What question did this study set out to answer?

To develop a multimodal deep learning framework for aircraft engine prognostics that incorporates uncertainty quantification.

Multimodal spatiotemporal graph attention network with uncertainty quantification for aircraft engine prognostics

View Full Paper

Ask AI

Bookmark

Authors

Randomized trial demonstrates uncertainty-aware prognostics in aircraft engines, indicating improved fleet safety management.

To develop a multimodal deep learning framework for aircraft engine prognostics that incorporates uncertainty quantification.
Introduced the Multimodal Spatiotemporal Graph Attention Network with Uncertainty Quantification (MST-GAT-UQ).
Integrated CNN-LSTM for temporal patterns and graph attention mechanisms for spatial correlations.
Validated the model on an operational dataset of 84 sensors with 43,492 samples, focusing on RUL and health state predictions.
Achieved R^2 = 0.911, RMSE = 0.087, MAE = 0.072 for normalized RUL scale predictions.
Synthetic health-state replication accuracy of 99.5% confirmed via auxiliary tasks.
Demonstrated positive R^2 across benchmark datasets with native C-MAPSS training (R^2 = 0.607–0.863).

Sławomir Szrama (2026) studied this question.

View Full Paper

Ask AI

Bookmark