Fault diagnosis of power transformer based on improved differential evolution-neural network

The proposed model combining improved differential evolution(IDE) algorithm with BP algorithm is applied to fault diagnosis of power transformer in the paper. Despite for its simplicity and high-efficiency, differential evolution (DE) algorithm has the problem of parameters difficult to dynamical adjustment. Based on it, IDE algorithm adopts adaptive control parameters according to swarms´ distribution condition. It has a strong global searching capability and can quickly find the global optimal point. The algorithm can effectively overcome defects of conventional BP algorithm, such as the slow convergence of weight and threshold learning, premature result. And it achieves the two kinds of algorithms from each other. Its application in power transformer fault diagnosis is simulated, Comparing with other algorithms. Results show that the proposed method possesses following advantages of good convergence performance, good robustness and high classification accuracy.