Keyhole mapping to enable closed-loop weld penetration depth control for remote laser welding of aluminium components using optical coherence tomography

Remote Laser Welding (RLW) combines the positive features of tactile laser welding with additional benefits such as increased processing speed, reduced operational cost and service as well as higher process flexibility. A leading challenge preventing the full uptake of RLW technology in industry is the lack of efficient Closed Loop In-Process (CLIP) monitoring and weld quality control solutions. This underpins the need to fuse multiple sensor technologies, data analytics along with predictive engineering simulations. Although the development and integration of a variety of sensors, covering the radiation spectrum from ultra-violet to farinfrared, the flawless deployment of CLIP solutions is still challenged by the need for: signal de-noising in case of process instability; real-time data analytics; adaptive control engineering architecture to cope with process variations induced by manufacturing tolerances. This paper focuses on the aspect of the Weld Penetration Depth Control (WPDC) using Optical Coherence Tomography (OCT) as necessary step to enable adaptive penetration depth control during RLW of aluminium components in fillet lap joint configuration in consideration of part-to-part gap variation. The approach is decoupling the welding process parameters in two sub-sets: (1) in-plane control of the heat input on the upper part to facilitate the droplet formation; (2) outof-plane heat management to achieve the desired level of penetration control in keyhole mode. The paper presents the results of the keyhole mapping with variable part-to-part gap, that provide the insights for future research to enable the fully automatic closed-loop weld penetration depth control. Current limitations and next phases of research and development are highlighted based on the experimental study.