基于IGA的板形板厚神经网络分散解耦PID控制  

Strip Flatness and Gauge Neural Network Decentralized-decoupling PID Control Based on IGA

作　　者：王莉[1] 王学秀[2] 马云[3] 

机构地区：[1]北京科技大学信息工程学院,北京100083 [2]新疆工业高等专科学校电子信息系,新疆830091 [3]包头钢铁学院,内蒙古包头014010

出　　处：《计算机仿真》2003年第12期82-85,57,共5页Computer Simulation

摘　　要：针对板带材轧制是一个复杂的非线性过程 ,板形控制 (AFC)和板厚控制 (AGC)又是相互耦合的一个综合系统等特点 ,该文首先采用神经网络分散解耦方法 ,对此板形板厚多变量耦合系统进行解耦 ,而后再应用基于免疫遗传算法的PID控制对解耦后的已近似成为两个独立的SISO系统的广义对象进行控制 ,从而建立了基于免疫遗传算法的板形板厚神经网络分散解耦PID控制系统。仿真结果证明了此AFC -AGC控制系统具有良好的自适应跟随和抗扰性能 ,其控制效果优于传统的解耦PID控制。In the paper, because strip rolling is a very complicated nonlinear process, and characterized by couple between automatic flatness control (AFC) and automatic gauge control(AGC), the authors firstly make this coupling multivariable system decouple using neural network decentralized-decoupling method. Secondly, authors adopt PID controller based on immune genetic algorithm to control the decoupled generalized object which can be approximately equal to two SISO systems. Thus, construct strip flatness and gauge neural network decentralized-decoupling PID control system based on immune genetic algorithm and simulate on it. The simulation results show that this kind of new controller has good performances of adaptively tracking target and resisting disturbances and is superior to the conventional decoupled PID control in terms of improving the strip flatness and gauge accuracy.

关 键 词：板形 板厚 分散解耦PID控制 神经网络 IGA 板材轧制 

分 类 号：TG334.9[金属学及工艺—金属压力加工] TP273[自动化与计算机技术—检测技术与自动化装置]