An intelligent welding robot system with visual sensors is developed in order to realize full automatic welding of thin mild steel plates including automatic seam tracking and automatic control of welding conditions. A system to detect the shape and dimension of molten pool using CCD camera and a penetration control system using Neural Network in TIG arc welding are investigate. In order to characterize the shape of molten pool, width, length and area of the molten pool were measured, and are used to form the contour of the molten pool as a shape parameters. These parameters are input to the neural network, which outputs optimum welding condition to control the penetration of the molten pool. Consequently, if unexpected change occurs in welding conditions, such as root gap, welding speed and so on, the welding system can optimumly control the welding condition. The constructed system is tested and found to be effective for penetration control in automatic butt welding of thin mild steel plates.
Recently, mechanization and robotization of welding has been widely investigated and many new intelligent welding robots have some sensors, such as arc-sensor andvision sensor[1]. Of particular interest is the adaptive control of welding conditions, in which the welding arc processes and the molten pool condition are monitored during welding, and the welding conditions are controlled in real-time. For instance, a method has been investigated to control the penetration of weld, in which the shape and dimensions of the molten pool are visually monitored and welding conditions are controlled so as to keep them optimum[2]- [6]. On the other hand, in automatic butt welding of relatively thin plates, it is important to control welding conditions to obtain a good full penetration weld.
