An estimation method of sound speed profile based on grouped dilated convolution informer model
IntroductionThe accurate determination of the ocean sound speed profile (SSP) is essential for oceanographic research and marine engineering. Traditional methods for acquiring SSP data are often time-consuming and costly. Machine learning techniques provide a more efficient alternative for SSP inver...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-02-01
|
| Series: | Frontiers in Marine Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmars.2025.1484098/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832087408810655744 |
|---|---|
| author | Siyuan Qin Yi Zhang Zhou Chen |
| author_facet | Siyuan Qin Yi Zhang Zhou Chen |
| author_sort | Siyuan Qin |
| collection | DOAJ |
| description | IntroductionThe accurate determination of the ocean sound speed profile (SSP) is essential for oceanographic research and marine engineering. Traditional methods for acquiring SSP data are often time-consuming and costly. Machine learning techniques provide a more efficient alternative for SSP inversion, effectively addressing the limitations of conventional approaches.MethodsThis study proposes a novel SSP inversion model based on a grouped dilated convolution (GDC) Informer architecture. By replacing the standard one-dimensional convolution in the Informer model with GDC, the proposed model expands its receptive field and improves computational efficiency. The model was trained using Argo profile data from 2008 to 2017, incorporating empirical orthogonal function (EOF) decomposition data, geographic location, temporal information, and historical SSP data, enabling SSP inversion across diverse regions and time periods.ResultsThe model’s performance was evaluated using mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE) metrics. Experimental results demonstrate that the Informer-GDC model achieves evaluation metrics of 0.355 m/s and 0.611 m/s for MAE, 0.241 m/s and 0.394 m/s for RMSE, and 0.018% and 0.025% for MAPE compared with measured data from 2018.DiscussionCompared to the LSTM and Informer models, the proposed model improves MAE, RMSE, and MAPE by 46.51% and 29.66%, 51.65% and 39.28%, and 51.25% and 37.08%, respectively. These findings highlight the superior accuracy, stability, and efficiency of the Informer-GDC model, marking a significant advancement in SSP inversion methodologies. |
| format | Article |
| id | doaj-art-a506f9cd507d474792b3a9eaba34b86d |
| institution | Kabale University |
| issn | 2296-7745 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Marine Science |
| spelling | doaj-art-a506f9cd507d474792b3a9eaba34b86d2025-02-06T05:21:53ZengFrontiers Media S.A.Frontiers in Marine Science2296-77452025-02-011210.3389/fmars.2025.14840981484098An estimation method of sound speed profile based on grouped dilated convolution informer modelSiyuan Qin0Yi Zhang1Zhou Chen2School of Civil Engineering, Zhengzhou Professional Technical Institute of Electronic & Information, Zhengzhou, ChinaSchool of Marine Sciences, Sun Yat-Sen University, Zhuhai, ChinaCollege of Marine Science and Engineering, Nanjing Normal University, Nanjing, ChinaIntroductionThe accurate determination of the ocean sound speed profile (SSP) is essential for oceanographic research and marine engineering. Traditional methods for acquiring SSP data are often time-consuming and costly. Machine learning techniques provide a more efficient alternative for SSP inversion, effectively addressing the limitations of conventional approaches.MethodsThis study proposes a novel SSP inversion model based on a grouped dilated convolution (GDC) Informer architecture. By replacing the standard one-dimensional convolution in the Informer model with GDC, the proposed model expands its receptive field and improves computational efficiency. The model was trained using Argo profile data from 2008 to 2017, incorporating empirical orthogonal function (EOF) decomposition data, geographic location, temporal information, and historical SSP data, enabling SSP inversion across diverse regions and time periods.ResultsThe model’s performance was evaluated using mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE) metrics. Experimental results demonstrate that the Informer-GDC model achieves evaluation metrics of 0.355 m/s and 0.611 m/s for MAE, 0.241 m/s and 0.394 m/s for RMSE, and 0.018% and 0.025% for MAPE compared with measured data from 2018.DiscussionCompared to the LSTM and Informer models, the proposed model improves MAE, RMSE, and MAPE by 46.51% and 29.66%, 51.65% and 39.28%, and 51.25% and 37.08%, respectively. These findings highlight the superior accuracy, stability, and efficiency of the Informer-GDC model, marking a significant advancement in SSP inversion methodologies.https://www.frontiersin.org/articles/10.3389/fmars.2025.1484098/fullsound speed profileinversiongrouped dilated convolutioninformer modelempirical orthogonal decomposition |
| spellingShingle | Siyuan Qin Yi Zhang Zhou Chen An estimation method of sound speed profile based on grouped dilated convolution informer model Frontiers in Marine Science sound speed profile inversion grouped dilated convolution informer model empirical orthogonal decomposition |
| title | An estimation method of sound speed profile based on grouped dilated convolution informer model |
| title_full | An estimation method of sound speed profile based on grouped dilated convolution informer model |
| title_fullStr | An estimation method of sound speed profile based on grouped dilated convolution informer model |
| title_full_unstemmed | An estimation method of sound speed profile based on grouped dilated convolution informer model |
| title_short | An estimation method of sound speed profile based on grouped dilated convolution informer model |
| title_sort | estimation method of sound speed profile based on grouped dilated convolution informer model |
| topic | sound speed profile inversion grouped dilated convolution informer model empirical orthogonal decomposition |
| url | https://www.frontiersin.org/articles/10.3389/fmars.2025.1484098/full |
| work_keys_str_mv | AT siyuanqin anestimationmethodofsoundspeedprofilebasedongroupeddilatedconvolutioninformermodel AT yizhang anestimationmethodofsoundspeedprofilebasedongroupeddilatedconvolutioninformermodel AT zhouchen anestimationmethodofsoundspeedprofilebasedongroupeddilatedconvolutioninformermodel AT siyuanqin estimationmethodofsoundspeedprofilebasedongroupeddilatedconvolutioninformermodel AT yizhang estimationmethodofsoundspeedprofilebasedongroupeddilatedconvolutioninformermodel AT zhouchen estimationmethodofsoundspeedprofilebasedongroupeddilatedconvolutioninformermodel |