Research On Torque Ripple Reduction Based On Current And Torque Injection Method For Switched Reluctance Motor

Switched reluctance motor(SRM)has a promising prospect due to its merits such as simple structure,high safety and no consumption of rare earth materials.However,the double-saliency pole constructure of SRM leads to a large torque ripple,especially in lowspeed condition,which limits its further development and wider application in the field of new energy vehicles.Therefore,this paper proposes two control strategies for the purpose of torque ripple reduction of the SRM.(1)The current injection method control strategy based on Fourier neural network(FNN)is proposed to minimize the torque ripple of the SRM.The current waveform is adjusted online by current injection method to achieve torque ripple minimization.In the SRM control system based on the torque-sharing function,the current waveform is periodic and each phase of current has the same shape.The constructed FNN current model which reflects the periodicity of the SRM is built according to the mathematical structure of the Fourier series.The output of FNN compensates the control current through injection method,and the ideal current curve under constant torque is obtained by online training procedure of FNN.Simulation results shows that,the proposed current injection method compensates the control current of SRM accurately,the corresponding control current under constant torque is obtained,and torque ripple reduces effectively.(2)The control strategy for SRM torque ripple reduction based on torque injection feedforward compensation method is proposed in this paper.The torque ripple of SRM control system is periodic which meets the Dirichlet condition.Hence,the FNN torque model is built to obtain the harmonic information of electromagnetic torque.Considering the historical harmonic information of torque waveform,a signal generator based on FNN is utilized,and the compensation signal is injected into the reference torque to eliminate the harmonics which causes torque ripple.The simulation results verify the effectiveness of the control strategy based on torque injection feedforward compensation.The results shows that the proposed method can significantly reduce the torque ripple without precise SRM model parameters.On the SRM experimental platform based on Infineon XC2765-104 F microcontroller,the proposed torque injection feedforward compensation control strategy is tested and analyzed.Experimental result illustrates that the proposed torque ripple reduction control strategy reduces torque ripple effectively and its feasibility was validated.