基于微粒群单神经元的BLDCM调速算法研究

Speed Control Algorithm Study on BLDCM System Based on Single-neuron and PSO

无刷直流电机(BLDCM)是一种多变量、非线性的系统,传统控制算法难以满足系统的控制要求,针对这一现状,提出了一种基于微粒群单神经元的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自适应控制,以微粒群优化算法对单神经元的学习步长进行在线优化,提高了单神经元的自学习、自适应能力。仿真实验表明,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静态特性和鲁棒性。研究结果为分析和设计BLDCM控制策略提供了有效的平台,具有一定的理论和工程实际意义。

Brushless DC motor（BLDCM） is a multivariable and nonlinear system, traditional controller can hardly meet the controlling requirement. Aiming at this problem, an adaptive speed control algorithm based on single neural model and particle swarm optimization （PSO） was proposed. The single neural can adjust weight on-line which makes the adaptive control come true and the step pace of the learning rule of single neural was optimized by PSO, so as to improve the self-study and adaptive ability of single neural. The simulation result proves that the overshoot of the system is small and the speed response is fast with a little fluctuation. The algorithm has better dynamic characteristic and robustness than traditional PID control. The re- sult provides an effective platform of analysissing and designing the control strategy of BLDCM. It has actual meanings both in theory and engineering.