Advances of Digital Twins for Predictive Maintenance
Yingchao You, Chong Chen, Fu Hu, Ying Liu, Ze Ji
Abstract
Digital twins (DT), aiming to improve the performance of physical entities by leveraging the virtual replica, have gained significant growth in recent years. Meanwhile, DT technology has been explored in different industrial sectors and on a variety of topics, e.g., predictive maintenance (PdM). In order to understand the state-of-the-art of DT in PdM, this paper focuses on the recent advances of how DT has been deployed in PdM, especially on the challenges faced and the opportunities identified. Based on the relevant research efforts recognised, we classify them into three main branches: 1) the frameworks reported for application, 2) modelling methods, and 3) interaction between the physical entity and virtual replica. We intend to analyse the techniques and applications regarding each category, and the perceived benefits of PdM from the DT paradigm are summarized. Finally, challenges of current research and opportunities for future research are discussed especially concerning the issue of framework standardisation for DT-driven PdM, needs for high-fidelity models, holistic evaluation methods, and the multi-component, multi-level model issue.