Sparse-Sensor Three-Dimensional Thermal-State Reconstruction for Black Tea Fermentation Using a CFD-Prior-Constrained Physics-Informed Neural Network
Fermentation · 24.07.2026 · 12 (8) · 345
DOI: 10.3390/fermentation12080345 ↗Kurz gefasst
Fachartikel in der Zeitschrift Fermentation, online erschienen am 24.07.2026. Verfasst von Yingjie Liang und 3 weiteren Autor:innen. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Hubei Provincial Department of Education, Hubei University for Nationalities.
Abstract (Originalsprache)
Internal temperature distributions in black tea fermentation regulate enzymatic oxidation, heat accumulation and fermentation uniformity, but continuous three-dimensional measurements remain difficult in practical processing. We developed a CFD-prior-constrained physics-informed neural network (PINN + CFD) to reconstruct the three-dimensional thermal state of a 1.35 m × 0.96 m × 0.08 m fermentation bed under sparse sensing. Nine sensors at z = 0.04 m were used for training, and six held-out depth-wise sensors at z = 0.02 m and z = 0.06 m were reserved for depth-wise validation. The model integrated measured temperatures, transient heat-transfer physics, convective boundary conditions and a CFD-derived volumetric soft spatial prior, which guided spatial extrapolation rather than serving as ground-truth temperature data. Although the multilayer perceptron achieved the lowest fitting error at the instrumented z = 0.04 m plane, PINN + CFD showed better depth-wise extrapolation, with RMSEs of 0.128, 0.129 and 0.296 °C across the three stages. However, its advantage was stage- and validation-target-dependent: the baseline PINN was slightly better in part of the dynamic-stage validation, and standalone CFD had the lowest surface infrared error in the constant-temperature stage, indicating that PINN + CFD mainly improved spatial extrapolation rather than uniformly minimizing all error metrics. The inferred apparent process-level heat-source index Qreact(t) varied continuously, and its cumulative trajectory showed a descriptive association with cumulative polyphenol loss. These results indicate that PINN + CFD enables physically consistent thermal-state reconstruction within the tested sparsely instrumented black tea fermentation bed.
Förderung
- Hubei Provincial Department of Education
- Hubei University for Nationalities
Zitieren
APA
Liang, Y., Li, W., Liu, C., & Xie, Z. (2026). Sparse-Sensor Three-Dimensional Thermal-State Reconstruction for Black Tea Fermentation Using a CFD-Prior-Constrained Physics-Informed Neural Network. Fermentation, 12(8), 345. https://doi.org/10.3390/fermentation12080345
Vancouver
Liang Y, Li W, Liu C, Xie Z. Sparse-Sensor Three-Dimensional Thermal-State Reconstruction for Black Tea Fermentation Using a CFD-Prior-Constrained Physics-Informed Neural Network. Fermentation. 2026;12(8):345. doi:10.3390/fermentation12080345
BibTeX
@article{Liang2026,
author = {Liang, Yingjie and Li, Weicheng and Liu, Chuangye and Xie, Zhiyin},
title = {Sparse-Sensor Three-Dimensional Thermal-State Reconstruction for Black Tea Fermentation Using a CFD-Prior-Constrained Physics-Informed Neural Network},
journal = {Fermentation},
year = {2026},
volume = {12},
number = {8},
pages = {345},
doi = {10.3390/fermentation12080345}
}
EUVIMED