Fuzzy Iterative Learning Synchronization Control For H-type Motion Platform Driven By Linear Motors

The object of this thesis is an H-type motion platform composed of three identical permanent magnet linear synchronous motors.Although on the two Y-axis chose the same linear motor,but due to the motor parameters,mechanical coupling,load disturbance and uncertainty factors such as friction will lead to two parallel linear motor motion are not synchronized,this not only affect the work-piece machining accuracy,serious can lead to system damage and even threatened the personal safety.In order to improve the machining accuracy of H-type moving platform and promote the development of the platform,a controller is designed between single axis and double axis to reduce the error.In this thesis,after considering the interference of beam motion to biaxial axis,the mathematical model of H-type moving platform is established.A recursive RBF neural network friction compensation sliding mode controller is designed on single axis to reduce the tracking error of single axis,a type-2 fuzzy iterative controller is designed on two-axis to reduce the synchronization error between two-axis and improve the precision of the system.Firstly,after consulting relevant data,this thesis summarizes the development status of the H-type motion platform driven by PMLSM at home and abroad and the research status of some control strategies.Since the load on the X-axis will generate torsional pendulum force on the linear motor on the Y-axis when it reciprocates along the X-axis,the force on the two linear motors on the Y-axis is not uniform.According to the basic principle of PMLSM,the mathematical model of the H-type motion platform is established.Second,according to the tracking error of the single axis problem,recursive RBF neural network sliding mode controller was designed,and the combination of neural network and sliding mode control effectively weaken the chattering of sliding mode control is easy to occur phenomenon,at the same time in order to further improve the single axis tracking precision,on the impact of friction compensation,then according to the Lyapunov theorem,prove the stability of the controller.The designed controller is simulated by MATLAB/Simulink and compared with the ordinary neural network sliding mode controller.Finally,synchronization controller was designed to reduce the synchronization error of the system between biaxial,but as a result of H-type motion platform do reciprocating motion disturbance,are prone to repeat in order to solve this problem,will control repeated disturbance is very effective iterative learning control and the range of type-2 fuzzy control,the combination of the two kinds of combined control strategy can speed up the convergence of iterative learning control and enhance the anti-jamming ability of controlled system,improve the system accuracy.The simulation and analysis of the two controllers are carried out by MATLAB/Simulink,and the simulation results of the controlled system are compared between the two controllers to verify the effectiveness of the controller designed in this chapter.