EUVIMEDEUVIMED entdecken
Crossref

Data-Driven RUL Prediction of CMAPSS Jet Engines: A Swarm Intelligence-Optimized Transformer Approach

Hao Wu, Tianle Yin

Aerospace Engineering Communications · 11.03.2026 · 1 (2) · 57-67

DOI: 10.62762/aec.2026.464396 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Aerospace Engineering Communications, online erschienen am 11.03.2026. Verfasst von Hao Wu und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Remaining useful life (RUL) prediction is a core task in prognostics and health management. While Transformers excel at modeling long-range temporal dependencies, their performance is highly sensitive to hyperparameters, and improper splitting of sliding-window samples can introduce data leakage. We propose a Sparrow Search Algorithm (SSA)-optimized Transformer for CMAPSS RUL prediction, adopting an engine-wise split for leakage-aware model selection and using validation RMSE as the fitness function to guide SSA-based hyperparameter optimization. On the FD001 test set, the model achieves RMSE $13.79$, MAE $10.00$, $R^2=0.88$, and a NASA score of $356.26$. The prediction curves and residual diagnostics show stable fitting, with only a few large-error cases.

Zitieren

APA

Wu, H., & Yin, T. (2026). Data-Driven RUL Prediction of CMAPSS Jet Engines: A Swarm Intelligence-Optimized Transformer Approach. Aerospace Engineering Communications, 1(2), 57-67. https://doi.org/10.62762/aec.2026.464396

Vancouver

Wu H, Yin T. Data-Driven RUL Prediction of CMAPSS Jet Engines: A Swarm Intelligence-Optimized Transformer Approach. Aerospace Engineering Communications. 2026;1(2):57-67. doi:10.62762/aec.2026.464396

BibTeX
@article{Wu2026,
  author = {Wu, Hao and Yin, Tianle},
  title = {Data-Driven RUL Prediction of CMAPSS Jet Engines: A Swarm Intelligence-Optimized Transformer Approach},
  journal = {Aerospace Engineering Communications},
  year = {2026},
  volume = {1},
  number = {2},
  pages = {57--67},
  doi = {10.62762/aec.2026.464396}
}
Kostenlos starten →