Research On Sliding Mode Control With Anti-Windup For Permanent Magnet Synchronous Motor Speed Control System

Permanent magnet synchronous motor(PMSM)have been applied in the numerical control machine tools,industrial production,ship industry and so on because of its high power density,high reliability,simple structure and small size,and in these fields,mo-tor speed control plays an important role,but PMSM is a complex nonlinear system with physical limitation and strong coupling.In order to get high precision speed control per-formance,a deeper research in the control methods has to be done and also the methods must consider the current coupling,system saturation and system nonlinearity.Sliding mode control(SMC)has the advantages of overcoming the system nonlinearity and can combines with the Anti-Windup method,therefore,SMC has been applied widely for PMSM speed control system.First of all,the mechanism analysis of the windup phenomenon cause by satura-tion limit in the PMSM speed control system is given,and then the thesis proposes the improved measures based on the traditional Anti-reset Windup method.The new Anti-reset Windup introduces exponential gain in the integral part of the controller to weaken the integral impact before the system reaches steady state,thereby reducing the adverse effect of saturation limit on the speed control system.Simulations prove that the im-proved Anti-reset Windup can effectively weaken the adverse impact that comes from the saturation limit of the PMSM speed control system.Secondly,the chattering produced by the SMC will affect the PMSM speed control performance,in order to solve the problem,this thesis first analyses the contradiction between the system approach speed and chattering level of the traditional SMC reach-ing laws,such as the exponential reaching law.And then new siding mode control with Anti-Windup(NSMCAW)based on the convergent exponential reaching law is proposed,after analysis the switching layer width between the different reaching laws,we can find that the convergent exponential reaching law can effectively reduce the chattering of the system,at the same time the method can also retains the quick re-sponse advantage of the traditional exponential reaching law.Two different types of simulations are used to respectively verify the great performance of the proposed SMC and NSMCAW in the motor speed control system.Again,in view of the existence of the switching control of SMC causing the sys-tem chattering problem,the thesis proposes the neural network sliding mode control method with Anti-reset Windup(NNSMCAW).The new method replaces the switch-ing control by neural network control to avoid the generation of chattering,at the same time the parallel control is introduced to the method overcome the local convergence of the neural network,simulation results show that in the PMSM speed control system,the proposed NNSMCAW can not only retain the advantages of quick response and high speed control precision of NSMC,but also avoid chattering problem.Finally,experiments are used to prove the validity of the NSMCAW,firstly,the thesis introduces the hardware platform and software processes of PMSM speed control system,then digital implementation of the incremental SMCAW is given,at last,the dynamic and steady performances of the proposed method is proved by the PMSM start and speed regulation experiments,also comparison and analysis of the algorithm time complexity is given at the end of the thesis.