Machine learning-based bridge damage detection using cross-correlation of train-borne acceleration signals

This study presents a novel machine learning-based approach for damage detection using train-borne measurements, under operational conditions (speed > 50 kph, rail irregularities and noise). To this end, an optimised two-dimensional convolution neural networks (CNN) with the network-in-network architecture is built, trained, and tested to detect damage of various severity and location in a bridge, using train-borne measurements only. As an input, cross-correlation of signals from two train bogies is used as a damage-sensitive feature for the first time. The proposed method in this study is applied to a cohort of simulated acceleration measurements on a nominal RC4 power car passing over a 25 m simply-supported reinforced concrete bridge. The presented method has shown great accuracy in detecting damage under operational condition. The sensitivity and robustness of the approach are tested and validated for 18 damage severity and location scenarios and 100 random vehicle speeds, ranging between 70 to 130 kph. This is of particular value as speed defines the length of the train-borne signal whilst passing over the bridge, hence the amount of information manifested in a single passing. The results demonstrate the feasibility of the approach for data-driven damage detection using train-borne measurement only.