数字孪生工艺模型驱动的加工质量动态控制方法研究

数字孪生技术作为解决智能制造信息物理融合难题，和践行智能制造理念与目标的关键使能技术，正成为智能制造领域前沿的研究热点。加工质量控制技术是机械产品智能制造的关键技术，已成为制造企业“提质、增效、降本”的有效途径之一。经过长期研究，加工质量在误差产生机理、传递规律等相关技术已逐渐被攻克，但其控制方法存在假设性前提和主观因素，已无法满足智能制造过程对加工质量动态控制的要求。.为此，本项目综合采用数字孪生、信息物理融合、知识建模、特征识别等领域相关理论与技术，以快速响应动态演变加工过程的工艺设计为切入点，基于加工过程几何特征与工艺知识的融合机理，探究工艺知识的演变规律和工艺模型的变迁机制；研究在制状态下多源工艺数据的感知与融合方法，揭示加工过程工件实时状态、运行历史、虚拟仿真等数据间映射机制，构建虚实同步迭代的数字孪生工艺模型；研究加工质量与工艺元素的关联机制，提出基于智能算法的加工质量预测方法以探究零件在制状态下加工质量的变化趋势，形成基于工艺元素优化配置的加工质量动态控制方法；最后开发数字孪生工艺模型驱动的加工质量动态控制平台，以船用柴油机复杂零件为对象验证项目关键技术。.通过本项目提出的数字孪生工艺模型建模技术、加工质量仿真方法以及软件平台对加工过程工艺元素的分析、决策与优化，实现加工质量实时监控、主动预测、动态控制的需求，具有良好的理论与技术引领性和原创性特征。项目研究成果将产品加工质量控制从静态转向动态、从局部转向系统、从表面转向本质的根本转变，为形成加工质量动态控制基础理论与方法体系提供支撑，对提高产品质量、缩短产品研制周期、提升企业核心竞争力和推动智能制造发展具有广泛的推广价值和应用前景。

Digital twin technology, as a key enabling technology to solve the fusion problem of information and physical in intelligent manufacturing, and to implement the concepts and goals of intelligent manufacturing, is becoming a hotspot research in the field of intelligent manufacturing. Processing quality control technology is a key technology for intelligent manufacturing of mechanical products, and it has become one of the effective ways to "improve quality, increase efficiency and reduce costs" of manufacturing enterprises. After a long-term research, the technical obstacle of processing quality such as the error generating mechanism, transmission law have been gradually overcome..Firstly, taking the process design that quickly responds to the dynamic evolution in processing as the starting point, this project probes the evolution law of process knowledge and the change mechanism of process models based on the integration mechanism between the geometric characteristics of the processing process and the process knowledge. Secondly, the perception and fusion method of multi-source process data in the in-process state, the mapping mechanism among the real-time state of the workpiece, operation history, virtual simulation and other data, and the construction method of digital twin process model of virtual and real synchronization iteration are studied. Then, the correlation mechanism between processing quality and process elements is researched, and the processing quality prediction method which explore the changing trend of processing quality is proposed based on intelligent algorithm, and then a dynamic control method of processing quality is formed based on the optimal configuration of process elements. Finally, a digital twin process model-driven dynamic processing quality control platform is developed to verify the proposed key technologies with the complex parts of marine diesel engines as objects..According to the modeling technology of digital twin process model, processing quality simulation method and the developed software platform, the processing elements are analyzed, decided and optimized, this can meet the needs of real-time monitoring, active prediction and dynamic control of processing quality, and the proposed method has good theoretical and technological leadership and original characteristics. The research results can transform the processing quality control mode from static to dynamic, from local to system, from surface to essential. And it provides support for the formation of the basic theory and method system of processing quality dynamic control, and has extensive promotion value and application prospects for improving product quality, shortening product development cycles, enhancing enterprise core competitiveness and promoting the development of intelligent manufacturing.