Observations on the damage behaviors of corrosion fatigue in steel strands based on image analysis

Abstract

In this work, the corrosion behaviors of steel strands exposed to acidic rain under alternating stress, static stress, and no stress were modeled based on experimental data. The results are based on an image corrosion feature analysis of steel strands at different corrosion times and operating conditions, and the corrosion fatigue behaviors indicate that the steel strands display different degrees of corrosion under different loading conditions. The corrosion fatigue behaviors of the steel strands were analyzed based on the corrosion features in the images of the steel strands subjected to different corrosion times and operating conditions. The results showed that different loading conditions lead to different degrees of corrosion in the steel strands. The samples’ surface images and corrosion damage continuously changed as the corrosion time increased. Given the same corrosion time, the degree of corrosion was the highest in samples subjected to alternating stress, followed by that in samples under static stress. Unstressed samples had the lowest degree of corrosion. The variations of the corroded surface and the degree of corrosion in the material were particularly reflected by three image features pitting rate, pit depth, and pit radius. No significant changes occurred in the samples when the corrosion time was 120–360 h, regardless of the type of stress they were subjected to. This finding indicated that pitting has a minimal effect on the force-bearing capacity of steel strands during pit nucleation and expansion. However, a substantial decrease in the fracture strength occurred under alternating stresses when the pits expanded into cracks (very deep cracks were found at the 720-h mark in the samples subjected to alternating loads). Conventional damage tolerance analysis methods and software can then be used to perform crack propagation life predictions by treating pits as quantizations of a crack. Fatigue life estimation based on the corrosion images with the appropriate combination of corrosion damage levels and load-bearing capacities is a viable technical pathway. The findings of this work will provide new ideas for the assessment of structural reliability and safety via the processing of inspection photographs.