Physics-constrained deep learninig for solving the Eikonal equation

S. Grubas, G. Loginov, A. Duchkov

Результат исследования: Публикации в книгах, отчётах, сборниках, трудах конференцийстатья в сборнике материалов конференциинаучнаярецензирование

Аннотация

The Eikonal equation is a non-linear PDE that is used for modeling seismic traveltimes. Here we test the idea of using neural networks for solving the 2D Eikonal equation. The concept of the physics-informed neural networks implies including the PDE and boundary conditions into the loss functions. Then no labeled data are required for training the network. While testing this approach we show that it is not sufficient to include only the equation and the boundary condition into the loss function as the training procedure may converge to solutions corresponding to various source terms. We propose supplementing the loss function with additional physics constraint promoting monotonic behavior (time increasing away from the source location). We were testing various neural-network architectures for several inhomogeneous velocity models: with linear vertical gradient, with a smooth high-velocity anomaly, the two-layered models. In the tests, the physics-informed neural network was able to reproduce the behavior of propagating fronts with the mean absolute relative error of about 5 % for all the considered tests. Further development of the training strategy is necessary for further accuracy improvement.

Язык оригиналаанглийский
Название основной публикации82nd EAGE Conference and Exhibition 2021
ИздательEuropean Association of Geoscientists and Engineers, EAGE
Страницы2252-2256
Число страниц5
ISBN (электронное издание)978-171384144-9
СостояниеОпубликовано - 2021
Событие82nd EAGE Conference and Exhibition 2021 - Amsterdam, Virtual, Нидерланды
Продолжительность: 18 окт. 202121 окт. 2021

Серия публикаций

Название82nd EAGE Conference and Exhibition 2021
Том3

Конференция

Конференция82nd EAGE Conference and Exhibition 2021
Страна/TерриторияНидерланды
ГородAmsterdam, Virtual
Период18.10.202121.10.2021

Предметные области OECD FOS+WOS

  • 1.05 НАУКИ О ЗЕМЛЕ И СМЕЖНЫЕ ЭКОЛОГИЧЕСКИЕ НАУКИ

Fingerprint

Подробные сведения о темах исследования «Physics-constrained deep learninig for solving the Eikonal equation». Вместе они формируют уникальный семантический отпечаток (fingerprint).

Цитировать