Magnetic Properties Simulation of Electrical Steel Sheet Based on Recurrent Neural Network

In this work, a Play model based on recurrent neural network (RNN) is proposed to predict hysteresis characteristics of electrical steel sheets under complex excitation conditions. The proposed model combines the neural network with the Play hysteresis operator, replacing the distribution function in the hysteresis model with the trained neural network parameter structure. An automatic selection method for neural network hyperparameters based on genetic algorithm (GA) is suggested, which improves the prediction accuracy. Comparing the experimental with the model prediction results, the proposed model can accurately predict the hysteresis characteristics of materials under sinusoidal and harmonic conditions.