Abstract: Welding process parameters are the key factors affecting the quality of welding. Since the relationship between process parameters and the mechanical properties of welded joints is multi-dimensional and implicit， an optimization algorithm combining Kriging model and particle swarm optimization is proposed to optimize the process parameters of 3.5mm magnesium alloy sheet in AC_TIG welding. Firstly， the sample set is constructed by Taguchi orthogonal method. Secondly， the Kriging surrogate model is established between output and input， and then the optimal combination of process parameters and its mechanical properties are obtained by the proposed algorithm. The results show that such optimal process parameters as tensile strength， yield strength and average micro-hardness of the welded joints reach 97.6%， 98% and 91.5% of the base metal respectively. The proposed algorithm not only reduces economic and time costs， but also improves the welding process design capabilities.

Key words: AC_TIG welding, magnesium alloy, welding process parameters, Kriging model, particle swarm optimization