基于径向基函数网络的永磁同步电机直接转矩控制  

Permanent Magnet Synchronous Motor Direct Torque Control Based on Radial Basis Function Network

作　　者：张春梅[1] 刘贺平[1] 王玉锋[1] 

机构地区：[1]北京科技大学信息学院,北京100083

出　　处：《控制与决策》2005年第7期803-806,共4页Control and Decision

摘　　要：对永磁同步电机直接转矩控制系统特性进行了深入研究,提出一种新的径向基函数神经网络控制器,给出了神经网络控制器的结构设计、样本选取及训练方法.利用系统中的开关表作为导师对径向基函数神经网络控制器进行训练,实现了永磁同步电机直接转矩控制的径向基函数神经元网络输出矢量选择.该控制器可以简化获得输出电压矢量的过程,并具有并行计算速度快、转矩响应迅速的性能.仿真结果验证了该控制器的有效性.The basic theory of direct torque control (DTC) of permanent magnet synchronous motor (PMSM) is analyzed and then a novel radial basis function neural networks controller is presented. The neural networks structure design, learning samples and training algorithms are expounded. A conventional switching table of permanent magnet synchronous motor direct torque control is used to train radial basis function neural network. The controller could quickly select an output vector and the system using this neural networks controller has quickly parallel speed and high torque response. The simulation of the permanent magnet synchronous motor shows the effectiveness of the control scheme.

关 键 词：永磁同步电机 径向基函数 神经网络 直接转矩控制 

分 类 号：TM921.5[电气工程—电力电子与电力传动]