Research on a Data-Driven and Lightweight Structure Monitoring and Early Warning System for Integrated Subway Hubs

With the urban travel increasing, there is a growing demand for subway systems. This has led to the emergence of deep and complex integrated subway hubs with multiple line transfers in some mega cities. These hubs pose safety risks related to the building of deep foundation pits not just during the construction of new lines, but also due to the effects on existing operating lines. Given the significant economic losses and potential casualties caused by accidents, it is necessary to implement effective monitoring and early warning management. In this study, a data-driven approach has been developed to create a lightweight structural monitoring and early warning system for integrated subway hub construction and operational phases. By integrating various data sources, such as three-dimensional model data and structural monitoring data, and using data-driven methods to extract data features, the optimized Whale Optimization Algorithm-Long Short-Term Memory (WOA-LSTM) model is utilized to produce prediction results. Additionally, the PKPM-FPS + cloud platform has been employed for further development to provide a lightweight and visual solution, enabling real-time structural monitoring and effective early warning for both ongoing construction segments and existing operational lines in integrated subway hubs. Lastly, the study utilizes a case analysis approach to validate the effectiveness and feasibility of the proposed system, offering valuable insights and practical recommendations for a robust, integrated structural monitoring and early warning management system to prevent catastrophic incidents in complex subway hub.