Microgrid fault diagnosis based on whale optimization algorithm optimizing BP neural network

Aiming at the microgrid (MG) fault diagnosis problem, this paper proposes a new microgrid fault diagnosis method that comprehensively utilizes wavelet feature extraction and whale optimization algorithm to optimize BP neural network. First, by applying wavelet transform to extract features of the current signals in the microgrid system, a representation with time-frequency domain characteristics was obtained. Subsequently, a failure mode library was constructed with the help of these features, which provided a key basis for fault diagnosis. On this basis, the whale algorithm is optimized for the weights and biases of the BP neural network, which effectively improves the training speed and generalization performance. By applying the method proposed in this article to the microgrid system, we make an in-depth comparison with BP neural network, RNF neural network to obtain better results. The experimental results clearly show that this method greatly enhances the accuracy and efficiency of microgrid fault diagnosis and provides strong support to ensure the safe operation of the microgrid system