• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.8
• /
• pp.133-138
• /
• 2019

This study evaluated the tensile properties of SAPH440, a hot-rolled steel for automotive structural applications, based on GMAW lap welding, the welding current, the welding voltage, and the feed rate. Tensile tests were performed according to the joint parameters of the GMAW process, for which specimens were fabricated according to KS B ISO 9018 by lap welding. The bead appearance was observed in each condition, and the weldability was evaluated by the tensile test. Higher the welding current resulted in a deeper weld, but the tensile strength was not significantly different from when the parameter was fixed due to the fracture of the base material. When the current was higher than the voltage, as in the case of a welding current of 200 A and welding voltage of 17 V, a large amount of spatter is generated, the welding is unstable, and the welded part breaks. Higher the voltage resulted in the bead not causing defects in general, and it also affected the weldability. If the current and voltage were too low, the welding was not performed normally, and the tensile strength could not be measured. However, as the current increased, the increase of the voltage and the feed rate did not affect the tensile strength.

• Journal of the Korean Society of Industry Convergence
• /
• v.5 no.2
• /
• pp.111-118
• /
• 2002

In GMAW(Gas Metal Arc Welding) processes, bead geometry (penetration, bead width and height) is a criterion to estimate welding quality, Bead geometry is affected by welding current, arc voltage and travel speed, shielding gas, CTWD (contact-tip to workpiece distance) and so on. In this paper, welding process variables were selected as welding current, arc voltage and travel speed. And bead geometry was reasoned from the chosen welding process variables using neuro-fuzzy algorithm. Neural networks was applied to design FLC(fuzzy logic control), The parameters of input membership functions and those of consequence functions in FLC were tuned through the method of learning by backpropagation algorithm, Bead geometry could he reasoned from welding current, arc voltage, travel speed on FLC using the results learned by neural networks. On the developed inference system of bead geometry using neuo-fuzzy algorithm, the inference error percent of bead width was within ${\pm}4%$, that of bead height was within ${\pm}3%$, and that of penetration was within ${\pm}8%$, Neural networks came into effect to find the parameters of input membership functions and those of consequence in FLC. Therefore the inference system of welding quality expects to be developed through proposed algorithm.