Academic Journal
Single Reference Frequency Loss for Multi-frequency Wavefield Representation using Physics-Informed Neural Networks
| العنوان: | Single Reference Frequency Loss for Multi-frequency Wavefield Representation using Physics-Informed Neural Networks |
|---|---|
| المؤلفون: | Huang, Xinquan, Alkhalifah, Tariq Ali |
| المساهمون: | Earth Science and Engineering Program, Physical Science and Engineering (PSE) Division, Physical science and engineering division, KAUST, Seismic Wave Analysis Group |
| بيانات النشر: | IEEE |
| سنة النشر: | 2022 |
| المجموعة: | King Abdullah University of Science and Technology: KAUST Repository |
| مصطلحات موضوعية: | Physics-informed neural network, partial differential equation, multi-frequency wavefield, single reference frequency loss |
| الوصف: | Physics-informed neural networks (PINNs) can offer approximate multidimensional functional solutions to the Helmholtz equation that are flexible, require low memory, and have no limitations on the shape of the solution space. However, the neural network (NN) training can be costly and the cost dramatically increases as we train for multi-frequency wavefields by adding frequency as an additional input to the NN multi-dimensional function. In this case, the often large variation of the wavefield features (specifically wavelength) with frequency adds more complexity to the NN training. Thus, we propose a new loss function for the NN multidimensional input training that allows us to seamlessly include frequency as a dimension. We specifically utilize the linear relation between frequency and wavenumber (the wavefield space representation) to incorporate a reference frequency scaling to the loss function. As a result, the effective wavenumber of the wavefield solution as a function of frequency remains almost stationary, which reduces the learning burden on the NN function. We demonstrate the effectiveness of this modified loss function on a layered model. ; The authors would like to thank KAUST for its support and the SWAG group for the collaborative environment. This work utilized the resources of the Supercomputing Laboratory at KAUST, and the authors are grateful for that. |
| نوع الوثيقة: | article in journal/newspaper |
| وصف الملف: | application/pdf |
| اللغة: | unknown |
| تدمد: | 1558-0571 |
| Relation: | https://ieeexplore.ieee.org/document/9779736/; https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9779736; Huang, X., & Alkhalifah, T. (2022). Single Reference Frequency Loss for Multi-frequency Wavefield Representation using Physics-Informed Neural Networks. IEEE Geoscience and Remote Sensing Letters, 1–1. https://doi.org/10.1109/lgrs.2022.3176867; IEEE Geoscience and Remote Sensing Letters; http://hdl.handle.net/10754/678177 |
| DOI: | 10.1109/LGRS.2022.3176867 |
| الاتاحة: | http://hdl.handle.net/10754/678177 https://doi.org/10.1109/LGRS.2022.3176867 |
| Rights: | Archived with thanks to IEEE Geoscience and Remote Sensing Letters |
| رقم الانضمام: | edsbas.F967A42F |
| قاعدة البيانات: | BASE |
| تدمد: | 15580571 |
|---|---|
| DOI: | 10.1109/LGRS.2022.3176867 |