Intelligent wave height prediction based on multitrunk fusion neural networks with bidirectional long short-term memory network DOI Creative Commons

Chenyue Fan,

Xingyi La,

Franck Aurel Likeufack Mdemaya

и другие.

Physics of Fluids, Год журнала: 2025, Номер 37(4)

Опубликована: Апрель 1, 2025

This article introduces a multi-backbone neural network fusion model that integrates Bidirectional Long Short-Term Memory with residual connections, while incorporating convolutional layers, fully connected and attention mechanisms. The uses real-world marine environment data as its primary input, analyzes the nonlinear relationships between different backbones through framework, jointly predicts future changes in ocean wave height. By training model, predicted values demonstrate minimal error range compared to actual values, highlighting strong predictive capabilities accurate results. Furthermore, of models offers novel optimization approach for predicting environments. proposed provides innovative solutions forecasting simulated intelligently ship motion responses.

Язык: Английский

CFD Study of Submarine Hydrodynamics near the Free Surface in Snorkel Conditions DOI Open Access
Doojin Jung, Sunho Park

Water, Год журнала: 2025, Номер 17(5), С. 734 - 734

Опубликована: Март 3, 2025

Submarines are primarily designed for optimal performance while operating submerged, as they spend the majority of their operational time below free surface. However, also navigate at various depths near surface, such during snorkel conditions or other shallow-water operations. Under conditions, sail depth decreases and distance between surface top hull is reduced, a suction effect occurs, inducing an upward force on submarine. Consequently, comprehensive assessment hydrodynamic forces different speeds essential design phase to ensure stability optimization. In this study, computational fluid dynamic (CFD) simulations were performed analyze heave surge acting generic Joubert BB2 (BB2) The computed forces, well pitch moment, validated against experimental data, showing discrepancies within approximately 12%. influence these moments was investigated, demonstrating trends consistent with both measurements numerical predictions. These findings confirm that CFD serve reliable tool predicting free-surface effects submarines, offering valuable insights process.

Язык: Английский

Процитировано

0
Intelligent wave height prediction based on multitrunk fusion neural networks with bidirectional long short-term memory network DOI Creative Commons

Chenyue Fan,

Xingyi La,

Franck Aurel Likeufack Mdemaya

и другие.

Physics of Fluids, Год журнала: 2025, Номер 37(4)

Опубликована: Апрель 1, 2025

This article introduces a multi-backbone neural network fusion model that integrates Bidirectional Long Short-Term Memory with residual connections, while incorporating convolutional layers, fully connected and attention mechanisms. The uses real-world marine environment data as its primary input, analyzes the nonlinear relationships between different backbones through framework, jointly predicts future changes in ocean wave height. By training model, predicted values demonstrate minimal error range compared to actual values, highlighting strong predictive capabilities accurate results. Furthermore, of models offers novel optimization approach for predicting environments. proposed provides innovative solutions forecasting simulated intelligently ship motion responses.

Язык: Английский

Процитировано

0