用FCGA优化的四辊冷轧机恒张力模糊控制系统

Intelligence Control Based on RBF Neural Network in Nonlinear System

为了保证冷轧机轧制中带钢恒张这一特点,设计了四辊冷轧机恒张力模糊控制系统。该系统含有电压、电流和速度三个内环,最外环为张力环;电压环和电流环采用一维模糊控制器控制,速度环和张力环采用二维模糊控制器控制。为了加强模糊控制的自适应性,采用了一种模糊控制的遗传算法将外张力模糊控制器的隶属参数进行优化。理论分析和仿真结果都表明该系统对带钢恒张控制具有很强的鲁棒性和实时性,即使在变工况下(大范围变负荷下)也保持了良好的控制性能。

A constant tension fuzzy control system is designed for four-roller cold rolling mill to ensure that strap steel tension is constant. This system contains inside three loops as voltage loop ,current loop and velocity loop, the outside loop is tension one.One-dimension fuzzy controllers are used in the voltage loop and the current loop,two-dimensions fuzzy controllers are used in the velocity loop and the tension loop. The tension fuzzy controller＇s membership function is optimized by a kind of fuzzy control genetic algorithm in order to enhance adaptability of the tension fuzzy controller. Theoretical analysis and simulation results show that the controlling system has stronger robustness and high-speed for constant tension control,and has better control efficiency even under bigger vary loads.