| The basic conditions required for the application of digital twin have been met through the rapid and high-level development of Io T sensing,data transmission,modeling and simulation method and also big data,cloud computing,artificial intelligence and other technology,which makes digital twin the center of information technology and the focus of research and development in major countries,developed cities,large enterprises and researchers in related areas all over the world.Meanwhile,the continuous advancement of the new round of industrial revolution makes industrial intelligent equipment such as welding robots occupy an increasingly important position in industrial production and the national economy,thus there will be many benefits of conducting digital twin research on welding robots.At the aspect of technological development,it will help drive the further development of relevant basic support technologies,so that the entity will have an advantage in future scientific and technological research;at the aspect of application effects,it will help to achieve test before real welding,remote monitoring,fault diagnosis,operation prediction and auxiliary decision-making in process and production and other aspects;besides,it can also help to reduce the harm of welding to the human body by reducing the necessity of people appearing on the production site.This paper mainly studies the construction and evolution of digital twin model for welding robots.Firstly,digital twin related theories are analyzed,and the research framework of digital twin for welding robots is designed.According to relevant work,four basic characteristics of digital twin are refined,five basic supporting technologies required for digital twin research are analyzed,and the functions of digital twin are clarified,and then the concepts of digital twin and its similar concepts are distinguished.On this basis,the designed digital twin research framework for welding robots is proposed,which lays the foundation for the development of the following research.Secondly,the collection,transmission and storage of the data required for the running and evolution of the welding robots digital twin are studied.The data types needed for digital twin are analyzed as well as the requirements for transmission and storage of different data.For the data required for running,a high-speed data service framework based on Redis and My SQL is designed and implemented,which can realize "one-time collection,multiple distribution,and reuse available".For the data required for evolution,orthogonal experiment is designed to obtain representative welding parameters and welding quality data.The data required for the operation and evolution of the digital twin of the welding robots is prepared through these work.Then,the digital twin model of welding robots is built in a digital twin platform.Based on the analysis of the modeling requirements of digital twin for welding robot,Blender is accepted as the digital twin platform,in which a high fidelity 3D model of welding robots is constructed.Through the design and implementation of digital twin data-driven engine,a bridge from external data to twin model is built,which makes the 3D model evolve into digital twin model by data driving have the ability to describe physical devices in real time.Finally,the evolution of digital twin of welding robots is studied.The concepts of pre-evolution and runtime evolution of digital twin are proposed.The preevolution enables the digital twin to predict welding quality according to given welding parameters,and quickly deduce the optimal parameters by repeatedly adjusting and prediction,which makes the digital twin get the ability of prediction and assistant decision-making;whereas runtime evolution enables digital twin to verify the running state of welding robots from two levels namely micro validity and macro correctness,so that digital twin has the ability of state monitoring and operation diagnosis,and eventually forms the digital twin of welding robots that meets the definition requirements. |