Digital twin technology for continuously welded turnout on high-speed railway bridges based on improved MOPSO algorithm

The high-speed railway (HSR) in China has entered the stage of large-scale operation, and there is an urgent need for the intelligent and refined management of track structures. However, due to operating requirements and railroad clearance of HSR, it is difficult to fully grasp the service status of track structures in real time. The digital twin theory, which emphasizes the interconnection of physical and virtual worlds, provides a novel perspective for managing track structures. Therefore, this study proposes using digital twin technology to evaluate the status of track structures, specifically focusing on the continuously welded turnout (CWT) on HSR bridges. Via the refined finite element method (FEM), a virtual entity (VE) model is established to mirror the behavior and status of the physical entity (PE), a continuously monitored CWT on the Beijing-Shanghai High-speed Railway. To ensure an efficient and accurate mapping relationship between VE and PE, an improved MOPSO algorithm based on the surrogate model technique and global ranking criterion has been developed. The proposed method is applied to evaluate the performance evolution and damage status of the CWT. The results demonstrate good consistency with insitu manual inspections, suggesting that this method can accurately locate local defects in track structures, assisting in efficiently managing HSR.