基于神经网络的SRM直接转矩控制系统仿真研究  

Simulation Research on Switched Reluctance Motor Using Direct Torque Control Based on Neural Network

作　　者：王勉华[1] 王瑞[1] 刘春元[2] 

机构地区：[1]西安科技大学电气与控制工程学院,西安710054 [2]河西学院物理与机电工程学院,甘肃张掖734000

出　　处：《微电机》2013年第1期55-58,共4页Micromotors

基　　金：陕西省教育厅专项科研计划项目(09JK586)

摘　　要：文中采用直接转矩控制(DTC)方法有效的抑制了开关磁阻电机的转矩脉动。并将基于神经网络自适应PID控制的电机调速系统应用于该直接转矩控制系统,解决了常规PID控制器难以取得良好控制效果的问题。其中主要介绍了BP神经网络自适应PID以及RBF神经网络自适应PID控制策略的基本原理以及算法,并对比了二者的响应速度和鲁棒性。仿真结果表明,在SR电机调速系统中,利用神经网络收敛迅速的优点,两种控制器均能实现对给定转速快速、稳定的跟踪,并能适应系统参数的变化,具有良好的适应性和鲁棒性。但RBF-PID调速系统响应速度更快,更适用于实时控制的系统。This paper adopted Direct Torque Control （DTC） method to reduce the torque ripple of switched reluctance motor. It was difficult to achieve good control effect was the main problem of the conventional PID controller. In order to solve this problem, motor speed control system based on the neural network adaptive PID control was applied to this SRD system, and mainly introduced the basic principle and algorithm of BP neural network adaptive PID and RBF neural network adaptive PID. The simulation results show that, BP- PID and RBF-PID controller are both able to fast and stably track the given speed, and can adapt to the vari- ation of parameters in SR motor speed control system, and these two methods both have good adaptability and robustness. But compared with BP-PID, the speed of response of RBF-PID speed control system is faster, so it＇s more suitable for the real time control system.

关 键 词：开关磁阻电机 直接转矩控制 神经网络 BP—PID RBF—PID 

分 类 号：TM352[电气工程—电机]