A hardware-centered digital twin framework for real-time monitoring of railway vehicle bogies

Railway transportation offers high convenience, environmental sustainability, and strategic importance. Bogies, serving as the core mechanical structure of a railway train, play a vital role in ensuring safe and efficient operations. However, conventional bogie maintenance practices remain constrained by fixed schedules and reactive fault responses, lacking real-time monitoring and predictive capabilities. While previous digital twin (DT) studies have focused primarily on data analytics and system modeling, this study emphasizes a foundational yet often overlooked aspect: the development of sensor hardware and data acquisition systems essential for any DT platform. Leveraging the customized hardware, a rapid deployment DT platform is proposed in this study, integrating various sensors, microcontrollers, wireless communication protocols, and cloud databases to support real-time monitoring of train bogie conditions. A three-layered architecture, comprising the physical, digital, and service layers, is proposed to enable seamless data flow and predictive diagnostics. The designed predictive functionality of the platform is validated through a case study involving vibration-based fault detection of a high-speed train gearbox. All experiments are conducted in a laboratory setting to facilitate data acquisition. This work provides a practical framework for digital twin system implementation and underscores the critical role of hardware development in advancing intelligent railway maintenance solutions.