Monitoring of weld joint penetration during variable polarity plasma arc welding based on the keyhole characteristics and PSO-ANFIS

Abstract

In this paper, a simple and flexible vision sensor system was developed for acquiring the keyhole images from the back-side of the work plate. The keyhole variation mechanism based on a thermal-force model was described and demonstrated by a series of acquired keyhole images. The keyhole characteristics specified by area and tilt angle were extracted based on part-based tree model. To investigate the relationships, which could be complicated, between the weld penetration and keyhole characteristics with different welding conditions, a novel hybrid approach, combining particle swarm optimization (PSO) and an adaptive-network-based fuzzy inference system (ANFIS) was proposed to recognize the penetration status. Extensive experiments were conducted to predict the joint penetration from the keyhole images by incorporating keyhole size and shape parameters. The work shows that PSO-ANFIS model can be used to relate the keyhole characteristics to the weld joint penetration.