Abstract: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, and 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 and data analytics with predictive engineering simulations. Although the development and integration of a variety of sensors covers the radiation spectrum from ultraviolet to far-infrared, the flawless deployment of CLIP solutions is still challenged by the need for the following: signal denoising in the case of process instability; real-time data analytics; and adaptive control engineering architecture to cope with process variations induced by manufacturing tolerances. This paper focuses on the aspect of weld penetration depth control using optical coherence tomography (OCT) as a necessary step to enable adaptive penetration depth control during RLW of aluminum components in the fillet lap joint configuration with consideration to part-to-part gap variation. The approach entails decoupling the welding process parameters in two subsets: (1) in-plane control of the heat input on the upper part to facilitate the droplet formation; and (2) out-of-plane heat management to achieve the desired level of penetration control in the keyhole mode. This paper presents the results of finding the optimal placement of the OCT beam with variable part-to-part gap conditions. Results have shown that statistical signal processing of the raw OCT signal gives insight not only into the depth of the keyhole but can infer the shape of the keyhole itself. Current limitations and next phases of research and development are highlighted based on the experimental study.

Optical coherence measurement-based penetration depth monitoring of stainless steel sheets in laser lap welding using long short-term memory network

In-process Monitoring of Penetration Depth and Contact Width During Laser Lap Welding Using Fused OCT Depth and Coaxial High Speed Image

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

Measurement of pulsed laser welding penetration based on keyhole dynamics and deep learning approach

A penetration depth monitoring method for Al-Cu laser lap welding based on spectral signals

Penetration State Recognition during Laser Welding Process Control Based on Two-Stage Temporal Convolutional Networks

Penetration prediction of narrow-gap laser welding based on coaxial high dynamic range observation and machine learning

A data-driven time-sequence feature-based composite network of time-distributed CNN-LSTM for detecting pore defects in laser penetration welding

Novel approach for weld depth determination using optical coherence tomography measurement in laser deep penetration welding of aluminum and steel

DBN-based online identification of porosity regions during laser welding of aluminum alloys using coherent optical diagnosis

Machine Learning-Based In-Process Monitoring for Laser Deep Penetration Welding: A Survey

Prediction of penetration based on plasma plume and spectrum characteristics in laser welding

An Efficient Method for Laser Welding Depth Determination Using Optical Coherence Tomography

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

Penetration Depth Modeling and Process Parameter Maps for Laser Welds Using Machine Learning

Monitoring of the Weld Pool, Keyhole Morphology and Material Penetration State in Near-Infrared and Blue Composite Laser Welding of Magnesium Alloy

Penetration Monitoring and Control of CO 2 Laser Welding with Coaxial Visual Sensing System

Processing of Keyhole Depth Measurement Data during Laser Beam Micro Welding

Real-time monitoring of high-power disk laser welding statuses based on deep learning framework

Weld Penetration in Situ Prediction from Keyhole Dynamic Behavior under Time-Varying VPPAW Pools Via the OS-ELM Model

Investigating the ability of deep learning to predict Welding Depth and Pore Volume in Hairpin Welding