Maintenance of bogie components through vibration inspection with intelligent wireless sensors: A case study on axle-boxes and wheel-sets using the empirical mode decomposition technique

Abstract

The maintenance of bogie components, a critical aspect of railway maintenance, is difficult due to the confined underframe space. This makes it difficult to install traditional monitoring equipment, resulting in a labour-intensive process. Thus, a lot of time has to be expended to conduct these tests, which makes the process both tedious and expensive. Moreover, this approach is somewhat inadequate, since the tests can only be conducted at the depot and thus only when the trains are out of service. We have developed and deployed a non-intrusive solution based on a small wireless sensor network that can be easily installed on the different parts of the bogie and along the whole train. We have worked out a technique to discriminate between the various sources of vibration and can thus monitor the state of several components using only a few sensors. In this paper, we present a case study on how to maintain an axle-box and a wheel-set by attaching a single intelligent sensor to the bogie frame or the bearing cover and using the empirical mode decomposition technique to analyse the generated data. In light of the promising results obtained in this study, we suggest that the proposed approach can lead to a value-added predictive maintenance strategy as long as the test conditions are kept under control. However, we do highlight that the generalization of the approach relies on the flexibility of the system to adapt to new environments and operational scenarios.