Fusion-Based Remaining Useful Life Prediction of SGCMG Considering Measured and Hidden Parameters
IEEE Transactions on Instrumentation and Measurement · 2026 · 75 · 3518833-3518833
DOI: 10.1109/tim.2026.3712930 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Instrumentation and Measurement, 2026. Verfasst von Baoding Liu und 6 weiteren Autor:innen. Die Arbeit nennt 38 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Postdoctoral Research Funding of Hangzhou International Innovation Institute, Beihang University.
Abstract (Originalsprache)
Single-gimbal control moment gyroscopes (SGCMGs) are critical actuators for spacecraft attitude control, and the prediction of their remaining useful life (RUL) is essential for mission reliability. However, accurate RUL prediction is challenging due to the complex electromechanical coupling and the limited information in telemetry data. To address these challenges, this article proposes a fusion-based RUL prediction method that integrates physical knowledge with a deep-learning-based approach. First, an electromechanical coupling model was established to link measured parameters with unobservable system hidden parameters. Next, a two-stage identification strategy combining the nonlinear least squares (NLS) method with a Wiener-process-driven Adaptive extended Kalman filter (EKF) was developed to accurately track the hidden parameters. Finally, a fusion-based deep learning network with a three-phase training strategy was designed. In the proposed network, the adversarial learning was employed to extract operating condition (OC)-independent features from raw measured parameters, and an SE-Attention module was incorporated to effectively fuse these features with the identified hidden parameters. Experimental results demonstrated that the proposed method outperforms other baselines, reducing the RUL prediction RMSE by over 52 % and MAE by over 55 %, achieving a coefficient of determination ( $R^{2}$ ) of 0.98. Ablation studies confirmed the effectiveness of each component within the proposed framework.
Förderung
- National Natural Science Foundation of China
- Postdoctoral Research Funding of Hangzhou International Innovation Institute, Beihang University
Zitieren
APA
Liu, B., Yu, J., Liu, Z., Kong, L., Tang, D., Tian, L., & Han, D. (2026). Fusion-Based Remaining Useful Life Prediction of SGCMG Considering Measured and Hidden Parameters. IEEE Transactions on Instrumentation and Measurement, 75, 3518833-3518833. https://doi.org/10.1109/tim.2026.3712930
Vancouver
Liu B, Yu J, Liu Z, Kong L, Tang D, Tian L, et al. Fusion-Based Remaining Useful Life Prediction of SGCMG Considering Measured and Hidden Parameters. IEEE Transactions on Instrumentation and Measurement. 2026;75:3518833-3518833. doi:10.1109/tim.2026.3712930
BibTeX
@article{Liu2026,
author = {Liu, Baoding and Yu, Jinsong and Liu, Zhigang and Kong, Lingkun and Tang, Diyin and Tian, Limei and Han, Danyang},
title = {Fusion-Based Remaining Useful Life Prediction of SGCMG Considering Measured and Hidden Parameters},
journal = {IEEE Transactions on Instrumentation and Measurement},
year = {2026},
volume = {75},
pages = {3518833--3518833},
doi = {10.1109/tim.2026.3712930}
}
EUVIMED