[1]郝 坤,张天坤,史振威.基于时空特征的热带气旋强度预测方法[J].南京师范大学学报(工程技术版),2019,19(03):001.[doi:10.3969/j.issn.1672-1292.2019.03.001]
　Hao Kun,Zhang Tiankun,Shi Zhenwei.An Tropical Cyclone Intensity Prediction MethodBased on Spatial-Temporal Features[J].Journal of Nanjing Normal University(Engineering and Technology),2019,19(03):001.[doi:10.3969/j.issn.1672-1292.2019.03.001]
点击复制

基于时空特征的热带气旋强度预测方法
分享到：

南京师范大学学报（工程技术版）[ISSN:1006-6977/CN:61-1281/TN]

卷:
19卷
期数:
2019年03期
页码:
001
栏目:
计算机工程
出版日期:
2019-09-30

文章信息/Info

Title:
An Tropical Cyclone Intensity Prediction MethodBased on Spatial-Temporal Features
文章编号:
1672-1292(2019)03-0001-07
作者:
郝 坤; 张天坤; 史振威
北京航空航天大学宇航学院,北京 100083
Author(s):
Hao Kun; Zhang Tiankun; Shi Zhenwei
School of Astronautics,Beihang University,Beijing 100083,China
关键词:
热带气旋; 时空特征; 强度预测; 深度学习
Keywords:
tropical cyclone; spatial-temporal features; intensity prediction; deep learning
分类号:
TP391
DOI:
10.3969/j.issn.1672-1292.2019.03.001
文献标志码:
A
摘要:
热带气旋是一种极具破坏力的天气系统,我国每年都深受其带来的灾害困扰. 目前热带气旋的强度预报业务以统计预报方法为主,通过利用气候持续因子对热带气旋未来的强度建立回归模型,不仅需要进行复杂的特征选择,而且缺乏对周围环境信息的利用,预报精度多年以来都未能有显著提升. 提出了一个能同时提取时序特征与空间特征的热带气旋强度预测网络,针对环境场物理量因子对热带气旋的影响,使用卷积层学习其空间信息,结合循环神经单元对热带气旋的历史时间序列进行建模,实现端到端的预测输出. 在对西北太平洋的热带气旋样本进行测试后,结果表明该24 h强度预测网络显著优于上海台风研究所公布的相应时段其他预报方法,故可作为一种新的智能预测模型,为预报员提供有价值的客观参考.
Abstract:
Tropical Cyclone(TC)is a destructive weather system,which causes disasters every year in China. At present,researchers usually develop statistical forecast methods for TC intensity forecast. They use the climatic persistence factors to develop a regression model for the future intensity of TC. Such model,however,needs a complex procedure of feature selection and lacks the use of information in the surrounding environment. So the forecast accuracy has not been significantly improved over the recent years. This paper proposes a TC intensity prediction model that can extract spatial and temporal features simultaneously. As for the influence of environmental physical factors on TC,the convolutional layer is used to learn its spatial information,and the recurrent neural unit is used to model the historical time series of tropical cyclone to achieve an end-to-end prediction. Having tested the TC samples in the Northwest Pacific,the results show that our spatial-temporal intensity prediction network is superior to other forecast methods published by Shanghai Typhoon Institute(STI)in the corresponding period,so it can be used as a new intelligent prediction model to provide valuable objective reference for forecasters.

参考文献/References:

[1] CAGIGAL G L,MéNDEZ I F J,RUEDA A,et al. On the prediction of tropical cyclones wind wave footprint using satellite data and neural networks[C]//AGU Fall Meeting. Washington,DC:American Geophysical Union,2018.
[2]董祎杰. 1998—2017年夏季西北太平洋台风活动特征研究[J]. 中国高新科技,2019(1):124-126.
DONG Y J. A study on characteristics of typhoon activity in summer in the Northwest Pacific Ocean from 1998 to 2017[J]. China High-Tech,2019(1):124-126.(in Chinese)
[3]KNAFF J A,SAMPSON C R,DEMARIA M. An operational statistical typhoon intensity prediction scheme for the Western North Pacific[J]. Weather and forecasting,2005,20(4):688-699.
[4]李晴岚. 西北太平洋热带气旋强度预报研究[C]//中国气象学会. 第35届中国气象学会年会 S1 灾害天气监测、分析与预报. 合肥:中国气象学会,2018.
LI Q L. A study on tropical cyclone intensity forecast in the northwest Pacific[C]//Chinese Meteorological Society. Meeting S1 of Annual Meeting of Chinese Meteorological Society on Disaster Weather Monitoring,Analysis and Forecast. Hefei,China:Chinese Meteorological Society,2018.(in Chinese)
[5]张守峰,余晖,向纯怡. 中央气象台台风强度综合预报误差分析[J]. 气象,2015,41(10):1278-1285.
ZHANG S F,YU H,XIANG C Y. Comprehensive forecast error analysis of typhoon intensity of central meteorological station[J]. Meteorological monthly,2015,41(10):1278-1285.(in Chinese)
[6]周聪. 基于偏最小二乘法的西北太平洋热带气旋强度预报模式[D]. 青岛:中国海洋大学,2014.
ZHOU C. Intensity forecasting model of tropical cyclone over the Northwestern Pacific based on partial least squares method[D]. Qingdao:Ocean University of China,2014.(in Chinese)
[7]陈彩珠. 西北太平洋热带气旋强度预报方法的评估[D]. 南京:南京信息工程大学,2011.
CHEN C Z. Evaluating the performance of tropical cyclone intensity guidance over the Northwest Pacific[D]. Nanjing:Nanjing University of Information Science & Technology,2011.(in Chinese)
[8]FITZPATRICK P J. Understanding and forecasting tropical cyclone intensity change with the typhoon intensity prediction scheme(TIPS)[J]. Weather and forecasting,1997,12(4):826-846.
[9]宋金杰,王元,陈佩燕,等. 基于偏最小二乘回归理论的西北太平洋热带气旋强度统计预报方法[J]. 气象学报,2011,69(5):745-756.
SONG J J,WANG Y,CHEN P Y,et al. A statistical prediction method of tropical cyclone intensity in the Northwest Pacific based on partial least squares regression theory[J]. Acta meteorologica sinica,2011,69(5):745-756.(in Chinese)
[10]黄小燕. 模糊神经网络在热带气旋强度预报中的应用研究[C]//第26届中国气象学会年会热带气旋科学研讨会分会场论文集.北京:中国气象学会,2009.
HUANG X Y. Application of fuzzy neural network in tropical cyclone intensity forecast[C]//Proceedings of the 26th Annual Conference of Chinese Meteorological Society on Tropical Cyclone Science. Beijing,China:Chinese Meteorological Society,2009.(in Chinese)
[11]史旭明,金龙,黄小燕. 基于KPCA的台风强度神经网络集合预报方法研究[J]. 气象科学,2013,33(2):184-189.
SHI X M,JIN L,HUANG X Y. A new neural network ensemble forecast method based on KPCA for typhoon intensity[J]. Journal of the meteorological sciences,2013,33(2):184-189.(in Chinese)
[12]黄颖,金龙,黄小燕,等. 基于局部线性嵌入的人工智能台风强度集合预报模型[J]. 气象,2014,40(7):806-815.
HUANG Y,JIN L,HUANG X Y,et al. An artificial intelligence ensemble prediction scheme for typhoon intensity using the locally linear embedding[J]. Meteorological monthly,2014,40(7):806-815.(in Chinese)
[13]PAN B,XU X,SHI Z. Tropical cyclone intensity prediction based on recurrent neural networks[J]. Electronics letters,2019,55(7):413-415.
[14]YING M,ZHANG W,YU H,et al. An overview of the China meteorological administration tropical cyclone database[J]. J Atmos Oceanic Technol,2014,31:287-301.
[15]黄小燕,金龙. 条件数在台风移动路径预报中的应用[J]. 自然灾害学报,2007(3):35-40.
HUANG X Y,JIN L. Application of condition number to forecasting moving track of typhoon[J]. Journal of natural disasters,2007(3):35-40.(in Chinese)
[16]HOCHREITER S,SCHMIDHUBER J. Long short-term memory[J]. Neural computation,1997,9(8):1735-1780.
[17]陈国民,汤杰,曾智华. 2011年西北太平洋热带气旋预报精度评定[J]. 气象,2012,38(10):1238-1246.
CHEN G M,TANG J,ZENG Z H. Error analysis on the forecasts of tropical cyclones over Western North Pacific in 2011[J]. Meteorological monthly,2012,38(10):1238-1246.(in Chinese)
[18]陈国民,余晖,曹庆. 2012年西北太平洋热带气旋预报精度评定[J]. 气象,2013,39(10):1350-1358.
CHEN G M,YU H,CAO Q. Error analysis on the forecasts of tropical cyclones over Western North Pacific in 2012[J]. Meteorological monthly,2013,39(10):1350-1358.(in Chinese)
[19]陈国民,曹庆. 2013年西北太平洋热带气旋预报精度评定[J]. 气象,2014,40(12):1549-1557.
CHEN G M,CAO Q. Error analysis on the forecasts of tropical cyclones over Western North Pacific in 2013[J]. Meteorological monthly,2014,40(12):1549-1557.(in Chinese)
[20]陈国民,曹庆,白莉娜. 2014年西北太平洋热带气旋预报精度评定[J]. 气象,2015,41(12):1554-1561.
CHEN G M,CAO Q,BAI L N. Error analysis on the forecasts of tropical cyclones over Western North Pacific in 2014[J]. Meteorological monthly,2015,41(12):1554-1561.(in Chinese)
[21]陈国民,白莉娜,万日金. 2015年西北太平洋热带气旋预报精度评定[J]. 气象,2017,43(4):501-507.
CHEN G M,BAI L N,WAN R J. Error analysis on the forecasts of tropical cyclones over Western North Pacific in 2015[J]. Meteorological monthly,2017,43(4):501-507.(in Chinese)
[22]陈国民,张喜平,白莉娜,等. 2016年西北太平洋和南海热带气旋预报精度评定[J]. 气象,2018,44(4):582-589.
CHEN G M,ZHANG X P,BAI L N,et al. Error analysis on the forecasts of tropical cyclones over Western North Pacific and South China Sea in 2016[J]. Meteorological monthly,2018,44(4):582-589.(in Chinese)
[23]陈国民,张喜平,白莉娜,等. 2017年西北太平洋和南海热带气旋预报精度评定[J]. 气象,2019,45(4):577-586.
CHEN G M,ZHANG X P,BAI L N,Et al. Error analysis on the forecasts of tropical cyclones over Western North Pacific and South China Sea in 2017[J]. Meteorological monthly,2019,45(4):577-586.(in Chinese)

相似文献/References:

[1]王宇环,靳 诚,杜家禛.基于多源数据的成都市居民出行热点时空特征分析[J].南京师范大学学报(工程技术版),2020,20(02):080.[doi:10.3969/j.issn.1672-1292.2020.02.012]
　Wang Yuhuan,Jin Cheng,Du Jiazhen.Temporal and Spatial Characteristics of Traveling Hotspots ofChengdu Residents Based on Multi-source Data[J].Journal of Nanjing Normal University(Engineering and Technology),2020,20(03):080.[doi:10.3969/j.issn.1672-1292.2020.02.012]

备注/Memo

备注/Memo:
收稿日期:2019-07-05.
基金项目:国家重点研发计划(2017YFC1405605).
通讯联系人:史振威,博士,教授,博士生导师,研究方向:图像处理、机器学习. E-mail:shizhenwei@buaa.edu.cn
更新日期/Last Update: 2019-09-30