Cellular Automaton Simulation and Static Mechanical Property Degradation of Corroded High-Strength Steel Wires

High-strength steel wires in bridge stay cables are susceptible to chloride attack and wet–dry cycling during long term service. Furthermore, the resulting spatially non-uniform evolution of corrosion morphology can lead to the degradation of their static mechanical properties. In this study, high-strength steel wires with a nominal diameter of 7 mm were subjected to copper-accelerated acetic acid salt spray (CASS) exposure for 0–80 d, followed by 3D laser scanning and uniaxial tensile testing. Corrosion depth, surface roughness, and cross-sectional area loss were first quantified. Based on these morphological indicators, a 3D cellular automaton (CA) model incorporating an axially stationary correlated random field was developed. The maximum cross-sectional area loss ratio was subsequently adopted as the characteristic damage parameter for evaluating mechanical degradation, and a quantitative relationship between the CA simulation results and the degradation of static mechanical properties was established. The results show that, with increasing corrosion duration, the corroded regions on the wire surface progressively expanded, coalesced, and deepened, producing a spatially non-uniform morphology characterized by alternating zones of severe and mild corrosion along the axial direction. From 20 to 80 d, the maximum cross-sectional area loss ratio increased from 13.307% to 31.175%. Subsequently, the developed CA model successfully reproduced the evolution trends of the corrosion morphology indicators, with relative errors of less than 19% compared with the 3D scanning measurements. By incorporating the maximum cross-sectional area loss ratio obtained from the CA simulations into the mechanical degradation relationships, the mean prediction errors for Young’s modulus, yield strength, and ultimate tensile strength were 2.788%, 2.581%, and 3.198%, respectively, relative to the experimental results. These results demonstrate that coupling the CA model outputs with the mechanical degradation relationships provides an effective approach for predicting the residual mechanical properties of high-strength steel wires subjected to different corrosion durations.