Principal component analysis for multiple quality characteristics optimization of metal inert gas welding aluminum foam plate

Abstract

Principal component analysis (PCA) coupled with Taguchi methods are employed in the study for multiple quality characteristics optimization of metal inert gas (MIG) arc welding aluminum foam plates. The quality characteristics investigated are the micro-hardness and the bending strength of the weldments. Eight control factors selected are the type of filler material, MIG current, welding speed, MIG gas flow rate, workpiece gap, MIG arcing angle, groove angle, and electrode extension length. It is shown by the experimental results that the optimal parameter combination of the MIG welding process is A 2 (filler material: Type No. 5356), B 3 (MIG current: 100 A), C 1 (welding speed: 80 mm/min), D (MIG gas flow rate: 13 L/min), E 2 (workpiece gap: 1.7 mm), F 3 (MIG arcing angle: 50°), G 3 (groove angle: 20°), and H 1 (electrode extension length: 15 mm). Moreover, it is ascertained from the analysis of variance (ANOVA) results that B (MIG current), C (welding speed), and E (workpiece gap) are the most important control factors in the process design, and thus strict control must be applied to the three factors. The experimental results likewise show that the optimal process design could indeed improve the multiple quality characteristic values of the MIG-welded aluminum foam plates.