| 摘要: |
| 【目的】为保证气象预报结果的准确性。【方法】提出了考虑随机误差传递的高空气象秒级数据快速校验方法。对高空气象秒级数据进行预处理，包括数据中心化处理、特征点提取。数据中心化的处理方式为实施各组数据的加权平均处理，获得新的数据组作为基准数据；特征点提取使用的处理方法为遍历法，需要遍历全部数据点。考虑数据获取过程中的随机误差传递情况，对预处理后的高空气象秒级数据实施空间一致性检查、内部一致性检查。应用集合经验模态分解EEMD（Ensemble Empirical Mode Decomposition）算法与对比源反演CSI(Contrast Source Inversion)算法构建基于CSI-EEMD的高空气象秒级数据快速校验模型，实施数据得到快速校验。【结果】测试结果表明，该方法能够实现4个气象站数据的快速校验，校验结果的均方根误差与平均绝对误差均低于0.1，实现了数据的质量控制。【结论】将该方法应用于气象站数据校验中，可以实现精准度的数据校验，具有一定的实用价值。 |
| 关键词: 随机误差传递；特征点提取；高空气象秒级数据；一致性检查；数据校验 |
| DOI： |
| 投稿时间：2023-02-24 |
| 基金项目:广西气象局气象科研计划项目（桂气科〔2023〕CG03、桂气科〔2023〕CG01）资助 |
|
| A Fast Verification Method for Upper-Air Meteorological Second-Level Data by Random Error Transmission |
| WEI Liying,LIANG Jianping,XIE Fei,HUANG Meiyan,QIN Xiaoling |
| (Baise Meteorological Office of Guangxi Zhuang Autonomous Region, Baise 533000 , China;Guilin Meteorological Office of Guangxi Zhuang Autonomous Region , Guilin 541000 , China;Hechi Meteorological Office of Guangxi Zhuang Autonomous Region, Hechi 547000 , China) |
| Abstract: |
| To ensure the accuracy of meteorological forecasts, this article puts forward a fast verification method for upper-air meteorological second-level data considering random error transmission. It can preprocess upper-air meteorological second-level data, including data centralization processing and feature point extraction. The processing method of data centralization is to implement weighted average processing of each group of data, and obtain new data groups as benchmark data, while the processing method used for feature point extraction is the traversal method, which requires traversing all data points. Then, the fast verification method is to consider the random error transmission during the data acquisition process, and implement spatial consistency check and internal consistency check on the preprocessed upper-air meteorological second-level data. A fast verification model of upper-air meteorological second-level data based on CSI-EEMD is constructed using the Ensemble Empirical Mode Decomposition (EEMD) algorithm and the Contrast Source Inversion (CSI) algorithm, and the experiment data can get quickly verified. The test results show that this new method can realize the rapid verification of the data of four weather stations, and the root mean square error and the mean absolute error of the verification results are both lower than 0.1, achieving the quality control of observation data. Applying this method to the verification of meteorological station data can achieve accurate data verification and has a certain practical value. |
| Key words: random error transmission； feature point extraction； upper-air meteorological second-level data； consistency check； data verification |
