Process control based on double-side image sensing of the keyhole in VPPA welding

The double-side image sensing of the keyhole puddle in the variable polarit y plasma arc welding (VPPAW) of aluminum alloys has been investigated in th is paper to extract the characteristic geometrical size of the keyhole and to realize feedback controlling for weld formation in the welding process. Some geometrical sizes of the visible keyhole in the front and back image s uch as the width, height, area, etc., can be used to monitor both the keyho le puddle and the welding formation in the welding process, Under the condi tions of varied heat sink, varied gap and misalignment the trend from norma l welding to cutting can be reflected from the variations of geometrical si zes of the keyhole puddle. The keyhole area, the keyhole height and the sha pe parameters of the keyhole puddle are the key parameters which reflect th e trend from normal welding to cutting when meeting the conditions of the v aried heat sink, varied gap and misalignment. The algorithm for the image p rocess of the keyhole puddle and the periphery extracting of the visible ke yhole developed in the paper can be used to determine real-time, the geomet rical sizes of the visible keyhole when the welding process is conducted. A n artificial neural network is applied to establish the steady model for pr edicting the geometrical sizes of the back keyhole puddle. The inputs of th e model are the geometrical sizes of the front keyhole puddle and the weld parameters, and the outputs of the model are the geometrical sizes of the b ack keyhole puddle. The model can be used to control the stability of the k eyhole and the weld formation. (C) 2001 Elsevier Science B.V. All rights re served.