EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Lightweight Transformer with Corrosion Gating and Physical Embeddings for Pipeline Corrosion Growth Prediction

Fangchao Kang, Zeguang Zhang, Hang Zhang, Guan Chen, Shuqian Shen, Gaoshen Cai, Xiaoqing Lu, Wenkai Chen, Maodong Li

Coatings · 17.07.2026 · 16 (7) · 854

DOI: 10.3390/coatings16070854 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Coatings, online erschienen am 17.07.2026. Verfasst von Fangchao Kang und 8 weiteren Autor:innen. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Guangdong Provincial Education Research, State Key Laboratory Cultivation Base for Gas Geology and Gas Control.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Pipeline corrosion critically threatens the safe operation of chemical industrial park pipeline networks, making accurate corrosion growth prediction essential for preventing catastrophic failures. Mechanistic models assume steady states, which conflict with in-service corrosion dynamics; data-driven approaches, however, presume complete datasets but frequently face missing parameters due to sensor failures and limited samples from costly inspections, increasing the risk of noise and overfitting. In this paper, the TinyTransCorrosion model was proposed, which is a lightweight Transformer-based corrosion growth prediction model specifically designed for small-sample scenarios. A cross-validation residual analysis is employed for data cleaning, while five physical embedding features are constructed to encode domain knowledge and compensate for missing parameters. A compact Transformer encoder containing only 6433 parameters was adopted, and a corrosion gating mechanism along with a classification (CLS) token was introduced to achieve efficient feature interaction. Evaluated on a real-world pipeline inspection dataset with 215 records, TinyTransCorrosion attains an R2 of 0.6411 and an MAE of 0.1608 mm, outperforming nine conventional baseline models, including Mean predictor, Linear Regression, Ridge, SVR-RBF, Random Forest, XGBoost, LSTM, MLP, and CNN. While these results are limited to a single-site dataset and require external validation on independent multi-source data, the proposed lightweight physics-guided architecture demonstrates promising predictive capability for small-sample pipeline corrosion assessment, with model error approaching the metrological limit imposed by field ultrasonic gauge accuracy. It provides an acceptable pathway for prioritizing inspection intervals and optimizing maintenance scheduling in resource-constrained industrial settings.

Förderung

  • National Natural Science Foundation of China
  • Guangdong Provincial Education Research
  • State Key Laboratory Cultivation Base for Gas Geology and Gas Control
  • Science and Technology Plan of Maoming

Zitieren

APA

Kang, F., Zhang, Z., Zhang, H., Chen, G., Shen, S., Cai, G., Lu, X., Chen, W., & Li, M. (2026). A Lightweight Transformer with Corrosion Gating and Physical Embeddings for Pipeline Corrosion Growth Prediction. Coatings, 16(7), 854. https://doi.org/10.3390/coatings16070854

Vancouver

Kang F, Zhang Z, Zhang H, Chen G, Shen S, Cai G, et al. A Lightweight Transformer with Corrosion Gating and Physical Embeddings for Pipeline Corrosion Growth Prediction. Coatings. 2026;16(7):854. doi:10.3390/coatings16070854

BibTeX
@article{Kang2026,
  author = {Kang, Fangchao and Zhang, Zeguang and Zhang, Hang and Chen, Guan and Shen, Shuqian and Cai, Gaoshen and Lu, Xiaoqing and Chen, Wenkai and Li, Maodong},
  title = {A Lightweight Transformer with Corrosion Gating and Physical Embeddings for Pipeline Corrosion Growth Prediction},
  journal = {Coatings},
  year = {2026},
  volume = {16},
  number = {7},
  pages = {854},
  doi = {10.3390/coatings16070854}
}
Kostenlos starten →