ABSTRACT
The actual response of a bridge deck under heavy vehicles is of great interest for bridge assessment. To better understand this, strain and displacement monitoring technologies have been deployed on many bridges around the world to measure their responses, particularly for stress histories under heavy vehicle events. For orthotropic steel decks (OSD) of highway bridges, it remains a key technical challenge to process and interpret these monitored stress responses. This is due to the uncertainties in vehicle weave (transverse axle position), axle load, axle type and vehicle configuration. Fatigue performance governs the life of OSD, and efficient processing of monitoring data would significantly improve the ease of assessing its health. In this paper, the outputs from detailed finite element modelling (FEM) are used to interpret the monitored heavy vehicle responses from the OSD of a long span suspension bridge. The FEM investigates the effects of different weave positions and axle arrangements on stress histories at locations equivalent to those monitored on the bridge. Responses from heavy vehicle events are also extracted from the high frequency strain gauge monitoring data. Through engineering interpretation, key features of the generated responses under different types and locations of heavy vehicles are identified. It is envisaged that this interpretation could be automated through machine learning, which ultimately could be linked back to local fatigue assessments and bridge inspections to correlate processed monitoring data with fatigue health.
