基于蚁群算法PID神经网络的热连轧活套系统自适应解耦控制

带钢热连轧机活套系统是一个耦合的多输入多输出非线性系统。针对活套系统的解耦控制问题，通过对活套系统动态耦合过程的分析，在工作点附近建立控制对象的动态数学模型。利用基于蚁群优化多个单神经元和RBF神经网络相结合的自适应控制策略以减弱系统的耦合影响。最后的仿真结果验证了本方法的有效性，表明解耦后的活套控制系统可获得更好的控制效果。

Adaptive Decoupling Control for Looper System of Hot Strip Mill Based on ACA- PID Neural Network

Looper system in the hot strip finishing mill is a nonlinear system with coupling multiple inputs and multiple outputs. With analyzing the coupling process of looper system,a dynamic mathematical model of the controlled object is given and a decoupling control method is proposed aiming at looper system. The control strategy that multi-PID neutral network based on ACA and RBF is introduced to complete decoupling control of the looper' height and tension system. Simulation results show the effectiveness of this algorithm.