Neuro-fuzzy Modeling of Human Welder's Response to 3D Weld Pool Surface in GTAW

Understanding and modeling of welder response to 3D weld pool surface help develop intelligent welding robotic systems and better train welders. In this paper, a welder's adjustment on the welding current as a response to the 3D weld pool surface as characterized by its width, length, and convexity is studied. A vision sensing system is developed to real-time measure the specular 3D weld pool in gas tungsten arc welding (GTAW). Experiments are designed to produce random changes in the welding speed resulting in fluctuating 3D weld pool surface. Adaptive Neuro Fuzzy Inference System (ANFIS) model is developed for the human welder response in order to correlate the current adjustment to the 3D weld pool surface. To justify the use of the neuro-fuzzy model, linear model has also been fitted and compared. It is found that the proposed ANFIS modeling can derive detailed correlation between the human welder's responses and the weld pool geometry and help better understand the nonlinear response of the human welder to 3D weld pool surfaces.