基于电流注入法的开关磁阻电机转矩脉动抑制  

Torque Ripple Reduction of Switched Reluctance Motor Based on Current Injection Method

作　　者：党选举[1] 潘登 DANG Xuan-ju;PAN Deng(School of Electronic and Automation,Guilin University of Electronic Technology,Guilin Guangxi 541004,China)

机构地区：[1]桂林电子科技大学电子工程与自动化学院,广西桂林541004

出　　处：《组合机床与自动化加工技术》2021年第5期115-119,共5页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金(61863008;61863007);广西自然科学基金(2016GXNSFDA380001)。

摘　　要：开关磁阻电机(SRM)在低速区转矩脉动比较严重,制约了其广泛应用。利用SRM电流波形具有周期重复性的特点,提出基于傅里叶神经网络(FNN)电流注入法抑制开关磁阻电机转矩脉动的控制策略,在线自适应调整电流波形,实现转矩脉动的抑制。在基于转矩分配函数的SRM控制系统中,电流信号有周期性且无间断点,对其进行傅里叶分解,依据此傅里叶分解式的数学结构,搭建能够反映SRM周期性的电流傅里叶神经网络,神经网络的输出通过注入的方式补偿相电流,以最小化转矩脉动为目标设计神经网络训练算法,在线训练得到理想的电流曲线,进而减小转矩脉动。仿真结果表明,与转矩分配控制相比,所提出的电流注入法能够精确补偿SRM控制电流曲线,得到恒定转矩下对应的控制电流,有效抑制转矩脉动。The large torque ripple of switched reluctance motor(SRM)at low speeds limit its further application,benefit from the periodicity of current waveform of SRM,this paper proposed a novel control strategy based on the Fourier neural network(FNN)current injection method for SRM torque ripple reduction by adjusting current Online and adaptively.The current signal of SRM under torque-sharing function control is periodic without any discontinuity point,a current Fourier neural network,which reflected the periodicity of SRM,is built According to the mathematical structure of the Fourier series decomposed from the current signal.Then by injecting the output of FNN,the phase current was compensated,furthermore,an Online neural network training algorithm taken lowest torque ripple as target is proposed for obtaining ideal current curve,which lead to a corresponding minimization of torque ripple.Simulation results show that the proposed control strategy reduces torque ripple effectively,compared with torque-sharing function control,the proposed current injection method accurately compensate the SRM command current for maintaining constant torque.

关 键 词：开关磁阻电机 傅里叶神经网络 转矩脉动 电流注入法 

分 类 号：TH16[机械工程—机械制造及自动化] TG506[金属学及工艺—金属切削加工及机床]