Transfer learning for modeling and prediction of marine buoy motion characteristics

Deng Hongying; Li Xintian; Zhu JialiangLiu Yi

首页> 外文期刊>Ocean engineering >Transfer learning for modeling and prediction of marine buoy motion characteristics

【24h】

Transfer learning for modeling and prediction of marine buoy motion characteristics

机译：Transfer learning for modeling and prediction of marine buoy motion characteristics

获取原文

获取原文并翻译 | 示例

获取外文期刊封面封底 >>

开具论文收录证明 >>

文献代查 >>

页面导航

摘要
著录项
相关主题

摘要

To ensure the reliability of marine buoy systems, it is necessary to predict the buoy motion characteristics in different environment conditions. Unfortunately, an accurate model describing the complex characteristics is still not available. In this work, a domain adaptation transfer learning method is proposed to enhance the prediction performance by integrating the informative data of different conditions. First, a simple evaluation criterion is designed to automatically divide modeling data into two subsets, corresponding to the normal and extreme conditions. Sequentially, the domain adaptation transfer learning model is constructed by delivering useful information between normal conditions and extreme conditions. Moreover, an efficient strategy is developed to select suitable model parameters. Consequently, the prediction domain can be enlarged to capture different marine buoy motion characteristics in normal and extreme conditions. The prediction results show the superiority of the proposed method as compared to two traditional models.

著录项

来源
《Ocean engineering》 |2022年第5期|113158.1-113158.8|共8页
作者
Deng Hongying; Li Xintian; Zhu JialiangLiu Yi;
展开▼
作者单位

Zhejiang Univ Technol;

Hangzhou Appl Acoust Res Inst||Hangzhou Ruili Marine Equipment Co LTD;

展开▼
收录信息
原文格式 PDF
正文语种英语
中图分类
关键词
Ocean environment; Buoy motion characteristic; Prediction model; Transfer learning; Domain adaptation;

Transfer learning for modeling and prediction of marine buoy motion characteristics

摘要

著录项

相关主题

期刊订阅