网络环境下的迭代学习控制及应用研究

项目名称： 网络环境下的迭代学习控制及应用研究

项目编号： No.61203065

项目类型： 青年科学基金项目

立项/批准年度： 2013

项目学科： 自动化学科

项目作者： 卜旭辉

作者单位： 河南理工大学

项目金额： 25万元

中文摘要： 迭代学习控制的研究结果主要针对点对点方式的控制系统给出的。实际系统中，控制系统的结构和运行模式正在发生较大的变化，网络技术的广泛应用使得网络控制方式在多个领域中逐步替代点对点的控制方式。本项目研究网络环境下迭代学习控制方法的分析和设计问题，主要研究内容包括：(1) 针对通信网络存在数据传输延时、数据包丢失以及信道传输随机迭代相关扰动的问题，研究迭代学习控制的理论收敛性分析方法，并分析上述因素对迭代学习控制系统的影响。(2) 研究网络环境下鲁棒迭代学习控制器的设计方法、数据传输延时和数据包丢失的补偿方法以及迭代相关扰动的观测和滤波方法，以降低网络基本参量约束对迭代学习控制系统的影响，提高控制方法的鲁棒性。(3) 通过基于网络环境实现的城市快速路交通系统入口匝道控制，验证本项目中的理论分析结果和鲁棒控制器设计方法。本项目的研究既是理论研究课题，同时也是应用问题，具有重要的学术和应用价值。

中文关键词： 迭代学习控制；网络控制系统；数据包丢失；二维系统理论；随机稳定

英文摘要： The existing contributions of iterative learning control are devoted to focusing mainly on traditional centralized point-to-point control systems. However, in many practical control systems, the configuration and operation mode of control systems are changing due to the development of network technology. Control loops that are closed over a communication network have become more and more common, which results in point-to-point control systems replaced by networked control systems in many control fields. This work considers the network-based iterative learning control and applications, and the main contents include as follows: (1) The stabilization problem of network-based iterative learning control system are studied where the focus are networked-induce delay, packet dropout and communication channel iteration varying disturbance issues. The influences of these factors for ILC systems are also analyzed. (2) The design of robust ILC controller, compensation algorithms for networked-induce delays and packet dropouts, and observers for disturbances are studied. The improved ILC schemes can enhance the robustness against network induced issues. (3) The results of stability analysis and controller design are verified through the urban express traffic iterative learning control in network-based environment with ramp m

英文关键词： iterative learning control；networked control systems；data dropouts；2-D system thoery；stochastic stablility