Monitoring Vertical Acceleration of Railway Wagon using Machine Learning Technique

Wireless communications and modern machine learning techniques have jointly been applied in the recent development of vehicle health monitoring (VHM) systems. The performance of rail vehicles running on railway tracks is governed by the dynamic behaviors of railway bogies especially in the cases of lateral instability and track irregularities. In this study we have proposed a system to monitor the vertical displacements of railway wagons attached to a moving locomotive. The system uses a classical linear regression machine learning technique with real wagon body acceleration data to predict vertical displacements of vehicle body motion. The system is then able to generate precautionary signals and system status which can be used by the locomotive driver for necessary actions. This VHM system provides forward-looking decisions on track maintenance that can reduce maintenance costs and inspection requirements of railway systems.