Expert System for Power Transformer Condition Monitoring and Diagnosis

In this paper a prototype expert system is proposed. This expert system utilizes dissolved gas in oil analysis techniques to monitor and diagnose for fault condition of transformer. The Roger\´s four ratio method, NTT flag point method, generation rate ratio method and total dissolve combustible gas method is used in this expert system. Both the Roger\´s four ratio and the NTT flag point method are implemented. These two methods are fuzzified. Which is increasing the accuracy of both methods, as the generation rate ratio method is used to analyze the trend of fault gases. All the four methods are used together into this expert system, to achieve greater accuracy in analyzing a fault condition monitoring and diagnose of transformers. The "MATLAB" based expert system is developed. This expert system tested with some faulty transformer cases. 96% accuracy is obtained.