Multiple SVMs Modelling Method for Fault Diagnosis of Power Transformers

For enhancing the accuracy of fault diagnosis for power transformers, a multiple SVMs scheme is proposed in this paper. In this scheme, SVM is used to establish the base classifier for its good performance and fast learning speed. Secondly, the several base classifiers based on single SVM will be combined by consulting ensemble techniques. And then a multiple SVM s method is obtained. The real gas records data from a power company is used to establish fault diagnosis system for power transformers based on the new multiple SVM s method. For comparison, the conventional methods are used to build fault diagnosis models by the same data. The experiments demonstrate the new multiple SVMs method has the best performance in both learning ability aspect and generalization ability aspect for fault diagnosis of power transformers. Streszczenie. Zaproponowano schemat SVM (support vector machine) w celu poprawy dokładności diagnostyki transformatorów mocy. W pprównaniu do metod konwencjonalnych proponowana metoda ma możliwość uczenia się i efektywnego wykorzystania bazy danych. (Metoda wykorzystująca technikę SVM do diagnostyki transformatorów mocy)