[1]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(01):41-47.[doi:10.19849/j.cnki.CN45-1356/P.2020.1.09]
　Liang Zhenqing,Chen Sheng.Accuracy evaluation of nowcasting in South China based on deep learning and radar observation[J].Journal of Meteorological Research and Application,2020,41(01):41-47.[doi:10.19849/j.cnki.CN45-1356/P.2020.1.09]
点击复制

基于深度学习和雷达观测的华南短临预报精度评估()
分享到：

气象研究与应用[ISSN:1673-8411/CN:45-1356/P]

卷:
第41卷
期数:
2020年01期
页码:
41-47
栏目:
新技术应用
出版日期:
2020-03-31

文章信息/Info

Title:
Accuracy evaluation of nowcasting in South China based on deep learning and radar observation
作者:
梁振清1;  陈生2; 3
1. 南宁师范大学地理科学与规划学院, 南宁 530001;
2. 中山大学大气科学学院, 广东 珠海 519082;
3. 广东省气候变化与自然灾害研究重点实验室, 广州 510275
Author(s):
Liang Zhenqing1;  Chen Sheng2; 3
1. School of Geography and Planning, Nanning Normal University, Nanning 53001;
2. School of Atmospheric Sciences, Sun Yat-sen University, Zhuhai Guangdong 519082;
3. Guangdong Province Key Laboratory of Climate Change and Natural Disasters, Guangzho
关键词:
深度学习; 神经网络; 降水短临预报; 精度评估
Keywords:
deep learning; neural network; short-time precipitation forecast; accuracy assessment
分类号:
P457.6
DOI:
10.19849/j.cnki.CN45-1356/P.2020.1.09
摘要:
利用最新的深度学习算法，即卷积长短期记忆（Convolution Long-Short Term Memory）神经网络，构建基于深度学习的人工智能短临预报系统，以广州地区2019年3-5月雷达观测的数据为输入进行训练，然后进行短期1h内的降水预报。利用常用的统计评分指标（探测率POD、误报率FAR、临界成功指数CSI，相关系数CC）检验模型。结果表明，预报结果与实际观测的相关系数在1h内预报均保持在0.6以上，在1h内预报探测率均保持在80%以上，临界成功指数在降水强度为10mm·h-1时，基本保持在60%，误报率均小于40%。
Abstract:
The latest deep learning algorithm, namely convolution long-term short-term memory neural network, is used to construct an artificial intelligence short-time prediction forecast system. The radar observation data from March to May 2019 in Guangzhou is used as input for training, and then the short-time precipitation forecast within 1 hour is carried out. The commonly used statistical scoring indicators(POD, FAR, CSI, CC) were used to test the model. The results show that:1) the CC between the prediction results and the actual observation are kept above 0.6; 2) All the POD are above 80%; 3) the CSI are basically kept at 60% when the precipitation intensity is 10mm/h; 4) All FAR are less than 40%.

参考文献/References:

[1] 张小玲,杨波,盛杰,等.中国强对流天气预报业务发展[J].气象科技进展, 2018,8(3):8-18.
[2] 张鹏程,贾旸旸.一种基于多层感知器的动态区域联合短时降水预报方法[J].计算机应用与软件,2018,35(11):159-164+189.
[3] 郑永光,张小玲,周庆亮,等.强对流天气短时临近预报业务技术进展与挑战[J].气象, 2010, 36(7):33-42.
[4] 韩雷,王洪庆,谭晓光,等.基于雷达数据的风暴体识别、追踪及预警的研究进展[J].气象, 2007,33(1):5-12.
[5] NAGARAJAN, Aditya.Explorations into Machine Learning Techniques for Precipitation Nowcasting[DB/OL].https://scholarworks.umass.edu/masters_theses_2/480,2017.
[6] 段婧,苗春生.人工神经网络在梅雨期短期降水分级预报中的应用[J].气象, 2005, 31(8):31-36.
[7] Han L, Dai J, Zhang W, et al. A deep belief network approach using VDRAS data for nowcasting[C]//International Conference on Graphic&Image Processing. Ninth International Conference on Graphic and Image Processing (ICGIP 2017), 2018.
[8] Shi E, Li Q, Gu D, et al. A Method of Weather Radar Echo Extrapolation Based on Convolutional Neural Networks[M]//MultiMedia Modeling Springer Cham 2018.
[9] Jiang L, Zhang W, Han L. Strong convective storm nowcasting using a hybrid approach of convolutional neural network and hidden Markov model[C]//Ninth International Conference on Graphic and Image Processing,2018.
[10] Pan B, Hsu K, AghaKouchak A, et al. Improving precipitation estimation using convolutional neural network[J]. Water Resources Research, 2019, 55(3):2301-2321.
[11] Zhang W, Han L, Sun J, et al. Application of Multi-channel 3D-cube Successive Convolution Network for Convective Storm Nowcasting[C].2019 IEEE International Conference on Big Data (Big Data),2017:1705-1710.
[12] Ayzel G, et al. All convolutional neural networks for radar-based precipitation nowcasting[J]. Procedia Computer Science,2019,150:186-192.
[13] Xingjian S H I, Chen Z, Wang H, et al. Convolutional LSTM network:A machine learning approach for precipitation nowcasting[C]//Advances in neural information processing systems. 2015:802-810.
[14] [1] 李琨,武麦凤.基于SWAN数据的定量降水预报(QPF)研究[C].//中国气象学会:第30届中国气象学会年会论文集,2013:1-5.
[15] 韩丰,沃伟峰. SWAN2.0系统的设计与实现[J].应用气象学报, 2018, 29(1):27-36.
[16] 黄奇章.广东降水气候特征及其成因分析[J].热带地理, 1990, 10(2):113-124.
[17] 伍红雨,邹燕,刘尉.广东区域性暴雨过程的定量化评估及气候特征[J].应用气象学报, 2019(2):233-244.
[18] Graves A. Generating Sequences With Recurrent Neural Networks[J]. Computer Science, 2013.

相似文献/References:

[1].基于神经网络的广州市能见度预报[J].气象研究与应用,2014,35(01):17.
[2].基于MATLAB神经网络对大同市气溶胶浓度预测[J].气象研究与应用,2014,35(01):50.
[3]覃卫坚,黄志,李耀先.广西寒露风开始期短期气候预测方法研究[J].气象研究与应用,2014,35(03):11.
[4]金龙,黄颖,姚才,等.人工智能技术的热带气旋预报综述（之一）——BP神经网络和集成方法的热带气旋预报研究和业务应用[J].气象研究与应用,2020,41(02):1.[doi:10.19849/j.cnki.CN45-1356/P.2020.2.01]
　Jin Long,Huang Ying,Yao Cai,et al.A review of tropical cyclone forecast based on artificial intelligence (part 1)——BP neural network and ensemble method for tropical cyclone forecast research and operational application[J].Journal of Meteorological Research and Application,2020,41(01):1.[doi:10.19849/j.cnki.CN45-1356/P.2020.2.01]
[5]金龙,黄颖,姚才,等.人工智能技术的热带气旋预报综述（之二）——流形学习、智能计算及深度学习的热带气旋预报方法[J].气象研究与应用,2020,41(04):5.[doi:10.19849/j.cnki.CN45-1356/P.2020.4.02]
　Jin Long,Huang Ying,Yao Cai,et al.Summary of tropical cyclone forecasting based on artificial intelligence technology (part 2)——tropical cyclone forecasting methods based on manifold learning, intelligent calculation and deep learning[J].Journal of Meteorological Research and Application,2020,41(01):5.[doi:10.19849/j.cnki.CN45-1356/P.2020.4.02]
[6]何慧,陆虹,覃卫坚,等.人工神经网络在月降水量预测业务中的研究和应用综述[J].气象研究与应用,2021,42(01):1.[doi:10.19849/j.cnki.CN45-1356/P.2021.1.01]
　He Hui,Lu Hong,Qin Weijian,et al.Research and application of artificial neural network in monthly precipitation forecast[J].Journal of Meteorological Research and Application,2021,42(01):1.[doi:10.19849/j.cnki.CN45-1356/P.2021.1.01]
[7]李宗飞,陈凯华,赵玉娟.卷积神经网络和传统算法的雷达面雨量计算效果对比研究[J].气象研究与应用,2021,42(04):89.[doi:10.19849/j.cnki.CN45-1356/P.2021.4.16]
　Li zongfei,Chen Kaihua,Zhao Yujuan.A comparative study of convolution neural network and traditional algorithm in radar area rainfall calculation[J].Journal of Meteorological Research and Application,2021,42(01):89.[doi:10.19849/j.cnki.CN45-1356/P.2021.4.16]
[8]黄友菊,韦强,罗恒,等.基于SAR影像的广西东北地区2022年“龙舟水”洪涝智能监测[J].气象研究与应用,2023,44(01):94.[doi:10.19849/j.cnki.CN45-1356/P.2023.1.16]
　Huang Youju,Wei Qiang,Luo Heng,et al.Intelligent monitoring of flooding of a dragon-boat precipitation process in northeast Guangxi in 2022 based on SAR images[J].Journal of Meteorological Research and Application,2023,44(01):94.[doi:10.19849/j.cnki.CN45-1356/P.2023.1.16]
[9]黄颖,陆虹,黄小燕,等.基于EOF和LSTM的广西月降水量预测模型研究[J].气象研究与应用,2023,44(02):20.[doi:10.19849/j.cnki.CN45-1356/P.2023.2.04]
　Huang Ying,Lu Hong,Huang Xiaoyan,et al.Study on monthly precipitation prediction model in Guangxi based on EOF and LSTM[J].Journal of Meteorological Research and Application,2023,44(01):20.[doi:10.19849/j.cnki.CN45-1356/P.2023.2.04]

备注/Memo

备注/Memo:
收稿日期:2020-01-02。
基金项目:国家自然科学基金项目（41875182）、广州科技局计划项目（201904010162）、中山大学"百人计划"项目（74110-18841203）、广西自然科学基金项目（2018JJA150110）、南宁师范大学-高校高层次人才和教师素质提升（6020303890216）、广西自然科学基金项目（2018GXNSFAA050130）
作者简介:梁振清(1994-),男,在读硕士研究生,研究方向:地图学与地理信息系统。E-mail:liangzhenqi
更新日期/Last Update: 1900-01-01