直接转矩控制系统定子电阻的神经网络辨识

在研究异步电机直接转矩控制的基础上 ,提出了基于神经网络方法的定子电阻辨识 为了缩短学习时间 ,保证系统收敛 ,采用了自适应调整学习率 选取实时递归网络 ,对不同隐含层单元数和训练次数进行比较 ,获得合适的单元数和训练次数 仿真结果证明 ,应用该辨识可以进一步改善直接转矩控制系统的动态性能

Stator Resistance Identification of Direct Torque Control System Using Neural Networks

In this paper, a theoretical analysis of the direct torque control has been carried out. A novel approach to estimating the induction motor stator resistance is presented. This method is based on the theory of the neural networks identification algorithm. In order to save learning time and keep system convergent, a learning method with adaptive adjusting and a current recursive network are used. Comparing different neurons in the hidden layer and training epochs, appropriate results are obtained. From the computer simulation it is shown that the direct torque control system using this method has good performance.