Innovative 3D cellular automata simulation of corrosion evolution and mechanical property in weathering steel and butt welds

Abstract

Weathering steel is increasingly used in bridge engineering due to its excellent atmospheric corrosion resistance. However, corrosion damage and mechanical property degradation, especially in marine environments, remain critical concerns. This study conducted accelerated corrosion tests on weathering steel and butt weld specimens under three corrosion levels, characterizing surface corrosion morphology and pit depth distribution using 3D laser scanning. An innovative 3D cellular automata (CA) model combined with a genetic algorithm was developed to reconstruct corrosion morphology based on actual corrosion parameters. Mechanical properties of uncorroded and corroded specimens were obtained by monotonic tensile tests and elastoplastic finite element analyses. The results show that the weld metal zone exhibits the highest corrosion resistance, while the heat-affected zone shows the lowest. The improved 3D CA model predicts pit depths with an error margin of less than 5 %, demonstrating high accuracy. Pit depths for base metal and weld specimens follow lognormal and normal distributions, respectively. As the corrosion degree increases, the elastic modulus, yield strength, tensile strength, and elongation decrease, with maximum reductions of 27.21 %, 9.56 %, 9.75 %, and 48.28 %, respectively. A linear relationship is observed between the degradation rate and the corrosion degree, except for elongation in weld specimens. Stress concentration and mechanical property degradation are more pronounced in weld specimens compared to base metal specimens. The proposed 3D CA model is validated for accurately simulating both corrosion processes and mechanical property degradation.