无刷直流电机单神经元自适应智能调速系统

本文针对无刷直流电机设计了一种可在线学习的单神经元自适应PID智能控制器,通过有监督的Hebb学习规则来调整权值,每一次采样都直接根据反馈误差对神经元权值进行调整,实现PID三个参数的自适应,采用分段线性化方法建立无刷直流电机反电动势梯形波,采用滞环电流控制器实现电流调节,并在Matlab/Simulink仿真平台上搭建仿真模型,结果表明智能PID控制效果和鲁棒性都优于常规PID,大大提高了系统的跟随性,能满足无刷电机系统对实时性的要求。

Single Neuron Self-adaptive Intelligent Speed Control System of Brushless DC Motor

According to brushless DC motor, this paper designed a kind of single neuron self-adaptive PID intelligent controller which can be learned online, the weights can be adjusted by supervised Hebb learning rule, the neuron weights are adjusted according feedbackerror in every sampling, so as to realize the self-adaptation of three PID parameters. Brushless DC motor back EMF wave was built by using piccewise linearization method, the current regulation was realized by using hysteresis current controller and simulation model wasstructured based on Matlab / Simulink simulation platform. The results show that the intelligent PID control effect and robustness are better than the conventional PID, so it greatly improved the following characteristic of the system, and it can satisfy the real-time requirements of the brushless motor system.