Reduced Order Modeling (ROM) based method for the two-dimensional water exit problem using snapshot Proper Orthogonal Decomposition (POD) and CFD simulations
Reduced Order Modeling (ROM) based method for the two-dimensional water exit problem using snapshot Proper Orthogonal Decomposition (POD) and CFD simulations
In this paper, the concepts of snapshot Proper Orthogonal Decomposition (POD) and Reduced Order Modeling (ROM) are combined (referred to the POD-ROM method) to solve the two-dimensional (2D) water exit problem. Attention is paid to the pressure distribution along the wetted surface of the body. Computational Fluid Dynamics (CFD) simulations are employed to obtain high-fidelity data on pressure distribution. After applying snapshot POD, it is found that two POD basis modes for the wedge model and three modes for the ship section model are adequate to capture dynamic features of the pressure distribution without losing too much detail. It can also be observed that neither the body motion state nor the initial immersion condition influences all POD functions of the wedge model, but the temporal POD functions of the ship section model are significantly dependent on the initial immersion height. A group of empirical formulae is provided to deal with this issue. The validity and reliability of our POD-ROM method are assessed by investigating water exit cases with both constant and time-varying body accelerations. In this context, after deriving POD functions of any given 2D body from a single CFD simulation, predictions of the pressure distribution along the body can be facilitated for further water exit cases.
Computational fluid dynamics, Reduced order modeling, Snapshot POD, Water exit
Sui, Xupeng
b4651a2c-d0f9-492f-b7fb-e8214a6f1168
Djidjeli, Kamal
94ac4002-4170-495b-a443-74fde3b92998
Sun, Zhe
80e5a9b0-493f-4b95-85ea-1813bb57f022
Xing, Jing T
d4fe7ae0-2668-422a-8d89-9e66527835ce
9 July 2025
Sui, Xupeng
b4651a2c-d0f9-492f-b7fb-e8214a6f1168
Djidjeli, Kamal
94ac4002-4170-495b-a443-74fde3b92998
Sun, Zhe
80e5a9b0-493f-4b95-85ea-1813bb57f022
Xing, Jing T
d4fe7ae0-2668-422a-8d89-9e66527835ce
Sui, Xupeng, Djidjeli, Kamal, Sun, Zhe and Xing, Jing T
(2025)
Reduced Order Modeling (ROM) based method for the two-dimensional water exit problem using snapshot Proper Orthogonal Decomposition (POD) and CFD simulations.
Applied Ocean Research, 161, [104697].
(doi:10.1016/j.apor.2025.104697).
Abstract
In this paper, the concepts of snapshot Proper Orthogonal Decomposition (POD) and Reduced Order Modeling (ROM) are combined (referred to the POD-ROM method) to solve the two-dimensional (2D) water exit problem. Attention is paid to the pressure distribution along the wetted surface of the body. Computational Fluid Dynamics (CFD) simulations are employed to obtain high-fidelity data on pressure distribution. After applying snapshot POD, it is found that two POD basis modes for the wedge model and three modes for the ship section model are adequate to capture dynamic features of the pressure distribution without losing too much detail. It can also be observed that neither the body motion state nor the initial immersion condition influences all POD functions of the wedge model, but the temporal POD functions of the ship section model are significantly dependent on the initial immersion height. A group of empirical formulae is provided to deal with this issue. The validity and reliability of our POD-ROM method are assessed by investigating water exit cases with both constant and time-varying body accelerations. In this context, after deriving POD functions of any given 2D body from a single CFD simulation, predictions of the pressure distribution along the body can be facilitated for further water exit cases.
Text
1-s2.0-S0141118725002834-main
- Version of Record
More information
Accepted/In Press date: 2 July 2025
e-pub ahead of print date: 9 July 2025
Published date: 9 July 2025
Keywords:
Computational fluid dynamics, Reduced order modeling, Snapshot POD, Water exit
Identifiers
Local EPrints ID: 504560
URI: http://eprints.soton.ac.uk/id/eprint/504560
ISSN: 0141-1187
PURE UUID: 9e26d2d9-7db3-4528-a9c9-8828ab1f2d91
Catalogue record
Date deposited: 15 Sep 2025 16:42
Last modified: 17 Sep 2025 16:50
Export record
Altmetrics
Contributors
Author:
Xupeng Sui
Author:
Zhe Sun
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics