Virtual Axle Detector: Train Axle Localization based on Bridge Vibrations

Abstract Infrastructure worldwide is facing the challenge of aging bridges and increasing traffic loads. Prolonged serviceability safety these structures can be enabled by Structural Health Monitoring (SHM) methods. Knowledge actual operating loads critical for evaluation remaining service life. However, direct measurement challenging requires a significant financial investment. Bridge Weigh‐In‐Motion (BWIM) methods use structural response bridge to determine loads, but generally rely on accurate knowledge position as function time. Positions determined using conventional axle detectors, their lifetime limited, installation expensive. To avoid problems, we propose an improved Virtual Axle Detector (VAD) with Enhanced Receptive field (VADER) that detect axles all types accelerometers placed anywhere along bridge. The same data set 3787 train passages recorded steel trough railway under real conditions was used. Our results show that, in comparison VAD, VADER reduces number undetected over 79% detects 99.5% average spatial accuracy 4.6 cm.