随机非线性切换系统的智能优化控制器设计及性能分析

For the stochastic nonlinear switched systems, this project will research the following objective: First, on the basic of Type-I or Type-II fuzzy systems, we will give new modelling methods which can reflect all dynamic characteristics of the controlled stochastic nonlinear switched systems and have few calculation amount. Then, based on the exsiting methods, such as intelligent optimization control method, stochastic system theorem, Lyapunov stability theorem, et al, we will research new intelligent predictive optimization control methods with simple structure, quick convergence and high accuracy, in order to improve the ability which the controlled system deals with nonlinear and stochastic disturbance, and to reduce effect of environment uncertainty and model mismatch. The main research contents are as follows: 1.For the stochastic nonlinear switched systems with iISS iverse dynamics, we will research the methods of modeling with simple structure which can reflect the whole dynamic of the controlled systems will given, and design new stable intelligent optimization algorithms which has quick rate of convergence； 2.For nonlinear switched systems with stochastic disturbance, we will research and design new modelling methods and stable intelligent optimization algorithms which can stable the controlled systems and has quick rate of convergence； 3.To research and design new modelling mothods and controller for general stochastic nonlinear switched systems； 4. Giving the switched law which can ensure the subsystems and controllers switch steadily each other； 5.Giving the performance analysis of closed-loop subsystems and the whole closed-loop systems.

本项目拟针对随机非线性切换系统，借鉴已有的T-S模糊建模方法，建立能完整反映系统动态特性的系统模型；借鉴已得到广泛应用的智能优化控制方法、自适应方法等，研究和探索收敛速度快、搜索精度高的新型智能预测优化控制方法，以提高控制系统处理非线性、随机干扰等的能力，降低对象和环境不确定性对控制系统带来的不利影响，并逐渐形成一种面向实际控制需要的系统化的建模和优化控制理论体系。具体内容：(1) 首先针对具有iISS逆动态的随机切换系统，研究能完整体现系统动态的建模方法和能给出显式控制律的智能优化控制算法；(2) 针对带随机干扰的非线性切换系统，设计完善的建模方法和实时性强且稳定的智能优化控制算法；(3) 针对一般的随机非线性切换系统进行研究，并对上两步得出的建模方法和智能优化算法进一步完善；(4) 设计在各子系统及各控制律间平稳切换的切换律；(5) 闭环子系统及整个闭环系统的性能分析。

在国家自然科学基金（No.61374004）的资助下，课题组在随机非线性切换系统的智能预测优化控制设计及最优估计方面进行了研究，较好地完成了预期任务。具体内容包括：1）针对不同类型的非线性系统（包括带变量约束的、时滞、随机干扰的非线性系统或切换非线性系统、不对称系统），通过引入幂次函数、Lyapunov函数等方法，研究了智能预测优化控制设计问题，以保证闭环系统的稳定性，得到了一些结果；2）基于Lyapunov-Krasovskii定理及改进加幂积分器、设计初始稳定域等方法，研究了非线性系统的有限时间优化稳定控制问题、全局自适应镇定问题等，得到了一系列结果；3）利用新息重组、状态重构等方法，研究了网络控制模式下非线性系统的最优估计和稳定优化控制问题，得到了一些结果。这些结果有助于推动非线性系统稳定优化控制的进一步发展。.研究成果发表在国内外刊物上共27篇，其中SCI论文13篇，北大中文核心期刊1篇，国内外重要会议论文1篇。主要刊物有：Automatica，Int. Journal of Robust Nonlinear Control，Journal of the Franklin Institute，Int. Journal of Computer Mathematics，Int. Journal of Control Automation and Systems，系统科学与数学等。