TY  - JOUR
AU  - Wu, Hao
AU  - Yin, Tianle
PY  - 2026
DA  - 2026/03/11
TI  - Data-Driven RUL Prediction of CMAPSS Jet Engines: A Swarm Intelligence-Optimized Transformer Approach
JO  - Aerospace Engineering Communications
T2  - Aerospace Engineering Communications
JF  - Aerospace Engineering Communications
VL  - 1
IS  - 2
SP  - 57
EP  - 67
DO  - 10.62762/AEC.2026.464396
UR  - https://www.icck.org/article/abs/AEC.2026.464396
KW  - remaining useful life
KW  - prognostics and health management
KW  - CMAPSS
KW  - hyperparameter optimization
AB  - 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.
SN  - pending
PB  - Institute of Central Computation and Knowledge
LA  - English
ER  - 

@article{Wu2026DataDriven,
  author = {Hao Wu and Tianle Yin},
  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},
  url = {https://www.icck.org/article/abs/AEC.2026.464396},
  abstract = {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.},
  keywords = {remaining useful life, prognostics and health management, CMAPSS, hyperparameter optimization},
  issn = {pending},
  publisher = {Institute of Central Computation and Knowledge}
}