In this paper, a novel approach for extracting the characteristics of ocean wave height in real time is investigated. This model is an integrated model of an autoregressive (AR) model, a weighted least squares, linear and nonlinear support vector machines (SVMs) and the HF radar system. To develop the proposed algorithm, ocean wave data are collected from two Wellen radar systems that are installed in Samcheok City, Gangwon-do in the East Coast of Korea. A random data set is used for constructing the proposed algorithm. The trained model is tested using other data sets that are not used. It is shown from the numerical testing that the proposed algorithm is very effective in extracting ocean wave characteristics.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

AR-SVM for Ocean Wave Characteristics Identification

  • Nuoyi Zhu,
  • Yeesock Kim

摘要

In this paper, a novel approach for extracting the characteristics of ocean wave height in real time is investigated. This model is an integrated model of an autoregressive (AR) model, a weighted least squares, linear and nonlinear support vector machines (SVMs) and the HF radar system. To develop the proposed algorithm, ocean wave data are collected from two Wellen radar systems that are installed in Samcheok City, Gangwon-do in the East Coast of Korea. A random data set is used for constructing the proposed algorithm. The trained model is tested using other data sets that are not used. It is shown from the numerical testing that the proposed algorithm is very effective in extracting ocean wave characteristics.