数字孪生驱动的制造服务可靠协作方法及演化理论研究

制造服务协作是共享和整合制造资源、能力，以满足个性化复杂任务需求的必然趋势，但随着物理世界制造资源和能力与虚拟世界逐步互联互通及融合，服务协作内涵和本质，也从“虚拟空间封装的静态服务能力协作”发展到为“以高层应用服务为对象，数据、虚实对象耦合交互连接为支撑，预测与迭代优化逼近制造任务需求为目的，在虚实空间开展的各生产要素集成共享与均衡配置的新服务协作模式”，同时伴随着虚实空间内服务是否可信、如何协作、失效后如何维护等难题，统一归纳为基于数字孪生的制造服务可靠协作方法及演化理论研究，通过可信特征提取、模型定义的数字孪生体建模、全局及局部孪生体网络拓扑特征，要素交叉耦合及孪生体共生迭代优化、关键孪生体节点识别、失效原因追溯定位，依次探明可信服务生成机理、可靠协作机理、蓄意攻击下的网络系统失效分析及鲁棒控制方法，为实现全流程/全范围/全要素的制造服务可靠协作提供理论依据。

Manufacturing service collaboration have become an inevitable trend of sharing and integrating manufacturing resources and capabilities to meet personalized and complex task requirements. With gradually interconnection and fusion of manufacturing resources and capabilities between physical and cyber space, the connotation and essence of collaboration have also developed from static collaboration of encapsulated service capability in cyber space to a new mode, in which services is high-level application objective, supported by interactive coupling among data, virtual and physical object, production factors in physical and cyber space are balanced configurated to iteratively forecast and approaching target. However, the reliable service is unknown, the reliability of collaborative service is not high, and maintainability under intentional attack is low. In order to solve above problem, it can be summarized as the methodology and evolutionary theory on digital twin driven manufacturing service reliable collaboration. Through credible feature extraction, model definition of digital twin modeling, global and local network topological analysis of digital twin structure, element cross-coupling, symbiosis iterative optimization, key node recognition, failure reason tracing, location and robust control. It would provide theoretical basis for realizing reliable collaboration of whole process/whole scope/whole elements.