Research On Multi-step Optimal Model Predictive Vector Control Of Permanent Magnet Synchronous Motor

Permanent Magnet Synchronous Motor(PMSM)has the advantages of low inertia and low noise,which is widely used.With the improvement of China’s industrial level,the traditional control algorithm has been difficult to meet the increasingly high control accuracy of PMSM requirements.Model Predictive Control has many advantages,such as simple control structure,fast dynamic response and the ability to achieve multi-objective cooperative Control,which has great research value.However,the MPC strategy has some disadvantages,such as delay of predictive control,large amount of multi-step prediction calculation,and relatively dependent algorithm on motor prediction model,which will increase prediction error under the influence of model parameter mismatch and lead to decline of prediction accuracy,which affects the control performance of PMSM.In view of the above problems,the following aspects are studied in this thesis:Firstly,the development status of PMSM control strategy is briefly introduced.According to the research situation of MPC algorithm in PMSM control field at home and abroad in recent years,the advantages and disadvantages of MPC algorithm are summarized.The mathematical model of PMSM in two-phase coordinate system is derived,and the model is discretized by Euler discrete method.The vector control principle and voltage space pulse width modulation strategy are described,which lays the foundation for the research of MPC controller and its optimization in the following thesis.Secondly,the closed loop control model of PMSM predictive vector control is built,and the model predictive current controller is designed.In order to solve the control delay problem existing in the model prediction algorithm,the cost function was redesigned by using the model prediction "multi-step prediction" control strategy.To solve the problem that the multi-step prediction calculation of MPC was too large,proposing a fast optimization method of the cost function.The control effect of the MPC controller proposed in this thesis on PMSM is verified in the simulation.Thirdly,in view of the problem that the MPC algorithm is easily influence by model parameter mismatch disturbance.Theoretically,the influence of parameter disturbance on the prediction error is analyzed,and the theory is verified and supplemented through simulation.To improve the ability of the MPC controller to resist the model parameter mismatch,a disturbance observer was designed based on Kalman filter.A simulation model is built to verify the disturbance observer’s ability to suppress parameter mismatch disturbance.At last,the DSP28335 chip of TI Company is used as the control core to build the PMSM hardware experiment platform.Firstly,the feasibility and availability of the designed MPC controller were verified by motor starting and load experiments.Then,experiments are carried out to verify that the disturbance observer can resultful suppress the adverse effects caused by parameter mismatching.Motor control performance and improve the robustness of the system when the motor parameters mismatch.