一种混沌优化的双模糊控制器———倒立摆系统的设计

来源期刊：中南大学学报(自然科学版)2006年第3期

论文作者：邹恩 刘俊萍 李祥飞 张泰山

文章页码：567 - 571

关键词：模糊控制器;倒立摆;混沌;参数最优化

Key words：fuzzy controller; inverted pendulum; chaos; optimal parameters

摘    要：将单级倒立摆的4维输出分解为2个2维模糊控制器的输入量,与倒立摆组成双闭环控制,内环调节摆杆的角度,外环控制小车的位移。采用混沌算法优化控制器的参数,首先将混沌因子引入模糊控制器参数域的优化搜索中并在全局范围内直接寻优,当获得全局近似最优解后,再缩小寻优区间,在近似最优解的附近继续寻优。对倒立摆系统在不同情况下进行仿真,结果表明:该方法能提高搜索效率,能较快搜索到全局最优解,为解决多输入快速系统的模糊控制器优化设计提供了一种较好的实现方法。

Abstract: The 4D output of inverted pendulum system was directly decomposed into the inputs of two of 2D fuzzy controllers (FC). The double-loop controlled system was composed of the inverted pendulum and two fuzzy controllers. The inner-loop regulated the angle of pendulum, while the outer-loop was used to control the displacement of the cart. The parameters of FC were optimized by chaos method. Firstly, chaos factors were applied to searching for parameters of FC and direct optimization within global range. Then the searching space of variables was continually reduced when approximate global optimal solutions were obtained. The simulation results of inverted pendulum system in different states show that the method can search quickly global optimal values and solve effectivly the optimization design of FC in multi-input and speediness system.