A Novel Method in Light-Rail Condition Monitoring Using Smartphones

Abstract

Tram and track interaction inevitably generates undesirable vibration, and the magnitude of this vibration may depend on various factors, including track quality, rail type, rail profile, tram type, ride quality, and speed. To investigate the impact of track quality on vibration, all other factors should be eliminated in experiments. To remove the effects of ride quality, tram type, rail type, and rail profile, various measurements were obtained and analyzed independently for similar trams with the same driver. Another significant factor which plays a crucial role in vibration is speed. The selection of a method which can detect the influence of speed on vibration is very demanding since this evaluation needs to be based on various measurements in similar settings. Therefore, various experiments and data from the same arrangement should be obtained to evaluate the impact of speed on vibration. This paper reports on several practice tests and data acquisition via smartphone built-in sensors on a specific route to investigate the effects of speed on vibration. Two different statistical and non-statistical methods are then proposed to monitor the condition of rail tracks and schedule preventive maintenance for light rail based on Melbourne tram network data. These approaches will optimize the total investment in tram track maintenance for managing authorities.