基于微粒群算法的无刷直流电机单神经元自适应控制  

Particle Swarm Optimization Based Single Neuron Adaptive Control for Brushless DC Motor

作　　者：代睿 曹龙汉[2] 何俊强[2] 唐超[2] 刘小丽[2] 

机构地区：[1]海军驻兰州地区军事代表室,兰州730070 [2]重庆通信学院控制工程重点实验室,重庆400035

出　　处：《电工技术学报》2011年第4期57-63,70,共8页Transactions of China Electrotechnical Society

基　　金：科技部国际科技合作(2007DFR10420);重庆理工大学汽车零部件制造及检测技术教育部重点实验室开放基金(2009-10)资助项目

摘　　要：为提高无刷直流电机速度控制性能,提出一种基于微粒群优化算法的单神经元自适应速度控制算法,该算法利用单神经元在线调整连接权值的能力,实现无刷直流电机速度的自适应控制。针对传统单神经元权值调整规则容易陷入局部最优等不足,利用微粒群优化算法良好的全局和局部寻优能力对单神经元连接权值进行在线调整,提高了单神经元的自学习、自适应能力。Matlab仿真和实验结果表明,系统超调量小、转速响应快、转速波动小,比传统PID速度控制具有更好的动态特性和鲁棒性。A single neuron adaptive speed control algorithm based on particle swarm optimization（PSO） is proposed in order to improve the speed control performance of brushless DC motor（BLDCM）.The ability of the single neuron which can adjust conjunction weights on-line is used to make adaptive speed control.Aimed at the disadvantage that the weight-adjusting rule of the single neuron is prone to be trapped in local optimum,it was proposed to use PSO which has good ability to make both global and local optimization to adjust the weights of single neuron on-line.As a result,the algorithm improves the self-learning and adaptive ability of the single neuron.Matlab simulation and experimental results indicate that under the proposed algorithm the overshoot of the system is small and the speed response is fast with a little fluctuation.The proposed algorithm has better dynamic characteristic and robustness than traditional PID control.

关 键 词：无刷直流电机 微粒群算法 单神经元 自适应控制 

分 类 号：TM383[电气工程—电机]