Predictive Current Control Of Switched Reluctance Motors Based On State-Space Model

Switched Reluctance Motor has received extensive attention in electric vehicles application due to its simple structure,low production cost,high starting torque,high reliability,strong regenerative braking capability and so on.However,the characteristics of the double salient pole structure result in the instantaneous torque ripple of the motor especially during the commutation process.Since the reduction of torque ripple depends to a large extent on the control accuracy of the motor current,this paper takes advantage of good rolling optimaization of model predictive control and designs the current loop controller of the switched reluctance motor speed control system based on model predictive control,which improves the motor current tracking capability and thus effectively reduces the motor torque ripple.The traditional finite control set predictive current control is introduced.Aiming at the large computational complexity of traditional finite control set predictive current control,an improved finite control set predictive current control method is proposed to reduce the computational burden while ensuring control performance is not affected.Simulation studies show that compared to hysteresis current control the improved finite control set predictive current cont rol improves the current tracking capability,but the current and torque ripple suppression effects are not ideal,and there is also a problem that the switching frequency is not fixed.In light of deficiencies of large current ripple and unfixed switching frequency in the improved finite control set predictive current control method simulation results,predictive current control method based on continuous control set is proposed.The current tracking error and control quantity are selected to form the obje ctive function,and the precise duty cycle required for the reference current on the actual current tracking is directly calculated by solving the optimization target,which is applied to the current loop control of the switched reluctance motor speed cont rol system so as to realize the control of current and torque ripple.This paper also proposes a simplified nolinear phase inductance model,which uses the measured motor incremental inductance datas to obtain the phase inductance required in predictive current control.This method improves the accuracy of the phase inductance model and eliminates the need for motor flux linkage datas.MATLAB/SIMULINK simulation results show that the model predictive current control method based on continuous control set can realize the accurate tracking of the actual current to the reference current,and has good torque ripple reduction effect.Finally,a 1.5kW three-phase 12/8 switched reluctance motor for electric vehicles is taken as the experimental object,and a DSP-based switched reluctance motor speed control system experimental platform is built to verify the feasibility and effectiveness of the proposed predictive current control based on the continuous control set.