Model Predictive Torque Control Of Permanent Magnet Synchronous Motor Systems

Permanent magnet synchronous motors(PMSM)are widely used in aerospace industry,transportation,shipping industry,etc.,due to their advantages of simple structure,high power density,high efficiency and high torque current ratio.Model predictive control(MPC)has huge potential to be applied in PMSM drives with stable running and high efficiency.However,to be the essential technology in PMSM control area,MPC still has many problems to be address immediately.Therefore,this dissertation focuses on 2 level voltage source inverter fed permanent magnet synchronous motor(2L-VSI-PMSM)drive system,carries out a series of researches about the problem of high torque ripple,low parameter robustness in MPC-PMSM.Two kinds of model predictive torque control strategies,that is,finite control set model predictive torque control(FCS-MPTC)and continuous control set model predictive torque control(CCS-MPTC),are discussed from three aspects,ie control set,predictive model,the optimal vector or control law.On this basis,the main problems and difficulties of MPC-PMSM are dissected.In the traditional FCS-MPTC,the cost function consists of torque and flux magnitude errors,and they are different in unit and magnitude.Thus a suitable weighting factor is required to balance the control performance between torque and flux.However,there are not enough theoretical design guidelines for the tuning of weighting factor currently.Therefore,the determination of weighting factor is an exploratory process determined by a large number of simulations and experimental results.To solve this problem,this dissertation proposes a FCS-MPTC without weighting factors for 2L-VSI-PMSM.In the proposed strategy,the reference vector of stator flux is constructed according to the relationship among the torque,flux magnitude references,and the phase angle of the stator flux.On this basis,the cost function that is only made up of the error between the prediction and reference of the stator flux vector is constructed;the predictive model based on stator flux vector is established.Combining the above two measures,the proposed strategy can be achieved to improve the control performance efficiently.In the traditional FCS-MPTC,only six active vectors with fixed amplitude and phase angle and two zero vectors are alternatives in the conventional finite control set,which leads to the main limitations of the torque and flux control precision.Furthermore,duty cycle modulation cannot be effectively used in traditional FCS-MPTC.To solve the above problems,this dissertation proposes an improved FCS-MPTC.Combining the discrete duty cycle technology,the candidate vectors are synthesized by two adjacent active vectors and zero vectors,meanwhile,the duty cycles of the two adjacent active vectors and zero vectors are generated directly.Afterwards,an extended control set is constructed and the corresponding vector selection is set up.According to the extended control set,a FCS-MPTC with discrete duty cycle control for 2L-VSI-PMSMis proposed in this dissertation.The performance of CCS-MPTC is directly related to the predictive model.If there are unknown parameter variations in the actual system,CCS-MPTC will be difficult to suppress the disturbances,thus results in worse control effects and lower level stabilization.To solve this problem,a robust nonlinear predictive torque control strategy is proposed.In the proposed strategy,the influence of the parametrical nonlinearity on the torque and stator flux is taken into account and the nonlinear mathematical model is established.On this basis,a nonlinear disturbance observer is designed to estimate the disturbances caused by parameter uncertainty.At the same time,a nonlinear predictive torque controller is designed to improve the performance of torque.The experimental results show that,in the proposed approach when the values of the electrical parameters can not match with the actual values,the proposed robust nonlinear predictive torque control strategy will strengthen the robustness in a wide range.