Research On Motion Control Of Chain Automatic Magazine Based On Inversion Method

Automatic chain magazine is an essential part of artillery automatic filling system.The motion accuracy of it is an important guarantee to improve the efficiency and the reliability of artillery automatic filling system.Therefore,it is of great significance to study the motion control of the automatic chain magazine.In this paper,the chain type automatic magazine in an artillery automatic filling system is researched.Based on the existing automatic chain magazine platform in the laboratory,utilizing the dynamic analysis of the automatic chain magazine,the ADAMS virtual prototype modeling,the MATLAB simulation analysis,and combining the adaptive method,back-stepping method,neural network theory and other control methods,the motion control of the automatic chain magazine is studied.The main contents of this paper are as follows:(1)The structure and working principle of chain type automatic magazine are introduced.According to the mechanical system equation,the electrical system equation and the actual working situation of the magazine,the mathematical model of the magazine motion system with nonlinear term is derived,considering the electromagnetic torque change caused by the change of the motor parameters,the system parameter change caused by the change of the magazine load and the unknown external disturbance in the process of the magazine motion as the main factors affecting the magazine motion accuracy.The virtual prototype of the magazine is established by ADAMS,and the co-simulation platform of ADAMS-MATLAB is established.(2)The simulation results of the co-simulation platform demonstrate that the PID control method has some limitations in the magazine motion control.In order to improve the motion accuracy,an adaptive back-stepping controller is designed according to the characteristics of the automatic chain magazine.The differential explosion problem in the back-stepping method is solved using the dynamic surface control method.The parameters are estimated by the adaptive algorithm,compensating for the changes of system parameters when the magazine moves.Using the ability of RBF neural network to approach the nonlinear unknown term,the unknown disturbance outside the magazine and the nonlinear change of the electromagnetic torque of the motor are approached,meaning the influence of the external disturbance on the motion accuracy is reduced.The co-simulation results show that the neural adaptive backstepping control has higher control accuracy than PID control and adaptive dynamic surface back-stepping control.(3)The automatic chain magazine experiment platform is built to apply the control method designed in this paper.The experimental results illustrate that the proposed neural adaptive back-stepping control method is less affected by the load change.And it compensates the adverse effect caused by the unknown nonlinear external disturbance and the nonlinear change of the motor parameters.The proposed neural adaptive back-stepping control method has advanced control effect and improves the accuracy of the magazine motion control.