Intelligent Adaptive Backstepping Control For Permanent Magnet Linear Synchronous Motor

With the rapid development of science and technology,permanent magnet linear synchronous motor(PMLSM)is widely used in modern manufacturing industries of industrial robot,CNC machining and semiconductor manufacturing system,etc.Because of its advantages such as high thrust,fast response and high precision.The PMLSM servo system directly drives the feed mechanism to do linear motion,which omits the intermediate transmission device in the structure.However,the tracking accuracy of the system is easily influenced by the uncertainty factors such as parameter change,friction force and load disturbance,etc.Therefore,in order to improve the tracking and robust performance of the system.In this thesis,the combination of neural network and backstepping control is adopted to control the PMLSM servo system.Firstly,the research status of PMLSM servo system and backstepping control method is summarized.The structure and working principle of PMLSM are introduced.And the causes of uncertainties in the system and their effects on tracking performance are analyzed.A mathematical model of PMLSM with uncertainties is established.The PMLSM vector control system is established based on the vector control principle.Then,aiming at the position tracking problem of PMLSM servo system,a backstepping control scheme is designed.The virtual control function is devised by gradually modifying the algorithm.In each backstepping stage,a new virtual control law is generated by using the control law of the previous stage until the actual control input is obtained.At the same time,the stability theory of Lyapunov function is used to guarantee the stability of the system at each stage of the backstepping.The simulation results show that the proposed control scheme improves the tracking performance of the system significantly.Finally,in order to further improve the tracking performance and robust performance of the PMLSM servo system,for the problem which the uncertain factors are unknown and difficult to obtain in the design process,the static adaptive radial basis function(RBF)neural network backstepping control scheme and the dynamic adaptive Elman neural network backstepping control scheme are adopted respectively,and the dynamic equations of the system are comprehensively analyzed and deduced theoretically.The neural network is used to estimate the uncertainty to overcome the influence of uncertainty on the system and to improve the robust performance of the system.Compared with the simulation results,the intelligent backstepping control can further improve the robust performance and tracking performance of the system,while the dynamic Elman neural network backstepping control has faster convergence speed and better tracking performance than the static RBF neural network backstepping control.