摘要
针对矢量控制系统提出了一种转子电阻在线辨识方案。应用神经网络理论,以转子磁链电压模型的输出为参考值,神经网络模型的输出为估计值,通过反向传播算法不断调节神经网络的权值,使转子磁链的估计值跟踪参考值,间接辨识出转子电阻。在MATLAB6.5/SIMULINK下,对无速度传感器感应电机矢量控制系统在电阻变化时的情况进行了仿真。仿真结果表明,辨识算法具有较好的静动态性能。
This paper proposes an adaptive scheme for identification of rotor resistance based on the Artificial Neural Networks. By using the BP algorithm theory, the rotor flux error between the voltage model and the neural network model is back propagated to adjust the weights of the neural network model which can be used to calculate the rotor resistance. The simulation scheme of speed sensorless induction motor using vector control is built in the Matlab6.5/Simulink, and the results of simulation indicate that the neural network observer has good steady and dynamic performance.
出处
《变流技术与电力牵引》
2007年第2期6-10,共5页
Converter Technology & Electric Traction
关键词
无速度传感器
神经网络
电阻辨识
矢量控制
speed sensorless
artificial neural network
resistance estimation
vector control
