摘要
网络化分布式系统由多个相互耦合的分布式子系统共同作用下的网络化系统,为达到整体系统行为的一致性,通过构建信息网络实现子系统动态行为的协调与优化,称为CPS(cyber-physical system)系统的优化.传统的集中式优化方法难以实现高效性、实时性的问题求解,研究分布式优化与决策的理论方法具有重要的理论意义,在当今以智能电网、传感器网络等为代表的网络化分布式系统中有着广泛的应用需求.针对网络化环境下复杂系统分布式优化与决策,通过研究子系统间的信息交互机制与物理耦合关联,形成子系统动态耦合约束下的目标优化、多任务冲突情况下的子系统共识决策与拓扑结构变化下的自适应协同问题,通过调整动态耦合约束下各子系统优化问题的可行域,分析并预测子系统动态的行为,调控系统在不同拓扑结构下的多模态切换,发展一类适用于网络化系统的分布式实时优化与决策理论,以提高整体系统的全局性能.本文结合专刊所取得的成果对此进行讨论.
Networked CPS systems are a class of complicated physical systems with inter-connected and coupled sub-systems. By exchanging information via networks,each sub-system can coordinate and optimize their behaviors and make local decisions in order to achieve global tasks. For this class of systems,it is difficult to solve the optimization problems efficiently and adaptively using the classical centralized approaches. Therefore,it is necessary and important to develop novel distributed optimization methods for networked systems. In particular,there is a great demand for such distributed methods in modern distributed networked systems,including smart-grids and sensor networks. In this project,we tackle the difficulties in real-time optimization and decision making of networked systems. By investigating the information exchange mechanisms and the physical couplings between each sub-system,three key problems are formulated： the constrained optimization problem when dynamical coupling,the decision making problem for consensus under multi-task conflict and the adaptive cooperation problem under network topology changes. By adjusting the feasible region of the dynamically-coupled constrained optimization problem,analyzing and predicting the dynamic behavior of each sub-system,and switching the operation mode of the system,we develop a comprehensive framework for realtime optimization and decision making of networked systems. Much progress in the CPS optimization and control has been made,in this paper,some further problems are discussed based on the results of this special issue.
作者
李少远
夏元清
程鹏
LI Shaoyuan;XIA Yuanqing;CHENG Peng(School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China;School of Automation, Beijing Institute of Technology, Beijing 100091, China;College of Control Science and Engineering, Zhejiang University, Hangzhou 310058, China)
出处
《信息与控制》
CSCD
北大核心
2018年第1期1-4,共4页
Information and Control
基金
国家自然科学基金资助项目(61590920)
关键词
信息物理融合系统
分布式实时优化
协同策略
自适应优化
cyber- physical systems（CPS）
distributed real-timeoptimization
coordination strategy
adaptive optimization
