基于Matlab人机界面的无刷直流电机自适应控制

Adaptive Control for BLDCM System Based on Matlab in Human-machine Interface

提出了基于改进的RBF神经网络的无刷直流电机自适应控制新方法。该方法首先利用由Matlab中的RBF神经网络函数设计出的人机界面平台对无刷直流电机进行离线辨识,确定RBF神经网络的网络结构及初始权值;再采用RBF神经网络在线算法在线辨识无刷直流电机模型,获得PID参数在线调整信息,并由单神经元PID控制器参数的在线自整定,实现系统的智能控制。由于该算法具有自适应确定网络结构和无需人为确定网络初始权值的优点,因此减少了网络训练的随机性,提高了训练精度。实验结果表明,该控制方法具有较高的鲁棒性和控制精度。

A novel approach based on an improved RBF neural network was proposed, and through this approach, the adaptive control for BLDCM was realized. In the off-line method, the network architecture and initial weight values of the RBF neural network was obtained by using an interactive human-machine platform which was designed based on Matlab neural network toolbox to identify BLDCM~ then, RBF neural network i- dentified BLDCM model by on-line means and acquired on-line tuning information of PID parameters, and the self-tuning of controller parameters was implemented by the single neural controller, and the intelligence con- trol of system was achieved. Because Matlab-based RBF realization method has such advantages as self-adap- ting network architecture and non-artificially selected initial values, the random of network training was de- creased. Thus, the accuracy is improved. The results show that with the proposed method, and higher control precision and robustness is obtained.