Research On Continuous Control Set Model Predictive Control Of Permanent Magnet Synchronous Motor

Permanent Magnet Synchronous Motor(PMSM)is a high-performance AC motor.Because it uses permanent magnet excitation rather than excitation winding,it does not generate excitation current loss.It has the characteristics of high efficiency and low rotor heat.Compared with the induction motor,PMSM can save up to 50%electric energy.In addition,the permanent magnet has a higher air-gap flux density than the excitation current,which also makes higher power density of PMSM.PMSM has smaller volume,lighter weight,high torque current ratio,fast dynamic response speed,higher reliability and better low-speed operation performance compared with induction motor at the same power level.Therefore,PMSM has been widely used in many industrial fields,and the development of its control system has attracted more and more attention.In recent years,due to the rapid development of processors,model predictive control has been widely used in PMSM control system.PMSM drive system based on model predictive control has some problems,such as mismatch between system constraints and model parameters,frequent steady-state current fluctuation,and great influence of load disturbance.Therefore,this paper takes PMSM as the research object and the control system based on continuous control set model predictive control is studied.The main contents include:A cascade control method combining model predictive current controller and sliding mode control is proposed to solve the problems of poor robustness and large steady-state current fluctuation of PID cascade control in traditional vector control.In this method,vector control technology is adopted,and the sliding mode function is designed according to the control objective of the speed loop.The chattering phenomenon and steady-state error are reduced by using exponential law to approach the sliding mode surface.The current increment model of q-axis current loop is designed to cancel the interference items,and then the model predictive current controller is designed by using the weighted sum of tracking trajectory error and control increment as the cost function,and the stability of the algorithm is discussed.Finally,the simulation results prove that the speed loop has good speed tracking performance and anti-interference ability,and the steady fluctuation of the current loop is reduced.To handle the unavoidable current and voltage constraints in PMSM system,a noncascaded PMSM explicit model predictive control method based on multi-point linearization is proposed.Firstly,a non-cascade PMSM model is constructed,and the multi-point linearization form of the model is given to adapt to the multi-operating conditions of the system.Then,model selection and switching modules are designed according to the actual working conditions to achieve accurate ad justment of the prediction model.The designed controller takes the error between the system output and tracking value as well as the sum of the error weight between the input voltage and the steady state voltage as the objective function,and transfers the solving process of the optimization problem with constraints to the offline.Online computation is only about to look up the table to obtain the optimal control law,which greatly reduces the online calculation time.Finally,the simulation results show that the proposed method can solve the system constraint problem,and the dynamic response speed and steady-state performance of the system are improved compared with PI algorithm.In order to solve the problem that the motor is susceptible to various external disturbances,especially load disturbances,an improved explicit model predictive control with load observer is proposed considering multiple working conditions.In this method,load torque is taken as the independent extended state variable.Motor speed,q-axis current and the extended variable are taken as the state variables of the improved model,and daxis current is separately controlled by PI.The explicit model predictive controller based on this model takes the error between the set value of speed and the steady state value and the sum of the increment of current and voltage as the objective function to reduce the current fluctuation.Based on this,the load disturbance is estimated by using the load observer to further reduce the negative impact of the load disturbance on the system performance.Finally,the simulation results show that the method can reduce the offline storage space without multiple offline optimizations.The dynamic response speed and steady-state performance under the condition of disturbance are also ensured.Furthermore,a fast model predictive control based on parameter identification is proposed to solve the problem that model parameters mismatch due to the change of external environment or long running time.This method uses the interior point method to transform inequality constraints into equality constraints,and then uses the infeasible Newton method to solve the problem.Block elimination,warm start and fixed barrier parameters are used to accelerate the optimization process,and then the algorithm complexity is analyzed to verify the feasibility for fast sampling system.The inductance parameters are identified by the least square method,so the PMSM controller parameters are updated online and the model accuracy is improved.Finally,the inductance mismatch is simulated to verify the effectiveness of the proposed method.Finally,the experimental platform of PMSM drive control system based on continuous control set model predictive control is built.The experiments are carried out to verify the feasibility and effectiveness of the proposed method.The Dynamic performance,steady-state performance,disturbance resistance and parameter robustness of PMSM control system are analyzed.The experimental results show that the proposed method can improve the dynamic performance,steady-state accuracy,anti-interference ability and parameter robustness of PMSM system.The research results of this paper provide solutions to the key problems in PMSM drive control system,and can promote the application of continuous control set model predictive control in practical engineering.Therefore,the research is with high-level academic value and a great potential application perspective.