焊接
结构工程
概率逻辑
材料科学
疲劳极限
几何学
工程类
冶金
数学
计算机科学
人工智能
作者
Gustav Hultgren,Rami Mansour,Zuheir Barsoum
标识
DOI:10.1016/j.ijfatigue.2022.107364
摘要
• Probabilistic framework investigating the effect of variability in weld geometry. • Detailed finite element simulations of the laser-measured geometry. • Model calibration using comprehensive fatigue testing. • Comparing local fatigue properties with the weld geometry from the Winteria system. • The reduction of fatigue scatter enables the effect of weld quality to be separated. Progress in developing digital quality assurance systems for welded joints has made it possible to accurately measure the local geometry and its variation, making it possible to derive new relations between the geometric variations and fatigue. A probabilistic model for the fatigue strength is here presented based on the actual weld geometry. The novelty lies in that representative stresses can be determined for both the complete weld and sections of the weld. Calibration of the model using 105 fatigue-tested specimens shows a reduced variation in SN-diagrams compared with the nominal stress methods when substantial weld geometry variations are present.
科研通智能强力驱动
Strongly Powered by AbleSci AI