The pressure wave generated by an urban-rail vehicle when passing through a tunnel affects the comfort of passengersand may even cause damage to the train and related tunnel structures.Therefore,controlling the trainspeed in the tunnel is extremely important.In this study,this problem is investigated numerically in the frameworkof the standard k-εtwo-equation turbulence model.In particular,an eight-car urban rail train passingthrough a tunnel at different speeds(140,160,180 and 200 km/h)is considered.The results show that the maximumaerodynamic drag of the head and tail cars is most affected by the running speed.The pressure at selectedmeasuring points on the windward side of the head car is very high,and the negative pressure at the side windowof the driver’s cab of the tail car is also very large.From the head car to the tail car,the pressure at the same heightgradually decreases.The positive pressure peak at the head car and the negative pressure peak at the tail car aregreatly affected by the speed.
Rail defects can pose significant safety risks in railway operations, raising the need for effective detection methods. Acoustic Emission (AE) technology has shown promise for identifying and monitoring these defects, and this study evaluates an advanced on-vehicle AE detection approach using bone-conduct sensors—a solution to improve upon previous AE methods of using on-rail sensor installations, which required extensive, costly on-rail sensor networks with limited effectiveness. In response to these challenges, the study specifically explored bone-conduct sensors mounted directly on the vehicle rather than rails by evaluating AE signals generated by the interaction between rails and the train’s wheels while in motion. In this research, a prototype detection system was developed and tested through initial trials at the Nevada Railroad Museum using a track with pre-damaged welding defects. Further testing was conducted at the Transportation Technology Center Inc. (rebranded as MxV Rail) in Colorado, where the system’s performance was evaluated across various defect types and train speeds. The results indicated that bone-conduct sensors were insufficient for detecting AE signals when mounted on moving vehicles. These findings highlight the limitations of contact-based methods in real-world applications and indicate the need for exploring improved, non-contact approaches.
Lei JiaJee Woong ParkMing ZhuYingtao JiangHualiang Teng
