约束不确定非线性系统的鲁棒优化镇定.PDFVIP

第 30卷第 5期 2013年 5月 控 制 理 论 与 应 用 Control Theory Applications Vol. 30 No. 5 May 2013 约束不确定非线性系统的鲁棒优化镇定 DOI: 10.7641/CTA.2013.20524 何德峰†, 宋秀兰, 俞 立 (浙江工业大学信息工程学院,浙江杭州 310023) 摘要:针对状态和输入约束不确定非线性仿射系统,提出一种鲁棒镇定的优化控制器设计方法. 基于弱鲁棒控制 Lyapunov函数概念,构造一个参数可调控制器. 再利用LaSalle定理和逆优化理论,验证该控制器的鲁棒镇定性和逆 最优性. 进一步,采用滚动优化原理在线计算控制器的可调参数,实现闭环系统的鲁棒优化镇定. 最后对一个开环 不稳定振荡器系统进行鲁棒优化镇定,其结果验证了文中方法的有效性. 关键词: 非线性仿射系统;约束控制;鲁棒镇定;鲁棒控制Lyapunov函数;滚动优化 中图分类号: TP273 文献标识码: A Optimization-based robust stabilization of constrained nonlinear uncertain systems HE De-feng†, SONG Xiu-lan, YU Li (College of Information Engineering, Zhejiang University of Technology, Hangzhou Zhejiang 310023, China) Abstract: The design method of optimization-based controllers with robust stabilization is presented for nonlinear affine systems subject to uncertainties and constraints on the state and input. A controller with adjustable parameters is constructed based on the concept of weak robust control Lyapunov functions. Then the LaSalle’s theorem and inverse optimization theory are employed to validate the robust stabilization and inverse optimization for the controller with respect to the uncertainty. Furthermore, the receding optimization principle is exploited to compute online the adjustable parameters in the controller, which leads to the optimization-based robust stabilization of the closed-loop system. The robust stabilization method has been applied to an open-loop unstable oscillator system and the successful results validates the effectiveness of the method presented here. Key words: nonlinear affine systems; constrained control; robust stabilization; robust control Lyapunov functions; receding optimization 1 引言(Introduction) 许多工业控制系统面临同步优化与镇定问题.对 于线性系统, LQ控制、µ-综合和H∞控制等理论提供 了良好的优化镇定方法[1–2],但如烯烃聚合过程、换 热过程和空分过程等复杂工业过程大都具有强非线 性,线性化控制方法很难取得好的控制效果,甚至稳 定性都难以保证[3–4]. 同时,实际控制系统通常存在 一定约束限制和不确定性影响.因此,约束不确定非 线性系统的同步优化与镇定问题成为了近年来国内 外先进控制理论研究的热点课题[5–6]. 对于标称非线性系统,可利用控制Lyapunov函数 (CLF)工具构造一镇定控制器,该控制器对特定的性 能指标具有最优性[7–9]. 进一步,为降低CLF设计难