Parameter Optimization of Hybrid-Tandem Gas Metal Arc Welding Using Analysis of Variance-Based Gaussian Process Regression

In this paper, the parameter optimization of hybrid-tandem gas metal arc welding (GMAW) process was studied. The GMAW uses an additional filler-wire with opposite polarity in contrast to conventional tandem process. process, more parameters and relationship between causing strong nonlinearity should be considered. analysis variance-based Gaussian regression (ANOVA-GPR) method implemented construct surrogate modeling, which can express including uncertainty weld quality. Major among several extracted by use novel method. quality used as a cost function is defined characteristics related penetration bead shape, sequential quadratic programming (SQP) determine optimal condition. This approach enabled sound at high travel speed 1.9 m/min, difficult achieve