Multifractal analysis for gas metal arc welding

The gas metal arc welding (GMAW) is a complicated chaotic dynamic process and its arc electrical signals have the characteristics of both chaos and fractals. The multifractal can accurately depict the geometric characteristics and local scale behaviors of time series signals. Through local and global self-similarity, we can determine the overall characteristics from the local dynamic behaviors of a dynamic system. Here, we used the multifractal theory to obtain the multifractal spectra under different welding parameters, and defined a multifractal spectrum index that described the multifractal spectrum of welding current signals. When the shielding gas flow increases under the short-circuiting mode, the welding process becomes more stable and the multifractal spectrum index decreases. As the welding current increases, the droplet mode gradually changes from the stable short-circuiting mode to the stable spray droplet mode, while the multifractal spectrum index first increases and then decreases. It is confirmed the droplet transfer behavior of GMAW is a complicated chaotic dynamic process, and the current signal has the multifractal characteristics. The multifractal spectrum distribution is closely related to the droplet transition mode and stability.