Physics informed machine learning based applications for the stability analysis of breakwaters

Research output: Contribution to journal › Journal article › Research › peer-review

Standard

Physics informed machine learning based applications for the stability analysis of breakwaters. / Saha, Susmita; De, Soumen; Changdar, Satyasaran.

In: Ships and Offshore Structures, 2024, p. 1-13.

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Saha, S, De, S & Changdar, S 2024, 'Physics informed machine learning based applications for the stability analysis of breakwaters', Ships and Offshore Structures, pp. 1-13. https://doi.org/10.1080/17445302.2024.2344929

APA

Saha, S., De, S., & Changdar, S. (2024). Physics informed machine learning based applications for the stability analysis of breakwaters. Ships and Offshore Structures, 1-13. https://doi.org/10.1080/17445302.2024.2344929

Vancouver

Saha S, De S, Changdar S. Physics informed machine learning based applications for the stability analysis of breakwaters. Ships and Offshore Structures. 2024;1-13. https://doi.org/10.1080/17445302.2024.2344929

Author

Saha, Susmita ; De, Soumen ; Changdar, Satyasaran. / Physics informed machine learning based applications for the stability analysis of breakwaters. In: Ships and Offshore Structures. 2024 ; pp. 1-13.

Bibtex

@article{01505ab3e1524994b6a538bf64746dae,

title = "Physics informed machine learning based applications for the stability analysis of breakwaters",

abstract = "One of the key aspects in designing and stability analysis of breakwater structures is predicting the stability number of their armour blocks. This study presents a novel approach called physics informed deep neural network, for the stability analysis of rubble-mound breakwaters. The present work makes two main contributions. Firstly, it proposes a method for creating hybrid combinations of theoretical models or physical models and deep neural network architectures, leveraging the advantages of both physics and data. This framework incorporates the output of physics-based simulations and observational features into a hybrid modelling setup. Secondly, the framework employs physics-based loss functions in the learning objective of these deep neural networks, which not only demonstrate lower errors on the training set but also adhere to the established physical relations. The proposed study may have the potential to address the existing limitations in this field and provide better accuracy in estimating the stability number.",

keywords = "breakwaters, deep neural network, Machine learning, physics-informed neural network, stability number",

author = "Susmita Saha and Soumen De and Satyasaran Changdar",

note = "Publisher Copyright: {\textcopyright} 2024 Informa UK Limited, trading as Taylor & Francis Group.",

year = "2024",

doi = "10.1080/17445302.2024.2344929",

language = "English",

pages = "1--13",

journal = "Ships and Offshore Structures",

issn = "1744-5302",

publisher = "Taylor & Francis",

}

RIS

TY - JOUR

T1 - Physics informed machine learning based applications for the stability analysis of breakwaters

AU - Saha, Susmita

AU - De, Soumen

AU - Changdar, Satyasaran

PY - 2024

Y1 - 2024

N2 - One of the key aspects in designing and stability analysis of breakwater structures is predicting the stability number of their armour blocks. This study presents a novel approach called physics informed deep neural network, for the stability analysis of rubble-mound breakwaters. The present work makes two main contributions. Firstly, it proposes a method for creating hybrid combinations of theoretical models or physical models and deep neural network architectures, leveraging the advantages of both physics and data. This framework incorporates the output of physics-based simulations and observational features into a hybrid modelling setup. Secondly, the framework employs physics-based loss functions in the learning objective of these deep neural networks, which not only demonstrate lower errors on the training set but also adhere to the established physical relations. The proposed study may have the potential to address the existing limitations in this field and provide better accuracy in estimating the stability number.

AB - One of the key aspects in designing and stability analysis of breakwater structures is predicting the stability number of their armour blocks. This study presents a novel approach called physics informed deep neural network, for the stability analysis of rubble-mound breakwaters. The present work makes two main contributions. Firstly, it proposes a method for creating hybrid combinations of theoretical models or physical models and deep neural network architectures, leveraging the advantages of both physics and data. This framework incorporates the output of physics-based simulations and observational features into a hybrid modelling setup. Secondly, the framework employs physics-based loss functions in the learning objective of these deep neural networks, which not only demonstrate lower errors on the training set but also adhere to the established physical relations. The proposed study may have the potential to address the existing limitations in this field and provide better accuracy in estimating the stability number.

KW - breakwaters

KW - deep neural network

KW - Machine learning

KW - physics-informed neural network

KW - stability number

UR - http://www.scopus.com/inward/record.url?scp=85191346277&partnerID=8YFLogxK

U2 - 10.1080/17445302.2024.2344929

DO - 10.1080/17445302.2024.2344929

M3 - Journal article

AN - SCOPUS:85191346277

SP - 1

EP - 13

JO - Ships and Offshore Structures

JF - Ships and Offshore Structures

SN - 1744-5302

ER -

ID: 392213104

Forskning