Solution-Guided Machine Learning for Physical Field Prediction in Complex Geometries
Journal of Computing and Information Science in Engineering · 01.11.2026 · 26 (11)
DOI: 10.1115/1.4072215 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Computing and Information Science in Engineering, online erschienen am 10.07.2026. Verfasst von Yuchen Du und 6 weiteren Autor:innen. Die Arbeit nennt 70 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
The high computational expense associated with conventional numerical solvers and the limited generalizability of purely data-driven machine learning (ML) models present significant challenges for predicting physical fields in complex geometries. To address this limitation, we propose a solution-guided machine learning (SGML) framework that synergistically embeds a physics-informed analytical approximation within a convolutional neural network (CNN). This methodology extracts universal physical features from explicit solutions and generalizes them to arbitrary domains through a boundary distance function, thereby constructing a physics-guided approximate field. This approximation serves as a structured supplementary input to the network, substantially diminishing the learning complexity and enhancing extrapolative capability. Validation on irregular pipe flow demonstrates that the CNN-SGML framework exhibits superior data efficiency, robust generalization, and enables the use of more lightweight network architectures. In contrast to physics-informed neural networks (PINNs), the proposed approach exemplifies a synergistic coupling of data-driven learning and physical principles. Subsequent applications, including stress distribution in an elastic matrix with an Eshelby inclusion, stress concentration around an irregular cavity in a nonlinear material, and flow velocity and pressure fields surrounding an irregular obstacle, confirm the versatility of the method. Consequently, this framework establishes an efficient predictive tool for geometry-sensitive applications, such as fluidic device optimization and composite material design, particularly where high-fidelity simulation data are computationally prohibitive.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Du, Y., Wang, R., Ying, T., Deng, Y., Wu, J., Wang, B., & Chang, T. (2026). Solution-Guided Machine Learning for Physical Field Prediction in Complex Geometries. Journal of Computing and Information Science in Engineering, 26(11). https://doi.org/10.1115/1.4072215
Vancouver
Du Y, Wang R, Ying T, Deng Y, Wu J, Wang B, et al. Solution-Guided Machine Learning for Physical Field Prediction in Complex Geometries. Journal of Computing and Information Science in Engineering. 2026;26(11). doi:10.1115/1.4072215
BibTeX
@article{Du2026,
author = {Du, Yuchen and Wang, Ruijin and Ying, Tianquan and Deng, Yang and Wu, Jianzhao and Wang, Bofu and Chang, Tienchong},
title = {Solution-Guided Machine Learning for Physical Field Prediction in Complex Geometries},
journal = {Journal of Computing and Information Science in Engineering},
year = {2026},
volume = {26},
number = {11},
doi = {10.1115/1.4072215}
}
EUVIMED