Computational weld-mechanics assessment of welding distortions in a large beam structure

Abstract Unwanted distortions are typically observed in components after the welding process. Physical trial tests and extra post-treatments are being widely utilized in industries to minimize and correct the out of tolerance distortions. These methods are time-consuming and costly. There has been growing interest in digital tools which have great potential to minimize the physical test loops and corrections. In this study welding distortions analysis has been carried out on a large beam structure experimentally and numerically using computational welding mechanics (CWM) techniques such as the inherent strain (local–global) method and the shrinkage method, together with the lumping approach. The estimated distortions from the shrinkage together with lumping approaches were in good agreement with the experimental measurements and the computational time affordable. The inherent strain (local–global) method captured the trend of distortion with an underestimation of distortions. The accuracy of the estimated residuals stresses from the inherent strain (local–global) approach is higher than the one from shrinkage together with lumping approaches. Moreover, the effects of various welding process parameters (i.e. welding sequence, fixture, and weld pool size) on welding distortions were investigated. It is found that following the proper welding sequence could minimize the welding distortion of the beam structure. Increasing the constraints of fixtures can prevent welding distortion effectively and reducing weld pool size results in less welding distortions of the beam structure.