DNN modeling of partial differential equations with incomplete data

被引：2

作者：

Churchill, Victor ^{[1
]}

Chen, Yuan ^{[2
]}

Xu, Zhongshu ^{[2
]}

Xiu, Dongbin ^{[2
]}

机构：

[1] Trinity Coll, Dept Math, Hartford, CT 06106 USA

[2] Ohio State Univ, Dept Math, Columbus, OH 43210 USA

来源：

JOURNAL OF COMPUTATIONAL PHYSICS | 2023年 / 493卷

关键词：

Data driven modeling; Deep neural networks; Reduced PDE systems; Incomplete data; UNIVERSAL APPROXIMATION; GOVERNING EQUATIONS; DYNAMICS;

D O I：

10.1016/j.jcp.2023.112502

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

We present a computational technique for modeling the evolution of partial differential equations (PDEs) with incomplete data. It is a significant extension of the recent work of data driven learning of PDEs, in the sense that we consider two forms of partial data: data are observed only on a subset of the domain, and data are observed only on a subset of the state variables. Both cases resemble more realistic data collection scenarios in real -world applications. Leveraging the recent work on modeling partially-observed dynamical systems, we present a deep neural network (DNN) structure that is suitable for PDE modeling with such kinds of incomplete data. In addition to the mathematical motivation for the DNN structure, we present an extensive set of numerical examples in both one -and two-dimensions to demonstrate the effectiveness of the proposed DNN modeling. In one example, the method can accurately predict the solution when data are only available in less than half (40%) of the domain.(c) 2023 Elsevier Inc. All rights reserved.

引用

页数：20

共 34 条

[21]

Raissi M., 2017, arXiv

[22]

Raissi M, 2018, Arxiv, DOI arXiv:1801.01236

[23]

Raissi M, 2017, Arxiv, DOI [arXiv:1711.10561, DOI 10.48550/ARXIV.1711.10561]

[24]

Raissi M, 2018, J MACH LEARN RES, V19

[25] Deep learning of dynamics and signal-noise decomposition with time-stepping constraints [J].