Data-driven nondegenerate bound-state solitons of multicomponent Bose–Einstein condensates via mix-training PINN
In this paper, by modifying loss function MSE and training area of the physics-informed neural network (PINN), we proposed two neural network models: mix-training PINN and prior information mix-training PINN. We demonstrated the advantages of these models by simulating nondegenerate bound-state soli...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2023-09-01
|
Series: | Results in Physics |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S2211379723006356 |
_version_ | 1797682835353501696 |
---|---|
author | Shifang Tian Chenchen Cao Biao Li |
author_facet | Shifang Tian Chenchen Cao Biao Li |
author_sort | Shifang Tian |
collection | DOAJ |
description | In this paper, by modifying loss function MSE and training area of the physics-informed neural network (PINN), we proposed two neural network models: mix-training PINN and prior information mix-training PINN. We demonstrated the advantages of these models by simulating nondegenerate bound-state solitons (NDBSSs) of multicomponent Bose–Einstein condensates (BECs). Numerical experiments showed that our proposed models are not only simulate the NDBSSs of multicomponent BECs, but also significantly improve the simulation capability. Compared with original PINN, the prediction accuracy of our proposed models are improved by one to three orders of magnitude. By testing the inverse problem of multicomponent BECs, it is also proved that these models have good performance. |
first_indexed | 2024-03-12T00:06:36Z |
format | Article |
id | doaj.art-d98dcbfabcec41b6be10bf680d8f5ba0 |
institution | Directory Open Access Journal |
issn | 2211-3797 |
language | English |
last_indexed | 2024-03-12T00:06:36Z |
publishDate | 2023-09-01 |
publisher | Elsevier |
record_format | Article |
series | Results in Physics |
spelling | doaj.art-d98dcbfabcec41b6be10bf680d8f5ba02023-09-17T04:56:27ZengElsevierResults in Physics2211-37972023-09-0152106842Data-driven nondegenerate bound-state solitons of multicomponent Bose–Einstein condensates via mix-training PINNShifang Tian0Chenchen Cao1Biao Li2School of Mathematics and Statistics, Ningbo University, Ningbo 315211, PR ChinaCorresponding author.; School of Mathematics and Statistics, Ningbo University, Ningbo 315211, PR ChinaSchool of Mathematics and Statistics, Ningbo University, Ningbo 315211, PR ChinaIn this paper, by modifying loss function MSE and training area of the physics-informed neural network (PINN), we proposed two neural network models: mix-training PINN and prior information mix-training PINN. We demonstrated the advantages of these models by simulating nondegenerate bound-state solitons (NDBSSs) of multicomponent Bose–Einstein condensates (BECs). Numerical experiments showed that our proposed models are not only simulate the NDBSSs of multicomponent BECs, but also significantly improve the simulation capability. Compared with original PINN, the prediction accuracy of our proposed models are improved by one to three orders of magnitude. By testing the inverse problem of multicomponent BECs, it is also proved that these models have good performance.http://www.sciencedirect.com/science/article/pii/S2211379723006356Physics-informed neural networkMix-trainingPrior informationNondegenerate bound-state solitonsMulticomponent Bose–Einstein condensates |
spellingShingle | Shifang Tian Chenchen Cao Biao Li Data-driven nondegenerate bound-state solitons of multicomponent Bose–Einstein condensates via mix-training PINN Results in Physics Physics-informed neural network Mix-training Prior information Nondegenerate bound-state solitons Multicomponent Bose–Einstein condensates |
title | Data-driven nondegenerate bound-state solitons of multicomponent Bose–Einstein condensates via mix-training PINN |
title_full | Data-driven nondegenerate bound-state solitons of multicomponent Bose–Einstein condensates via mix-training PINN |
title_fullStr | Data-driven nondegenerate bound-state solitons of multicomponent Bose–Einstein condensates via mix-training PINN |
title_full_unstemmed | Data-driven nondegenerate bound-state solitons of multicomponent Bose–Einstein condensates via mix-training PINN |
title_short | Data-driven nondegenerate bound-state solitons of multicomponent Bose–Einstein condensates via mix-training PINN |
title_sort | data driven nondegenerate bound state solitons of multicomponent bose einstein condensates via mix training pinn |
topic | Physics-informed neural network Mix-training Prior information Nondegenerate bound-state solitons Multicomponent Bose–Einstein condensates |
url | http://www.sciencedirect.com/science/article/pii/S2211379723006356 |
work_keys_str_mv | AT shifangtian datadrivennondegenerateboundstatesolitonsofmulticomponentboseeinsteincondensatesviamixtrainingpinn AT chenchencao datadrivennondegenerateboundstatesolitonsofmulticomponentboseeinsteincondensatesviamixtrainingpinn AT biaoli datadrivennondegenerateboundstatesolitonsofmulticomponentboseeinsteincondensatesviamixtrainingpinn |