LIE Home • Issue Contents

Parameter Optimization of the Laser T-joint Welding of Aluminium Alloy with Low Carbon Steel Using Numerical and Statistical Methods
W-H. Huang, J-Q. Long and J-W. Xiang

Welding of aluminium alloy with low carbon steel is a difficult work due to the formation of brittle intermetallic compounds, and the use of laser welding technology by strictly control the welding parameters is a possible solution to surpass this problem. In other words, the laser welding input parameters has always played a very important role in welding aluminium alloy with low carbon steel process, especially in determining the quality of the weld joint. The quality of the weld joint can be evaluated in terms of properties such as weld bead distortion, von Mises stress and heat affected zone (HAZ) area. In this study, a three-dimensional (3-D) finite element method (FEM) model is used to simulate the steel to aluminium alloy laser welding process. The laser welded panel stress, HAZ and distortions can be calculated with the simulation FEM software (SYSWELD), taking into account all relevant physical phenomena. Orthogonal design is used to perform the experimental design and the selection of the experiment factors, levels and indexes are also revealed. The method of orthogonal polynomial regression is used for regression modelling to establish the relationship between the laser welding input parameters like laser power, welding speed and beam angle. Response surface methodology (RSM) was adopted for optimizing the process parameters. The optimal condition to have a quality weld is found at 1700 W of laser power, 40 mm/s of welding speed and 30o of beam angle. The study also observed that the effect of welding speed is more significant than other welding parameters. Finally, a confirmatory simulation experiment has been carried out to verify the optimal setting so obtained.

Keywords: Laser welding, aluminium alloy, low carbon steel, finite element method (FEM), response surface methodology (RSM), orthogonal design, orthogonal polynomial regression

Full Text (IP)