Statistical Characterization of Stress Concentrations along Butt Joint Weld Seams Using Deep Neural Networks

Abstract

In order to ensure high weld qualities and structural integrity of engineering structures, it is crucial to detect areas of high stress concentrations along weld seams. Traditional inspection methods rely on visual inspection and manual weld geometry measurements. Recent advances in the field of automated measurement techniques allow virtually unrestricted numbers of inspections by laser measurements of weld profiles; however, in order to compare weld qualities of different welding processes and manufacturers, a deeper understanding of statistical distributions of stress concentrations along weld seams is required. Hence, this study presents an approach to statistically characterize different types of butt joint weld seams. For this purpose, an artificial neural network is created from 945 finite element simulations to determine stress concentration factors at butt joints. Besides higher quality of predictions compared to empirical estimation functions, the new approach can directly be applied to all types welded structures, including arc- and laser-welded butt joints, and coupled with all types of 3D-measurement devices. Furthermore, sheet thickness ranging from 1 mm to 100 mm can be assessed.

Comparison of machine learning and stress concentration factors‐based fatigue failure prediction in small‐scale butt‐welded joints

“…One condition for meaningful analyses, for example, with respect to explainability, is a good performance of the models. The performance of the SCF ANN model has been confirmed in an earlier study 9 . The performance of the fracture location classifier was checked once for the SCF being known and once for it being unknown.…”

“…In the current study, a deep neural network (DNN) for butt joints with weld toe failure presented in Braun et al 9 is applied to determine the SCFs (for tension and bending loading) along the weld seams of the fatigue test specimens. A DNN is a subtype of an ANN with several layers.…”

“…SCF values were replaced with NaN (not a number) when they were below the 5th percentile and above the 95th percentile with respect to all values. This was done since it is expected that there are some outliers in the laser scan measurements 9 . All other feature values were replaced with NaN if they were outside of realistic bounds.…”

“…This median geometry data were also added to the butt joint dataset. The median was taken instead of the mean as it is less sensitive to outliers 9,13 …”

“…Recent studies show that fatigue cracks often initiate at locations with high stress concentrations 5–7 . A representative quantity to describe the magnitude of local stress concentrations is the stress concentration factor (SCF), 8 which is the quotient of local notch stress and nominal stress 9 …”