面向交互动态服务协作计算的信息物理融合系统关键技术研究

The future computer intelligent control systems will require more interaction between cyber & physical world, posses stronger ability for dynamic self-adaptive service, and persue closer collaboration among entities in the systems. Cyber-Physical Systems(CPS) are integrations of computing, communication and control, which aim at operating in a interactive and collaborative way, with joint cooperation between cyber & physical subsystems..This project research on an interactive dynamic service oriented & collaborative computing based Cyber-Physical Systems. In this research project, interactive time synchronization based collaborative measurement methods are proposed to solve real-time information sensing problem;To deal with magnanimity information collected by the systems, a multi-model based self-adaptive distributed progressive heterogeneous data fusion measure is introduced; Taking advantages of collaboration among physical entities and the characteristics of fused system dynamic information, research is carried out to predict the variation tendency of the systems; Towards interactive collaboratve service CPS infrastructure, an intelligent learning model a Petri net based cyber-physical managing archetecture will be developed; To test the performance of the proposed theoretical methods, an electricty smart managing testbed will be established..The main research work of this project is carried out under CPS infrastructure, and mainly focus on interactive theory & mechannism, as well as collaborative computing based self-adaptive dynamic servicing methods. The theoretical achievements will be introduced to an electricity smart managing testbed. Thus, the work of this project will creat great research significance & value.

未来计算机智能控制系统的信息物理更加交互、动态自适应服务能力更强、内部实体协作更加紧密。信息物理融合系统（CPS）是一种集成计算、通信和控制，并将信息与物理进行紧密结合的交互协作控制系统。研究基于CPS的信息交互、动态服务和系统协作方法将进一步推动计算机智能控制系统的发展.本项目研究面向交互动态服务协作计算的CPS关键技术。针对实时信息感知问题，研究CPS交互同步协作测量方法；面向系统实时测量海量信息，研究多模型自适应分布渐进式异构信息特征融合方法；为了解决系统动态预测问题，研究基于物理实体协作的信息物理协作互动预测方法；针对面向服务的智能控制，研究基于改进Petri网的CPS控制模型；建立基于智能用电的交互协作信息物理融合系统验证平台。.本项目基于CPS框架，研究信息物理交互理论机制、基于协作计算的自适应动态服务理论方法，并将理论成果应用于智能用电验证平台，具有重要的理论和实用价值。

本项目面向交互式信息物理融合系统的安全稳定需求，研究了分布式交互动态服务协作计算方法。研究了信息物理融合系统的交互同步协作测量方法和基于信息物理交互协作的分布渐进式信息融合方法，分析了面向交互协作服务的信息物理融合系统能效优化学习模型，提出了信息物理协作互动预测方法，建立了基于智能用电的信息物理协同网络分布式测量控制软硬件测试平台。提高了信息物理融合系统交互动态服务的安全性、智能性，实现了交互式动态信息物理融合系统在节能、自愈控制和长期安全运行的目标，具有重要的研究意义和应用价值。