---
_id: '66541'
abstract:
- lang: eng
  text: Expulsion in resistance spot welding (RSW) causes weld quality fluctuations
    and increases quality-control effort in high-volume manufacturing. Existing data-driven
    studies have mainly addressed post-occurrence expulsion detection, process-end
    classification, or the identification of influencing factors, whereas online monitoring
    requires short-term risk estimation before the event occurs. In this study, expulsion
    prediction is formulated as a sliding-window-based pre-expulsion risk estimation
    task for the currently welded spot. A physics-guided hybrid GRU-XGBoost ensemble
    is developed to combine temporal learning from dynamic resistance and electrode-force
    signals with process-physics-related scalar features describing heat input, resistance
    state, and force response. The framework was evaluated on 2730 valid welds, including
    588 expulsion and 2142 non-expulsion welds, using weld-grouped five-fold cross-validation
    with fold-level working-point selection. The ensemble achieved an area under the
    ROC curve of 0.945 ± 0.004 and a weld-level recall of 90.6 ± 3.7% at an average
    false alarm rate of 9.8 ± 0.2%, outperforming both individual branches. For the
    533 correctly warned expulsion welds, the median early-warning lead time was 56
    ms. These results indicate that online, physically interpretable pre-expulsion
    risk prediction is feasible under low-false-alarm constraints within the investigated
    RSW configuration and provide a basis for future adaptive monitoring and control
    studies.
article_type: original
author:
- first_name: Keke
  full_name: Yang, Keke
  id: '65085'
  last_name: Yang
  orcid: 0000-0001-9201-9304
- first_name: Chong
  full_name: Li, Chong
  last_name: Li
- first_name: Robert
  full_name: Beck, Robert
  id: '38279'
  last_name: Beck
  orcid: 0000-0001-9056-4528
- first_name: David
  full_name: Hein, David
  id: '7728'
  last_name: Hein
- first_name: Gerson
  full_name: Meschut, Gerson
  id: '32056'
  last_name: Meschut
  orcid: 0000-0002-2763-1246
citation:
  ama: Yang K, Li C, Beck R, Hein D, Meschut G. A physics-guided hybrid framework
    for online pre-expulsion prediction in resistance spot welding. <i>Journal of
    Manufacturing Processes</i>. 2026;174:135-153. doi:<a href="https://doi.org/10.1016/j.jmapro.2026.07.042">10.1016/j.jmapro.2026.07.042</a>
  apa: Yang, K., Li, C., Beck, R., Hein, D., &#38; Meschut, G. (2026). A physics-guided
    hybrid framework for online pre-expulsion prediction in resistance spot welding.
    <i>Journal of Manufacturing Processes</i>, <i>174</i>, 135–153. <a href="https://doi.org/10.1016/j.jmapro.2026.07.042">https://doi.org/10.1016/j.jmapro.2026.07.042</a>
  bibtex: '@article{Yang_Li_Beck_Hein_Meschut_2026, title={A physics-guided hybrid
    framework for online pre-expulsion prediction in resistance spot welding}, volume={174},
    DOI={<a href="https://doi.org/10.1016/j.jmapro.2026.07.042">10.1016/j.jmapro.2026.07.042</a>},
    journal={Journal of Manufacturing Processes}, publisher={Elsevier BV}, author={Yang,
    Keke and Li, Chong and Beck, Robert and Hein, David and Meschut, Gerson}, year={2026},
    pages={135–153} }'
  chicago: 'Yang, Keke, Chong Li, Robert Beck, David Hein, and Gerson Meschut. “A
    Physics-Guided Hybrid Framework for Online Pre-Expulsion Prediction in Resistance
    Spot Welding.” <i>Journal of Manufacturing Processes</i> 174 (2026): 135–53. <a
    href="https://doi.org/10.1016/j.jmapro.2026.07.042">https://doi.org/10.1016/j.jmapro.2026.07.042</a>.'
  ieee: 'K. Yang, C. Li, R. Beck, D. Hein, and G. Meschut, “A physics-guided hybrid
    framework for online pre-expulsion prediction in resistance spot welding,” <i>Journal
    of Manufacturing Processes</i>, vol. 174, pp. 135–153, 2026, doi: <a href="https://doi.org/10.1016/j.jmapro.2026.07.042">10.1016/j.jmapro.2026.07.042</a>.'
  mla: Yang, Keke, et al. “A Physics-Guided Hybrid Framework for Online Pre-Expulsion
    Prediction in Resistance Spot Welding.” <i>Journal of Manufacturing Processes</i>,
    vol. 174, Elsevier BV, 2026, pp. 135–53, doi:<a href="https://doi.org/10.1016/j.jmapro.2026.07.042">10.1016/j.jmapro.2026.07.042</a>.
  short: K. Yang, C. Li, R. Beck, D. Hein, G. Meschut, Journal of Manufacturing Processes
    174 (2026) 135–153.
date_created: 2026-07-18T12:03:15Z
date_updated: 2026-07-18T12:07:50Z
ddc:
- '600'
department:
- _id: '157'
doi: 10.1016/j.jmapro.2026.07.042
file:
- access_level: closed
  content_type: application/pdf
  creator: kekeyang
  date_created: 2026-07-18T12:04:02Z
  date_updated: 2026-07-18T12:04:02Z
  file_id: '66542'
  file_name: 1-s2.0-S1526612526007012-main.pdf
  file_size: 10864983
  relation: main_file
  success: 1
file_date_updated: 2026-07-18T12:04:02Z
has_accepted_license: '1'
intvolume: '       174'
keyword:
- Resistance spot welding
- Expulsion prediction
- Physics-guided machine learning
- Hybrid ensemble modelling
- Process monitoring
language:
- iso: eng
main_file_link:
- open_access: '1'
oa: '1'
page: 135-153
publication: Journal of Manufacturing Processes
publication_identifier:
  issn:
  - 1526-6125
publication_status: published
publisher: Elsevier BV
quality_controlled: '1'
status: public
title: A physics-guided hybrid framework for online pre-expulsion prediction in resistance
  spot welding
type: journal_article
user_id: '65085'
volume: 174
year: '2026'
...