Improvement Of Kinetic-Molecular Theory Optimization Algorithm And Application In Parameters Identification Of Solar Cell Model

The rapid and accurate identification of photovoltaic cell model parameters is of great engineering significance in the field of photovoltaic array output power prediction,maximum power point tracking and the characteristic research of battery fault model.The accuracy of traditional numerical analysis algorithm in system parameter identification is greatly affected by the initial value of parameters,but the intelligent optimization algorithm is widely considered to be an effective method to solve difficult optimization problems because it usually does not rely on specific conditions of certain issue.Kinetic molecular theory optimization algorithm(KMTOA)belongs to a class of swarm intelligent optimization algorithms with simulated physical law.It takes the thermal motion of molecules as the theoretical basis and converts the attraction,repulsion motion process between molecules into the exploitation-exploration optimization mechanism of the algorithm.In this paper,according to the complex characteristics of solving nonlinear equations in the parameters of photovoltaic cell model,the kinetic molecular theory optimization algorithm is selected for parameter identification.The main work is as follows:Firstly,the importance of solar cell parameter estimation is expounded,and the solar cell model and parameter identification method are introduced in detail.The advantages and disadvantages of the traditional parameter identification method and the intelligent optimization method are compared and analyzed,and the advantages and characteristics of the algorithm in this paper are compared with other algorithms of different optimization mechanisms.Finally,KMTOA is proposed to solve the problem of solar cell parameter identification.Secondly,Based on the kinetic molecular theory optimization algorithm,this paper proposes a strategy based on opposition-based learining and variable acceleration motion.This paper probes into the clustering phenomenon of the optimization algorithm.In view of the advantages and disadvantages of the algorithm,the opposition-based learining strategy and the concept of variable accelerated motion are combined.In the experiment of single、multi-mode and shift test function,the better optimization results show that the cooperation of the two search strategies can help the algorithm to achieve an effective balance between exploitation and exploration.Thirdly,two solar cell models were selected again,and the improved kinetic molecular theory optimization algorithm was applied to the parameter identification of solar cells.The results show that the improved algorithm can effectively optimize the parameter identification of single diode and double diode models of photovoltaic cells,and get better battery model parameters.Finally,an experimental platform for parameter identification of photovoltaic cell model based on GUI interface is designed to provide users with a simple and convenient tool in the form of visual interface and improving their work efficiency.