A comparison of regression models for the ice loads measured during the ice tank test
To evaluate the time-domain positioning performance of arctic marine structures, it is necessary to generate an ice load appropriate for the current position and heading of the structure. The position and orientation angle of a floating body continuously change with time. Therefore, an ice load is r...
Published in: | Brodogradnja |
---|---|
Main Authors: | , , , |
Format: | Article in Journal/Newspaper |
Language: | English Croatian |
Published: |
Faculty of Mechanical Engineering and Naval Architecture
2023
|
Subjects: | |
Online Access: | https://doi.org/10.21278/brod74301 https://doaj.org/article/208a2a4a768d4efda4913aa2a42b533a |
id |
ftdoajarticles:oai:doaj.org/article:208a2a4a768d4efda4913aa2a42b533a |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:208a2a4a768d4efda4913aa2a42b533a 2023-08-20T04:04:41+02:00 A comparison of regression models for the ice loads measured during the ice tank test Seung Jae Lee Kwang Hyo Jung Namkug Ku Jaeyong Lee 2023-01-01T00:00:00Z https://doi.org/10.21278/brod74301 https://doaj.org/article/208a2a4a768d4efda4913aa2a42b533a EN HR eng hrv Faculty of Mechanical Engineering and Naval Architecture https://hrcak.srce.hr/file/434503 https://doaj.org/toc/0007-215X https://doaj.org/toc/1845-5859 0007-215X 1845-5859 doi:10.21278/brod74301 https://doaj.org/article/208a2a4a768d4efda4913aa2a42b533a Brodogradnja, Vol 74, Iss 3, Pp 1-15 (2023) machine learning regression ice load power spectral density mean squared error Naval architecture. Shipbuilding. Marine engineering VM1-989 article 2023 ftdoajarticles https://doi.org/10.21278/brod74301 2023-07-30T00:39:06Z To evaluate the time-domain positioning performance of arctic marine structures, it is necessary to generate an ice load appropriate for the current position and heading of the structure. The position and orientation angle of a floating body continuously change with time. Therefore, an ice load is required for any attitude in the time-domain simulation. In this study, we present a fundamental technique for analyzing ice loads in the frequency domain based on data measured at various angles in the ice-water tank experiment. We perform spectral analysis instead of general FFT to analyze the ice load, which has the characteristics of a random signal. To generate the necessary ice load in the time domain, we must first interpolate the measured data in the frequency domain. Using the Blackman-Tukey method, we estimate the spectrum for the measured data, then process the data to generate the training set required for machine learning. Based on the results, we perform regression analysis by applying four representative techniques, including linear regression, random forest, or neural network, and compare the results with MSE. The deep neural network method performed best, but we provide further discussion for each model. Article in Journal/Newspaper Arctic Directory of Open Access Journals: DOAJ Articles Arctic Brodogradnja 74 3 1 15 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English Croatian |
topic |
machine learning regression ice load power spectral density mean squared error Naval architecture. Shipbuilding. Marine engineering VM1-989 |
spellingShingle |
machine learning regression ice load power spectral density mean squared error Naval architecture. Shipbuilding. Marine engineering VM1-989 Seung Jae Lee Kwang Hyo Jung Namkug Ku Jaeyong Lee A comparison of regression models for the ice loads measured during the ice tank test |
topic_facet |
machine learning regression ice load power spectral density mean squared error Naval architecture. Shipbuilding. Marine engineering VM1-989 |
description |
To evaluate the time-domain positioning performance of arctic marine structures, it is necessary to generate an ice load appropriate for the current position and heading of the structure. The position and orientation angle of a floating body continuously change with time. Therefore, an ice load is required for any attitude in the time-domain simulation. In this study, we present a fundamental technique for analyzing ice loads in the frequency domain based on data measured at various angles in the ice-water tank experiment. We perform spectral analysis instead of general FFT to analyze the ice load, which has the characteristics of a random signal. To generate the necessary ice load in the time domain, we must first interpolate the measured data in the frequency domain. Using the Blackman-Tukey method, we estimate the spectrum for the measured data, then process the data to generate the training set required for machine learning. Based on the results, we perform regression analysis by applying four representative techniques, including linear regression, random forest, or neural network, and compare the results with MSE. The deep neural network method performed best, but we provide further discussion for each model. |
format |
Article in Journal/Newspaper |
author |
Seung Jae Lee Kwang Hyo Jung Namkug Ku Jaeyong Lee |
author_facet |
Seung Jae Lee Kwang Hyo Jung Namkug Ku Jaeyong Lee |
author_sort |
Seung Jae Lee |
title |
A comparison of regression models for the ice loads measured during the ice tank test |
title_short |
A comparison of regression models for the ice loads measured during the ice tank test |
title_full |
A comparison of regression models for the ice loads measured during the ice tank test |
title_fullStr |
A comparison of regression models for the ice loads measured during the ice tank test |
title_full_unstemmed |
A comparison of regression models for the ice loads measured during the ice tank test |
title_sort |
comparison of regression models for the ice loads measured during the ice tank test |
publisher |
Faculty of Mechanical Engineering and Naval Architecture |
publishDate |
2023 |
url |
https://doi.org/10.21278/brod74301 https://doaj.org/article/208a2a4a768d4efda4913aa2a42b533a |
geographic |
Arctic |
geographic_facet |
Arctic |
genre |
Arctic |
genre_facet |
Arctic |
op_source |
Brodogradnja, Vol 74, Iss 3, Pp 1-15 (2023) |
op_relation |
https://hrcak.srce.hr/file/434503 https://doaj.org/toc/0007-215X https://doaj.org/toc/1845-5859 0007-215X 1845-5859 doi:10.21278/brod74301 https://doaj.org/article/208a2a4a768d4efda4913aa2a42b533a |
op_doi |
https://doi.org/10.21278/brod74301 |
container_title |
Brodogradnja |
container_volume |
74 |
container_issue |
3 |
container_start_page |
1 |
op_container_end_page |
15 |
_version_ |
1774715061028257792 |