Bayesian neural network-optimized single-mode sapphire ring Bragg grating for extreme high-temperature sensing
Optica · 20.08.2026 · 13 (8) · 1630
DOI: 10.1364/optica.603291 ↗Kurz gefasst
Fachartikel in der Zeitschrift Optica, online erschienen am 18.08.2026. Verfasst von Chao He und 2 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 47 Literaturangaben. Gefördert von: National Key Research and Development Program of China, National Natural Science Foundation of China, China Scholarship Council.
Abstract (Originalsprache)
Sapphire fiber Bragg gratings (SFBGs) offer great potential for extreme high-temperature sensing, but multimode-induced spectral complexity remains a major obstacle to stable single-peak interrogation. To address this challenge, a single-mode sapphire ring Bragg grating (SRBG) is developed through Bayesian neural network (BNN)-guided inverse optimization. An L 32 orthogonal design combined with a regression-rounding strategy is used to train the BNN with limited experimental data, establishing an accurate mapping between fabrication parameters and reflection-peak number (R=0.996). Model-driven virtual inference and global sensitivity analysis clarify the nonlinear influences of laser energy, writing speed, ring diameter, and grating length on modal characteristics, and identify the optimal process window for single-mode operation. Three-dimensional beam-propagation simulations confirm effective higher-order-mode suppression and good mode-field matching with standard single-mode fibers. The fabricated SRBG exhibits a high-signal-to-noise single-peak reflection spectrum and a symmetric Gaussian mode profile. High-temperature tests demonstrate excellent repeatability and long-term thermal stability up to 1200°C, with a maximum temperature sensitivity of 32.8 pm/°C. This work not only delivers high-performance sensing devices for extreme-environment monitoring, but also establishes an efficient paradigm for intelligent manufacturing to address multidimensional parameter optimization in complex photonic device fabrication.
Förderung
- National Key Research and Development Program of China
- National Natural Science Foundation of China
- China Scholarship Council
Zitieren
APA
He, C., Wu, J., & Shu, X. (2026). Bayesian neural network-optimized single-mode sapphire ring Bragg grating for extreme high-temperature sensing. Optica, 13(8), 1630. https://doi.org/10.1364/optica.603291
Vancouver
He C, Wu J, Shu X. Bayesian neural network-optimized single-mode sapphire ring Bragg grating for extreme high-temperature sensing. Optica. 2026;13(8):1630. doi:10.1364/optica.603291
BibTeX
@article{He2026,
author = {He, Chao and Wu, Jiaming and Shu, Xuewen},
title = {Bayesian neural network-optimized single-mode sapphire ring Bragg grating for extreme high-temperature sensing},
journal = {Optica},
year = {2026},
volume = {13},
number = {8},
pages = {1630},
doi = {10.1364/optica.603291}
}
EUVIMED