Computational study demonstrates accurate neural network prediction of mixed-mode stress intensity factors in corrugated steel webs, indicating enhanced fatigue assessment for bridge girders.
Corrugated steel webs (CSWs) are widely used in bridge girders, where weld‐toe cracks are prone to mixed‐mode fracture under service loading. This study presents a finite element and data‐driven framework for predicting stress intensity factors (SIFs) of semielliptical weld‐toe cracks in CSWs. ABAQUS and Franc3D were coupled to calculate mixed‐mode SIFs for 2880 parameter combinations covering crack depth ratio, crack aspect ratio, web corrugation angle, thickness ratio, weld‐toe angle, and weld size. Results indicate that crack depth ratio and crack aspect ratio dominate the SIF distribution, whereas the web corrugation angle modifies the crack‐driving force through the local stress‐transfer path. A polynomial regression model and an artificial neural network (ANN) surrogate model were developed for rapid SIF prediction. The ANN model achieved higher prediction accuracy and stronger interpolation capability within the investigated parameter range, thereby supporting fracture assessment, structural optimization, and fatigue‐life evaluation of CSW welded details.
No takes yet. Share an insight, caveat, or question.
Su et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: