基于柔性神经网络和转矩分配函数的SRM控制  

Direct Instantaneous Torque Control of Switched Reluctance Motor Based on Flexible Neural Network and Torque Distribution Function

作　　者：张锟 李红伟[1] 唐学建 胡其杰 陈琦 ZHANG Kun;LI Hongwei;TANG Xuejian;HU Qijie;CHEN Qi(School of Electrical Information, Southwest Petroleum University, Chengdu 610500, China)

机构地区：[1]西南石油大学电气信息学院,成都610500

出　　处：《微电机》2020年第7期55-60,共6页Micromotors

基　　金：四川省科技厅重点研发项目基金(2019YFG0334)。

摘　　要：开关磁阻电机(SRM)的应用越来越普及,但运行时存在转矩脉动是其较为突出的缺点,特别是电机处于低速转动时。针对这一问题,本文提出了一种柔性神经网络结合转矩分配函数的直接瞬时转矩控制方法(DITC)。采用PID环节对转速差外环进行调节,柔性神经网络结合转矩分配函数对转矩内环进行调节。该方法结合了柔性神经网络控制和转矩分配函数的优点,使电机能正常运行且能最大程度抑制转矩脉动,响应时间也有改善。通过Matlab/Simulink仿真,并与柔性神经网络和转矩分配函数单独与DITC结合的系统对比,验证了该控制效果更优。The application of switched reluctance motor(SRM)becomes much common,but the torque ripple is its main disadvantage,especially when the motor is operating with a low speed.To solve this problem,a direct instantaneous torque control(DITC)method based on flexible neural network and torque distribution function was proposed.The PID link was adopted in the control system to adjust the outer ring of rotation speed difference,and the flexible neural network combined the torque distribution function to adjust the inner ring of torque.This method combined the advantages of flexible neural network control and torque distribution function,so that the motor can run normally and restrain the torque ripple to the maximum extent,and the response time was also improved.Through Matlab/Simulink simulation,and compared with the system of flexible neural network and torque distribution function combined with DITC alone,the control effect was verified to be better.

关 键 词：柔性神经网络 开关磁阻电机(SRM) 转矩分配函数 直接瞬时转矩控制(DITC) 

分 类 号：TM352[电气工程—电机] TP273[自动化与计算机技术—检测技术与自动化装置]