设为首页
加入收藏
Thunderstorm Gale Identification Method Based on Support Vector Machine
-
摘要: 基于北京市观象台雷达基数据和加密自动气象站数据，利用支持向量机算法建立了雷暴大风天气的有效识别模型。首先确立了9个用于识别雷暴大风的预报因子：回波顶高、最大反射率因子、最大反射率因子所在高度、垂直积分液态水含量、垂直积分液态水含量随时间变率、垂直积分液态水含量密度、雷暴大风发生前最大反射率因子下降高度、风暴移动速度、速度谱宽。通过计算各预报因子在大风和非大风样本中的概率分布，得到对应的各项预报因子雷暴大风的隶属度，利用得到的隶属度函数对样本进行归一化处理。确立核函数和模型参数，利用支持向量机建立雷暴大风天气的提前识别和临近预警模型。通过对北京2017年7月7日飑线和2012年5月19日块状回波引起的灾害大风典型个例的识别效果检验，得到两个个例预测的命中率、误判率和临界成功指数分别为92.0%，22.1%，73.0%和99.1%，40.5%，59.2%，对于提高雷暴大风预警预报的准确率有一定帮助。Abstract: A thunderstorm gale recognition model is established using support vector machine based on data of radar and automatic weather stations from Beijing Weather Observatory. Firstly, 18 thunderstorms in Beijing during 2010-2014 are analyzed quantitatively in terms of the statistical method and 9 forecast factors are selected, i.e., the height of the echo top, the maximum albedo, the height of the maximum reflectivity, the total vertical liquid water content, the time rate change of total vertical liquid water content, the total vertical liquid water content density, the height of the maximum reflectivity factor, the storm moving speed and the width of the velocity spectrum. 451 non-high wind samples and 425 high wind samples are selected by matching the time and place of automatic weather stations with the value of the quantitative index of the PUP storm monomer recognition product in all the cases. Secondly, the probability distribution of prediction factors in the wind and non-wind samples are calculated, and relationships with corresponding forecast factors are obtained, and then sample data are normalized by the obtained membership function. Finally, the kernel function and model parameters are established, and the thunderstorm gale recognition model is established using support vector machine. Two typical cases in Beijing are analyzed and tested, one caused by a line thunderstorm which happened on 7 July 2017, and the other caused by an isolated single-cell storm which happened on 19 May 2012. Results show that the identified wind range is consistent with reality, and the hit rate, the false alarm rate and the critical success index are 92.0%, 22.1%, 73.0% and 99.1%, 40.5%, 59.2%, respectively. It will help to improve the accuracy of thunderstorm gale warning and forecasting. However, sometimes multiple thunderstorm cells can be misjudged as one according to these forecast factors. In this case, it is necessary for forecasters to conduct manual intervention in combination with the overall radar base reflectivity and weather conditions, to reduce the misjudgment rate of gale. In the future, long time series radar data should be used to carry out "large sample census" research and an automatic thunderstorm identification system based on weather radar can be established.
-
图 1 基于支持向量机的雷暴大风识别模型建立流程
Fig. 1 Flow chart of the thunderstorm gale identification model based on SVM
图 3 2017年7月7日实况大风位置与识别出的大风风暴单体分布
Fig. 3 The position of gale and identified thunderstorm cells with gale on 7 Jul 2017
图 4 2017年7月7日15:11(a)及15:23(b)反射率因子(填色)、实况大风位置与风暴单体
Fig. 4 Base reflectivity(the shaded), the position of gale and identified thunderstorm cells at 1511 UTC(a) and 1523 UTC(b) on 7 Jul 2017
图 5 2012年5月19日实况大风位置与识别出大风的风暴单体分布
Fig. 5 The position of gale and identified thunderstorm cells with gale on 19 May 2012
图 6 2012年5月19日12:47(a)及12:53(b)反射率因子(填色)、实况大风位置与风暴单体
Fig. 6 Base reflectivity(the shaded), the position of gale and identified thunderstorm cells at 1247 UTC(a) and 1253 UTC(b) on 19 May 2012
表 1 7个误判样本所对应的9个预报因子及自动气象站极大风速对比
Table 1 Comparision of forecast factors of misjudged samples with the maximum instantaneous wind speed at automatic weather stations
预报因子 样本1 样本2 样本3 样本4 样本5 样本6 样本7 回波顶高/km 8.6 6.4 6.4 7.9 6.7 7.9 11.1 最大反射率因子/dBZ 52 50 58 51 54 52 62 最大反射率因子所在高度/km 2.4 4.6 1.7 3.0 3.7 3.4 4.6 最大反射率因子下降高度/km 2.9 2.1 4.4 3.1 0.9 1.2 1.7 垂直积分液态水含量/(kg·m-2) 17 10 31 10 18 18 54 垂直积分液态水含量随时间变率/(kg·m-2·(6 min)-1) 2 17 21 4 8 1 21 垂直积分液态水含量密度/(g·m-3) 2.1 2.8 5.5 1.6 6.0 2.6 5.6 风暴移动速度/(km·h-1) 5 6 7 5 6 7 4 速度谱宽/(m·s-1) 0 0 0 0 14 7 13 匹配站点风速/(m·s-1) 6.2 5.9 3.4 2.8 5.9 5.0 2.0 
下载: 导出CSV
-
[1] 东高红, 吴涛.垂直积分液态水含量在地面大风预报中的应用.气象科技, 2007, 35(6):877-881. doi: 10.3969/j.issn.1671-6345.2007.06.024 [2] 梁爱民, 张庆红, 申红喜.北京地区雷暴大风预报研究.气象, 2006, 32(11):73-80. http://d.old.wanfangdata.com.cn/Periodical/qx200611012 [3] 刁秀广, 张新华, 朱君鉴.CINRAD/SA雷达风暴趋势产品在冰雹和大风预警中的应用.气象科技, 2009, 37(2):230-233. doi: 10.3969/j.issn.1671-6345.2009.02.022 [4] 王彦, 吕江津, 王庆元, 等.一次雷暴大风的中尺度结构特征分析.气象, 2006, 32(2):75-80. http://d.old.wanfangdata.com.cn/Periodical/qx200602014 [5] 王福侠, 俞小鼎, 裴宇杰, 等.河北省雷暴大风的雷达回波特征及预报关键点.应用气象学报, 2016, 27(3):342-351. http://qikan.camscma.cn/jams/ch/reader/view_abstract.aspx?file_no=20160309&flag=1 [6] 应冬梅, 许爱华, 黄祖辉.江西冰雹、大风与短时强降水的多普勒雷达产品的对比分析.气象, 2007, 33(3):48-53. http://d.old.wanfangdata.com.cn/Periodical/qx200703007 [7] 姚建群, 戴建华, 姚祖庆.一次强飑线的成因及维持和加强机制分析.应用气象学报, 2005, 16(6):746-754. doi: 10.3969/j.issn.1001-7313.2005.06.005 [8] 慕熙昱, 党人庆, 陈秋萍, 等.一次飑线过程的雷达回波分析与数值模拟.应用气象学报, 2007, 18(1):42-49. doi: 10.3969/j.issn.1001-7313.2007.01.006 [9] 谢健标, 林良勋, 颜文胜, 等.广东2005年"3.22"强飑线天气过程分析.应用气象学报, 2007, 18(3):321-329. doi: 10.3969/j.issn.1001-7313.2007.03.008 [10] 朱敏华, 俞小鼎, 夏峰, 等.强烈雹暴三体散射的多普勒天气雷达分析.应用气象学报, 2006, 17(2):215-223. doi: 10.3969/j.issn.1001-7313.2006.02.012 [11] 俞小鼎, 张爱民, 郑媛媛, 等.一次系列下击暴流事件的多普勒天气雷达分析.应用气象学报, 2006, 17(4):385-393. doi: 10.3969/j.issn.1001-7313.2006.04.001 [12] 俞小鼎, 周小刚, 王秀明.雷暴与强对流临近天气预报技术进展.气象学报, 2012, 70(3):311-337. doi: 10.3969/j.issn.1004-4965.2012.03.003 [13] 廖晓农, 于波, 卢丽华.北京雷暴大风气候特征及短时临近预报方法.气象, 2009, 35(9):18-28. http://d.old.wanfangdata.com.cn/Periodical/qx200909003 [14] 王福侠, 俞小鼎, 裴宇杰, 等.河北省雷暴大风的雷达回波特征及预报关键点.应用气象学报, 2016, 27(3):342-351. http://qikan.camscma.cn/jams/ch/reader/view_abstract.aspx?file_no=20160309&flag=1 [15] 廖玉芳, 潘志祥, 郭庆.基于单多普勒天气雷达产品的强对流天气预报预警方法.气象科学, 2006, 26(5):564-571. doi: 10.3969/j.issn.1009-0827.2006.05.015 [16] 周金莲, 魏鸣, 吴涛, 等.对流性大风天气的多普勒雷达资料识别方法研究//第28届中国气象学会年会论文集.北京: 中国气象学会, 2001. [17] 李国翠, 刘黎平, 张秉祥, 等.基于雷达三维组网数据的对流性地面大风自动识别.气象学报, 2013, 71(6):1160-1167. http://d.old.wanfangdata.com.cn/Periodical/qxxb201306013 [18] 周康辉, 郑永光, 王婷波, 等.基于模糊逻辑的雷暴大风和非雷暴大风区分方法.气象, 2017, 43(7):781-791. http://d.old.wanfangdata.com.cn/Periodical/qx201707002 [19] 赫英明, 王汉杰.支持向量机在积雪检测中的应用.南京气象学院学报, 2009, 32(1):134-139. doi: 10.3969/j.issn.1674-7097.2009.01.018 [20] 杨璐, 陈明轩, 孟金平, 等.北京地区雷暴大风不同生命期内的雷达统计特征及预警提前量分析.气象, 2018, 44(6):802-813. http://www.cnki.com.cn/Article/CJFDTOTAL-QXXX201806008.htm [21] Courant R, Hilbert D.Method of Mathematical Physics.New York:Springer Verlag, 1953. [22] 陈永义, 俞小鼎, 高学浩, 等.处理非线性分类和回归问题的一种新方法(Ⅰ)——支持向量机方法简介.应用气象学报, 2004, 15(3):345-353. doi: 10.3969/j.issn.1001-7313.2004.03.011 [23] 冯汉中, 陈永义.处理非线性分类和回归问题的一种新方法(Ⅱ)——支持向量机方法在天气预报中的应用.应用气象学报, 2004, 15(3):355-365. doi: 10.3969/j.issn.1001-7313.2004.03.012 [24] 冯汉中, 陈永义.支持向量机回归方法在实时业务预报中的应用.气象, 2005, 31(2):41-44. http://d.old.wanfangdata.com.cn/Periodical/qx200501009 -
计量
- 摘要浏览量: 4251
- HTML全文浏览量: 1538
- PDF下载量: 395
- 被引次数: 0
