A RBF-FD Physics-Informed Machine Learning Approach to Air Pollution Source Estimation

被引：0

作者：

Georges, Didier ^{[1
]}

Leirens, Sylvain ^{[2
]}

Lopez-Ferber, Roman ^{[2
]}

机构：

[1] Univ Grenoble Alpes, Grenoble INP Inst Engn & Management, GIPSA Lab, CNRS, F-38000 Grenoble, France

[2] Univ Grenoble Alpes, Leti, CEA, F-38000 Grenoble, France

来源：

IFAC PAPERSONLINE | 2024年 / 58卷 / 25期

关键词：

air pollution; source term estimation; advection-diffusion PDE; physics-informed machine learning; RBF-FD approximation;

D O I：

10.1016/j.ifacol.2024.10.245

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

In this paper, we propose a source term estimation approach for air pollution monitoring based on a physics-informed machine learning approach using radial basis function-generated finite differences (RBF-FD) approximations, rather than using neural network-based approximations. This approach looks promising for detecting a static pollution source, at a particularly low computing cost and based on a network of fixed or mobile sensors. A 3D case study demonstrates the effectiveness of the approach.

引用

页码：102 / 107

页数：6

共 20 条

[11] Solving 3D magnetohydrostatics with RBF-FD: Applications to the solar corona
Mathews, Nat H.
Flyer, Natasha
Gibson, Sarah E.
[J]. JOURNAL OF COMPUTATIONAL PHYSICS, 2022, 462
[12] QUASI-RANDOM SEQUENCES AND THEIR DISCREPANCIES
MOROKOFF, WJ
CAFLISCH, RE
[J]. SIAM JOURNAL ON SCIENTIFIC COMPUTING, 1994, 15 (06) : 1251 - 1279
[13] Oliver M., 2022, Spatialized epidemiological forecasting applied to Covid
[14] LSQR - AN ALGORITHM FOR SPARSE LINEAR-EQUATIONS AND SPARSE LEAST-SQUARES
PAIGE, CC
SAUNDERS, MA
[J]. ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE, 1982, 8 (01): : 43 - 71
[15] Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
Raissi, M.
Perdikaris, P.
Karniadakis, G. E.
[J]. JOURNAL OF COMPUTATIONAL PHYSICS, 2019, 378 : 686 - 707
[16] Septier F, 2020, ATMOS ENVIRON, V242, DOI [10.1016/j.atmosenv.2020.117793, 10.1016/j.atmosenv.2020.117733]
[17] An unfitted radial basis function generated finite difference method applied to thoracic diaphragm simulations
Tominec, Igor
Villard, Pierre-Frederic
Larsson, Elisabeth
Bayona, Victor
Cacciani, Nicola
[J]. JOURNAL OF COMPUTATIONAL PHYSICS, 2022, 469
[18] An unfitted RBF-FD method in a least-squares setting for elliptic PDEs on complex geometries
Tominec, Igor
Breznik, Eva
[J]. JOURNAL OF COMPUTATIONAL PHYSICS, 2021, 436
[19] World-Health-Organization, 2020, Ambient (Outdoor) Air Pollution
[20] Zannetti P., 1990, Air Pollution Modeling: Theories, Computational Methods and Available Software, DOI [10.1007/978-1-4757-4465-1, DOI 10.1007/978-1-4757-4465-1]

← 1 2 →