Research On Dynamic Analytical Inverse Preisach Hysteresis Model And Its Parameter Identification Algorithm

At present,China ’s power industry is developing rapidly.Transformers and other electrical equipment are increasingly demanding in stable operation and perfect performance.Ferromagnetic materials are widely used in generators,transformers and other electrical equipment as the generation and transmission medium of electric energy.As one of the most basic characteristics of ferromagnetic materials,whether the simulation results are accurate or not is related to the performance of the overall product of the transformer and the subsequent optimization design.At the same time,hysteresis characteristics are also one of the research hotspots of transformer magnetic characteristics.The study of hysteresis characteristics not only improves the hysteresis model,but also improves the method of determining the parameters of the hysteresis model in order to accurately simulate the hysteresis characteristics.At the same time,considering the increase of frequency in engineering practice,the construction of dynamic hysteresis model is also of great significance.The main research results of this paper are as follows :(1)Preisach model is a classical hysteresis model,which is widely used because of its high simulation accuracy of hysteresis characteristics.As a basic model,it has been further improved and simplified by many scholars.Through the understanding of the classical Preisach model,the distribution function of the Preisach model is the key to accurately simulate the hysteresis characteristics.Therefore,this paper first understands the improved Preisach model for constructing Everett function,which simplifies the calculation process of double integral,but still does not improve the Preisach distribution function.Therefore,in this paper,the Lorentz function is used to approximate the Preisach distribution function,and the reversible magnetization component is introduced into the model to construct an analytical Preisach model based on Lorentz function considering the reversible component.In addition,the parameters of the analytical Preisach model considering the reversible component are preliminarily calculated by analytical method,and the static hysteresis characteristics are simulated by using the parameters.Finally,the calculated hysteresis loop is compared with the experimental data.The results show that the analytical Preisach model considering the reversible component can simulate the hysteresis characteristics of ferromagnetic materials.However,due to the low accuracy of the analytical calculation parameters,the error increases.(2)In order to solve the problem of inaccurate parameters of analytical Preisach model considering reversible components,a parameter identification algorithm is introduced to establish a new hybrid optimization algorithm and identify the static model.Firstly,in order to improve the convenience and versatility of the analytical Preisach model of the reversible component,the control variable method is used to analyze the influence of each parameter on the irreversible component and reversible component of the hysteresis loop.Then,the genetic algorithm,simulated annealing algorithm and particle swarm optimization algorithm are introduced to identify the parameters of the irreversible component,and the determination method of the reversible component is given by the relationship between the permeability of the descending section of the hysteresis loop and the reversible component and the irreversible component.By combining the loop iteration method and the particle swarm optimization algorithm,a hybrid optimization algorithm for the characteristic parameters of the analytical Preisach model considering the reversible component is proposed.Finally,the hysteresis characteristics are simulated by combining the calculation parameters of the proposed hybrid optimization algorithm with the analytical Preisach model of the reversible component,and the simulation results are compared with the analytical results and experimental data.The results show that the hybrid optimization algorithm makes the analytical Preisach model of the reversible component have higher accuracy,and the maximum average relative error is 4.49 %,which verifies the accuracy and effectiveness of the method.(3)In view of the high frequency in engineering practice,a dynamic analytical inverse Preisach model based on R-L fractional derivative is constructed.Firstly,in order to simplify the construction process of dynamic hysteresis model,it is necessary to construct an inverse model with magnetic flux density B as input and magnetic field strength H as output.The classical Preisach model and the improved Preisach model that constructs the Everett function can be improved to the inverse model.Therefore,this chapter improves the static analytical Preisach model considering the reversible component by the difference method,and constructs the static analytical inverse Preisach model.Then,the R-L fractional derivative is used to improve the expression of eddy current loss component,and the improved dynamic analytical inverse Preisach hysteresis model is obtained by combining the static analytical inverse Preisach model,the improved eddy current loss component and the residual loss component through loss statistical theory and field separation technology.Finally,in order to verify the effectiveness of the dynamic model,the quantum genetic algorithm is introduced to globally optimize the fractional derivative parameters.The dynamic hysteresis loop simulation results are compared with the experimental data,and the maximum average relative error is 5.857 %,which verifies the accuracy and effectiveness of the proposed dynamic model.