基于模糊径向基函数神经网络的PID算法球磨机控制系统研究  

PID Control System for Ball Mill Based on Fuzzy Radial Basis Function Neural Network

作　　者：程启明[1] 程尹曼[1] 郑勇[2] 汪明媚[1] 

机构地区：[1]上海电力学院电力与自动化学院,上海市杨浦区200090 [2]上海华为技术有限公司,上海市浦东新区200040

出　　处：《中国电机工程学报》2009年第35期22-28,共7页Proceedings of the CSEE

基　　金：上海市教育委员会重点学科建设项目(J51301);上海市教委重点科研项目(06ZZ69)

摘　　要：针对球磨机制粉系统的多变量、强耦合、非线性和时变性等特点,提出应用于球磨机对象控制的基于模糊径向基函数神经网络的PID控制算法。在这种控制系统中,PID控制器的控制参数采用模糊径向基函数神经网络进行自适应整定,系统控制参数采用混合优化算法,即首先采用混沌粒子群优化(particle swarm optimization,PSO)算法进行离线粗调,再采用BP算法进行在线细调,从而快速全局收敛得到最佳的PID控制参数。Matlab仿真结果表明,该控制系统有效地解决了球磨机这种复杂对象的控制问题,该系统控制参数的优化算法收敛快、不易陷入局部极小点,系统控制跟踪快、超调小、解耦好、鲁棒性和适应性强,控制品质优于传统PID解耦控制方法。As to the multi-variable, strong coupling, nonlinear and time-varying characteristics of the ball mill pulverizing system, the PID control system for ball mill based on fuzzy radial basis function （RBF） neural network was proposed. In this control system, the control parameters of PID controller were adaptively adjusted by fuzzy RBF neural network, the control parameters of system were optimized by the hybrid learning methods integrating the offline particle swarm optimization （PSO） algorithm combined with chaos strategies of global searching ability, with the online BP algorithm of local searching ability. The Matlab simulation results show that the control system solve the problems of complex object control like the ball mill effectively, its parameter optimization algorithms can quickly converged and can not be easily fallen to partial minimum, the control method has fast tracking ability, small overshoot, good decoupling ability, strong robust and adaptive ability, and it has better quality than the traditional PID coupling control method.

关 键 词：球磨机 模糊径向基函数神经网络 混合优化算法 早熟判据 PID控制 

分 类 号：TP273.5[自动化与计算机技术—检测技术与自动化装置]