Anindito Purnowidodo, Redi Bintarto, M.A. Choiron
This study presents an artificial neural networks (ANN) Model for predicting fatigue crack growth behavior under variable amplitude loads, specifically for negative and zero stress ratios with a single tensile overload. Using the effective stress intensity factor (ΔKeff) as the primary input, the ANN Model estimates crack growth and fatigue life based on known crack length and increments. Results show that the Model provides accurate predictions of ΔKeff and crack growth behavior across various loading conditions. The ANN approach offers a practical tool for assessing fatigue life in engineering applications, even with limited datasets. © 2025 Elsevier Ltd
Department of Mechanical Engineering, University of Brawijaya, Jl. MT. Haryono 167, Jawa Timur, Malang, Indonesia
