• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Conference on Global Oceans : Singapore – U.S. Gulf Coast >A Combination Forecasting Model Based on AdaBoost_GRNN in Depth-Averaged Currents Using Underwater Gliders
【24h】

A Combination Forecasting Model Based on AdaBoost_GRNN in Depth-Averaged Currents Using Underwater Gliders

机译：基于adaboost_grnn在水下滑翔机深度平均电流中的组合预测模型

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

In this paper, a novel combination method based on the AdaBoost algorithm and the general regression neural network (GRNN) is proposed to predict the depth-averaged current (DAC) for the next profile using the underwater gliders (UGs). Firstly, considering changes of the seawater density and the hull deformation, the velocity calculation model of UGs and the calculation method of DAC are given. Secondly, two DAC forecasting models are established with historical data using the support vector machine (SVM) and the back-propagation neural network (BPNN) methods, respectively. Then, the forecasting models are combined nonlinearly by adopting the GRNN. Finally, outputs from the GRNN are combined using the AdaBoost algorithm, which further taken for the DAC forecasting. A sea trial was conducted using Petrel-II glider in the South China Sea, which verified the accuracy of the three forecasting models. Results demonstrate that the proposed combination method has a better performance in DAC forecasting than the other two single forecasting models.
机译：在本文中，提出了一种基于ADABOOST算法和一般回归神经网络（GRNN）的组合方法，以预测使用水下滑轮（UGS）的下一个轮廓的深度平均电流（DAC）。首先，考虑到海水密度和船体变形的变化，给出了UGS的速度计算模型和DAC的计算方法。其次，使用支持向量机（SVM）和后传播神经网络（BPNN）方法的历史数据建立了两个DAC预测模型。然后，通过采用GNN来非线性地组合预测模型。最后，使用进一步为DAC预测的Adaboost算法组合来自GRNN的输出。使用南海的Petrel-II滑翔机进行了海洋试验，验证了三种预测模型的准确性。结果表明，所提出的组合方法在DAC预测中具有比其他两个单一预测模型更好的性能。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号