两种估算近海面大气光学湍流强度方法的比较

发布时间：2018-08-29 16:32

【摘要】：选取随船观测的三亚地区2016-01-06至2016-01-09连续4天的折射率结构常数C_n~2及温度、风速、相对湿度三种常规气象参数,基于后向传播神经网络和逐步回归理论,分别建立两种模型并对C_n~2进行了连续3天的估算。结果显示,两种模型估算的结果在变化趋势及量级上均符合近海面光学湍流的一般特征和变化规律,并且可以表现出C_n~2的基本日变化特征,整体相关系数分别为0.8661和0.8496。选取了平均绝对误差、平均相对误差、均方根方差以及相关系数等统计量来衡量估算结果。分析表明,两种模型均能准确地估算出近海面的C_n~2,但在夜间弱湍流发生时估算值略高于测量值。为进一步提高估算的准确度,需要改进模式在夜间的估算效果。
[Abstract]:Three conventional meteorological parameters, C_n~2, temperature, wind speed and relative humidity, were selected for four consecutive days from 2016-01-06 to 2016-01-09 in Sanya area observed by ship, based on backward propagation neural network and stepwise regression theory. Two models were established and the C_n~2 was estimated for 3 consecutive days. The results show that the results of the two models are consistent with the general characteristics and the law of variation of optical turbulence near the sea surface in the order of magnitude, and can show the basic diurnal variation of C_n~2. The overall correlation coefficients are 0.8661 and 0.8496, respectively. The average absolute error, average relative error, root mean square variance and correlation coefficient were selected to measure the estimation results. The analysis shows that both models can accurately estimate CSTN _ 2 near the sea surface, but the estimated value is slightly higher than the measured value when the weak turbulence occurs at night. In order to further improve the accuracy of the estimation, it is necessary to improve the estimation effect of the model at night.
【作者单位】：中国科学院安徽光学精密机械研究所大气成分与光学重点实验室;中国科学技术大学研究生院科学岛分院;
【分类号】：O436

【相似文献】