The Design And Implementation Of Transformer Fault Monitoring Expert System

In this paper,we designed and implemented the transformer fault monitoring expert system using the relationship between the composition of dissolved gas in transformer oil and transformer fault type.In view of the construction of knowledge base and reasoning machine in the system,we identified production rules as the form of a knowledge base,a relational database as the storage way of the knowledge base and forward reasoning as control strategy of reasoning machine reasoning by analyzing the common knowledge base and reasoning machine in expert system and the characteristics of the industry expert experience in transformer fault diagnosis and expert reasoning process.Considering the form of production rules are not unified in the process of building a knowledge base,in order to facilitating the storage of the relational tables in the relational database,we put forward a form of production rule algorithm according to the conversion principle of production rules.In this algorithm,we implement the unification of the production rule premise format by converting production rules to clauses of connection operation;we implement the unification of the number of the premise and conclusion of production rules.In this paper,we present a knowledge base of redundant rules automatic elimination algorithm,in order to solve the redundancy problem when adding a new production rule to the system in the actual operation.We realized the judgement of redundant rules by calculating specific production rules set.Combined with the scheme of realization of the knowledge base and industry experts thinking way,this paper builds a reasoning machine based on forward reasoning control strategy.Then we complete reasoning by matching production rules of conditions and the specific fact data.The author completed a transformer fault monitoring expert system according to the overall design scheme of the system and combined with the field data acquisition and transmission mode.This system can automatic report output transformer fault diagnosis according to the composition of dissolved gas in transformer oil.The actual case analysis shows the effectiveness of the system.