摘 要:考虑系统的参数变化和包括摩擦力在内的非线性及时变的外界干扰情况下,永磁无刷直流电动机位置伺服控制系统是多变量和非线性的时变系统。针对传统PID控制方法的不足,提出了一种TSK型模糊神经网络控制器的设计方法,并用于永磁无刷直流电动机伺服控制系统的位置控制;可同时动态在线进行结构学习和参数学习,以提高位置控制静态精度和动态跟踪性能。仿真结果表明,所设计的TSK型模糊神经网络位置控制器响应速度快、跟踪性能好、输出精度高、动态和静态性优能于传统PID控制方法。 关键词:无刷直流电动机;TSK型递归模糊神经网络;位置控制器;PID控制
Abstract: Position control of a brushless DC motor is a multi-variable and non-linear system because of system parameters change and nonlinear friction and time-varying interference.This paper presented a novel approach of a takagi-sugeno-kang(TSK)type recurrent fuzzy neural network control reference adaptive control for position control of a brushless DC motors in virtue of the disadvantage of conventional PID control.It used TSK-type recurrent fuzzy neural network.The structure and the parameter learning phases are performed concurrently and on line in the proposed servo control systems in order to improve position control and tracking performance.Simulation results indicate that TSK type recurrent fuzzy network is proved the rapid response,the high accuracy of dynamic tracking.The dynamic and static performance are better than traditional PID control. Key Words: Brushless DC motor;TSK type recurrent fuzzy neural network;Position control;PID control
|