[1]黄启桥,麦雄发,李玲,等.基于ConvLSTM的广西短临降水预报[J].气象研究与应用,2021,42(04):44-49.[doi:10.19849/j.cnki.CN45-1356/P.2021.4.08]
　Huang Qiqiao,Mai Xiongfa,Li Ling,et al.Forecast of short-term precipitation in Guangxi based on ConvLSTM[J].Journal of Meteorological Research and Application,2021,42(04):44-49.[doi:10.19849/j.cnki.CN45-1356/P.2021.4.08]
点击复制

基于ConvLSTM的广西短临降水预报()
分享到：

气象研究与应用[ISSN:1673-8411/CN:45-1356/P]

卷:
第42卷
期数:
2021年04期
页码:
44-49
栏目:
研究论文
出版日期:
2022-01-27

文章信息/Info

Title:
Forecast of short-term precipitation in Guangxi based on ConvLSTM
作者:
黄启桥1;  麦雄发1;  李玲2;  唐菁2;  唐飞笼1
1. 南宁师范大学数学与统计学院，南宁 530001;
2. 南宁师范大学地理科学与规划学院，南宁 530001
Author(s):
Huang Qiqiao1;  Mai Xiongfa1;  Li Ling2;  Tang Jing2;  Tang Feilong1
1. School of Mathematics and Statistics, Nanning Normal University, Nanning 530001, China;
2. School of Geography and Planning, Nanning Normal University, Nanning 530001, China
关键词:
短临预报; 雷达回波; ConvLSTM; 光流法
Keywords:
nowcasting; radar echo; ConvLSTM; optical flow
分类号:
P457.6
DOI:
10.19849/j.cnki.CN45-1356/P.2021.4.08
摘要:
针对传统雷达回波外推算法在快速增长或消散降水过程预报精度较低的问题，利用华南雷达回波拼图资料数据，建立ConvLSTM回波外推模型，对广西区域范围进行短临降水预报研究。采用气象业务中的正确率（POD）、临界成功指数（CSI）及误报率（FAR）评判标准检验预报模型，并将ConvLSTM与光流法的预报结果进行对比分析。结果表明，ConvLSTM模型的CSI、POD分别比光流法提高0.06和0.059，而FAR下降了0.058。ConvLSTM方法比光流法的回波外推预报准确率高，该方法可为广西短临降水预报提供新的参考。
Abstract:
Aiming at the problem of low prediction accuracy of traditional radar echo extrapolation algorithms in the process of rapid growth or dissipation of precipitation, the paper established a ConvLSTM echo extrapolation model by using the South China radar echo mosaic data to study short-term precipitation in Guangxi. The accuracy rate(POD), critical success index(CSI)and false alarm rate(FAR)evaluation models in meteorological services were used, and ConvLSTM was compared with the optical flow method. The experimental results show that the CSI and POD of the ConvLSTM prediction result are 0.06 and 0.059 higher than that of the optical flow method, respectively, while the FAR decreases by 0.058. This shows that ConvLSTM has a higher accuracy of echo extrapolation than the optical flow method, which can provide a new method reference for the forecast of short-term precipitation in Guangxi.

参考文献/References:

[1] 施恩, 李骞, 顾大权, 等. 基于卷积神经网络的雷达回波外推方法[J]. 计算机应用, 2018, 38(3): 661-665+676.
[2] Johnson J T, MacKeen P L, Witt A, et al. The Storm Cell Identification and Tracking Algorithm: An Enhanced WSR-88D Algorithm[J]. Weather and forecasting, 1998, 13(2): 263-276.
[3] Rinehart R E, Garvey E T. Three-dimensional storm motion detection by conventional weather radar[J]. Nature, 1978, 273(5660): 287-289.
[4] Luo C, Li X, Wen Y, et al. A Novel LSTM Model with Interaction Dual Attention for Radar Echo Extrapolation[J]. Remote Sensing, 2021, 13(2): 164.
[5] Shi X, Gao Z, Lausen L, et al. Deep Learning for Precipitation Nowcasting: A Benchmark and a New model[J]. Advances in Neural Information Rocessing Systems, 2017, 30: 5617-5627.
[6] Shi E, Li Q, Gu D, et al. A method of Weather Radar Echo Extrapolation Based on Convolutional Neural Networks[C]//International Conference on Multimedia Modeling. Springer, Cham, 2018: 16-28.
[7] Pan B, Hsu K, AghaKouchak A, et al. Improving Precipitation Estimation Using Convolutional Neural Network[J]. Water Resources Research, 2019, 55(3): 2301-2321.
[8] Chen B F, Huang T S, Chen B, et al. Accurate and Clear Precipitation Nowcasting with Consecutive Attention and Rain-map Discrimination[J]. arXiv preprint arXiv: 2102. 08175, 2021.
[9] 路志英, 任一墨, 孙晓磊, 等. 基于深度学习的短时强降水天气识别[J]. 天津大学学报(自然科学与工程技术版), 2018, 51(2): 111-119.
[10] 吴香华, 华亚婕, 官元红, 等. 基于CNN-Attention-BP的降水发生预测研究[J/OL]. 南京信息工程大学学报(自然科学版): 1-13[2021-09-24].
[11] 梁振清, 陈生. 基于深度学习和雷达观测的华南短临预报精度评估[J]. 气象研究与应用, 2020, 41(1): 41-47.
[12] 黄骄文, 蔡荣辉, 姚蓉, 等. 深度学习网络在降水相态判识和预报中的应用[J]. 气象, 2021, 47(3): 317-326.
[13] Agrawal S, Barrington L, Bromberg C, et al. Machine Learning for Precipitation Nowcasting From Radar Images[J]. arXiv e-preprints arXiv: 1912.12132, 2019.
[14] Ayzel G, Scheffer T, Heistermann M. RainNet v1. 0: a Convolutional Neural Network for Radar-based Precipitation Nowcasting[J]. Geoscientific Model Development, 2020, 13(6): 2631-2644.
[15] 方巍, 庞林, 张飞鸿, 等. 对抗型长短期记忆网络的雷达回波外推算法[J]. 中国图象图形学报, 2021, 26(5): 1067-1080.
[16] 吴卓升, 张巍, 林艳, 等. 动态概率卷积神经网络在雷达回波外推中的应用[J]. 计算机应用研究, 2021, 38(7): 2125-2129.
[17] Shi X, Chen Z, Wang H, et al. Convolutional LSTM network: A Machine Learning Approach for Precipitation Nowcasting[J]. Advances in Neural Information Processing Systems, 2015(28): 802-810.
[18] Ayzel G, Heistermann M, Winterrath T. Optical Flow Models as an Open Benchmark for Radar-based precipitation Nowcasting(Rainymotion v0.1)[J]. Geoscientific Model Development Discussions, 2018: 1-23.

相似文献/References:

[1]屈梅芳,古文保.广西一次连续性暴雨过程分析[J].气象研究与应用,2016,37(04):41.
　Qu Mei-fang,Gu Wen-bao.A continuous torrential rain process analysis in Guangxi[J].Journal of Meteorological Research and Application,2016,37(04):41.
[2]李宜爽,刘蕾,王艺.广西柳州2016年4月一次致灾冰雹天气过程分析[J].气象研究与应用,2016,37(03):52.
　Li Yi-shuang,Liu Lei,Wang Yi.The Analysis of a Hail Weather on April 9th 2016 in Liuzhou, Guangxi[J].Journal of Meteorological Research and Application,2016,37(04):52.
[3]陈锐,刘洲荣.柳州市强对流暴雨的闪电和雷达回波特征及相关性研究[J].气象研究与应用,2016,37(02):38.
　Chen Rui,Liu Zhou-rong.Characteristics of Lighting and Radar Echoes of Strong Convective Rainstorm and Their Relations in Liuzhou[J].Journal of Meteorological Research and Application,2016,37(04):38.
[4]陈关清,方标.贵州铜仁暴雨和冰雹雷达回波特征对比分析[J].气象研究与应用,2015,36(01):72.
　Chen Guanqing,Fang Biao.Comparative analysis on radar echo of heavy rain and hail in Tongren Guizhou[J].Journal of Meteorological Research and Application,2015,36(04):72.
[5]方标.贵州梵净山区域两次大暴雨天气综合分析[J].气象研究与应用,2017,38(04):39.
　Fang Biao.The analysis of the two heavy rainstorms in the Fanjing Mountain of Guizhou Province[J].Journal of Meteorological Research and Application,2017,38(04):39.
[6]唐佐阳,吕巍伟,吕校华,等.衡邵盆地7-8月对流云特征与人工增雨作业效果分析[J].气象研究与应用,2018,39(01):51.
　Tang Zuo-yang,Lv Wei-wei,Lv Xiao-hua,et al.Analysis of convective cloud characteristics in Hengshao Basin from July to August and artificial precipitation enhancement effect[J].Journal of Meteorological Research and Application,2018,39(04):51.
[7]卓健,廖胜石,陈少斌,等.雷达回波特征向量应用初探[J].气象研究与应用,2018,39(03):83.
　Zhou Jian,Liao Shengshi,Chen Shaobin,et al.A primary study on the application of radar echo eigenvectors[J].Journal of Meteorological Research and Application,2018,39(04):83.
[8]郭彬,卓健,周冬静,等.基于智能计算的广西大风短临预报预警系统的产品检验[J].气象研究与应用,2021,42(01):80.[doi:10.19849/j.cnki.CN45-1356/P.2021.1.14]
　Guo Bin,Zhuo Jian,Zhou Dongjing,et al.Product inspection of Guangxi gale short-term forecast and early warning system based on intelligent computing[J].Journal of Meteorological Research and Application,2021,42(04):80.[doi:10.19849/j.cnki.CN45-1356/P.2021.1.14]
[9]钟有亮,李勋,张诚忠,等.NUDGING方法同化雷达反射率因子个例试验研究[J].气象研究与应用,2021,42(03):76.[doi:10.19849/j.cnki.CN45-1356/P.2021.3.13]
　Zhong Youliang,Li Xun,Zhang Chengzhong,et al.A case study of assimilated radar reflectivity factor by NUDGING method[J].Journal of Meteorological Research and Application,2021,42(04):76.[doi:10.19849/j.cnki.CN45-1356/P.2021.3.13]

备注/Memo

备注/Memo:
收稿日期:2021-07-05。
基金项目:广西自然科学基金项目（2018GXNSFAA294079）、广西教育厅科研基础能力提升项目（2021KY1751）、北部湾环境演变与资源利用教育部重点实验室（南宁师范大学）开放基金项目（NNNU-KLOP-K2103）
作者简介:黄启桥（1991—），男，在读硕士研究生，研究方向：人工智能应用研究。E-mail：huangqiqiao163@163.com
通讯作者:麦雄发（1974—），男，副教授，研究方向：智能计算与机器学习。E-mail：maixf@nnnu.edu.cn
更新日期/Last Update: 1900-01-01