In order to solve the problem of crossing-zero faulty phase voltage and fault line selection difficulty in distribution network with high resistance ground,a single-phase ground fault line selection method based on learn vector quantization(LVQ)neural network optimized by genetic algorithm(GA)is proposed. Firstly,the modulus maxima of line zero sequence current signals is extracted by wavelet analysis method,which is considered as the input vector of LVQ. On the basis of the improved GA with local search operator to optimize the initial wei...