Innovative sensor-node hardware architecture for on-board heavy haul wagon monitoring

Onboard real-time condition monitoring of wagons, as distinct from measuring track irregularities using wagons, is still not well developed for heavy haul applications. Track-side equipment detects a wide variety of component and vehicle faults, but is not able to diagnose the train at all times during the trip. Every year, a significant amount of resources are consumed on repairs and downtime caused by faults that could be detected early by equipment mounted on the vehicles. Furthermore, the absence of real-time information of the condition of each wagon leaves the train system at risk to occasional and catastrophic failures and derailments. A comprehensive study has been conducted to establish the current state of the art of on-board condition monitoring systems for unpowered vehicles. Advances in sensors, energy harvesting, and condition monitoring techniques were studied, to explore the possibility of installing health monitoring systems onboard each vehicle of the train. Several applications with the potential to be used in the heavy haul railway industry were found. Application cost, robustness and power supply were identified as still the main challenges to implement real-time condition monitoring using onboard sensors. A positive cost-benefit ratio will enable fleetwide onboard monitoring applications. Hence, the possibility exists for a real-time prognosis and advanced data analytics that enhances rail transport performance levels, and give real-time warnings of severe and fast developing vehicle faults. An innovative hardware architecture that enables low-power and low-cost sensornodes for real-time heavy haul wagon monitoring applications is presented.