Fault Location Method Based on QGA-Elman Neural Network

In order to solve the problem that the high frequency attenuation coefficient of the fault of UHVDC line is difficult to solve, the fault location model of Elman neural network optimized by QGA is proposed. The amplitude ratio of the first wave head between the rectifier side and the inverter side in different frequency bands is selected as the input set, and the distance from the fault point to the rectifier side is selected as the output set. The weight and threshold of Elman neural network are optimized by QGA algorithm. The optimized QGA-ELMAN fault location model is free from the interference of fault type and transition resistance, and the ranging accuracy is high.