Research On Control System Of Switched Reluctance Motor Based On Model Prediction Control

Switched reluctance motor(SRM)is simple,stable,reliable,and low-cost,so it has been popularly studied in recent years.Switched reluctance motor drive system(SRD)is developing rapidly,and now ti is widely used in mining,oil exploration,electric car driver and appliances and other industrial fields.However,due to the structure and the the control mode of SRM,the torque ripple is large,which limits the application of SRM in the field of high precision control.Therefore,how to effectively reduce the SRM torque ripple has become one of the hotspots for scholars.In this paper,based on the model predictive control method,the control system of switched reluctance motor is studied.After analyzing the advantages and disadvantages of the model predictive control method based on the finite switch condition,the predictive control based on the optimized current waveform is carried out to achieve the control of the torque ripple minimization of the SRM.Firstly,this paper summarize the research progress of SRM torque ripple optimization at home and abroad,then it introduce the basic working principle,mathematical equation and general control method and power circuit of control system of SRM.After that,in order to obtain the accurate flux model of the motor,an online experimental methods is given to obtain the parameters of the model,its essence is a SRM flux characteristic line detection method.From the controller 's own character,the error of the methord is estimated at the static or very low speed,and the flux estimation is corrected according to the equivalent resistance calibration method.Then,two kinds of on-line detection specific operation schemes are given,and the 18.5k WSRM flux linkage data is obtained by the scheme of speed open loop.Then it introduce a method to obtain the optimal reference current based on the case of torque minimization,and provide data support for the experiment and simulation part.Then,based on the introduction of the basic principle and basic characteristics of the model predictive control(MPC),the control system based on the limited switch condition model is realized,and the control of the torque ripple,the current and the switching frequency is realized.For the problem that the current fluctuation is large in the control system and the switching frequency is not fixed and the harmonic content is high,an improved model predictive control method is proposed,that is,the deadbeat predictive current control.Based on the optimal current profile obtained by minimizing the torque fluctuation obtained in the foregoing,the deadbeat predictive control of the reference current is achieved to reduce the fluctuate of the motor torque.Finally,this paper introduces the hardware composition and software structure of the control system,and completes the motor running test based on the model prediction control through the experimental platform.The experimental results show that the controller can realize the suppression of the reference current under torque ripple minization in the full speed range.The proposed method can effectively reduce torque ripple and follow the torque change rapidly.It achieves the high performance control of the SRM and has practicability and reference value.