Use this URL to cite or link to this record in EThOS:
Title: Evaluation of a fuzzy-expert system for fault diagnosis in power systems
Author: Park, Min Young
ISNI:       0000 0001 3468 2116
Awarding Body: Bournemouth University
Current Institution: Bournemouth University
Date of Award: 2001
Availability of Full Text:
Access from EThOS:
Access from Institution:
A major problem with alarm processing and fault diagnosis in power systems is the reliance on the circuit alarm status. If there is too much information available and the time of arrival of the information is random due to weather conditions etc., the alarm activity is not easily interpreted by system operators. In respect of these problems, this thesis sets out the work that has been carried out to design and evaluate a diagnostic tool which assists power system operators during a heavy period of alarm activity in condition monitoring. The aim of employing this diagnostic tool is to monitor and raise uncertain alarm information for the system operators, which serves a proposed solution for restoring such faults. The diagnostic system uses elements of AI namely expert systems, and fuzzy logic that incorporate abductive reasoning. The objective of employing abductive reasoning is to optimise an interpretation of Supervisory Control and Data Acquisition (SCADA) based uncertain messages when the SCADA based messages are not satisfied with simple logic alone. The method consists of object-oriented programming, which demonstrates reusability, polymorphism, and readability. The principle behind employing objectoriented techniques is to provide better insights and solutions compared to conventional artificial intelligence (Al) programming languages. The characteristics of this work involve the development and evaluation of a fuzzy-expert system which tries to optimise the uncertainty in the 16-lines 12-bus sample power system. The performance of employing this diagnostic tool is assessed based on consistent data acquisition, readability, adaptability, and maintainability on a PC. This diagnostic tool enables operators to control and present more appropriate interpretations effectively rather than a mathematical based precise fault identification when the mathematical modelling fails and the period of alarm activity is high. This research contributes to the field of power system control, in particular Scottish Hydro-Electric PLC has shown interest and supplied all the necessary information and data. The AI based power system is presented as a sample application of Scottish Hydro-Electric and KEPCO (Korea Electric Power Corporation).
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Computer Science and Informatics