基于机器学习短历时暴雨时空分布规律研究
|
摘要
城市内涝风险的精细化管理和防洪排涝市政工程的科学设计,需要对当地降雨的时空分布特征有深入的了解。而传统以单站雨型代表整个区域降雨特征的分析方法,不能满足这一要求。本文尝试将机器算法引入到暴雨时空分布特征研究中,以北京城区2004—2016年降雨资料为研究样本,利用动态聚类算法,提取北京城区短历时暴雨时空分布的动态特征。经分析,北京汛期的短历时暴雨时空分布特征,可以分为3种类型:(1)降雨自西北部山区移动到城中心区,逐渐扩散到城区;(2)降雨集中在城区西南部地区,逐渐向北部和城中心区扩散;(3)降雨集中在城区中心区和东部地区,基本不发生移动。研究结果表明,基于机器学习算法提取的暴雨时空分布特征,与实际暴雨时空动态发展趋势相符,并且有各自对应的降雨形成的不同物理机制,可为城区降雨设计、城市内涝风险管理等工作提供借鉴与参考。
It is necessary and important to understand the spatial and temporal distribution characteristics of rainfall in order to improve management of urban waterlogging risk and scientific design of municipal works for flood control and drainage. However,it is hard to derive accurate prediction based on monitoring data of a single station or the average of multiple stations. Availing of machine learning algorithm,this paper contributes to investigating rainfall spatial and temporal distribution characteristics. Specifically,the technique of dynamic clustering algorithm examines the spatial and temporal distribution characteristics of rainfall in Beijing urban area during recent 10 years. The result shows that there are three types of rainstorms in Beijing urban area. Rainstorms of type 1 move from the northwest to the center of Beijing,then spreads to the eastern part of the urban area;rainstorms of type 2 occurs in the southwestern region of the urban area,and gradually northward,but there is no rainfall in the mountainous northwest;rainstorms of type 3 are concentrated in the central and eastern regions,and basically does not move. The results are consistent with the actual rainstorm process. It provides references for urban rainfall design and urban waterlogging risk management.
It is necessary and important to understand the spatial and temporal distribution characteristics of rainfall in order to improve management of urban waterlogging risk and scientific design of municipal works for flood control and drainage. However,it is hard to derive accurate prediction based on monitoring data of a single station or the average of multiple stations. Availing of machine learning algorithm,this paper contributes to investigating rainfall spatial and temporal distribution characteristics. Specifically,the technique of dynamic clustering algorithm examines the spatial and temporal distribution characteristics of rainfall in Beijing urban area during recent 10 years. The result shows that there are three types of rainstorms in Beijing urban area. Rainstorms of type 1 move from the northwest to the center of Beijing,then spreads to the eastern part of the urban area;rainstorms of type 2 occurs in the southwestern region of the urban area,and gradually northward,but there is no rainfall in the mountainous northwest;rainstorms of type 3 are concentrated in the central and eastern regions,and basically does not move. The results are consistent with the actual rainstorm process. It provides references for urban rainfall design and urban waterlogging risk management.
引文
[1]王浩,梅超,刘家宏.海绵城市系统构建模式[J].水利学报,2017,48(9):1009-1014,1022.
[2]张炜,李思敏,时真男.我国城市暴雨内涝的成因及其应对策略[J].自然灾害学报,2012,21(5):180-184.
[3]程晓陶.城市型水灾害及其综合治水方略[J].灾害学,2010,25(S1):10-15.
[4] XIA Jun,SHI Wei. Perspective on water security issue of changing environment in China[J]. Journal of Hydraulic Engineering,2016,47(3):292-301.
[5] RAFIEEINASAB A,NOROUZI A. Toward high-resolution flash flood prediction in large urban areas-Analysis of sensitivity to spatiotemporal resolution of rainfall input and hydrologic modeling[J]. Journal of Hydrology,2015,531(2):370-388.
[6]吴彰春,岑国平,安智敏.坡面汇流的试验研究[J].水利学报,1995(7):84-89.
[7] PILGRIM D H,CORDERY I. Rainfall temporal patterns for design floods[J]. Journal of the Hydraulics Division,1975,101(1):81-95.
[8] KEIFER C J. CHU Henry Hsien. Synthetic storm pattern for drainage design[J]. Journal of the Hydraulics Division,1975,101(10):1371-1371.
[9] HUFF F A. Time distribution of rainfall in heavy storms[J]. Water Resources Research,1967,3(4):1007-1019.
[10] LECUN Y,BENGIO Y,HINTON G. Deep learning[J]. Nature,2015,521:436-444.
[11]王威,李郁,张文娟,等.深度学习技术在疾病诊断中的应用[J].第二军医大学学报,2018,39(8):852-858.
[12]张文海,李磊.人工智能在冰雹识别及临近预报中的初步应用[J].气象学报,DOI:10.11676/qxxb2019.014.
[13]杨明极,马池,王娅,等.一种改进的K-means聚类的FCMM算法[J].计算机应用研究,DOI:10.3969/j.issn.1001-3695.2017.12.0851.
[14]刘熙明,胡非,李磊,等.北京地区夏季城市气候趋势和环境效应的分析研究[J].地球物理学报,2006,49(3):689-697.
[15]北京市气象局.北京市气象灾害预警信号与防御指南[Z]. 2016.
[16]陈静,刘琳.2011年汛期北京城市暴雨特征及其灾害成因初步分析[J].暴雨灾害,2011,30(3):282-287.
[17]化全利,吴海山,白国营. 2004年7月10日北京城区暴雨分析及减灾措施[J].水文,2005,25(3):63-64.
[2]张炜,李思敏,时真男.我国城市暴雨内涝的成因及其应对策略[J].自然灾害学报,2012,21(5):180-184.
[3]程晓陶.城市型水灾害及其综合治水方略[J].灾害学,2010,25(S1):10-15.
[4] XIA Jun,SHI Wei. Perspective on water security issue of changing environment in China[J]. Journal of Hydraulic Engineering,2016,47(3):292-301.
[5] RAFIEEINASAB A,NOROUZI A. Toward high-resolution flash flood prediction in large urban areas-Analysis of sensitivity to spatiotemporal resolution of rainfall input and hydrologic modeling[J]. Journal of Hydrology,2015,531(2):370-388.
[6]吴彰春,岑国平,安智敏.坡面汇流的试验研究[J].水利学报,1995(7):84-89.
[7] PILGRIM D H,CORDERY I. Rainfall temporal patterns for design floods[J]. Journal of the Hydraulics Division,1975,101(1):81-95.
[8] KEIFER C J. CHU Henry Hsien. Synthetic storm pattern for drainage design[J]. Journal of the Hydraulics Division,1975,101(10):1371-1371.
[9] HUFF F A. Time distribution of rainfall in heavy storms[J]. Water Resources Research,1967,3(4):1007-1019.
[10] LECUN Y,BENGIO Y,HINTON G. Deep learning[J]. Nature,2015,521:436-444.
[11]王威,李郁,张文娟,等.深度学习技术在疾病诊断中的应用[J].第二军医大学学报,2018,39(8):852-858.
[12]张文海,李磊.人工智能在冰雹识别及临近预报中的初步应用[J].气象学报,DOI:10.11676/qxxb2019.014.
[13]杨明极,马池,王娅,等.一种改进的K-means聚类的FCMM算法[J].计算机应用研究,DOI:10.3969/j.issn.1001-3695.2017.12.0851.
[14]刘熙明,胡非,李磊,等.北京地区夏季城市气候趋势和环境效应的分析研究[J].地球物理学报,2006,49(3):689-697.
[15]北京市气象局.北京市气象灾害预警信号与防御指南[Z]. 2016.
[16]陈静,刘琳.2011年汛期北京城市暴雨特征及其灾害成因初步分析[J].暴雨灾害,2011,30(3):282-287.
[17]化全利,吴海山,白国营. 2004年7月10日北京城区暴雨分析及减灾措施[J].水文,2005,25(3):63-64.