The Digital Twin (DT) model within a Digital Twin System (DTS) serves as a real-time digital representation of its corresponding physical entity. It is a dynamic, interconnected model that enables real-time optimization in its application environment, allowing for the simulation, monitoring, evaluation, and control of the physical counterpart’s state and behavior while facilitating data-driven decision-making. In engineering practice, most scholars focus on data visualization and twin system construction. However, a complete digital twin system not only requires numerical representation of the real-time state of the physical entity but also sometimes requires real-time mechanical behavior analysis of the physical entity. Thus, a robust mechanical analysis module becomes essential within the DTS framework. Integrating a general-purpose mechanical analysis platform into the DTS offers an effective solution, thereby necessitating the development of novel fusion techniques for multi-source heterogeneous data. This study takes the integration of the Midas Civil mechanical analysis platform with a digital twin system as an example. By utilizing the API provided by Midas Civil, we develop a synchronization technique for virtual-physical systems, capable of handling and modeling multi-source heterogeneous data. This enables real-time mechanical computation and analysis within the DTS, facilitating the dynamic updating and aggregation of both simulation data from mechanical analysis and monitoring data from the physical entity. Consequently, the digital twin system can predict mechanical behaviors in the virtual domain, providing a more accurate representation of the real-world physical system’s state and dynamics.

Real-time mechanical analysis within digital twin (DT) systems requires high-fidelity models that synchronize with the “as-built” state of physical structures. This paper proposes a technical framework for constructing a “mechanical-core” DT by integrating computer vision (CV) sensing with automated finite element model (FEM) updating. Utilizing the Midas API, we developed a platform that automates data acquisition, modeling, and parameter inversion. A momentum-based optimization algorithm is implemented to invert the instantaneous elastic modulus of bridge segments during cantilever construction. The system was validated through a case study of a continuous box-girder bridge. Quantitative results indicate that the initial theoretical model, based on design-phase assumptions, exhibited a mean relative error of approximately 21.9% in vertical displacement. Following the real-time parameter inversion, this error was significantly reduced to less than 0.2% across all monitored construction stages. The rapid convergence (typically within three iterations) and the substantial increase in predictive accuracy demonstrate that the proposed framework effectively bridges the gap between raw sensing data and structural analysis, providing a reliable basis for proactive engineering decision-making.