| dc.description.abstract | In this study, we improve the digital twin model construction method and propose a methodology to solve the consistency problem of virtual model simulation. In this approach, we use parameter fitting, quantitative analysis and error comparison, combined with simulation metrics to develop digital twin technology. This methodology can give the manufacturing industry a suggestion for the development of digital twin, so that when facing digital twin consistency problems in industrial manufacturing in the future, they can refer to the suggestions in this study to troubleshoot the problems. In the research process, we first constructed a physical entity layer, and then constructed a digital twin model, including a virtual model, a data model, and a knowledge model. The virtual model was constructed using commercial digital simulation software, and the results of the data model tuning parameters were compared with those of the knowledge model for quantitative analysis, error and simulation metrics, in order to observe the work-time errors during the motion processing between the simulation software and the physical robotic arm. Finally, a case study and discussion are conducted to demonstrate the feasibility and effectiveness of this research methodology and to provide practical references for future applications. | en_US |