Mission Profiling of Railway Bridges using Low-Power Smart MEMS Sensors

Structural health monitoring of railway infrastructure, particularly bridges, is essential for ensuring long-term safety, reliability, and cost-effective maintenance. Traditional monitoring systems often rely on wired or high-power sensors, which limits large-scale deployment and long-term operation. Recent advancements in low-power micro-electro-mechanical systems (MEMS) sensors enable a new generation of smart, autonomous monitoring solutions tailored for mission profiling of railway bridges. By capturing acceleration data over extended periods, these sensors can provide insights into the cumulative loading patterns and degradation processes of bridges. Smart MEMS-based sensing networks enable energy-efficient data acquisition for improved prediction of the remaining useful lifetime of railway infrastructures, helping to reduce the need for frequent maintenance and manual inspections. Typically, for infrastructure monitoring at remote locations, external power supply is often unavailable and low power communication is required. Complex time series data must be evaluated and compressed locally on the sensor or edge device to reduce overall energy consumption and data traffic. This paper analyzes field data from railway bridges in the Czech Republic and demonstrates how key parameters like train passage, speed, length and number of wagons and axles can be extracted from acceleration data using signal processing methods. In addition, a low-power MEMS sensor system is investigated and evaluated by means of in-lab reproduction of field data. Suitable algorithms are developed with respect to the limited resources of such devices. Overall, this approach supports predictive maintenance strategies and enhances the understanding of load-induced structural behavior, contributing to safer and more sustainable railway infrastructure management.