The use of Taguchi and approximation methods to optimize the laser hybrid welding of a 5052-H32 aluminium alloy plate

Abstract

This study aims to identify the optimal welding conditions for laser hybrid welding on a 5052-H32 aluminium alloy. Non-linear transient thermal analysis is adopted to simulate the thermo-physical phenomena in the welding process. Taguchi's parameter design method is applied to establish the optimal welding parameters that minimize residual stress and strain. The results obtained via design of experiments (DOE) and Taguchi's parameter design using the welding simulations are compared with the experimental results. Using DOE data, approximation models such as the polynomial-based response surface method and the radial basis function neural network are constructed. Using such approximation models, the relationship between welding conditions and thermo-mechanical responses can easily be estimated according to the parameter variations.