模糊小脑模型与PID的复合控制器设计  

Compound Controller Design Based on FCMAC and PID

作　　者：冯超[1] 李兵[1] 

机构地区：[1]河北理工大学计算机与控制学院,河北唐山063009

出　　处：《控制工程》2009年第S3期179-181,共3页Control Engineering of China

基　　金：河北省唐山市科技局科技攻关计划基金资助项目(04360801B)

摘　　要：实际工业控制过程往往具有非线性、时变不确定性、难以建立精确的数学模型,在实际生产现场中,由于受到参数整定方法烦杂的困扰,常规PID控制器参数往往整定不良、性能欠佳,对运行工况的适应性很差。为改善这种情况,提出了模糊小脑模型与传统PID结合的控制方式,能够有效地对未知的非线性模型系统进行实时控制。将CMAC神经网络和模糊控制原理相结合,形成的FCMAC神经网络实现前馈控制,PID实现反馈控制,保证系统稳定性,且抑制扰动。仿真结果表明,该复合控制器具有响应快、超调量小、较强抑制干扰能力等良好特性。The actual industrial control process often has the nonlinearity,the time-variable uncertainty,to establish the precise mathematical model with difficulty,because of the parameter setting puzzles,the PID controller parameter setting often is bad,the performance is often unsatisfactory.In this thesis,a compound controller based on FCMAC and PID can solve the tracking problem of a class of nonlinear systems.In this scheme,FCMAC neural networks exploit a synergism between the CMAC neural networks and the theory of fuzzy logical controller.It is used for feed-forward control,PID is used for feed-back control,to ensure system stability,and restrain disorder.The simulation results show that this compound controller has faster response,smaller overshoots and better cross-coupling rejection etc.

关 键 词：模糊小脑模型 PID控制 非线性系统 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]