非线性系统的多模型鲁棒控制器设计

Multi-model robust control of nonlinear systems

摘要:

在互质分解框架下,定义了闭环系统中算子的图及其鲁棒边界的概念,并给出了反馈系统稳定的充要条件.通过对非线性系统中多个平衡点的线性化模型进行最优鲁棒控制器设计,使之具有最大的广义稳定裕度.将各线性闭环系统的鲁棒稳定边界用于对子控制器输出信号的加权,从而得到全局控制器的输出.仿真实验表明,该方法不仅可以较好地跟踪设定值,而且具有较强的抗干扰能力.

Abstract:

The graph of operator in closed-loop system and the robust stability margin are defined in the frame of the coprime factorization,and the sufficient and necessary conditions for feedback system are given by the graph of the plant and the inverse graph of the controller.A new multi-linear model approach is used to control a strongly nonlinear system,and reasonable operation points for nonlinear system are selected to design the corresponding optimal robust compensator with the largest generalized stability margin.Then the global robust controller is built from a weighted combination of local controllers whose outputs are weighted by the functions of generalized robust margin.The designed control system has good performance and robust stability for nonlinear system.