Research On Multi-objective Optimization Method Of Single-phase AC Contactor

In recent years,with the rapid development of new energy systems and the significant increase in power consumption,users have put forward new requirements for the performance and reliability of contactors.The optimal design of contactors draws the attention of people.Using methods such as orthogonal test and uniform experiment,traditional contactor optimization can only take the best parameters at the existing experimental level.However,it cannot search all uncertain factors in the optimization target.Therefore,it is necessary to introduce an intelligent optimization algorithm with global search capability.The contactor has many optimization goals,including static and dynamic characteristics,dimensional quality,power density,etc.It is a typical multi-objective optimization problem.In this paper,the single-phase AC contactor with more complicated structure is selected for optimization,and a multi-objective optimization method for complex electromagnetic mechanism contactor is proposed.The finite element method is currently the most effective method for analyzing the electromagnetic field of contactors,and it can also provide data for the establishment of an approximate model.This paper simulates the static and dynamic characteristics of the contactor and gets enough data points.The eddy current will have a great influence on the calculation results when calculating the dynamic characteristics of the contactor electromagnetic mechanism.Therefore,this paper uses FLUX to establish a finite element model of the dynamic characteristics of the contactor considering the eddy current.In the process of optimizing the contactor,due to the low computational efficiency of the finite element method,the need to solve the dynamic characteristics in the multi-objective optimization design process cannot be met.In order to improve the calculation efficiency of the dynamic characteristics of the contactor,a radial basis function neural network is used to establish an approximate model for the contactor.In order to further improve the accuracy of contactor approximation modeling,this paper proposes a fast algorithm for particle swarm optimization radial basis function neural network.In this paper,the contact time,suction speed,holding force and moving core quality of the contactor are the four objectives that need to be optimized.The orthogonal test is used to determine the key influencing factors of the optimization target.Using the improved fast algorithm of dynamic characteristics,this paper establishes an approximate model for single-phase AC contactor for multi-objective optimization design of contactors.Finally,this paper proposes a many-objective optimization algorithm NSGA?.In order to solve the problem of slow convergence of NSGA-? algorithm,this paper uses differential evolution algorithm to improve NSGA-?.While improving computational efficiency,it further enhances the convergence and diversity of the Pareto frontier.At the same time,the method of fuzzy membership is given to select the optimal compromise.Using the approximate modeling method and manyobjective algorithm proposed in this paper to optimize the optimization goal of single-phase AC contactor,the practicality of the proposed multi-objective optimization method on the contactor with complex electromagnetic mechanism is proved.