Study On The Control System Of Lithium Battery Electrode Thickness Based On GA-BP Neural Network

Due to the lithium battery pole piece affected by many factors such as the chemical composition of the electric slurry particle material,the initial density,the coating quality,the degree of semi-solid state,the rolling temperature,the rolling force,the rolling gap and so on,together with the characteristics of non-linearity,uncertainty and time-varying of the thickness control of the lithium battery pole piece.The traditional thickness control method can hardly meet the user's requirements for the thickness accuracy of lithium battery pole pieces.Therefore,it is necessary to study a better algorithm of thickness control.In view of the problem that the thickness of the lithium battery pole piece cannot reach the accuracy desired by the user,this paper presents a pole piece thickness control study based on genetic algorithm to optimize the BP neural network.The combination of the two algorithms can play their respective advantages and improve the predictive ability of the lithium battery pole piece thickness.The main work of this paper is as follows:First of all,from the lithium battery pole piece production process and system characteristics,the thickness of the lithium battery pole piece control system was studied.The mathematical model of control system of the thickness of lithium battery pole piece is briefly analyzed.In combination with the cause of the fluctuation of the thickness of the lithium battery pole piece,the influencing factors of the measured thickness of the lithium battery pole piece are analyzed.Secondly,aiming at the problem that the thickness of the measured thickness of the lithium battery pole piece is not too high,a prediction model of the pole piece thickness control based on the neural network is proposed.According to the basic principles and learning steps of BP neural network algorithm,the BP neural network is introduced into the control system of pole piece thickness.The topology and parameters of BP neural network are determined according to the main factors.Finally,the relevant data model is simulated by MATLAB software and the validity of the predicted thickness of the lithium battery pole piece is also analyzed.Thirdly,according to the large error of BP Neural Network in predicting the prediction value of lithium battery pole piece thickness,this paper analyzes the defects of BP neural network and proposes several improved methods for these problems.It is found that genetic algorithm can improve the initial connection weight and threshold of BP neural network to seek the best in the global scope to make up for the shortcomings of BP neural network.Subsequently it describes the mechanism of BP neural network of genetic algorithm.Finally,according to the requirement of thickness accuracy of lithium battery pole pieces,the prediction model of pole piece thickness control based on BP neural network optimized by genetic algorithm is designed.On the MATLAB simulation platform,the thickness of the lithium battery pole piece is simulated.The predicted result is very close to the expected thickness.At the same time,compared with the non-optimized BP neural network model simulation results,the effectiveness of the method is verified.