Abstract:
The analytical Preisach model based on the Lorentz function has the advantages of simple form and easy determination of distribution function, which can be applied to the simulation of hysteresis and loss characteristics of core. However, at present, analytical or neural network methods are usually used to identify its parameters. The above methods have some shortcomings in dealing with multi-dimensional complex problems, such as low accuracy and slow convergence speed. In addition, if the reversible magnetization component needs to be considered, the model complexity and model parameters will increase significantly, which aggravates the difficulty of parameter identification. Therefore, this paper first constructs an analytical Preisach model considering the reversible component by introducing the reversible magnetization component. Then, the influence of various characteristic parameters on the shape of hysteresis loop is analyzed, and a hybrid optimization algorithm for characteristic parameters considering the reversible component analysis Preisach model is proposed by combining the cyclic iterative method and particle swarm optimization algorithm. Finally, the calculated hysteresis loops obtained by the proposed hysteresis model and its parameter identification algorithm are compared with the measured values. The results show that the maximum average relative error of the model and its parameter identification algorithm is 4.49%, which verifies the accuracy and effectiveness of the proposed method.