Position Control Of An Artillery Magazine Based On Fuzzy Neural Network

Chain magazine is an important part of modern gun automatic loading system.Its positioning accuracy and motion state have a profound impact on the overall reliability of the automatic loading system.In this paper,based on sliding mode control,combined with fuzzy control theory and neural network control theory,the chain magazine is studied in depth.(1)The basic components and work flow of magazine system are introduced in detail,and the corresponding control requirements are put forward.According to the working characteristics of the system,the mechanical motion equation and electrical basic equation of the system are derived.At the same time,considering the accuracy of system modeling,Lu Gre friction model is introduced into the equivalent mathematical model of magazine system.(2)The corresponding equations of Lu Gre friction torque are obtained in the low-speed region and high-speed region of the system.The parameters are identified by recursive least square identification method,and the corresponding static parameters are obtained.The dynamic model of magazine system is established in ADAMS,and the model is checked.Finally,combined with MATLAB / Simulink software,ADAMS-MATLAB joint simulation experimental platform is built,and PID controller is designed for joint simulation.(3)In the case that the traditional PID control algorithm has some limitations,this paper designs an adaptive sliding mode controller based on the equivalent mathematical equation of the magazine system,which can not only meet the control requirements,but also estimate the dynamic parameters of Lu Gre friction model online.On this basis,the fuzzy adaptive sliding mode controller is designed to accelerate the convergence speed of the system and further improve the control accuracy.Finally,a fuzzy neural network adaptive sliding mode controller is designed,which can significantly improve the control accuracy and system stability.(4)The experimental platform of chain automatic magazine system is built,and the three control algorithms are verified in the platform.The experimental results show that the three algorithms can estimate the relevant parameters of Lu Gre friction model,and meet the control requirements of this paper,and have strong robustness.