EDA-ANN based transformer fault recognition with dissolved gas

Condition based maintenance and diagnosis technology plays an important role in power system reliability, because it is able to identify the faulty section in power system before the faults occur. With the technology development of smart grid which requires a more reliable power supply, a lot of researches have been focused on the transformer fault recognition. Based on this present situation, this paper introduces transformer fault recognition research status, and puts the current methods. Through the analysis of weakness of these current methods and the advantage of EDA-ANN method, a new method for the transformer fault recognition is designed to realize the fault recognition with dissolved gas. And through some real fault data, this proposed method is proven to be feasible and accurate.