Wednesday , December 19 2018

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

Liviu IONITA, Irina IONITA
Petroleum-Gas University of Ploieşti,

39, Bucureşti Blvd, Ploieşti, 100680, Romania
lionita@gmail.com; tirinelle@yahoo.com

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 threephase 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.

>>Full text
CITE THIS PAPER AS:
Liviu IONITA, Irina IONITA, Expert-GOSP – Expert System for Three-Phase Separator Diagnosis, Studies in Informatics and Control, ISSN 1220-1766, vol. 24 (3), pp. 293-300, 2015. https://doi.org/10.24846/v24i3y201506

  1. Introduction

Expert systems are software packages that emulate the interaction between a person with a human expert in order to obtain a piece of advice or a recommendation regarding solutions to a problem from a well-defined domain. Specialized knowledge about how to solve problems (i.e. the methods) is often limited and difficult to discover, which represents a key element among competitive companies. Expert systems gather this knowledge and allow its dissemination to other individuals. The majority of approaches referring to knowledge distribution are based only on the person’s capacity to understand and to transform knowledge into a usable form inside the organization. The role of expert systems is to directly transmit knowledge to the individuals, pieces of advice and recommendations more than information. As a result, difficult and complex decision problems are solved rapidly and efficiently without any additional training of human personnel involved in the process.

Application of expert systems is found in various domains such as: industry, medicine, telecommunication, banking, assurance etc. and aims at the diagnosis, prediction, analysis, situation or process planning. Oilfield production prediction, the assessment of fluid corrosion in oil and gas production wells, the selection of the candidate well for steaming up until the well is put on production, the detection of anomalies during industrial processes are just a few examples of expert system application.

An expert system is based on an inference engine, responsible with rationing (backward or forward chaining) and a knowledge base constituted from rules base and facts base. Knowledge representing the analyzed expert domain are implemented in two different ways: as data and as IF THEN ELSE rules.

In recent years, complex diagnosis problems have often been solved by means of intelligent diagnosis system [2]. Data mining techniques [10], multi-agent system [5][6], case-based reasoning systems [2], expert systems [1][11] are more and more applied for technical diagnosis in industry. The authors of this article focus their work in order to build an expert system as a diagnosis tool for a gas-oil separation plant (GOSP).

GOSP can be defined as an ensemble of industrial components used in the upstream oil industry. The role of GOSP is to separate the crude oil from sediments, solids and sand (below using a filter) and gases and then to allow the crude to be pumped on the pipeline system. A separator for petroleum production domain represents a large vessel designed to separate production fluids into their constituent components of oil, gas and water based on their different densities. When the fluid (crude oil) is moving slowly and the separation condition is achieved, the separation components are stratifying with gas on top, water on the bottom and oil in the middle (in a three-phase separator).

Expert-GOSP, the proposed system, collaborates with an intelligent agent named Diagnosis Agent (the supervisor agent from GOSP-MAS, a multi-agent system for diagnosis in oil industry [7]) in order to obtain diagnosis reports referring to the three-phase separator functioning.

This paper describes the designing phase, the implementing phase and the testing phase of Expert-GOSP realized with Exsys Corvid [15][16]. The authors used more than ten evaluation scenarios for testing the prototype, two of them are given in the current article. A diagnosis report obtained after expert system execution contains the alarms, the identified faults and the adequate recommendations to solve the diagnosis problems. The paper is organized as follows: Section 2 represents an overview of expert systems, as decision tools, applied in oil industry. Section 3 gives information about description, implementation and testing of Expert-GOSP. Section 4 discusses about related work and Section 5 contains the conclusions of authors regarding the proposed diagnosis system based on expert systems technology.

REFERENCES

  1. ANGELI, C., Diagnostic Expert Systems: From Expert’s Knowledge to Real-time Systems. Book: Advanced knowledge based systems: Model, Application &Research, Sajjia &Akerkar (eds), 2010, pp. 50-73.
  2. BALAKRISHNAN, K., V. HONAVAR, Intelligent Diagnosis Systems, J. of Intel. Sys., vol. 8(3-4), 1998, pp. 239-290.
  3. GHALLAB, S. A., N. BADR, A. B. SALEM, M. F. TOLBA, A Fuzzy Expert System For Petroleum Prediction, WSEAS, Croatia, 2, 2013, pp.77-82.
  4. IONITA, I., L. IONITA, A Knowledge- Based System for Diagnosis in Oil Industry, The 19th Intl. Conf. The Knowledge-Based Organization KBO 2013, Conference Proceedings 3 Applied Technical Sciences and Advanced Military Technologies, 2013, pp. 277-280.
  5. IONITA, L., I. IONITA, A Methodology for Agent-Oriented Systems Development Applied in Oil Industry, Studies in Informatics and Control, Vol. 23(3), 2014, pp. 239-248.
  6. IONITA, L., Multi-agent System Used for Technical Diagnosis in Gas-Oil Separation Plant, The 20th International Conference The Knowledge-Based Organization, Applied Technical Sciences and Advanced Military Technologies, Conference proceedings 3, „NicolaeBălcescu” Land Forces Academy Publishing House, Sibiu, 2014, pp. 42-47.
  7. IONITA, L., Sistem multi-agent pentru diagnoză tehnică aplicat în industria de petrol, teză de doctorat, Universitatea Politehnica din București, Școala doctorală Automatică și Calculatoare, 2014.
  8. KASZKUREWICZ, E., BHAYA, A., EBECKEN, N.F.F., A Fault Detection and Diagnosis Module for Oil Production Plants in offshore Platforms, Expert Systems with Applications, Vol.12, No.2, 1994, pp.189-194.
  9. KOOROSH, A., Final Report for the Petroleum Technology Research Center, Enhanced Oil Recovery (EOR) Knowledge-Based Expert System for Conformance Control in Saskatchewan Reservoirs, 2005, http://steps.ptrc.ca/concrete/files/5714/1029/3914/Enhanced_Oil_Recovery_EOR_Knowledge- based_Expert_Sytems_for_Conformance…_-_Final_Report.pdf
  10. NAMIKKA, E., G. J. GIBBON, Identification of Data Mining Techniques for Industrial Process Analysis and Control, School of Electrical and Information Engineering, University of the Witwatersrand, Johannesburg, 2002.
  11. NAN, C., F. KHAN, M. T. IQBAL, Realtime Fault Diagnosis using Knowledgebased Expert System, Process Safety and Environmental Protection, vol. 86(1), 2008, pp. 55-71.
  12. ODEDELE, T. O., H. D. IBRAHIM, Oil Well Performance Diagnosis System Using Fuzzy Logic Inference Models, Proceedings of the World Congress on Engineering 2014 Vol I, WCE 2014, July 2 – 4, 2014, London, U.K, ISBN: 978-988-19252-7-5, ISSN: 2078-0958 (Print); ISSN: 2078-0966 (Online).
  13. PIATETSKY-SHAPIRO, G., R. J. BRACHMAN, T. KHABAZA, W. KLOESGEN, E. SIMOUDIS, An Overview of Issues in Developing Industrial Data Mining and Knowledge Discovery Applications, In KDD, 96, 1996, pp.88-95.
  14. STANESCU, I. A., V. STEFAN, G. NEAGU, Renewable Energy Decision Support Systems: The Challenge of Data Integration, Studies in Informatics and Control, Volume 24, Issue 2, ISSN 1220- 1766, 2015, pp. 191-200.
  15. http://www.exsys.com/CorvidCore/CorvidCoreAdvancedTutorial.pdf
  16. http://www.exsys.com/PDF/CorvidManual.pdf