无刷直流电机调速智能优化控制

Intelligent Optimal Control of Speed Regulating for BLDCM

在无刷直流电机调速系统优化研究中,针对传统PID控制器存在跟踪性差、初始状态选取不准确等问题,在BP神经网络优化PID参数的基础上,设计了一种遗传算法和BP神经网络结合的PID控制器。利用遗传算法弥补BP神经网络的局部性缺点,首先对BP神经网络的权值和阈值优化,再对神经网络的初始拓扑结构进行优化设置,最后根据优化后BP神经网络训练得到PID参数。通过在MATLAB/Simulink建立调速系统模型,将优化后的PID控制参数应用到此调速控制中,实验结果显示优化后的调速控制系统动态响应快、对非线性干扰有很好的补偿,速度跟踪精度得到了明显提高。

Aim at the traditional PID controller problem of BLDCM speed regulating system, a new method was presented in the paper, which can optimize PID parameter based on Genetic Algorithm （GA） and BP Neural Network. Having researched the defects of the BP - PID control, both topology structure and network parameter of Neural Network were optimized by using GA. The optimal PID parameters were generated by this optimized network＇s train- ing. With MATLAB/Simulink, a simulation model was established, then the PID parameters were applied to the sys- tem, the experiment result proves that the response of speed regulating control is quickly, and can better make up the nonelinearity jam, and the speed tracking precision can be obviously improved.