Study On Modeling Method Of Hysteresis Characteristics Of Soft Magnetic Materials

Ferromagnetic materials are widely used in the manufacture of iron cores of various motors and transformers,which is of great significance to the safe,stable and efficient operation of electrical equipment and power systems.The results of materials physics study show that ferromagnetic materials have hysteresis characteristics,so the establishm ent of accurate hysteresis model has fundamental support for the electromagnetic optimization design of electrical machinery and appliances.Especially in engineering applications,more accurate hysteresis loss prediction can be obtained through hysteresis modeling analysis,thus providing effective theoretical basis for fine analysis and design of motors,transformers,reluctance actuators and related systems.However,at present,the sharing mechanism of magnetic measurement data has not been established at home and abroad,the large database of material characteristics on which hysteresis modeling depends is incomplete,and the mainstream commercial electromagnetic field software in the market mostly adopts linearization approximation method to simplify processing.Therefore,in order to further improve the electromagnetic analysis accuracy and efficiency of electrical devices such as motors and transformers,this paper has carried out in-depth research on modeling methods of magnetic hysteresis characteri stics of soft magnetic materials,and specifically completed the following contents:(1)Starting from the magnetic domain theory of magnetic materials and the spin of magnetic dipoles,the hysteresis characteristics of magnetic materials are analyzed.The significance and research status of hysteresis modeling are summarized,and three representative hysteresis models are introduced in detail,they are Preisach hysteresis model,Jiles-Atherton(J-A)hysteresis model and neural network model.(2)On the basis of classical Preisach hysteresis model,aiming at the problems of difficult determination of distribution function,large amount of calculation and low precision,cosh function is proposed to replace the distribution function of Preisach hysteresis model,the product of two unary functions is used to replace the distribution function,and Everett function is introduced to establish the analytical expression of hysteresis characteristics.The improved Preisach hysteresis model has three parameters.the influence of the parameters on the characteristic parameters of hysteresis loop is analyzed,which provides the basis for setting initial values for parameter identification and improves the computational efficiency of the algorithm.(3)The hysteresis loops of the commonly used silicon steel sheet 50WW600 are measured by Epstein square ring under three working conditions of different magnetic densities,different frequencies and different temperatures,and the effects of different physical factors on the hysteresis loops are studied.Levenberg-Marquardt algorithm is used to identify the parameters of the improved Preisach model.the optimal parameters obtained by identification are substituted into the model to fit the experimental measurement curve,and the fitting effect is good.