| 决策树算法在雷达反射率因子质控中的应用 |
| |
| 引用本文： | 袁凯,李明,庞晶,李武阶,邱盼柳戊弼.决策树算法在雷达反射率因子质控中的应用[J].新疆气象,2023,17(1):139-144. |
| |
| 作者姓名： | 袁凯 李明 庞晶 李武阶 邱盼柳戊弼 |
| |
| 作者单位： | 武汉市气象台,武汉市气象台,武汉市气象台,武汉市气象台,武汉河海云天慧科技有限公司;武汉市气象台 |
| |
| 基金项目： | 武汉市短时临近预警预报技术创新团队(2021年度)、武汉市气象局自立课题(WH202101) |
| |
| 摘 要： | 基于决策树算法，利用2020年武汉多普勒天气雷达探测资料，分析决策树算法对反射率因子的质控效果，得到以下主要结论：（1）决策树算法能有效滤除湖北东部地区由于山脉和城市建筑所导致的固定地物回波和超折射地物回波。（2）对于降水回波而言，决策树算法不改变雷达近距离范围内回波的位置、形状和强度，而远距离处（≥160 km）回波则存在一定程度的过度质控，主要是由于雷达波束随距离增加而抬高以及回波垂直梯度较大导致。（3）决策树算法对辐射状和弧状杂波的质控能力较好，其滤除率达到95%以上。
|
| 关 键 词： | 决策树算法 反射率因子 质量控制 地物杂波 |
| 收稿时间： | 2021/7/28 0:00:00 |
| 修稿时间： | 2021/12/1 0:00:00 |
Application of Decision Tree Algorithm in Quality Control for Radar Reflectivity |
| |
| yuan kai,and.Application of Decision Tree Algorithm in Quality Control for Radar Reflectivity[J].Bimonthly of Xinjiang Meteorology,2023,17(1):139-144. |
| |
| Authors: | yuan kai and |
| |
| Abstract: | Based on the decision tree algorithm and the Doppler radar data of Wuhan in 2020, this paper analyzes the quality control effect of the decision tree algorithm on the reflectivity factor in detail. The main conclusions are as follows: (1) the decision tree algorithm can effectively filter the fixed and hyperrefraction surface echoes caused by mountains and urban buildings in Eastern Hubei. (2) For the precipitation radar echo, the decision tree algorithm does not change the position, shape and intensity of the radar echo within the short distance and there is a certain degree of over-quality for the long distance radar echo (generally 160 km), which mainly caused by the radar beam rises with distance increase and the vertical gradient of echo is larger. (3) The statistical results show that the filtering rate of decision tree algorithm is more than 95% for radial and ARC echo, which shows good quality control effect for those echoes. |
| |
| Keywords: | decision tree algorithm radar reflectivity quality control anomalous propagation(AP) ground clutter(GC) |
|
| 点击此处可从《新疆气象》浏览原始摘要信息 |
| 点击此处可从《新疆气象》下载免费的PDF全文 |
|
