Sunteți pe pagina 1din 9

See discussions, stats, and author profiles for this publication at: https://www.researchgate.


Expert-GOSP - Expert System for Three-Phase Separator Diagnosis

Article  in  Studies in Informatics and Control · September 2015

DOI: 10.24846/v24i3y201506


4 22

2 authors, including:

Irina Ionita
Petroleum - Gas University of Ploiesti


Some of the authors of this publication are also working on these related projects:

Data Mining for Medical Diagnosis View project

Applying Data Mining in Industry View project

All content following this page was uploaded by Irina Ionita on 12 December 2017.

The user has requested enhancement of the downloaded file.

Expert-GOSP – Expert System for Three-Phase
Separator Diagnosis
Petroleum-Gas University of Ploieşti,
39, Bucureşti Blvd, Ploieşti, 100680, Romania,;
Abstract: Expert systems are known as artificial intelligence tools that store and implement human expert opinions,
beliefs, methods and rules to achieve accurate system results. With the goal to assist the management of a gas-oil three-
phase separation plant, the authors created Expert-GOSP, a diagnosis tool based on expert system technology. The
prototype is built in Exsys Corvid and the rules constituting the rules base of the expert system are provided by domain
experts and additional documentation regarding the functioning of a real gas-oil separation plant from Romania.
Keywords: expert system, diagnosis, gas-oil industry, three-phase separator

1. Introduction forward chaining) and a knowledge base

constituted from rules base and facts base.
Expert systems are software packages that Knowledge representing the analyzed expert
emulate the interaction between a person with a domain are implemented in two different ways:
human expert in order to obtain a piece of as data and as IF THEN ELSE rules.
advice or a recommendation regarding In recent years, complex diagnosis problems
solutions to a problem from a well-defined have often been solved by means of intelligent
domain. Specialized knowledge about how to diagnosis system [2]. Data mining techniques
solve problems (i.e. the methods) is often [10], multi-agent system [5][6], case-based
limited and difficult to discover, which reasoning systems [2], expert systems [1][11]
represents a key element among competitive are more and more applied for technical
companies. Expert systems gather this diagnosis in industry. The authors of this article
knowledge and allow its dissemination to other focus their work in order to build an expert
individuals. The majority of approaches system as a diagnosis tool for a gas-oil
referring to knowledge distribution are based separation plant (GOSP).
only on the person’s capacity to understand and
to transform knowledge into a usable form GOSP can be defined as an ensemble of
inside the organization. The role of expert industrial components used in the upstream oil
systems is to directly transmit knowledge to the industry. The role of GOSP is to separate the
individuals, pieces of advice and crude oil from sediments, solids and sand
recommendations more than information. As a (below using a filter) and gases and then to
result, difficult and complex decision problems allow the crude to be pumped on the pipeline
are solved rapidly and efficiently without any system. A separator for petroleum production
additional training of human personnel domain represents a large vessel designed to
involved in the process. separate production fluids into their
constituent components of oil, gas and water
Application of expert systems is found in based on their different densities. When the
various domains such as: industry, medicine, fluid (crude oil) is moving slowly and the
telecommunication, banking, assurance etc. and separation condition is achieved, the
aims at the diagnosis, prediction, analysis, separation components are stratifying with gas
situation or process planning. Oilfield on top, water on the bottom and oil in the
production prediction, the assessment of fluid middle (in a three-phase separator).
corrosion in oil and gas production wells, the
selection of the candidate well for steaming up Expert-GOSP, the proposed system,
until the well is put on production, the detection collaborates with an intelligent agent named
of anomalies during industrial processes are just Diagnosis Agent (the supervisor agent from
a few examples of expert system application. GOSP-MAS, a multi-agent system for
diagnosis in oil industry [7]) in order to obtain
An expert system is based on an inference diagnosis reports referring to the three-phase
engine, responsible with rationing (backward or separator functioning. This paper describes the

Studies in Informatics and Control, Vol. 24, No. 3, September 2015 293
designing phase, the implementing phase and This article presents an approach to design a
the testing phase of Expert-GOSP realized with fuzzy logic controller to detect incipient
Exsys Corvid [15][16]. The authors used more abnormal production of wells in a reservoir.
than ten evaluation scenarios for testing the Simulation results presented in the paper [12]
prototype, two of them are given in the current demonstrate how the designed controller
article. A diagnosis report obtained after expert detects well production fault, by using fuzzy IF
system execution contains the alarms, the THEN rules. Similarly, Expert-GOSP uses
identified faults and the adequate knowledge represented as heuristic rules in IF
recommendations to solve the diagnosis THEN ELSE form. The combination of all the
problems. The paper is organized as follows: heuristics allows the overall decision-making
Section 2 represents an overview of expert problem to be solved. The execution of the
systems, as decision tools, applied in oil rules in Expert-GOSP was made through
industry. Section 3 gives information about “Forward Chaining” which can be combined
description, implementation and testing of with Backward Chaining to run top level rules
Expert-GOSP. Section 4 discusses about related in a particular order.
work and Section 5 contains the conclusions of
In oil industry, human expertise is
authors regarding the proposed diagnosis
indispensable in the decision-making process of
system based on expert systems technology.
selecting the most adequate Enhanced Oil
Recovery (EOR) method [9]. The goal of the
2. Expert Systems –as Decision Tools project described in [9] is to develop an EOR
knowledge-based expert system that allows oil
In industry, decisions are crucial and errors that organizations to input new data and modify the
initially are considered without major effects existing data for a certain well in order to select
can disturb and cause disasters, involving a proper conformance control method in
equipment and people. Therefore, decision Saskatchewan oil fields. The designed EOR
tools are very useful in such cases to help expert system was formulated by representing
decision makers to do what is best referring to a the obtained knowledge using IF-AND-OR
situation/process. In literature [1][4][8][14] THEN-SELECT statements.
expert systems are often presented and
considered as decision tools, easy to understand Similar to paper [12], the authors of [3] analyze
and with logical rationing, based on knowledge oil well data chosen from distinct sources and
provided by experts in a certain domain. propose a Fuzzy Petroleum Prediction (FPP)
Motivated by the advantages of expert systems expert system. The main scope of fuzzy system
mentioned in the literature [2][3][11], the proposed in [3] is to help petroleum engineers
authors decided to build Expert-GOSP, a to detect areas adequate for the drilling process.
prototype system for diagnosis in oil industry The above-mentioned examples indicate that
by using knowledge-based systems technology. most applications of expert systems in
In the current section, there are discussed petroleum field fall into one of the following
several examples of expert systems applied in categories: interpreting and identifying,
gas-oil area as decision tools and a comparison predicting, diagnosing, planning, monitoring
with Expert-GOSP is made. and controlling.
In [8] the authors describe an expert system for In the next section of the current paper, the
process supervision and diagnosis of an authors present Expert-GOSP in details
offshore oil product plant. The software used to (description of the system, implementation of
implement the expert system was Gensym G2. knowledge base and testing phase).
In this paper the complete oil process is
considered, including gas-oil separation 3. Expert-GOSP: Description,
process. In comparison with this diagnosis
system, Expert-GOSP was built with Implementation and Testing
ExsysCorvid and provides a diagnosis report
with a list of faults, the causes of the faults Exsys Corvid represents a software for
and the recommendations only for a three- generating expert systems and enables the
phase separator. decision-making logic and process of the
domain expert to be converted into a structured
Another example of knowledge-based system form (IF THEN ELSE rules). The inference
application in the oil industry is given in [12]. engine dynamically drives interactive sessions

294 Studies in Informatics and Control, Vol. 24, No. 3, September 2015
that provide advice and recommendation to end 1. oil temperature inside the separator
users [15][16]. (oil_temp_inside_separator);
Expert-GOSP was designed to collaborate with 2. oil temperature at the outlet of the heat
Diagnosis Agent, the supervisor agent of the exchanger 41HE002 (oil_temp_outlet_heat
multi-agent system GOSP-MAS [7] to obtain _exchanger_41HE002);
diagnosis reports referring to the function of a
3. oil temperature at the outlet of the heat
three-phase separator. Expert-GOSP contains a
exchanger 41HE003 (oil_temp_outlet-
knowledge base, an inference engine and a GUI
that enables the communication with Diagnosis
Agent. The generated report after diagnosis is 4. pressure of the gases evacuated from the
taken over by the Diagnosis Agent which separator (gas_pressure_outlet_separator);
collaborates with the Operator Agent, 5. pressure of the water evacuated from the
responsible with interpreting and executing the separator
recommendations listed in the diagnosis report. (water_pressure_outlet_separator);
In Figure 1 there are represented the main 6. pressure of the oil evacuated from the
modules of Expert-GOSP. separator (oil_pressure_outlet_separator);
Expert-GOSP receives data referring to the 7. water level inside the separator
operating parameters of the separator through (water_level_separator);
the Diagnosis Agent. All the measurements of
the system parameters (oil temperature inside 8. oil level inside the separator
the separator, oil temperature at the outlet of (oil_level_separator).
the heat exchanger, water level inside the The diagnosis report of Expert-GOSP contains
separator, evacuated gas pressure etc.) received information regarding the alarms caused by
by the other agents of the multi-agent system abnormal operating states of different plant
GOSP-MAS are centralized in a database by sectors, the detected and identified faults and
the Diagnosis Agent. Expert-GOSP generates the solutions/recommendations proposed to
the diagnosis reports taking into account the remediate the functioning errors.
experience of the engineers and operators of the
gas-oil separation plant (representing the For expert system implementation there were
knowledge in gas-oil field) and the current used 14 attributes, including 8 attributes of
conditions/states of the analyzed system (three- Numeric type (the operating parameters of
phase separator). three-phase separator), 4 attributes of String
type (alarm – the generated alarm, cause – the
The parameters considered in the designing cause of alarm occurrence, solution – the
phase of Expert-GOSP and with significance solution/advice/recommendation and state – the
for the system diagnosis are:

Figure 1. Expert-GOSP – expert system for a three-phase separator diagnosis

Studies in Informatics and Control, Vol. 24, No. 3, September 2015 295
operating state) and 2 attributes of [water_pressure_outlet_separator] >=1.5
Collection/Report type (report– the diagnosis THEN:
report, state_report – the state of diagnosis [alarm] = " Water pressure in
separator HIGH"
report (elaborated/under development). Other 6 [cause] = " Cause: evacuation
attributes were added for building rules acting pump fault 44-P-002A"
at multiple levels (recommended_operation as [solution] = " Solution: verify the
Colelction/Report type and other 5 attributes as evacuation pump 44-P-002A and use by-pass
String type). pump 44-P-002S"
The rules stored in the rules base of expert ALARM"
SET [report.ADD] [alarm.VALUE]
system are presented in the well-known
SET [report.ADD] [cause.VALUE]
structure IF THEN ELSE, based on the SET [report.ADD] [solution.VALUE]
technical documentation of gas-oil separation
R5 IF:
plant and the experience of the personal [water_pressure_outlet_separator] <=0
(engineers, operators etc.). Expert-GOSP uses THEN:
the forward chaining to achieve the purpose, [alarm] = " Water pressure in
more exactly, to make the diagnosis of the separator LOW"
three-phase separator. [cause] = " Cause: control valve
upstream separator fault 21-SDV-006 or pump 44-
An example of a set of rules is given in table 1. P-002A locked on high speed"
[solution] = " Solution: verify
Table 1. Rules example control valve upstream separator 21-SDV-006 or
evacuation pump 44-P-002A"
[oil_temp_inside_separator] <=25 ALARM"
THEN: SET [report.ADD] [alarm.VALUE]
[alarm] = " Oil temperature inside the SET [report.ADD] [cause.VALUE]
separator LOW" SET [report.ADD] [solution.VALUE]
[cause] = " Cause: temperature control valve
fault TCV-41002 R6 IF:
or TCV-41003 or thermal insulation vessel ([water_pressure_outlet_separator] >0)
damage" &([ water_pressure_outlet_separator] <1.5)
[solution] = " Solution: verify the operating THEN:
state of temperature control valve TCV- [state] = " Water pressure in
41002/TCV-41003 or verify thermal isolation separator on normal range"
vessel" SET [report_state.ADD]
SET [report.ADD] [alarm.VALUE]
R7 IF:
SET [report.ADD] [cause.VALUE]
[oil_level_separator] >=1750
SET [report.ADD] [solution.VALUE]
R2 IF: [alarm] = " High-High Level in oil
[oil_temp_inside_separator] >=55 compartment of the separator"
THEN: [cause] = " Cause: damage of
[alarm] = " Oil temperature inside opening system of level control valve LCV-21007
the separator HIGH" (closed position) or accidentally closing of control
[cause] = " Cause: temperature valve downstream of the separator"
control valve fault TCV-41002 or TCV-41003 " [solution] = " Solution: close the
[solution] = " Solution: verify the safety control valve (SDV-21006) on the outlet
operating state of temperature control valve TCV- pipe of separator and open the safety control valve
41002/TCV-41003 " (POV-21005) on the by-pass pipe of separator"
ALARM" SET [report.ADD] [alarm.VALUE]
SET [report.ADD] [alarm.VALUE] SET [report.ADD] [cause.VALUE]
SET [report.ADD] [cause.VALUE] SET [report.ADD] [solution.VALUE]
SET [report.ADD] [solution.VALUE]
R8 IF:
R3 IF: [oil_level_separator] <=245
([oil_temp_inside_separator] THEN:
>25)&([oil_temp_inside_separator]<55) [alarm] = " Low-Low Level in oil
THEN: compartment of the separator"
[state] = " Oil temperature inside [cause] = " Cause: damage of
the separator in normal range" opening system of level control valve LCV-21007
SET [report_state.ADD] (opened position)"
[state.VALUE] [solution] = " Solution: verify the
level control valve LCV-21007"

296 Studies in Informatics and Control, Vol. 24, No. 3, September 2015
SET [report.ADD] [alarm.VALUE] oil_level_separator=1200 mm,
SET [report.ADD] [cause.VALUE] oil_pressure_outlet_separator=5.12 bar,
SET [report.ADD] [solution.VALUE] oil_temp_inside_separator=23.45 0C,
R9 IF:
([oil_level_separator] >245) &
alarm=” High-High Level in oil
([oil_level_separator] <1750) compartment of the separator”, state=” Oil
THEN: level in separator on normal range”.
[state] = " Oil level in separator on
normal range" A possible scenario used to test the expert
SET [report_state.ADD] system is SE1: all the considered parameters
[state.VALUE] have normal operating values (table 2).
R10 IF: Table 2. SE1 testing scenario
[alarm] = " Oil temperature inside
the separator LOW" AND Parameter Measured Measuring
[TCV-41002_status] = "80% open" value unit
[termal_isolation_sensor] = "normal" oil_temp_inside_separator 35.5 C
oil_temp_outlet_heat_exchanger 36.0 C
THEN: _41HE002
[recommended_operation] = "
Verify temperature control valve TCV-41002" oil_temp_outlet_heat_exchanger 36.1 C
SET [report.ADD] "TEMPERATURE _41HE003
SET [report.ADD] gas_pressure_outlet_separator 0.4 bar
[recommended_operation.VALUE] water_pressure_outlet_separator 1.2 bar
R11 IF: oil_pressure_outlet_separator 8.15 bar
[alarm] = "High-High Level in oil
compartment of the separator " AND oil_level_separator 1530 mm
[LCV-21007_status] = "75% open" water_level_separator 1200 mm
[recommended_operation] = "Verify the
control valve downstream of the separator "
The diagnosis report generated after Expert-
SET [report.ADD] " LEVEL ALARM" GOSP execution for SE1 scenario contains the
SET [report.ADD] values of the analyzed parameters and detailed
[recommended_operation.VALUE] information messages referring to the
functioning state of the three-phase separator.
The facts stored in the data base of expert As presented in Figure 2, all values are in the
system represent the measured values of normal range and the messages are highlighted
operating parameters (the initial values) and the with green color indicating a normal state of
obtained values for various attributes during functioning for the oil separation plant.
rationing process. An example of Expert-GOSP
facts is: The SE2 testing scenario considers the
following: the measurements of parameters

Figure 2. The diagnosis report for SE1 testing scenario

Studies in Informatics and Control, Vol. 24, No. 3, September 2015 297
oil_temp_outlet_heat_exchanger_41HE002, 3. the alarms, the causes and the solutions/
water_pressure_outlet_separator and pieces of advice/recommendations offered
oil_level_separator are out of normal range of for each case (temperature alarm, pressure
operation (table 3). alarm and level alarm).
Table 3. SE2 testing scenario An example of alarm is the following: the oil
Parameter Measured Measuring temperature at the outlet of the heat exchanger
value unit 41HE002 LOW, the water pressure inside the
oil_temp_inside_separator 35.8 0C separator HIGH, Maxim Maximum level inside
oil_temp_outlet_heat_exchanger 25 0C the oil compartment of separator. The identified
_41HE002 causes for the alarms’ occurrence are:
oil_temp_outlet_heat_exchanger 25.2 0C temperature controller TCV-41002 breakdown,
_41HE003 evacuation pump 44-P-002A breakdown,
gas_pressure_outlet_separator 0.67 Bar opening system for LCV-21007 (closed
water_pressure_outlet_separator 1.55 Bar position) breakdown or accidental closing of
oil_pressure_outlet_separator 6.71 Bar the valve downstream of the separator. The
oil_level_separator 1754 Mm solution given to remediate the functioning
water_level_separator 459 Mm faults/errors of system may be described as:
check the functioning state of the
The diagnosis report generated for the temperature controller TCV-41002, check the
measured values of parameters mentioned in pump 44-P-002 operating condition, close the
the SE1 testing scenario contains: protection valve (SDV-21006) situated on the
1. the values of the analyzed parameters; inlet pipe of the separator and open the
protection valve (POV-21005) on the by-pass
2. the detailed information messages about pipe of the separator.
normal values of the considered
parameters; The red sectors of the diagnosis report
represent the alarms, the causes and the

Figure 3. The diagnosis report for SE2 testing scenario

298 Studies in Informatics and Control, Vol. 24, No. 3, September 2015
adequate solutions for the SE2 testing scenario. message regarding to the alarms occurred and
The messages situated in the green sector the solutions given.
describe the normal operating states (Figure 3).
Expert-GOSP extends the KBS-Diag and
The human operator analyses the diagnosis represents an improved version with
reports received from the Operator Agent (the important features:
agent which is responsible with the
1. a simply user interface but useful, due to
communication between human operator and
the customized display areas for alarms and
GOSP-MAS) and executes operations
recommendations given (green area for
necessary to remediate the anomalies occurred
normal situation, red area for abnormal
inside the gas-oil separation plant, according to
situation), designed in Exsys Corvid;
the solutions/recommendations listed in the
diagnosis report provided by Expert-GOSP. 2. a data base with facts collected from real
situation of GOSP;
Exsys Corvid owns the tracing facility to view
the rationing method applied to solve the 3. a rules base with IF THEN ELSE rules
current problem (Run with Trace). The nodes of describing the multiple scenarios of system
expert system are tested, and the rules’ operating (extracted from process history),
expressions are evaluated in order to establish acting at multiple levels;
their value (TRUE/FALSE). 4. the ability of communication with an
In Exsys Corvid there is a type of variable intelligent agent, to improve the quality
named Confidence Variable referring to a of diagnosis;
variable that can be assigned as a confidence 5. the possibility of integration with other
value to reflect a degree of certainty. Various modules (for example with a Data
formulas can be used to combine the values Mining module for increase the accuracy
assigned for an overall confidence for the of diagnosis);
variable. The rules can include a “confidence
factor” for a particular answer. This enables 6. the possibility to add confidence factors for
expert systems to provide multiple optimizing the reasoning methods of
recommendations with differing degrees of expert system.
confidence. While in some cases it is possible The simulations of Expert-GOSP were
to give a specific recommendation/piece of executed on concrete facts provided by
advice with absolute precision, in other cases operating reports during a week. The current
multiple recommendations are simultaneously monitoring and diagnosis system implemented
possible and the system ranks them based on on the concerned GOSP on the oil field,
relative merit. The ability to handle confidence displays only the current alarms and a history
factors in Exsys Corvid provides a much more of alarms, with certain features: severity, time
effective way to design expert systems that (data and hour) of alarm occurrence, time (date
emulate the real world and give the type of and hour) of exiting alarm, alarm message,
solutions that human experts would. alarm state (normal, in alarm). Expert-GOSP
supplementary offers solutions to the human
4. Discussions operator to solve the failures, if it is possible.

The authors’ interest in knowledge-based 5. Conclusions

systems (KBSs) applied in oil industry was
found in previous work [4], were a prototype Expert-GOSP is an expert system developed for
named KBS-Diag was developed in Visual assisting the maintenance and operations
Prolog with the main goal of monitoring and departments with the diagnosis and correction
diagnosis of a GOSP. The authors used a group of failures within the gas-oil separation plant.
of parameters which are describing the status of The use of this expert system will help to
the plant at a certain moment. KBS-Diag reduce the down-time and repetitive start-ups
analyzed them and if a one or more parameters by diagnosing the cause of the shut-down
exceed the normal operating range, the before restarting the plant or other type of
inference engine starts the reasoning and try to failures. The overall reliability and availability
identify the causes of these abnormal situation. of the plant should improve the separation
As a result, on the user interface is displayed a process of crude oil and will increase the

Studies in Informatics and Control, Vol. 24, No. 3, September 2015 299
operating and maintenance costs. The data used 8. KASZKUREWICZ, E., BHAYA, A.,
to develop the database of Expert-GOSP, the EBECKEN, N.F.F., A Fault Detection and
rules base and the testing data were provided Diagnosis Module for Oil Production
by concrete operating reports of a gas-oil Plants in offshore Platforms, Expert
separation plant from Romania. Multiple Systems with Applications, Vol.12, No.2,
evaluation scenarios were used to analyze the 1994, pp.189-194.
behavior of the designed expert system, based
9. KOOROSH, A., Final Report for the
on functioning of GOSP during a week. Future
Petroleum Technology Research Center,
work will focus on adding confidence factors to
Enhanced Oil Recovery (EOR)
the rules of expert system and to combine data
Knowledge-Based Expert System for
mining techniques with expert system
Conformance Control in Saskatchewan
technology to increase the accuracy of results.
Reservoirs, 2005,
REFERENCES /3914/Enhanced_Oil_Recovery_EOR_Kno
1. ANGELI, C., Diagnostic Expert Systems: wledge-based_Expert_Sytems_for_
From Expert’s Knowledge to Real-time Conformance..._-_Final_Report.pdf
Systems. Book: Advanced knowledge 10. NAMIKKA, E., G. J. GIBBON,
based systems: Model, Application Identification of Data Mining Techniques
&Research, Sajjia &Akerkar (eds), 2010, for Industrial Process Analysis and
pp. 50-73. Control, School of Electrical and
2. BALAKRISHNAN, K., V. HONAVAR, Information Engineering, University of the
Intelligent Diagnosis Systems, J. of Intel. Witwatersrand, Johannesburg, 2002
Sys., vol. 8(3-4), 1998, pp. 239-290. 11. NAN, C., F. KHAN, M. T. IQBAL, Real-
3. GHALLAB, S. A., N. BADR, A. B. time Fault Diagnosis using Knowledge-
SALEM, M. F. TOLBA, A Fuzzy Expert based Expert System, Process Safety and
System For Petroleum Prediction, Environmental Protection, vol. 86(1),
WSEAS, Croatia, 2, 2013, pp.77-82. 2008, pp. 55-71.
4. IONIȚĂ, I., L. IONIȚĂ, A Knowledge- 12. ODEDELE, T. O., H. D. IBRAHIM, Oil
Based System for Diagnosis in Oil Well Performance Diagnosis System
Industry, The 19th Intl. Conf. The Using Fuzzy Logic Inference Models,
Knowledge-Based Organization KBO Proceedings of the World Congress on
2013, Conference Proceedings 3 Applied Engineering 2014 Vol I, WCE 2014, July 2
Technical Sciences and Advanced Military - 4, 2014, London, U.K, ISBN: 978-988-
Technologies, 2013, pp. 277-280. 19252-7-5, ISSN: 2078-0958 (Print); ISSN:
2078-0966 (Online).
5. IONIȚĂ, L., I. IONIȚĂ, A Methodology
for Agent-Oriented Systems 13. PIATETSKY-SHAPIRO, G., R. J.
Development Applied in Oil Industry, BRACHMAN, T. KHABAZA, W.
Studies in Informatics and Control, Vol. KLOESGEN, E. SIMOUDIS, An
23(3), 2014, pp. 239-248. Overview of Issues in Developing
Industrial Data Mining and Knowledge
6. IONIȚĂ, L., Multi-agent System Used for Discovery Applications, In KDD, 96,
Technical Diagnosis in Gas-Oil 1996, pp.88-95
Separation Plant, The 20th International
Conference The Knowledge-Based 14. STANESCU, I. A., V. STEFAN, G.
Organization, Applied Technical Sciences NEAGU, Renewable Energy Decision
and Advanced Military Technologies, Support Systems: The Challenge of Data
Conference proceedings 3, Integration, Studies in Informatics and
„NicolaeBălcescu” Land Forces Academy Control, Volume 24, Issue 2, ISSN 1220-
Publishing House, Sibiu, 2014, pp. 42-47. 1766, 2015, pp. 191-200.
7. IONIȚĂ, L., Sistem multi-agent pentru 15.
diagnoză tehnică aplicat în industria de CoreAdvancedTutorial.pdf
petrol, teză de doctorat, Universitatea 16.
Politehnica din București, Școala doctorală pdf
Automatică și Calculatoare, 2014.

300 Studies in Informatics and Control, Vol. 24, No. 3, September 2015

View publication stats