RULE-BASED EXPERT SYSTEM FOR ELECTRICAL EQUIPMENT TROUBLESHOOTING

Abstract

Enormous power of computers made it the most important tool . of the industrial world. Today computer is extending human capabilities to solve problems of increasing complexity. And this has been made possible by Artificial Intelligence technology. One of the major application areas of Al is knowledge Based Systems or Expert Systems. An expert system is a software which attempts to duplicate the function of an expert in a particular domain. Generally expert system consists of knowledge base, working memory, inference engine and user interface. The most successful corrunercial use of expert system is in diagnosing activities. Such systems infer what the problem is from the observable signs and symptoms that are presented to them. In this dissertation framework for developing a rule-based expert system for fault diagnosis of any equipment in general, and for some electrical equipments in particular, has been developed. The proposed system for electrical equipment troubleshooting has been implemented in GCLISP for IBM PC compatible computers. The proposed system is divided in three modules viz, consultation module, knowledge acquisition module and explanation module. The consultation module through its user interface communicates with the user and can infer about the problem. The inference mechanism is embedded in this module. It is (iv) further divided in submodules, each for a group of equipments. The knowledge acquisition module of the system provides the facility for modifying and updating the rulebase of the system. A password is required to have a session with this module to save the rule base from unauthorised access. The explanation module provides various explanation facilities about the inferences made by the system such as Why the system has asked a particular question, How the system has arrived at a conclusion, Whether a particular rule is used or not for deducing the conclusion, Showrule facility, and some other explanation facilities pertaining to the rulebase have been developed. The proposed system can be extended for major industrial utilization. It can be improved further by providing assistance module and self-learning ability. Though the modern technology has facilitated the development of such expert systems, human beings having the supreme brain power except some aliens, will always be the master.