基于神经网络的直流无刷电机控制系统

提出了一种直流无刷电动机的N-PI转速调节器的设计方法.在直流无刷电动机的高性能速度跟踪中,若仅采用传统的PI调节器,则难以克服系统超调和短时振荡问题.采用复合N-PI的控制方法,利用神经网络的自学习自适应功能在线调整PI控制参数.文中提出了一种模型参考自适应与神经网络相结合的控制策略,利用在线辨识技术,对参数变化实时补偿,及时修正神经网络权值的计算.最后,在Matlab/Simulink下进行了仿真,结果表明,运用这种设计方法很好地抑制了超调和振荡.

BLDCM Control System Based on Artificial Neural Network Method

In this paper,a neural network-PI controller is proposed for controlling the speed of BLDCM in high-performance drives environment.In such environment,if PI controller is only used,there must be over-regulating and instantaneous oscillation.In this paper,a multiple N-PI system is used,and online update the PI parameters with self-study and adaptive functions of a neural network.A model reference adaptive and neural network control strategy is proposed in this paper,an on-line indentification technique is used to compensate the variation of the parameters and modify the calculation of the neural network's weight.The result of this control method is satisfied in Matlab/Simulink.So this method help to reduce the over-regulating and eliminate oscillation.