Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note3 min read

A cross-process welding penetration status prediction algorithm based on unsupervised domain adaptation in laser and TIG welding

Source

arXiv

arxiv.org/abs/2606.26078v1

Read the source

Source

Plain-Language Brief

Researchers have developed a new approach to improve the performance of deep learning models in predicting weld penetration status, especially when transferring between welding processes with distinct physical mechanisms. This is achieved through an unsupervised domain adaptation (UDA) framework with a gradual source domain expansion (GSDE) strategy. The method significantly outperforms a supervised baseline in both same-process and cross-process tasks, and it reduces the cost of relabeling for new welding processes.

What the Document Says

  • The research focuses on improving deep learning models for weld penetration state classification.
  • The performance of these models often degrades significantly under domain shift, especially between welding processes with distinct physical mechanisms (e.g., arc-dominated TIG to keyhole-based laser welding).
  • To overcome this limitation, the authors propose an unsupervised domain adaptation (UDA) framework integrated with a gradual source domain expansion (GSDE) strategy.
  • The approach achieves high accuracy in both same-process and cross-process transfer tasks.
  • In same-process settings, it attains average accuracies of 90.65% on TIGFH and 90.72% on LSPS, surpassing a supervised baseline by 35.83% and 38.87%, respectively.
  • In cross-process scenarios, it reaches 80.48% for TIG to Laser and 81.13% for Laser to TIG, improving upon the baseline by 43.39% and 43.40%.
  • UMAP visualizations confirm that the model learns domain-invariant features while maintaining discriminative class boundaries.
  • The method reduces the relabeling cost for new welding processes and enhances the versatility of intelligent monitoring across different welding systems.

Key Concepts Mentioned

  • deep learning
  • weld penetration state classification
  • domain shift
  • unsupervised domain adaptation (UDA)
  • gradual source domain expansion (GSDE)
  • TIG welding
  • laser welding
  • accuracy
  • relabeling cost

Claims Made by the Document

  • The proposed approach achieves high accuracy in both same-process and cross-process transfer tasks.
  • The method outperforms a supervised baseline in both same-process and cross-process tasks.

Evidence Used

  • Average accuracies of 90.65% on TIGFH and 90.72% on LSPS in same-process settings.
  • In cross-process scenarios, it reaches 80.48% for TIG to Laser and 81.13% for Laser to TIG.
  • UMAP visualizations confirm that the model learns domain-invariant features while maintaining discriminative class boundaries.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Supervised deep learning has been widely used for weld penetration state classification; however, its performance often degrades significantly under domain shift, such as when transferring models between welding processes with distinct physical mechanisms:for instance, from arc-dominated tungsten inert gas (TIG) welding to keyhole-based laser welding. To overcome this limitation, we propose an unsupervised domain adaptation (UDA) framework integrated with a gradual source domain expansion (GSDE) strategy. Evaluated on dedicated TIG and laser welding datasets, our approach achieves high accuracy in both same-process and cross-process transfer tasks. Specifically, it attains average accuracies of 90.65% on TIGFH and 90.72% on LSPS in same-process settings, surpassing a supervised baseline by 35.83% and 38.87%, respectively. More notably, in cross-process scenarios, it reaches 80.48% for TIG to Laser and 81.13% for Laser to TIG, improving upon the baseline by 43.39% and 43.40%. UMAP visualizations verify that the model learns domain-invariant features while maintaining discriminative class boundaries. This method considerably lowers the relabeling cost for new welding processes and enhances the versatility of intelligent monitoring across different welding systems.