Hysteresis Parameter Identification of J-A Model Current Transformer Based on Improved Whale Algorithm

The accurate and rapid identification of model parameters is crucial to the simulation of the hysteresis characteristics of current transformer cores based on the J-A hysteresis model. With regard to the difficulties in measuring the input and output of the existing J-A model parameters in addition to the slow convergence speed and low accuracy of the extraction method, a parameter extraction method of the ψ-i JA hysteresis model based on the improved whale algorithm is proposed in this paper. The model is established in terms of flux linkage and current instead of flux density and magnetic field. The parameters are extracted using an improved whale algorithm with adaptive adjustment of weights and search strategies. By taking advantage of the proposed method, particle swarm algorithm, and whale algorithm respectively, the hysteresis curves of P-type current transformer and PR-type current transformer are identified and compared. The result comparison shows that the improved whale algorithm has higher accuracy and fewer iterations, such that the high efficiency of the proposed method is verified.