Quantitative Study of Predicting the Effect of the Initial Gap on Mechanical Behavior in Resistance Spot Welding Based on L-BFGS-B

The initial gap (IG) is frequently occurring in the process of resistance spot welding (RSW) for automotive body-in-white structures. It is an inevitable challenge that the RSW with IG can negatively impact the welding quality, subsequently reducing the structural integrity and safety of the vehicle. This research aims to study the influence of the IG on RSW mechanical behaviors based on the refined finite element model (FEM) of RSW with different IGs under tensile shear load. The influence of six types of IGs on the peak load and fracture modes of RSW of plates with similar thicknesses of 1.0 mm and 1.5 mm is investigated through FEM and experiments. To quantify the influence of the IG on the RSW's deformational behavior, a prediction model is introduced to predict the peak load of RSW with different IGs. The prediction model is locally optimized through the Limited memory Broyden Fletcher Goldfarb Shanno (L-BFGS-B) optimization algorithm. Based on the prediction model, the relationship among the peak load values of the tensile shear specimens, the IG, and the mechanical behavior of the RSW is revealed. The results show that the IG has an obvious influence on the peak load values of RSW under tensile shear load, and the fracture modes are both the pull-out fracture (PF) mode. The peak load values of the RSW's tensile shear specimens are decreased with the increment in the IG. Finally, the prediction model can accurately predict the peak load for various IGs, with errors of no more than 3%.