space is as much a effects as mentioned above. In other words,
political as a social process. certain spaces are used in the print media
without any critical inquiry about the origin of
Repeated use of such spatial categories in the such spaces or any evaluation about their
print media allows them to be accepted as potential social and political ramifications.
‘normal’, ‘given’ spatial entities. Thus, the Such uncritical endorsement of spaces
mere fact that certain spaces are frequently fundamentally conceals the fact that these
used in the print media can itself be a plausible spaces are, by all means, contested spaces. The
factor of their reproduction. The important repeated and frequent use together with the
point that should be highlighted here is that uncritical usage of space within the print media
such frequent use can ascertain an existence thus allows them to be considered as
that a particular space does not actually have unproblematic spatial categories. We thus
and policy decisions can be made on that basis. conclude that making decisions based on
uncritically publicized spaces and by
Further, it appears that not only the frequent implication disregarding their contested nature,
and repeated usage, but also uncritical usage of may lead to further social and spatial
such spaces in the print media and their consequences.
unquestioned acceptance by the
readership/audience can also produce similar
326
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Preparation of a Zoning Plan for the Madu Ganga Special Management Area
327
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
328
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
329
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Mahakanda Gal-bungalow). From the east it for Environmental Studies. This project has
covers the Hantana Mountain peak to the west integrated participatory mapping method to
up to the Gampola – Kandy railway line integrate local knowledge in the Virtual
(behind the Faculty of Engineering). Peradeniya University. The GIS mapping and
participatory mapping techniques include very
Methodology comprehensive spatial layouts of university
land information with infrastructure conditions.
Under the broad methodological approach of
Virtual GIS there are five methods that have Currently, there is no GIS mapping system
been adapted to develop the Virtual Peradeniya exist in the university to explore and access of
University. These methodologies are: the administrative or physical infrastructural
information. The lack of spatial information on
1. Introduction of the latest digital image the university limits many stakeholders to
technology on the World Wide Web of participate in the correct decision-making. The
Google Earth as base information for the study also highlights that it is not the lack of
Virtual University of Peradeniya. expertise or facilities which limited such model
2. Integration of Global Positioning Systems, development rather it is the lack of interests
GIS, and Google Earth for land use and VISION of people to build such robust
mapping. virtual model of Peradeniya University.
3. Integration of participatory GIS approach Therefore, it is suggested that it is vital to
into university land use mapping. develop a virtual model of the university where
4. Introduction of the technology of digital not only locally but also globally interested
image processing (GIS and Remote people can participate in higher education.
Sensing) for future land use monitoring.
5. Development of a Virtual GIS technology Result of Virtual Peradeniya University
to publish Faculty of Arts and its
departmental information through WWW The GIS mapping project has gathered
for wider audience data access such as necessary basic information from diverse
staff members, outsiders, international sources. The information includes remote
community, and students around the world sensing images using GOOGLE EARTH
using Web-GIS. technology, Global Positioning System for
locational information, Survey Department
Maps, ground truth information from the GIS
Project objectives team, administrative knowledge from the
There are a number of project objectives that university communities, university map
were identified. These are, to: archives, and information from maintenance
department. The project developed two large
1. Develop Virtual Peradeniya University scale digital maps for the wider community
Maps for the Faculty of Arts to support access and two other large scale analogue maps
monitoring future land-use changes, for university community use.
updating data of, and to enhance spatial
decision-making procedures within and Future training and discussion
outside the faculty
2. Develop an integrated database with GIS It is expected to have a preliminary training
for the entire university administrative and discussion with the key members from the
system Faculty of Arts to share ideas and comments
3. Develop spatial knowledge on location of for further development of the GIS mapping.
university premises and provide university Once we receive comments and suggestions
community and outside community to from the Arts Faculty, then the second level
search and explore university facilities project (university mapping) will be submitted
through Web-GIS. to the Vice Chancellor and other senate level
4. Expose Peradeniya University to a wider decision-makers’ opinions. Subsequently, with
community through WWW. a small training session the virtual GIS
mapping will be handed over to the Dean Arts
and Vice Chancellor to display in the relevant
Discussion, relevance and result of the study buildings for public access.
During the last one year four members team
engaged in a variety of map development, data
collection, and model development of the
Virtual Peradeniya University from the Center
330
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
331
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
K.W.G.R. Nianthi
Department of Geography, Faculty of Arts, University of Peradeniya
Introduction
Material and methods
Climate change (CC) may adversely affect the
water availability and management problems, To understand the reality of CC and its impact
thus in turn affecting agricultural production. on paddy cultivation in the DZ, a questionnaire
As the Dry Zone (DZ) of Sri Lanka is a water survey was carried out with particular reference
deficit area, its agriculture is highly vulnerable to farmers’ response to CC and its effects on
to CC. The economic disruption, social communities and their livelihood systems.
dislocation and human suffering caused by Discussions were also held with officers and
drought, among DZ communities have often the village communities. The 03 villages are
been described in historical chronicles. In the located close each other, are small in size and
modern time too, almost every year, there have have no remarkable differences among them
been media reports on rainfall failure in many and therefore, were analysed together with a
parts of the country especially in the DZ. total sample of 99 households. The sample was
Farmers` attitudes and perceptions towards CC drawn by using the random sampling method.
and adjustment of farm management practices The field survey was carried out during the
were examined in this paper. To understand the Maha season (MS) (October to March) 2003-
reality of CC and its impact on the paddy 2004. In addition to the questionnaire survey,
community in the DZ; Kekirawa division (with temperature and rainfall data were collected
03 villages) in the Anuradhapura district was from the Meteorology Department in Colombo
selected for this study. The main objectives of and the trends and fluctuation of rainfall and
the study are; to examine the present CC and temperature variation have been analyzed using
fluctuation of rainfall and temperature since linear trend analysis.
1960 and to study the responses of paddy
farming communities to perceived threats of Results and discussion
CC and its consequences. The study revealed
that the study area is heavily dependent on the Currently, 07 rain-fed small tanks are
availability of water and timely rainfall and the functioning in the study area; but invariably
dry spells are often caused by the failure of there is no water in the YS in all these tanks.
Northeast monsoon (NEM) rainfall. Such Therefore, none of the farmers has been able to
failure leads to a continuous decline in the cultivate paddy during the YS. Even in the MS,
availability of water resulting in the reduction during the period of the field survey, water had
of rice production. During the last two decades, dried up in all the tanks due to insufficient
rainfall has decreased drastically in the study rainfall from NEM. Therefore, the Maha
area. The irrigation system consisted of 07 cultivation was not possible due to inadequate
rain-fed tanks but these small tank systems do rainfall. The preceding YS cultivation had also
not protect the farmers from severe water failed due to inadequate rainfall. Next to paddy
shortages. Climatic factors have significantly cultivation, shifting cultivation (Chena slash
affected paddy production, its extent and yield, and burn cultivation) has been the main source
particularly under rain-fed conditions in the of livelihood for paddy farmers in this area.
field area. Due to the uncertainty of timely Due to the uncertainty of timely rainfall, there
rainfall, there is an increasing trend on the part is an increasing trend for farmrs to engage in
of farmers to engage in ‘Chena Cultivation’ ‘Chena cultivation’ than in paddy cultivation
than in paddy cultivation especially in the especially in the YS. None of the tanks are
‘Yala’ season Farmers have noted a variety of linked to any major irrigation schemes and
changes in climate affecting their paddy farms: therefore, water availability is vulnerable to
including an increasing general warming of the local rainfall failure. Although almost all the
climate with reduced rainfall and a shift in the farmers had cultivated paddy during the MS,
seasonal pattern of rainfall. the crop was completely damaged due to
inadequate rainfall during the NEM. The
332
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
333
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
A.S.M. Nawfhal
Department of Geography, Faculty of Arts, University of Peradeniya
334
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
buildings and monuments in the core/built up The challenges ahead for Kandy city can be
area gazetted by the government. A considered in terms of this increasing demand
systematically demarcated zoning plan has also for services, controlling traffic, waste
been gazetted to control the pressure on land management and minimizing natural disasters.
use. The need also has been recognized to In order to protect and safeguard the city and
protect paddy lands, unstable lands, stream and its socio-cultural heritage from the internal and
road reservations. These measures are expected external pressure or modernization, careful
to minimize the damage to the existing planning is required.
landscape.
335
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
A.M.P. Senanayake
Department of Archaeology, Faculty of Arts, University of Peradeniya
336
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
337
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
338
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
T. Wagalawatte
Department of Archaeology, Faculty of Arts, University of Peradeniya
Rock formations of the earth are categorized planes in all of the rock outcrops at the
into different types based on their natural Vessagiriya archaeological site.
characteristics. The rocks in Sri Lanka are also
classified by geologists based on the basis of The layering of rocks are formed during the
many factors. For an archaeologist, it is metamorphic or sedimentary processes and
important to have a scientific knowledge about layers are markedly shown be their different
rocks in making logical explanations and colors. Rocks having dark light colors had been
interpretations about environmental selection used for constructions at the Vessagiriya site.
of sites, historical constructions using rocks as For example, metamorphosed limestone was
raw material and the functional use of used to construct the moon-stone
technologies in relation to raw material. sandakadapahana, steps etc. This indicates that
the artist had applied his knowledge and
Vessagiriya archaeological site, which was technical skills for constructions considering
investigated since the 19th Century, is located at naturally available raw materials.
the southern end of the heritage city of
Anuradhapura. The central feature of the site is Ancient constructions of large dimensions
represented by three relatively large and which can be seen even today were exploited
scenically beautiful rock out crops, named by through simple techniques. These techniques
archaeologists as Rock A, B and C. Recent were applicable, since the raw material existed
investigations indicated that the rock outcrops in a controllable manner in the environment.
were used from the pre historic period. The pre For example, they made shila puwaru (stone
historic people used it as a protective safe slabs) from naturally layered rocks. A location
haven. The Early Iron Age folk used the lower near Rock C in the southern part of the
outcrops as a natural foundation for their Vessagiriya displays such shila puwaru that
settlements. The Early Historic (3rd Century had been separated from the parent rock.
BC) devotees chiseled the rocks and carved out
shelters for forest dwelling Buddhist monks. In order to separate the rock, it had been
They carved out drip-ledges to prevent rain broken along the holes created in the rocks.
water seeping into the caves and to engrave the This was evident by the drill holes found on the
earliest inscriptions of Sri Lanka. The advanced edges of the main rock. Metal equipment that
period of rock usage at Vessagiriya commences were used to separate the rocks by creating
from the Middle Historic period (4th to 10th pressure inside the holes were unearthed by
Century AC). It was noted during the 2007 archaeological investigations. Six of such
archaeological investigations that the ancient equipment are exhibited in the archaeological
people had an in-depth knowledge on the museum in Anuradhapura and two are
physical characteristics of rocks in selecting registered under Nos. C. 7.12 and C. 7.10. This
rocks as raw material, with special reference to method is yet utilized to break rocks. The only
the built environment. A whole range of difference between the ancient and present
architectural art representations were fashioned practices is the use of metal pegs and chemicals
out of rock during the Middle Historic Period. to create pressure at present.
Metamorphic rocks are the most abundant As described above, the main reason for the
rocks found in Sri Lanka. Due to the pressure built environment activity around Vessagiriya
and temperature changes, protoliths of these area (including Isurumuniya and Ranmasu-
rocks have been subjected to mineralogical and uyana) is the availability of natural resources
structural changes. Most of these rocks can be suitable for constructions and art
easily broken through the foliation plane which representations.
was formed from the metamorphic process.
We can identify well developed foliation
339
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
M. Gnanananda
Department of Pali and Buddhist Studies, Faculty of Arts, University of Peradeniya
340
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
M. Wijitadhamma
Department of Pali and Buddhist Studies, Faculty of Arts, University of Peradeniya
341
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
The Frankfurt School of Social Research and Critical Theory, with Special
Reference to Max Horkheimer
M. Rajaratnam
Department of Philosophy and Psychology, Faculty of Arts, University of Peradeniya
342
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
became evident in Horkheimer’s famous claim This study also identified in our country, the
that critical theory inherently involves the problems and the influences of orthodox
attempt to actualize the materialist content of Marxism and Positivism in the fields of
idealist philosophy. Critical theory in this sense education, research, social sciences, social and
continues the theoretical innovations political cultures undermine the genuine
undertaken by George Lukacs and Karl Korsch progress of our society in its practical fields.
in the early twenties. Both thinkers had also The social sciences and social research should
promoted a dialectical critique of crude have critical approach with new rational
materialism of orthodox Marxism and revolutionary focus in order to have
Positivism as well as fixed and finished progressive change in the society and in the
philosophical systems. Horkheimer’s critical field of political culture.
materialism anticipates what has now become
known as post metaphysical thinking. The use References
of ‘critical theory’ as code word become
evident in his early writings enabled a certain Horkheimer, M. (1937) Traditional and
interpretation of Marxism to enter academic Critical Theory: Selected Essays, Seabury,
discourse. His Italicise and Critical Theory New York 1972
focuses on the distinction of traditional and Horkheimer, M. and Ardorno, T. (1947)
critical theory. The Dialectic of Enlightenment Dialectic of Enlightenment, Verso, London,
is clearly a critique of enlightenment 1979.
undertaken from the standpoint of Horkheimer, M. (1947) Eclipse of Reason,
enlightenment itself and it remains a landmark New, Seabury, New York, 1974.
in radical thought. Horkheimer expressed that Horkheimer, M. (1967) Critique of
he could philosophically maintain his Instrumental Reason, Seabury, New York,
commitment to emancipation only by turning 1974.
away from the given historical reality.
343
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
344
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
and blaming others (21%) followed by results also suggest that high levels of distress
ventilation of emotions (31%). Increases in following a disaster which causes multiple
emotional distress were associated with disruptions in the child's life, may initially
significant increases in children's coping scores elicit a variety of coping strategies, both
(r(126) = .42, p < .001). The level of emotional positive and negative. These findings also
distress varied significantly among children indicate that the children are not passive
depending upon the type of coping strategy victims of adversity, but active survivors.
they used (t(124) = 20.85, p < .05). Blaming
others, acceptance, mental disengagement and Conclusions
humour were the coping strategies that were
least associated with emotional distress. There is a small minority of children who
Regression analyses found that use of continue to suffer from long lasting traumatic
individual coping strategies explained 34% of stress reactions; hence are at risk of developing
the variance in emotional distress. Distraction psychological problems. As all children are not
(I try to forget it) made a significant equally effective in choosing the coping
contribution of beta = .21(p < .05), while strategies which would effectively deal with
cognitive restructuring ((β =.23, p < .05) and distress, some would need guidance in
avoidance (β = .18, p < .05) also made a selecting the appropriate strategies. Findings of
significant contribution indicating that the current study would provide important
increases in the use of these three coping information to the service providers, hence
strategies were associated with increases in the could be used for the benefit of the tsunami
level of emotional distress. There was no affected children.
significant relationship between age and gender
with coping efforts. Acknowledgements
This research was made possible by a 2005
Discussion Dissertation Research Award from the National
Being affected by the loss of their familiar Centre for Advanced Studies (NCAS),
environment, disruption of the sense of Colombo.
protection and well-being, lack of care received
from the care givers might have led to elevated References
levels of emotional distress among most of the Carballo, M., Heal, B. and Horbaty, G. (2006)
affected children. The explanation behind the Impact of the Tsunami on Psychosocial
strong relationship between coping efforts and Health and Well-being, International
emotional distress seems to lie in the fact that Review of Psychiatry, 18 (3), 217-223.
the children who are highly distressed have put
more effort in coping than the children
experiencing a lesser amount of distress. These
345
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
A. Samaratunge
Department of Classical Languages, Faculty of Arts, University of Peradeniya
346
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
347
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
C.S.M. Wickramasinghe
Department of Classical Languages, Faculty of Arts, University of Peradeniya
348
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
offspring. Nonetheless, the contrast in the state my argument here it will be spelt out in
behaviour of Alkestis in these two scenes my paper.
shows us that she is psychologically disturbed.
Moreover, the behaviour of the rest of the In conclusion, what is clear is that her motherly
household does not show that her death was love predominates over her other concerns and
expected and even Pheres seems to consider this anxiety has persuaded her to offer her life
her death as a noble voluntary sacrifice but he on behalf of Admetos.
thinks it was offered on behalf of himself and
his son. Though space does not permit me to
This paper is an outcome of the research project I have undertaken with the help of a Research
Grant awarded by the University of Peradeniya, to translate the tragic-comedy Alkestis of Euripides
into Sinhala from the original Greek.
349
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Issues in Translating 'Sexuality' into Sinhala: Inadequacy of the Terms Used in the
Sinhala Translation of Lady Chatterley's Lover By D.H. Lawrence
D.M. Kumarasinghe
Department of Sinhala, Faculty of Arts, University of Peradeniya
The depiction of sexual and gender matters has English and further, it is believed to be a
been a controversial subject in Sinhala philosophic exploration of sex and sexuality. It
literature as in many other South Asian also offers provocative perspectives: female
literatures. Classical Sinhalese literature fantasy about male sexuality for the so-called
consists of translations of Buddhist texts from predominantly male audience. The citation
Pali. The writers being mainly monks who below shows the philosophical background of
emphasized the ideology of celibacy spurned Lady Chatterley's Lover:
sex and married life and described pleasures of
life as low, rude, worldly, ignoble and leading "I always labor at the same thing, to make
to ruination. This attitude that was primarily the sex relation valid and precious, instead
intended for the monastic community gained of shameful. And this novel is the furthest
ground among the laity. On the other hand, the I've gone. To me it is beautiful and tender
impact of Hindu and Sanskrit tradition that had and frail as the naked self is, and I shrink
kāmə (love) as one of the four ideals of life was very much even from having it typed.
elevated to a divine form as erotic sentiment, Probably the typist would want to
far above worldly, natural sex. Though some interfere-."1
Sinhala writers tried to emulate the Sanskritic
models, the general attitude was "The mind has to catch up, in sex. . . .
discouragement of sex or love as a literary Balance up the consciousness of the act,
experience. However, since the 19th century and the act itself. Get the two in harmony.
secular literature became established, It means having a proper reverence for sex,
apparently under the influence of western and a proper awe of the body's strange
literature, fiction being a newly introduced experience. It means being able to use the
genre. so-called obscene words, because these are
a natural part of the mind's consciousness
Except for a few early poetical works with of the body."2
descriptions of erotic scenes, and some erotic
poetical works of the 17th & 18th centuries Although the novel Lady Chatterley's Lover
A.D., we can not find many fictional works of became very popular around the 1950s in Sri
recent times that treat erotic themes. Lanka, due to the issues of gender and sexual
matters that the novel covers and its alleged
Sex is without a doubt, one of the most pornographic nature, no attempt was made to
intimate indicators of identity as it conjures up translate it into Sinhala until 1985. On the other
images of sexual activity, eroticism, pleasure, hand, contemporary attitudes of expressing sex
taboo, fantasies and desires. Language is matters in literature was very controversial in
considered to be the most intimate medium of the 1960s and 1970s; that is, “Sabhyə-asabhyə
expressing sentiments of sex. It is apparent that vādəyə” (the argument of courteousness and
the Sinhala language does not have sufficient discourteousness) was one of the main
vocabulary on sexuality and gender. However, arguments raised regarding this matter.
there is a hidden or vague vocabulary among However, the first translation (Wijesekara, M.)
Sinhala speakers for referring to sexual organs
and related physical acts. Especially as a
language with diaglossia, Sinhala has a 1
'Letters, ed. Harry T.Moore (London, 1962),
standard pattern in writing. Although many p. 972. The first typist did interfere; Nelly
taboo words are used in colloquial Sinhala such Morrison refused to go on after five chapters,
words are not used in the standard written and Lawrence had to farm it out to his friends.
language. 2
"Edmund Wilson, "Signs of Life: Lady
Chatterley's Lover" [19291, in Shores of Light
Lady Chatterley's Lover by D. H. Lawrence is (New York, 1961),
one of the most famous erotic novels written in p. 407
350
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
351
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
A. Velmurugu
Department of Tamil, Faculty of Arts, University of Peradeniya
352
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
353
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
354
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
N. Balamurali
Department of Economics and Statistics, Faculty of Arts, University of Peradeniya
355
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Table 3: The Granger causality test. contribute positively to economic growth. The
finding that FDI does not cause economic
Direction Lag F-Statistic P growth and economic growth causes FDI is
puzzling and may arise due to following
L(FDI) to L(GDP) 1 0.58986 0.449 reasons.
L(GDP) to L(FDI) 1 10.404* 0.003
Firstly, we used aggregate data which are in
Note: * denotes rejection of null hypothesis at monetary values and these data results shows
the 5% significance level. only a statistical relationship. Some impacts of
FDI in Sri Lanka quantitatively such as,
Empirical results show that all variables are knowledge acquisition technology,
stationary in second difference (Table 1) and international image can not be measured and it
suggest that the estimated coefficient for FDI may take a considerable time before these
does not significantly contribute to economic variables affect economic growth. Secondly,
growth though positively related. Other the extent of FDI inflows to Sri Lanka is
variables such as DI, OPENNESS, and LA relatively small and that may have contributed
significantly influence economic growth at the to the lack of significance impact on economic
5% significance level.6 growth. Thirdly, there may be other factors that
influence the relationship between FDI and
Based on the Granger causality test, the null economic growth, for example, firm level
hypothesis that FDI does not cause economic production and the value addition of firm to
growth is not rejected at the 5% level of domestic value addition. Other possible reasons
significance. The second null hypothesis that include the small sample size, the use of net
economic growth does not cause FDI is FDI instead of gross FDI and the possibility of
rejected at the 5% level of significance, which cointegration.
indicates there is one-way causation from
economic growth to FDI. Future research in this area may be able to
remedy these limitations and also analyze the
causal link in a multivariate VAR system to
Discussion take account of other vital determinants of FDI
Empirical evidence of recent studies is mixed. and economic growth.
For e.g. Frimopong (2006) found no causality
between FDI and economic growth. The results References
of the present study are similar to those
previously found for Sri Lanka by (Athukorala Athukorala, P.P.A.W and Karunaratna, R.M.
and Karunarathna, 2004).7 (2004) The Impact of Foreign Direct
Investments on Economic Growth in Sri
Lanka, Sri Lankan Economic Journal, 5,
Conclusions 88-119.
This study finds that FDI is not a significant Phillips, P.C.B. and Perron, P. (1988) Testing
determinant of economic growth but economic for Unit Root in Time Series Regression.
growth is necessary to attract FDI. Other Biometrica 75, 335-346.
variables such as DI, OPENNESS, and LA
6
The Durbin-Watson (DW) test for
autocorrelation in the error term indicates that
there is no autocorrelation
(value=1.87).Variance inflation factors (VIF)
of all explanatory variables are less than 10,
indicating that there is no severe
multicollinearity among the explanatory
variables.
7
Our contribution to the literature beyond
Athukorala and Karunarathne (2004) is in
using a longer time period and including
additional variables (e.g. labour force) in our
analysis.
356
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
The Causes of Domestic Violence and Its Impact on Rural Women Victims
J.A.P.S. Perera
Department of Sociology, Faculty of Arts, University of Peradeniya
Objectives of the study The findings of the study indicate that violence
has a huge impact on rural women victims. It
The purpose of this study is two fold. First, it affects their mental, emotional and physical
tries to identify the causes of domestic violence health and even results in death. It also
against rural women. Second, it attempts to prevents or restricts women’s participation in
understand the impact of it on women victims. day to day activities, in family life, cultural
activities and in the work force. Fear of
Research methodology violence restricts their movement and behavior,
A total number of 16 households afflicted with isolates them from friends and family and
domestic violence in a rural village in forces them into destitution. The abused suffer
Mawanella Divisional Secretariat area has been from deprivation, anxiety, psychosomatic
considered as the research sample of this small sympathy, sexual dysfunctions and
scale, in-depth study. The method of data reproductive disorders.
collection is qualitative and the information has
been collected through informal interviews and
the case study method. The obtained data has
357
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
• That women are victimized mainly by 5. Revising the existing laws and imposing
males while a few women are strict laws
victimized by both women and men is
an important revelation of the study. 6. Changing the negative attitudes of
women of womanhood will be an
This study brings some comprehensive effective attempt in addressing violent
recommendations for addressing and related issues. This can be done by
responding to this problem. This can be done mobilizing them through building their
through prevention, treatment and support. institutional capacities.
Prevention has long term aspects and support
and treatment can be either short term or long
term. These measures include:
358
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
M.M. Ihjas
Department of Political Science, Faculty of Arts, University of Peradeniya
359
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
360
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
Classroom Realities
S.K. Embakke
Department of Education, Faculty of Arts, University of Peradeniya
361
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
362
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
363
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
basis. (Cavadini et al., 1999; Narring and significant difference) in food habits and sport
Michaud, 1995). performance among girls and boys.
There is a significant difference in lunch habits Conclusions
among boys and girls (62% vs. 89% with
P<0.05). The research found no difference in The study revealed that there is a significant
frequency of consumption of meals and snacks difference in food habits at the pre exercise
between girls and boys, and as expected the period, during the exercise period and post
directory survey shows that the mean daily exercise period between girls and boys in some
energy intake is lower among girls than among food habits components. The association
boys (Narring and Michaud, 1995). between sport performance and food habits was
significantly higher among girls than boys.
Responses of students to open ended questions
revealed that the food habits of students were References
influenced by parents, teachers, coaches, peers, Cavadini, C., Decarli, B., Dirran, H., Candery,
cultural factors and their own psychological M., Narring, F. and Michaud, P.A. (1999)
factors. Meanwhile research revealed that the Assessment of adolescent food habits in
food habits of both boys and girls were found Switzerland, Appetite, 32 (1), 97-106.
to be quite traditional, whereby they tend to Narring, F. and Michaud, P.A. (1995)
consume traditional foods. Methodological issues in adolescent health
surveys: the case of the Swiss Multi centre-
Discussion adolescent Survey on Health, Soz
According to the chi square test and binominal Praventivmed, 40(3), 172-182.
probability test the results indicate mixed
results (both significant difference and no
364
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
M. Alawathukotuwa
Department of Fine Arts, Faculty of Arts, University of Peradeniya
Introduction
Each use of western techniques was confined to
Music can be classified as Eastern and these single forms (eg. The use of the primary
Western. While there are similarities between chords). However with time, more complex
them, the two systems employ certain chord changes were introduced and western
techniques peculiar to each one. These techniques came to be employed to highlight
techniques are used not only in classical music the meaning or emotion of the songs and lyrics.
but also in applied music such as film and Examples include the song Sandakada
drama scores and light songs. It is evident that pahanaka from the film Matara Achchi sung
western techniques are employed now a days in by Sunil Edirisingha, and music composed by
creative music whatever the fundamental music Victor Ratnayaka and the song named Sulan
system. However, only the simplest forms koda wehi sung by Neela Wicramasingha and
appear to be used. The aim of the present composed by Sena Weeerasekara . Both these
research is to study from a musicological point examples use counterpoint to illustrate the
of view the employment of musical techniques meaning of the lyric.
in light songs composed in Sri Lanka. These
include counterpoints, harmony, local and Modulation which is an important feature of
instrumental techniques. It is also possible to western music is also employed occasionally.
observe the use of techniques employed in folk Eg. Viyoo Gee Geyena hade sung by Neela
music of the west such as Yodelling and fiddle- Wicramasingha with music by Nimal Mendis,
playing. and the song Nimterak nodutu sung by Sunil
Edirisingha with music by Rohana
Methodology Weerasingha . Following the practice of some
The research methodology followed in the western songs these have their last chorus sung
study is a survey of relevant musical in a higher key (compare the song from the late
compositions. Library research includes book 50s We were two innocent hearts sung by Tony
collections and archives of recordings such as Brent).
those at the Sri Lanka Broadcasting
Cooperation. Information was sought through Vocal music plays an important part in the
interviews with artistes, composers and western system, and the voices are
producers who were influenced by western systematically classified and blended; but the
music. In this research I faced some difficulties use of multiple layers of voice in Sri Lankan
in finding certain types of applied music such music seems to be somewhat limited. Mr.
as recordings of music for drama, films and Premasiri Kemadasa’s songs often employed
tele-dramas. vocal arrangements suggestive of western
music. Eg- song Kanyawi sung by Amarasiri
Pieris and the Song Udumbara Hinehenawa.
Discussion
The use of western music techniques has Although local singers can be found whose
introduced a variety of forms in Sinhala light voices fall within the ranges of western voice
songs. The traditional Stay and Antara classification such voices are not employed
structures have been expanded through external methodically in Sri Lanka.
elements such as introduction, interlude and
coda. These elements are first evident in the In addition to the standard techniques of
songs of the gramophone era but became more western music, Sri Lankan songs also employ
complex in film songs. From the 1960s so- techniques found in western folk music such as
called “group songs” employed close harmony yodeling.
and Spanish Guitar type accompaniments.
Since the use of vocal harmony was the main The influence of western techniques is stronger
feature, these groups came to be known as in instrumental music. Eg- the positions,
harmony groups. systems of tuning and playing techniques for
365
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
the Violin such as Pizzicato and Vibrato are depth and ornamentation to the form of the
used in Sri Lankan songs. songs.
Results
Whatever the melodic basis of the song,
western techniques figure prominently. These
are introduced with the objective of giving
366
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
L. Seneheweera
Department of Fine Arts, Faculty of Arts, University of Peradeniya
367
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
S. Mantillake
Department of Fine Arts, Faculty of Arts, University of Peradeniya
368
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
is sacred and not to be worn by females. masculine dance form. Treating females as
Treating females as impure creatures by impure creatures by traditional artistes was one
traditional artistes is the basis of this idea. of the main reasons that kept females away
Restrictions and criticisms have been made by from Kandyan dance.
traditional dance masters against attempts by
females to wear traditional Kandyan dance References
costumes.
Nürnberger, M. (1998) Dance is the language
Conclusions of the Gods: The Chitrasena School and
traditional roots of Sri Lankan stage–
In this research some features of Kandyan dance, Amsterdam, Vrie University Press.
dance have been analyzed in order to explore Obeyesekere, G. (1984) The Cult of Goddess
the masculine characteristics of Kandyan Pattini, Chicago, University of Chicago
dance. It also included a discussion as to how Press.
this dance form can be determined as a
369
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
R.M.M. Chandraratne
Department of Archaeology, Faculty of Arts, University of Peradeniya
Introduction
Figure 1 shows that a premolar of Equus
The main objective of this study is to infer the species from the JPL is smaller in size to
presence of the mule (Equus hemionus) in the modern horses, but is metrically similar to that
ancient Sri Lanka. The first ever archaeological of Surkotada. The species at Surkotada are
identification of mule bones in Sri Lanka was identified as mule (Equus hemionus). The JPL
carried out by the author (Chandraratne, 2005). sample is similar to the mule hence it can be
The ancient Sri Lankan literature cites two identified as the mule. Literary sources provide
kinds of horses, horses (Assa/Aśva) and Sind the words Ājānīya Aśva, Sindhava and
horses (Saindhava) that are identified referring Aśvatara (Sindhava?) to identify the mule. In
to archaeological and literary sources. the Jatakapali (1932) and the Ummagga
Jataka, mule is referred to Sindhava, and it is
Material also mentioned that Ājānīya Aśva means an
The sample consists of an isolated premolar of excellent (true) horse, i.e. not hybrid with other
Equus from Jetavana Pilimage (JPL). Other species.
sources of data are skeletal measurements of
the modern horse, published archaeological Discussion
data on the mule (Meadow and Patel, 1997) The hybrid species of equids may be described
and ancient Sinhalese and Pali literature as a truly man-made animal. The mule’s
(Ummagga Jataka, 1966, Jatakapali, 1939). mother is mare (Equus caballus) while it is
perhaps as the father mule (Equus hemionus)
Methodology that it has made its greatest contribution to
New Archaeological Theory advocated human economies in a way similar to that of
‘generalization of data’ which came to be the donkey (Equus asinus). It is a perfect
regarded as archaeology’s supreme example of hybrid vigour as a beast of burden
achievement (Redman, 1973). This and is more sure-footed than either the ass or
methodology was used for the above literary the horse (Clutton-Brock, 1987). Its head is
evidence. The application of osteo- similar to the ass and the rest of the body is
archaeological measurements which is comparable with the horse. It is identified in
employed for species separation, size variation the Egyptian and Mesopotamian Civilizations
and sex determination (Uermann, 1978) was (1400 BC) and was also recorded from the
used for the species separation between horse Roman period (Clutton-Brock, 1987).Previous
and mule. Specifically, the morphological research on the horse remains found at Gedige
features were compared with the osteometric (69) by Deraniyagala (1972) suggested that the
data of equid teeth from the Jetavana, and teeth of the ancient small horse was metrically
Surkotada, and modern horses were used for between Mannar ponies and modern horses,
the confirmation of this research. and indicates two species of horses, local and
Sind. The local species might have derived
from the South Indian wild horse (Equus
Results nomadicus). The more recent research on
horses raises several opinions on the origin of
Fig. 1 Measurements of Premolars: the South Asian species. The wild form of
Pilimage Vs Modern Horse
horses did not exist in India during the Post-
34 36 Pleistocene period; consequently, the horse
32
30
34
32 L en gth could not be domesticated there. Rather,
28
26
30
28
Wid th domesticated horses had been introduced later
24 26 (Bőkőnyi, 1997). Besides, wild horses were
JPL SUR SUR Mo de rn extinct during the late Pleistocene period and
had not been domesticated in India.
Figure 1 Nevertheless, domestic horses were considered
370
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
as introduced species from outside (Meadow from the Persian Gulf region and subsequently
and Patel, 1997). In the light of the above spread all over India (Bökönyi, 1997).
research, it is difficult to accept the opinion
given by Deraniyagala (1972) on South Indian The author believes that local and Sind horses
wild horses. Domestic horses and hybrid refer to the mule. It is known as Aśvatarya in
species were probably introduced into India Sinhala and Assataro in Pali (Table 1).
Table 1. Selected literary evidence for Sindhavo and Aśvatara in Sri Lanka
Literary terms Sources Chronology
Sindhavo (mule?) Mahavamsa (23: 71) 5th Century AD
Susetha chatura sindhavam Mahavamsa (31: 38) data is related to 2nd
(four white mule?) century BC)
Assataro (mule) Jatakapali (in Pali) 5th Century AD
Astara (mule) Dampiya-Atuva-Geta-Padaya 10th Century AD
Aśvataraya (mule) UmmaggaJataka 14th Century AD
The early evidence goes back to the Deraniyagala, P.E.P. (1972) Faunal remains,
Anuradhapura Period (900 BC-1000 AD), The the citadel of Anuradhapura, in 1969, in the
Dampiya Atuva Geta Padaya (74/31) states Gedige area, Ancient Ceylon, 2, 155-159.
that it was born after interbreeding of ass and Clutton-Brock, J. (1987) A Natural History
mare. The Mahavamsa refers to horses that Society of Domesticated Mammals, British
were commonly used by kings and royal Museum (Natural History), Cambridge
officials for transportation and warfare. For University Press.
instance, Sena and Guttika (177-155 BC), had Jatakapali, (1939) Ed. Ven. Widurupola
brought Saindhava horses. The state chariot of Piyatissa Mahanayaka Thera, Maligakanda,
King Dutugemunu (101-77 BC) was drawn by Colombo, The Tripitaka Publicaton Press.
the white Sindhu horses while Velusumana Meadow, R.H. and Patel, A. (1997) A
tamed a wild Sindhu horse. Finally it appears comment on “Horse remains from
that a mule (Sindhava) was more useful than a Surkotada”, South Asian Studies, 13, 307-
horse at that time and was common in 315.
comparison to true horses because of its Redman, C.L. (1973) (Ed.) Research and
adaptability to the environment. Theory in Current Archaeology, New York,
Willey.
References Ummagga Jataka (14th Century AD) (1966)
(ed.) Sucharita Gamlath and
Bőkőnyi, S. (1997) Horse Remains from the E.A.Wicramasinghe, Colombo, S. Godage
Prehistoric Site of Surkotada, Kuth, Late 3rd and Brothers.
Millennium B.C., South Asian Studies 13, Uermann, H.P. (1978) Animal bone finds and
297-307. economic archaeology: A critical study of
Chandraratne, R.M.M. (2005) Horse Remains osteo-archaeological methods, World
from the Jetavana Pilimage Excavation Archaeology, 4, 307-322.
Central Cultural Fund Anuradhapura,
Proceedings of PURSE 2005, 10-35.
371
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part II, 30thNovember 2007
372
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part I, 30thNovember 2007
AUTHOR INDEX
A B
373
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part I, 30thNovember 2007
F I
374
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part I, 30thNovember 2007
Jeyaraman, U. 329 M
Jiffry, N. 133
Jin, L.J. 124 Mahatantila, K. 284
Mallawarachchi, A.S.P. 139
K Malmuthuge, N.S.P. 9
Manangoda, M.G.B.S 221
Kadawathaarachchi, L. 72 Manchanayake, G.S.M. 141
Kahanda, I.M.K 233 Mantillake, S. 368
Kalpage, C.S 193, 196, 199 Mapa, R.B. 337
Kalyanaratne, K.A.K. 120 Menike, A.M.W 196
Kanankearachchi, P.K.A.P. 141 Mikkelsen, L.L. 12
Kaneshalingam, T. 315 Mubarak, M.N.A. 301
Kanewela, U.G 233 Mudalige, A.P. 101
Karunananda, P.A.K 219 Mudalige, S.P. 141, 148
Karunaratne, D.G.G.P 202, 208 Mudiyanse, R. 133
Karunaratne, D.N. 318, 322 Munasinghe, D.M.S. 170
Karunaratne, D.N. 51, 57 Munasinghe, M.D. 157
Karunaratne, K.I. 139, 146 Munasinghe, S.R. 77
Karunaratne, S. 307
Karunaratne, S.H.P.P. 62, 98 N
Karunaratne, V. 57, 60, 312
Karunaratne, W.M.K.K. 74 Nanayakkara, D. 133
Karunathilake, A.A.K. 89 Nanayakkara, D.P. 105
Kasturi, K.M.G.A. 353 Nanayakkara, D.P. 131
Kodithuwakku, K.A.S.S. 24 Nanayakkara, D.P. 162
Kottegoda, N. 312 Nanayakkara, S.D.I. 146
Kudagama, N. 160 Nandadeva, D. 146
Kularathne, M.N. 153, 177 Nandasiri, W.T.D 181
Kumar, N.S. 51, 82, 89, 292 Nasir, H.M. 278, 312
Kumar, V. 101 Navaratne, A.N. 72
Kumara, K.B.J. 162 Navaratne, D.N 266
Kumara, W.R.B. 177 Nawfhal, A.S.M. 334
Kumarasinghe, D.M. 350 Nawodinee, M.H.V. 84
Kumarasiri, P.V.R. 141 Nazriya, N.F. 1
Kumbukgolla, W.W 82, 89 Nianthi, K.W.G.R. 332
Nilam, J. 133
L Nishshanka, S.P. 327
Nissanka, S.P. 33
Laheetharan, A. 281
Lalantha, N. 6 P
Lamawansa, M.D. 135, 144
Liyanage, G.L.A.D. 318 Pallegama, R.W. 118, 126
Liyanage, L.G.A 221 Palugaswawa, P.S 205
Panagoda, G.J. 55
Paranavitana, G.N. 190
Pathberiya, L.G. 60
375
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part I, 30thNovember 2007
376
Proceedings of the Peradeniya University Research Sessions, Sri Lanka, Vol.12, Part I, 30thNovember 2007
T wickramaarachchi, D. 22
Wickramagamage, P. 327
Tennakone, K. 275 Wickramanayake, H.B.P.W. 129
Tennakoon, D.T.B. 294 Wickramanayake, I.L. 129
Thevanesam, V. 82, 89 Wickramasinghe, A. 57
Thiruchelvam, S. 20 Wickramasinghe, C.S.M. 348
Tirimanne, D.N. 27 Wickramasinghe, W.A.L.D 193
Wickramasuriya, H.V.A. 39
U Widanapathirana, L.S. 322
Wijayakulasooriya, J.V 236
Udugala-Ganehenege, M.Y. 304 Wijayasinghe, I.B 249
Ukuwela, K.D.B. 67 Wijekoon, A. 133
Wijekoon, P. 281
V Wijekoon, W.M.A.M.B. 292
Wijerathne, K.M. 118
Velmurugu, A. 352 Wijerathne, M.A.D. 3
Vidanapathirana, E. 327 Wijeratne, S. 294
Vidanarachchi, J.K. 6, 9, 12 Wijesekara, P.N.K. 172
Wijesinghe, A.R.A.S. 44
W Wijesinghe, K.G.G. 70
Wijesinghe, S.J. 175, 177
Waasim, M.S.M. 278 Wijesundara, D.S.A. 57, 60
Wagalawatte, T. 339 Wijesundara, M. De S. 151
Wanigasekara, A. 164 Wijesundara, N.L.S. 151
Wanniarachchi, C.R. 148 Wijesuriya, W.A.S.S.K. 148
Warusawithana, L. 65 Wijetunge, D.D.S. 153
Weerahewa, J. 29 Wijewardana, T.G. 166
Weerakoon, S.B. 186, 188 Wijitadhamma, M. 341
Weeraratne, T.C. 98 Wimaladharma, W. M.G.I.K. 118
Weerasekara, A.I 263 Wimalasinghe, E.R.R. 160
Weerasinghe, W.P.T.T. 175 Wolseley, P. 60
Weerasundera, N.K. 160
Welagedara, A. 301 Y
Weliwegamage, U.S.K 193
Wellappuli, N.C. 120 Yakandawala, D.M.D. 70
Weragodetenna, D. 329 Yaparatna, Y.M.C.D. 37
Werellagama, D.R.I.B. 190 Yatigammana, S.K. 47, 65
Wettasinghe, K.A. 118 Young, S.M. 290
Wickrama, S.M.I 236
377