Hydrology and Water Resources

SEISMIC ANALYSIS OF ELEVATED RESERVOIRS

Assist. Prof. Dr. Kamila Kotrasov1 Assoc. Prof. Dr. Eva Kormankov2 Assoc. Prof. Dr. Ioan Sorin Leoveanu3
1 2

Technical University of Koice, Slovakia Technical University of Koice, Slovakia 3 Transilvania university of Brasov, Romania

ABSTRACT Elevated reservoirs are used to store a variety of liquids. During earthquake activity the liquid exerts impulsive and convective pressures (sloshing) on the walls and bottom of tank. This paper provides theoretical background for analytical calculating of elevated tanks during an earthquake and deals with comparing of simplified seismic design procedures for elevated tanks, and the applicability for subsoil classes. The analysis has been carried out considering four different subsoil classes A, B, C, D, given EC8. The design by simplified seismic procedures given EC8 and Housner model was compared. Keywords: elevated tank, simplified procedures, seismic, impulse mass, convective mass INTRODUCTION Seismic event is certainly one of the most critical external events regarding safety of industrial plants, as demonstrated by recent earthquakes. If industrial facilities store large amount of hazardous materials, accidental scenarios as fire, explosion or toxic dispersion may be triggered, thus possibly involving working people within the installation, population living in close surrounding or in urban area where the industrial installation is located. Liquid storage tanks are considered essential lifeline structures. Large-capacity ground-supported tanks are used to store a variety of liquids, e.g. water for drinking and fire fighting, petroleum, chemicals, and liquefied natural gas. Elevated tanks are used in military bases, industrial facilities, and nuclear reactor illustrations. These structures consist of two main parts: a tower and vessel. The former can be a steel braced frame, a multi-column assembly, or an axisymmetric pedestal shell. The vessel comes in a variety of geometric shapes such as cylinders, spheres, cones, ellipsoid, or a combination of any of these geometric shapes. Elevated tanks are vulnerable to earthquake excitation mainly because of the relatively small resistance that the supporting system can offer during seismic events. Most elevated tanks are regarded as essential facilities as they should remain functional even after a major earthquake. The seismic analysis and design of liquid storage tanks are complicated by many numbers of problems, for examples: dynamic interaction between contained fluid and vessel which is called fluid-structure interaction; sloshing motion of the contained fluid; and dynamic interaction between vessel and supporting structure. In addition, the
1

13th International Multidisciplinary Scientific GeoConference SGEM 2013

supporting tower may need to be analyzed in post-elastic state, and for special cases, a three-dimensional analysis may be required to take into account torsional effect on the supporting structure. The many different geometric shapes of both vessel and tower dictate different techniques and methods of analysis for each application, and finally, soil-tower interaction could under specific conditions have a significant effect on seismic response of the tower. Satisfactory performance of tanks during strong ground shaking is crucial for modern facilities. Tanks that were inadequately designed or detailed have suffered extensive damage during past earthquakes [2 8]. Knowledge of pressures and forces acting on the walls and bottom of containers during an earthquake and frequency properties of containers is important for good analysis and design of earthquake resistant structures/facilities tanks. SEISMIC DESIGN OF LIQUID STORAGE TANKS Seismic design of liquid storage tanks has been adopted in [4, 8, 10]. When a tank containing liquid vibrates, the liquid exerts impulsive and convective hydrodynamic pressure on the tank wall and the tank base, in addition to the hydrostatic pressure. The dynamic analysis of a liquid filled tank may be carried out using the concept of generalized single degree of freedom (SDOF) systems representing the impulsive and convective modes of vibration of the tank liquid system. For practical applications, only the first convective mode of vibration needs to be considered in the analysis (Fig. 1). The impulsive mass of liquid mi is rigidly attached to tank wall at height hi (or hi' ). Similarly convective mass mc is attached to the tank wall at height hc (or hc' ) by a spring of stiffness kc. The mass, height and natural period of each SDOF system are obtained by the methods described in [4, 8, 10]. For a horizontal earthquake ground motion, the response of various SDOF systems may be calculated independently and then combined to give the net base shear and overturning moment. The most tanks have slimness of tank , whereby 0,3 < < 3. Tanks slimness is given by relation = H/R, where H is the height of filling of fluid in the tank and R the tank radius [6 10].

= kc/2 mc
kc/2

H
z x

mi hc hi ' ( hc ) ( hi' )
z x

Fig. 1. Two single degree of freedom systems for ground supported cylindrical tank

Hydrology and Water Resources

THE SIMPLIFIED SEISMIC ANALYSIS PROCEDURES A simplified analysis procedure has been suggested by Housner, in 1963 [1], for fixed base elevated tanks. In this approach, the two masses (mi and mc) are assumed to be uncoupled and the earthquake forces on the support are estimated by considering two separate single-degree-of-freedom systems: The mass mc represents only the sloshing of the convective mass; the mass of consist of the impulsive mass of the fluid, the mass derived by the weight of container and by some parts self-weight of the supporting structure. This two-mass model suggested by Housner has been commonly used for seismic design of elevated tanks. The dynamic characteristic of these model are estimated by using the expressions given by equations (1b), (4b), (5)-(16). Similar equivalent masses and heights for this model based on the work of Velestos and coworkers [Malhotra], with certain modification that make the procedure simple, are also suggested in the Eurocode 8 ( next only EC8) are given in Table 1 and (1a), (4a) and (6). mi is the impulsive mass of fluid, mc is the convective mass of fluid, hi is height of wall pressure resultant for the impulsive component, hc is height of wall pressure resultant for the convective component, hi' is height resultant of pressures on the wall and on the base plate for the impulsive component, hc' is height resultant of pressures on the wall and on the base plate for the convective component, H is height to the free surface of the liquid; R is tanks radius; D is tanks diameter, C i is dimensionless coefficient and C c is the coefficient dimension of s/m1/2. Periods are necessary after determination of the two masses of mi and mc with their locations and stiffnesses. Base shear and overturning moment for design can be estimated using standard structural dynamic procedures. Except EC8 all international codes use rule Square of Sum of Squares (next only SRSS rule), (2a) and (3a), to combine response from impulsive and convective mode. In EC8 absolute summation rule is used (2b) and (3b), which is based on work of Malhotra. The basis for absolute summation is that the convective mode time period may be several times the impulsive mode period, and hence, peak response of impulsive mode will occur simultaneously when convective mode response is near its peak. However, recently through a numerical simulation for a large number of tanks, Malhotra shoved that SRSS rule gives better results then absolute summation rule, [9]. Eurocode-8 Slimness: = H R m i , mc , hi , hc , hi' , hc' : from Tab. 1. (1a) Housner

= H 2R

(1b) (7) - (15).

Total base shear at the bottom of staging is given

V = Vi 2 + Vc2 M ' = M i' + M c'

(2a) (3a)

V = Vi + V c . M ' = M i' + M c' .

(2b) (3b)

Total overturning moment at base of staging is given

The natural period of the convective mode of vibration Tc in [s]

13th International Multidisciplinary Scientific GeoConference SGEM 2013

(4b) C c = 2 / 3,68 tanh(3,68 H D ) (5) The natural period of the impulsive mode of vibration Ti in second for elevated tank is given by [10] Ti = 2 (mi + m s ) K s (6) where m s is mass of container and one-third mass of standing, K s is lateral stiffness of standing. Table 1 Recommended design values for first impulsive and convective modes of vibration as a function of a tank height-to-ratio (h/R) [9]

Tc = Cc R C c From Tab. 1.

(4a)

Tc = Cc D g

H /R
0.3 0.5 0.7 1.0 1.5 2.0 2.5 3.0

Ci
9.28 7.74 6.97 6.36 6.06 6.21 6.56 7.03

Cc
2.09 1.74 1.60 1.52 1.48 1.48 1.48 1.48

mi / m
0.176 0.300 0.414 0.548 0.686 0.763 0.810 0.842

mc / m
0.824 0.700 0.586 0.452 0.314 0.237 0.190 0.158

hi / H
0.400 0.400 0.404 0.419 0.439 0.448 0.452 0.453

hc / H
0.521 0.543 0.571 0.616 0.690 0.751 0.794 0.825

hi' / H
2.640 1.460 1.009 0.721 0.555 0.500 0.480 0.472

hc' / H
3.414 1.517 1.011 0.785 0.734 0.764 0.796 0.825

1 0,8 0,6 0,4 0,2 0

m c /m

3 2 1 0

h i /H

'

3 2 1 0

h c ' /H

H/R H/R 0 1 2 3 0 1 2 1 2 3 Figure 2: Impulsive and convective masses as fractions of the total liquid mass in the cylindrical tank, impulsive heights as fraction of the height of the liquid in the cylindrical tank, convective heights as fraction of the height of the liquid in the cylindrical tank

H/R 3 0

mi tanh(0.866 D H ) = m 0.866 D H m tanh(3.68 H D ) c = 0,23 m H D h i = 0.375 , pre H D 1.5 H

(7) (8) (9)

Hydrology and Water Resources

hc cosh (3.68 H D ) 1.0 = 1 (10) H 3.68 H L sinh (3.68 H D ) 0.09375 = 0.5 , pre H D >1.5(11) H D hi' 0.866 D H = 0.125 , pre D L 1.33 H 2 tanh (0.866 D H ) = 0.45 ,pre D L >1.33 ' hc cosh (3.68 H D ) 2.01 = 1 H 3.68 H D sinh (3.68 L D ) mg k c = 0.836 tanh 2 (3.68 H D ) H g 2 = 3.68 tanh (3.68 H D ) D (12) (13) (14) (15) (16)

SOLUTION, RESULTS AND DISCUSSION The reinforced concrete elevated tank on four different subsoil classes A, B, C, D given EC8 with a container capacity of 255 m3 is considered in seismic analyses (Fig. 3, 4). The elevated tank has frame supporting system in which six columns ( 0.65 m) are connected by the horizontal bracing at regular interval at 4 m 8 m and 12 m elevations. Youngs modulus and the weight of concrete per unit volume are selected as 32000 MPa and 25 kN/m3, respectively. The container is also filled with the water density of 1000 kg/m3 as seen from Fig. 3. Tank is located on hard soil in seismic zone 1, it is height seismic zone in Slovakia. The damping values for the reinforced concrete elevated tank are taken as 5% for the impulsive mode and 0.5% for convective mode, as recommended in most literature. [9,10]. Seismic analyses for the selected elevated tanks are carried out under three main groups are: model 1 and model 3 are using the dynamic characteristic: similar equivalent masses and heights of two-mass model suggested by Housner, model 2 and model 4 are using the dynamic characteristic: similar equivalent masses and heights from EC8. Total base shear at the bottom of staging tank and total overturning moment at base of staging of model 1 and model 2 are obtained by combining part of impulsive and convective modes through SRSS rule. Total base shear at the bottom of staging tank and total overturning moment at base of staging of model 3 and model 4 are obtained by combining part of impulsive and convective modes through absolute summation rule. The dynamic characteristic of fully reinforced concrete elevated tank: periods, equivalent masses and heights for models 1 and 3 are: Ti = 0.86 s, Tc = 3.14 s, mi = 141145.41 kg, mc = 109728.13 kg, hi' = 3.43 m, hc' = 3.43 m. About 55% of liquid mass is excited in impulsive mode while 43% liquid mass participates in convective mode. Sum of impulsive and convective mass is 250873.53 kg which is about 2% less than the total mass of liquid. The dynamic characteristic of fully elevated tank for models 2 and 4 are: Ti = 0.86 s, Tc = 3.15 s, mi = 140100.58 kg, mc = 115557.16 kg,

13th International Multidisciplinary Scientific GeoConference SGEM 2013

hi' = 3.17 m, hc' = 3.45 m. Time period of impulsive mode of empty elevated tank is

Ti = 0.66 s.

Fig. 3. Vertical cross section of the reinforced concrete elevated tank considered for the seismic analysis
2500

Fig. 4. Plan of staging

2083 2096 2000 1500 1000 500 0 469 468 496 496 905 903 987 989 1671 1678 1477 1476 1785 1785

V [kN]

model 1 model 2 model 3 model 4

Figure 5: Total base shear V [kN] immediately bellow the base plate in dependency on category of sub-soil

Hydrology and Water Resources

Table 2 Total base shear at the bottom of staging tank on four different subsoil classes A, B, C, D A V [kN] model 3r V [kN] model 4 % Deviation of V [kN] V [kN] empty tank 495.69 496.20 0.10 356.57 B 986.50 988.91 0.24 686.39 C 1671.35 1478.34 0.42 734.33 D 2083.04 2096.32 0.64 881.19

30000 25000 20000 15000 10000 5000 0 9267 9204 9762 19459 17772 19506 17888 22007 21942

24498 24295 18953 18951

M' [kNm]

18282 18204

model 1 model 2 model 3 model 4

9802

Figure 6: Total overturning moment M ' [kNm] immediately bellow the base plate in dependency on category of sub-soil Table 3 Total overturning moment at base of staging on four different subsoil classes A, B, C, D A M [kNm] model 3 M [kNm] model 4 % Deviation of M [kNm] M [kNm] empty tank CONCLUSION The seismic design of elevated tanks was using single lumped-mass models provides smaller base shears and overturning moments in both fixed-base and flexible soil conditions. It is evident from the Figures 5 and 6, that the values of total base shears an overturning moments are grooved with category of subsoil. It is seen, that using SRSS rule for calculating of total base shear at the bottom of staging tank and total overturning moment at base of staging are obtained smaller value as by combining of impulsive and convective modes through absolute summation rule recommended by EC8. Diferences between two-mass model suggested by Housner and model based on
7

B 19506.09 19458.63 -0.24 13164.98

C 21842.12 22007.26 0.30 14084.43

D 24295.23 24498.09 0.84 16901.31

9802.57 9762.41 -0.41 6823.28

13th International Multidisciplinary Scientific GeoConference SGEM 2013

the work of Velestos and co-workers [Malhotra], simple procedure, which are suggested in the EC8 for seismic design of elevated tanks, are seen from Table 2 and Table 3. Calculating of total base shear at the bottom of staging tank and total overturning moment at base of staging are seen from Figure 5 and Figure 6. ACKNOWLEDGEMENTS Preparation of the paper has been supported by the Scientific Grant Agency of the Ministry of Education of Slovak Republic and the Slovak Academy of Sciences under Project 1/0201/11 and by the Centre Progressive Constructions and Technologies in Transportation Engineering. The Centre was supported by the Slovak Research and Development Agency under the contract No. SUSPP-0013-09 and the companies Ininierske stavby and EUROVIA SK. REFERENCES [1] [2] Housner, G., W.: Earthquake pressures on fluid containers, California institute of technology, Pasadena, California, 1954. Mel-Rakabawy, M., M., El-Arabaty, H., A., El-Sherbiny, M., G.: Response of elevated water tanks yo seismic load. In: 11th ICSGE, 17.-19. May 2005, Cairo Egypt.
Juhsov, E., Benat, J., Kritofovi, V., Kolcn, ., 2002: Expected seismic response of steel water tank, In: 12th European Conference on Earthquake Engineering, Paper reference 595, London 2002. Krlik, J.: Dynamic analysis of soil-fluid-tsmk interaction due to earthquake even. In: Dynamic of ridig and deformable bodies 2012: st nad Labem. Univerzita J. E. Purkyn v st n. L.: Univerzita J. E. Purkyn v st n. L., ISBN 978-80-7414-500-0.

[3]

[4]

[5]

Livaolu, R., Doangn, A.: Simplified seismic analysis procedures for elevated tanks considering fluid-soil interaction. In: Journal of fluid and structures 22, 2006. p. 421-439. Malhotra, P. K., Wenk, T., Wieland, M.: Simple procedure for seismic analysis of liquid-storage tanks, Structural Engineering International, No. 3, 2000, s. 197-201. Melcer, J., Lajkov, G.: Dynamick vpotov model asfaltovej vozovky In: Stavebn a environmentlne ininierstvo Ro. 7, . 1 (2011), s. 2-12 ISSN: 13365835.
Jaiswal, O., R., Rai, D. C., Jain, S., K.: Review of code provisions on design seismic forces for liquid storage tanks. Kanpur, Indian Institute of Technology Kanpur, 2005 EN 1998-4: 2006 Eurocode 8. Design of structures for earthquake resistance. Part 4: Silos, tanks and pipelines. CEN, Brussels, 2006 with commentary and explanatory examples. Kanpur, Indian Institute of Technology Kanpur, 2005

[6] [7]

[8] [9]

[10] IITK-GSDMA, 2005: Guidelines for seismic design of liquid storage tanks provisions