EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Spatiotemporal Transformer–ConvLSTM Surrogate with Condition Embedding for Dynamic Three-Dimensional Temperature Field Prediction of XLPE HVDC Cables

Kang-Jie Ruan, Jia-Xun He, Ya Zhang, Hai-Yan Yang, Ming-Liang Cheng, Fan Xu, Shi-Li Liu, Wei Lu

Automation · 05.11.2026 · 7 (6) · 180

DOI: 10.3390/automation7060180 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Automation, online erschienen am 05.11.2026. Verfasst von Kang-Jie Ruan und 7 weiteren Autor:innen. Die Arbeit nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Science and Technology Project of China Electrical Equipment Group Co., Ltd..

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The conductor temperature of a high-voltage direct-current (HVDC) cable limits its current-carrying capacity and governs the aging of the cross-linked polyethylene (XLPE) insulation. Transient finite-element (FE) simulation provides the full temperature field but is too expensive for online rating and digital-twin applications. This paper presents an autoregressive deep-learning surrogate for the three-dimensional cable temperature field. Exploiting axisymmetry, the network predicts the next field on a two-dimensional radial–axial grid from the previous H frames and the operating condition, and the three-dimensional field is reconstructed by revolution. ConvLSTM layers capture spatiotemporal correlations, a temporal Transformer fuses the history states, and a continuous condition embedding lets one model cover the operating space. A scaled residual-increment formulation and a rollout curriculum stabilize the 600 min closed-loop prediction. On a COMSOL dataset of 49 operating conditions (ambient temperature −20 to 40 °C, load current 0 to 1200 A), the surrogate achieves RMSEs of 0.28, 0.27, and 0.49 °C for training, interpolation, and extrapolation conditions, the latter being held-out corner combinations within the sweep ranges. It reduces the wall-clock time from about 110 s per FE simulation to 1.5 s per case (73× speed-up; 139× batched) and achieves lower mean RMSE than three baseline models.

Förderung

  • Science and Technology Project of China Electrical Equipment Group Co., Ltd.

Zitieren

APA

Ruan, K., He, J., Zhang, Y., Yang, H., Cheng, M., Xu, F., Liu, S., & Lu, W. (2026). A Spatiotemporal Transformer–ConvLSTM Surrogate with Condition Embedding for Dynamic Three-Dimensional Temperature Field Prediction of XLPE HVDC Cables. Automation, 7(6), 180. https://doi.org/10.3390/automation7060180

Vancouver

Ruan K, He J, Zhang Y, Yang H, Cheng M, Xu F, et al. A Spatiotemporal Transformer–ConvLSTM Surrogate with Condition Embedding for Dynamic Three-Dimensional Temperature Field Prediction of XLPE HVDC Cables. Automation. 2026;7(6):180. doi:10.3390/automation7060180

BibTeX
@article{Ruan2026,
  author = {Ruan, Kang-Jie and He, Jia-Xun and Zhang, Ya and Yang, Hai-Yan and Cheng, Ming-Liang and Xu, Fan and Liu, Shi-Li and Lu, Wei},
  title = {A Spatiotemporal Transformer–ConvLSTM Surrogate with Condition Embedding for Dynamic Three-Dimensional Temperature Field Prediction of XLPE HVDC Cables},
  journal = {Automation},
  year = {2026},
  volume = {7},
  number = {6},
  pages = {180},
  doi = {10.3390/automation7060180}
}
Kostenlos starten →