基于改进BP神经网络PID的无刷直流电动机速度控制的研究

引进模糊归一化控制策略，在线实时地调整与收敛速度密切相关的学习速率和动量系数，克服了BP网络收敛慢和容易陷入局部最小的缺点,并将改进的BP神经网络PID算法成功应用于无刷直流电动机速度控制中。仿真结果表明,改进BP神经网络PID使收敛变得更快，而且系统具有较强的鲁棒性和自适应能力。

Research on the Speed Control of BLDCM based on Improved BP Neural Network PID

The strategy of fuzxzy control is proposed and the study speed and momentum coefficient can be regulated real-time to shorten the comvergence and overcome the local optimication. Tben the algorithm of improved BP neural network PID is used successfully in the speed control of the BLDCM. The emulational results show that the improved BP neural network PID enables the convergence to be faster and the system has strong robustness and self-adaptlve.