A calibration tool for weld penetration depth estimation based on dimensional and thermal sensor fusion

Abstract

Weld quality monitoring and assessment in industrial robotic arc welding processes is key to ensure suitability of a component for the intended application. In particular, weld penetration depth is as a major fabrication requirement that has to be addressed. Several alternatives have been proposed based on the use of individual monitoring techniques, but, due to the physical challenges of the welding process and accessibility restrictions to the weld root, multi-sensor approaches have been recently developed. These methods require the adoption of complex setups or calibration strategies. In this work, a multi-sensor approach is proposed to create a calibration tool for weld penetration depth assessment, overcoming physical accessibility restrictions and enabling depth evaluation instantaneously, opening the gate for online usage.

Laser Welding Penetration Monitoring Based on Time-Frequency Characterization of Acoustic Emission and CNN-LSTM Hybrid Network

“…To understand the keyhole dynamics and further improve the product quality, a series of in-situ monitoring approaches, including thermal [ 4 , 5 ], photoelectric [ 6 ], and optical sensors [ 7 , 8 ], have been proposed in industry and academia. The design idea of mainstream monitoring solutions lies in the utilization of the phenomenon of keyhole formation, which carries various types of valuable information about weld quality.…”