检索
E-mail
RSS

作者投稿 主编审稿 专家审稿 远程编辑
基于CNN-LSTM的高频地波雷达海流方向校正方法
许毅1(), 韦骏1(), 魏春雷2, 杨凡2
An ocean current direction correction method for high-frequency surface wave radars based on CNN-LSTM
XU Yi1(), WEI Jun1(), WEI Chunlei2, YANG Fan2

图5. 原始雷达数据与模式数据的EOF椭圆对比图(a)及校正后雷达数据与模式数据的EOF椭圆对比图(b)

数据点以经度0.15°、纬度0.10°的间隔绘制。a、b中蓝色分别是原始雷达数据和网络模型校正后数据的EOF椭圆, 红色是模式数据的EOF椭圆, 箭头表示平均流方向

Fig. 5. EOF ellipses contrast graph (a) between original radar data and model data and EOF ellipses contrast graph (b) between corrected radar data and model data. Data points are plotted at intervals of 0.15° longitude and 0.10° latitude. In (a) and (b), blue represents the EOF ellipses of the original radar data and the neural network corrected data, respectively, and red represents the EOF ellipses of the model data. The arrows indicate the direction of the mean flow