Application of fuzzy neural network optimized by MEA to transformer fault diagnosis

A new transformer fault diagnostic method based on fuzzy neural network and mind evolutionary algorithm was presented. According to the “similartaxis” and “dissimilation”, mind evolutionary algorithm has been used to optimize the membership function parameters and connection weights of fuzzy neural network, and it benefits to find the global optimal solution quickly. The analysis and experimental results showed that the method can improve processing ability of network, and the convergence of method is faster and diagnosis accuracy is higher than that of the GA-fuzzy neural network and PSO- fuzzy neural network. Therefore, the method can be used for the transformer fault diagnosis.