Abstract
This paper deals with the problem concerning the sensing and the controlling of weld pool. In order to obtain the high quality of the welding result, it is important to control the weld pool depth in the robotic welding regardless of the external disturbance such as an irregular change of the groove gap. The method of controlling the weld pool depth without mathematical model is discussed.
Since it is difficult to directly measure the depth, the depth is estimated from the surface shape of the weld pool, the groove gap, and the welding current. A neural network is used to estimate the depth without mathematical model. The weld pool depth is controlled from the output of the neural network by using the fuzzy controller. The validity of the neural network and of the fuzzy controller was verified by the welding experiments.
