Concept for an event-triggered wireless sensor network for vibration-based diagnosis in trams

Abstract

Trams are the most durable and resource efficient forms of public transportation. However, because of the varying wear in dependence on their operation mode and load levels, there is a need for condition monitoring and early maintenance. Vibration sensors provide interesting possibilities to monitor the relevant tram drive components. In this contribution we investigate their use under real conditions. On the basis of cable bound vibration measurements, the influence of the crossed track section, the tram speed and the tram condition is shown. Based on the investigation results, a concept is proposed in which a meshed and wireless sensor network, event-triggered, can acquire vibration measurement data, which are suitable for the diagnosis of tram drive components. The proposed concept has the potential to operate the sensor nodes in an energy efficient way through decentralized data evaluation taking place on the sensor node.