Towards discrete manufacturing workshop-oriented digital twin model: Modeling, verification and evolution

Abstract

Digital twin (DT) is one of key enabling technologies of intelligent transformation in discrete manufacturing workshop (DMW). The construction of digital twin system in DMW has greatly expanded the connotation and extension of smart manufacturing. Although previous studies have shown the success in DT, there is still a lack of clear and systematic methods of DT in DMW. To bridge this gap, this article focuses on how to systematically and effectively construct DT model methods for DMW. Taking the modeling, verification, and evolution of DT model as the main line, three key technologies related to digital twin in DMW are proposed, including model migration based matching modeling technology, graph convolutional network and temporal convolutional network based DT model verification and Adaboost based synchronous evolution of DT model, which will provide a systematic theory and method for the construction of DT model for DMW. The experiment demonstrates that the proposed methods have good performance for physical workshop in industrial environment and provide a holistic understanding of DT model modeling in DMW.