Predicting characteristic performance for arc welding process

predicting characteristic performance of the arc welding process is an underlying task for online weld quality control. As a continuous yet complex high-thermal welding process, conventional modelling method can be ineffective in the presence of big uncertainties and noise. Besides, experiments and evaluations could be costly and very inefficient because of resource utilization, energy consumption, and dedicated human labor. Hence, we investigate the welding process modelling problem in this paper to understand how to tune the welding parameters to achieve the desired characteristic performance. Gaussian Process Regression, a non-parametric modeling technique, is exploited to model the relationship between the welding parameters and characteristic performance. Cross-validation method is employed to avoid overfitting problem. Gas tungsten arc welding experiments were performed and the experimental data were collected and utilized to validate the proposed modelling method. The predicted characteristic performance is compared with the original data and it shows that the proposed method can accurately predict the weld bead geometry. This paper opens a door for online parameter tuning to achieve optimal performance, hence the proposed method will innovate the welding process.