Research On Process Planning Method Of Production Line Based On Digital Twin And Knowledge Graph

In the face of the growing market demand of small batch and customized machining production,the process planning efficiency of manufacturing enterprises has gradually become an important factor affecting their response speed.In the process planning,on the one hand,because the production line resources are in dynamic change,it is often difficult for the process designer to fully understand the situation of the production line resources in real time,and the subsequent process changes waste a lot of time and cost.On the other hand,due to the diverse forms and complex composition of process knowledge,the collection,management and decision-making of process knowledge cost a lot of manpower,which brings obstacles for manufacturing enterprises to improve the efficiency of production line process planning.In recent years,the emergence of digital twin technology and knowledge graph technology has brought new ideas to solve the above problems.Digital twin technology provides a means for production line system modeling virtual-real consistency and virtual-real dynamic mapping,and becomes a constraint data source for production line process planning.Knowledge graph technology can collect and manage heterogeneous and dispersed process knowledge,which provides a tool for efficient gathering,storage and decision-making of process knowledge.Based on digital twin technology and knowledge graph technology,a process planning method of production line based on digital twin and knowledge graph is proposed.The main research contents are as follows:(1)In view of the complex problems of sensing,updating and managing the time-varying complex data of the production line resources and environment,the construction method of the"resource-environment" digital twin model of the production line was designed,and the digital twin models of the device,tool,fixture,measure,workblank,temperature and humidity were constructed according to the criteria of multi-level,perceptability,renewability and reuse.Realize real-time perception and mapping of production line resources and environment,and provide constraint data source for process planning.(2)Aiming at the problem of heterogeneous,dispersed and difficult to manage process knowledge,the construction of "task-process" twin knowledge base of production line was completed by using knowledge graph technology.By sorting out the machining task and process knowledge contained in the process specification file,the pattern layer and data layer of the machining task and process are constructed.Finally,Neo4j graph database is used to construct,store and access the twin knowledge base.(3)In order to solve the problem of automatic review and decision of production line process planning,firstly,a machining task identification method based on twin knowledge base was designed to identify new processing tasks,and similar historical processing technologies were recommended to realize the reuse of process knowledge.Then,a process constraint review method based on the"resource-environment" digital twin model is designed,and the decision tree algorithm is used to automatically examine whether the process matches the production line resources,which reduces the dependence of the previous process planning on personnel and improves the efficiency of the enterprise process planning.In order to verify the above methods,a "resource-environment" digital twin modeling system was designed for production line managers to construct digital twin models of production lines.A"task-process" twin knowledge base system was designed for process designers to manage and decide process knowledge.Finally,for the piston production line of a company,the construction of the "resource-environment" digital twin model of the piston production line and the twin knowledge base of the piston machining process were completed,and the whole working flow of the process planning was verified,which proved the feasibility of the proposed method and provided a solution for manufacturing enterprises to implement intelligent process planning.