In automatic butt welding of relatively thin plates, it is important to control welding conditions in order to obtain a sound full penetration weld. Especially, it is difficult to obtain a full penetration weld, because the optimum welding condition change with groove conditions, such as gap of groove, shape of groove and so on. Recently, It was reported that there is an intimate relationship between the oscillation of the molten pool and penetration. The frequency of a molten pool decreases drastically with transformation from partial penetration to full penetration. Accordingly, estimation of penetration mode by detecting frequency of the molten pool is attempted, and, furthermore, real-time control for full penetration welding is performed. In this study, the principle of detection of penetration, detection method of molten pool oscillation and control system of welding condition according to the frequency of molten pool oscillation are investigated. As a result of experiment, it is demonstrated that the control system constructed is effective for automatic TIG arc welding of thin plates.
Recently, research and development for mechanization and robotization of welding has been carried out widely, and a number of intelligent welding robots which have some sensors, such as arc-sensors and visual sensors, have been developed. Several studies have addressed the adaptive control of welding conditions, in which the welding arc phenomena and the molten pool condition are observed during welding and furthermore the welding conditions are controlled in real-time. For instance, a method to control the penetration of weld, in which the shape and dimensions of the molten pool are visually detected and welding conditions are controlled to keep them optimum, has been proposed and the effectiveness confirmed.
