ABSTRACT
In chloride-laden environment, steel corrosion in RC members leads to their structural performance deterioration. Inspection methods using unmanned aerial vehicle (UAV) have attracted attention as an alternative method of classical visual inspection. Corrosion crack width detected by UAV could be an effective information to evaluate the structural condition. However, since the relationship between steel corrosion and crack widths is complex and uncertain due to associated parameters such as structural details, this observational information has not been used to evaluate the structural performance of in-situ concrete structures. In this study, the load-bearing capacity of deteriorated RC members is probabilistically evaluated using the observational corrosion cracks obtained by UAV, in addition to finite element analysis, stochastic field theory, and machine learning. A case study is presented to investigate the effect of the different measurement method of corrosion crack widths between UAV shooting and close-up shooting with digital camera on the load-bearing capacity.
