Residual life prediction of aluminum alloy plates under cyclic loading using an integrated prognosis method

Structures under cyclical loading are prone to fatigue-induced cracks. These cracks reduce structures’ residual life and increase their risk of failure. Historically, engineers have used degradation models like Paris’ law to estimate crack length when planning repair and maintenance on such structures. In this study, we developed a method to estimate Paris’ law parameters that are useful for predicting the fatigue life of 2024-T3 aluminum alloy plates under cyclic loading. A new optimization model is developed to estimate Paris’ law parameters’ mean and standard deviation. The research also considers the case in which the magnitude of the applied load is unavailable. Using Bayesian updating, the optimized parameters are further updated based on condition monitoring data to increase crack length estimation accuracy. The proposed method is validated with the help of Virkler crack propagation data for an aluminum alloy plate. According to the validation results, the average error in structure’s lifetime prediction based on crack length is 1.5% when the crack reaches 71% of the failure threshold.