| 粤港澳闪电定位系统对高建筑物雷电的探测 |
| |
| 引用本文： | 师正, 吴子珉, 孙京, 胡佳瑞, 管啸林. 2023: 基于三种聚类算法的雷暴系统识别效果对比研究. 暴雨灾害, 42(3): 334-345. DOI: 10.12406/byzh.2022-179 |
| |
| 作者姓名： | 师正 吴子珉 孙京 胡佳瑞 管啸林 |
| |
| 作者单位： | 1.南京信息工程大学应急管理学院，南京 210044;2.中国气象科学研究院灾害天气国家重点实验室，北京 100081;3.中国气象局武汉暴雨研究所 暴雨监测预警湖北省重点实验室，武汉 430205 |
| |
| 基金项目： | 国家自然科学基金项目(41805002)；中国气象科学研究院灾害天气国家重点实验室开放课题(2021LAW-B05)；南京信息工程大学人才启动项目(2016r042，2020r093) |
| |
| 摘 要： | 为了评估不同聚类算法对雷暴系统的识别效果，进一步提高雷电临近预报能力，本文采用地闪定位数据和雷达反射率数据，利用基于密度的空间聚类(Density-Based Spatial Clustering of Application with Noise，DBSCAN)、快速搜索和查找密度峰聚类(Clustering by Fast Search and Find of Density Peaks，CFSFDP)以及改进的快速搜索和查找密度峰聚类(Extended Clustering by Fast Search and Find of Density Peaks，E_CFSFDP)三种聚类算法，对2018年9月21日19∶15—20∶57(北京时)发生在(114°—117°E、27°—30°N)区域的一次雷暴过程进行了聚类识别计算，探讨了三类聚类算法在雷暴系统识别中的差异。结果表明：(1) DBSCAN算法在地闪数据分布清晰且不同数据簇之间有显著距离间隔时，分类识别的准确率较高；当各个闪电数据簇的簇间距离或密度相差很大时，分类识别的准确率较低；(2) 地闪数据“无密度峰值”分布时CFSFDP算法会分裂出错误类，每个闪电数据簇仅具备唯一的密度峰值点是CFSFDP算法识别准确的前提条件；(3) E_CFSFDP算法解决了CFSFDP算法的“无密度峰值”问题，受地闪数据分布影响较小，因此基于E_CFSFDP算法的雷暴系统识别准确率明显高于DBSCAN和CFSFDP算法。
|
| 关 键 词： | 雷暴 识别 闪电定位资料 聚类分析 机器学习 |
| 收稿时间： | 2022-09-09 |
Lightning Detection of High Buildings by Guangdong-Hong Kong-Macao Lightning Location System |
| |
| SHI Zheng, WU Zimin, Sun Jing, HU Jiarui, GUAN Xiaolin. 2023: Comparative research on thunderstorms identification based on three clustering methods. Torrential Rain and Disasters, 42(3): 334-345. DOI: 10.12406/byzh.2022-179 |
| |
| Authors: | SHI Zheng WU Zimin Sun Jing HU Jiarui GUAN Xiaolin |
| |
| Affiliation: | 1.Emergency Management College, Nanjing University of Information Science and Technology, Nanjing 210044;2.State key Laboratory of Severe Weather, Chinese Academy of Meteorological Sciences, Beijing 100081;3.Hubeil Key Laboratory for Heavy Rain Monitoring and Warning Research, Institute of Heavy Rain, China Meteorological Administration, Wuhan 430205 |
| |
| Abstract: | In order to evaluate the recognition effect of different clustering algorithms on thunderstorm system, further improve the ability of lightning nowcasting, in this study, with the ground-based lightning location data and radar reflectivity data, three different clustering algorithms for thunderstorm identification are analyzed for a thunderstorm case in the region of 114°-117°E, 27°-30°N and during the time window of 19∶15 BT to 20∶57 BT on 21 September, 2018. The performance of these three algorithms, which are density-based spatial clustering of application with noise (DBSCAN), clustering by fast search and find of density peaks (CFSFDP) and extended clustering by fast search and find of density peaks (E_CFSFDP) are compared and discussed. Here are the results. (1) The DBSCAN algorithm has a higher accuracy of classification and identification when the distributions of lightning location show clear patterns and different clusters are apart from each other, but the accuracy will be low when the density of each lightning data cluster is uneven or the distances between different clusters are very large. (2) the CFSFDP algorithm works when each cluster in the data sets has one single density peak, otherwise, when there are no density peaks, CFSFDP will be failed. (3) The E_CFSFDP algorithm can solve the |
| |
| Keywords: | thunderstorm identification lightning location data clustering analysis machine learning |
|
| 点击此处可从《暴雨灾害》浏览原始摘要信息 |
|
点击此处可从《暴雨灾害》下载全文 |
|
