Based on the parameters of long-term memory neural network in fault, which is determined by experience and has uncertainty, which leads to the reduction of the fault accuracy of the model, this paper optimizes the long-term memory neural network by using the improved gray wolf algorithm, and proposes an OLTC fault diagnosis method based on the LSTM network parameter optimization based on the improved gray wolf algorithm. First, multi-scale weighted arrangement entropy and energy entropy based on GOA-VMD joining are used as inputs to LSM, then LSTM's related overriding is iteratively optimized by using the improved gray wolf algorithm, and finally, the IGWO-LSTM combination model is constructed to classify the different faults in OLTC. The model overcomes the problem of selecting parameters according to experience, which leads to the angle of classification accuracy. The study shows that compared with the traditional classification algorithm, the proposed method can classify the different faults of OLTC, improve the classification accuracy, and provide further theoretical basis for OLTC online monitoring.
正在翻译中..