Probabilistic reliability evaluation of ice zone jacket platforms considering multi-region corrosion effects

The structural safety of offshore platforms is critical for the economic and environmental sustainability of marine oil and gas development. Quantifying the reliability of the ultimate bearing capacity, a key safety indicator, is essential to mitigate failure risks induced by complex marine environments, particularly ice loading. Focusing on a four-legged jacket platform in the Chengdao Oilfield, this study systematically investigates the degradation effects of multi-region corrosion on structural performance. An innovative reliability assessment framework that incorporates multi-region corrosion effects is proposed, utilizing the Response Surface Method (RSM). A simplified beam-element model is established in ABAQUS, and pushover analysis is performed to determine the ice-resistant ultimate bearing capacity under various corrosion durations. Considering the randomness of material properties and corrosion depths in critical regions, samples are generated via the Box-Behnken experimental design. Multivariate regression is then employed to derive an explicit response surface function relating the ultimate bearing capacity to the random variables. Finally, Monte Carlo Simulation (MCS) is applied to assess the platform's failure probability and probabilistic reliability under simulated 100-year return period ice loads. This framework provides a robust approach for quantifying the impact of corrosion on offshore platform safety.