Research on Fault Diagnosis for Power Transformer Based on HHGA-WNN

Aiming at some problems of the training algorithm of wavelet neural network(shorted for WNN) used for power trasfomer fault diagnosis, a new hybrid hierarchy genetic algorithm was introduced by combining hierarchy genetic algorithm and least-square method. The hybrid algorithm was able to determine the structure and parameters of the RBF neural network. Adaptive crossover and mutation probability could accelerate the genetic speed and avoid the occurrence of prematurity. A lot of data of power trasfomer fault diagnosis, and the normalized data were fed into the radial base function neural network to predict the fault diagnosis for power transformer. The simulation demonstrates that the wavelet neural network based on hybrid hierarchy genetic algorithm(shorted for HHGA─WNN) is robust, promising and converges very fast.