Fault Prediction of a Transformer Bushing Based on Entropy Weight TOPSIS and Gray Theory

Abstract

Transformer bushing insulation fault prediction can effectively guide transformer maintenance. The fault prediction method of a transformer bushing based on entropy weight TOPSIS and gray prediction theory is studied. This method uses the entropy theory instead of the subjective experience to determine the weight of the evaluation indicator. The TOPSIS evaluation method is used to transform the insulation evaluation problem into the vector space distance problem. Based on the Euclidean distance between the test data and standard grade data, the assessment problem is transformed into a distance problem of vector space. The state of a transformer bushing can be obtained from the TOPSIS calculations. Subsequently, the development trend of the fault state is predicted by using the insulation state data and the gray GM (1.1) prediction model. Prediction results are helpful for maintenance. The validity of the method was verified by the example.