人工智能在冰雹识别及临近预报中的初步应用
|
摘要
基于广东10部S波段多普勒天气雷达的三维拼图资料,利用机器学习技术开发了一种冰雹识别和临近预报的人工智能算法。算法设计时以雷达回波反射率的垂直和水平扫描数据为基础训练集,将冰雹云的雷达反射率扫描数据作为正样本,将其他雷达反射率扫描数据作为负样本,通过贝叶斯分类法对正、负样本数据集进行机器学习,训练人工智能识别冰雹云内在规律的能力。训练时以广东省2008—2013和2015—2016年的数据作为训练集,使用了2014年广东省12次冰雹过程的数据做检验。对比检验的结果表明,人工智能法比传统的概念模型法击中率高9个百分点。研究结果表明了人工智能对冰雹这类非线性强天气过程具有较强的识别能力。
Based on three-dimensional mosaic reflectivity data from ten S-band Doppler radars in Guangdong Province, an artificial intelligence(AI) algorithm for automatic hail detection and nowcasting is developed using the machine learning technology. The training dataset used to develop the algorithm includes vertical and horizontal slice data of the mosaic radar reflectivity, in which the slice data of hail clouds are taken as positive samples, and other data are used as negative samples. The Bayes classifier method is used during the training process of machine learning to establish the capability of AI on recognizing the characteristics of the hail cloud. The data during the period of 2008-2013 and 2015-2016 are taken as training sets, while the observed data during the 12 hail weather processes in 2014 are used to validate the capability of AI. The result of comparative validation is encouraging, and the AI method is 9% higher than the traditional Conceptual Model method on identifying the hit rate. The current study preliminarily shows the strong capability of AI on identifying the nonlinear strong weather processes.
Based on three-dimensional mosaic reflectivity data from ten S-band Doppler radars in Guangdong Province, an artificial intelligence(AI) algorithm for automatic hail detection and nowcasting is developed using the machine learning technology. The training dataset used to develop the algorithm includes vertical and horizontal slice data of the mosaic radar reflectivity, in which the slice data of hail clouds are taken as positive samples, and other data are used as negative samples. The Bayes classifier method is used during the training process of machine learning to establish the capability of AI on recognizing the characteristics of the hail cloud. The data during the period of 2008-2013 and 2015-2016 are taken as training sets, while the observed data during the 12 hail weather processes in 2014 are used to validate the capability of AI. The result of comparative validation is encouraging, and the AI method is 9% higher than the traditional Conceptual Model method on identifying the hit rate. The current study preliminarily shows the strong capability of AI on identifying the nonlinear strong weather processes.
引文
曹春燕,陈元昭,刘东华等.2015.光流法及其在临近预报中的应用.气象学报,73(3):471-480.Cao C Y,Chen Y Z,Liu D H,et al.2015.The optical flow method and its application to nowcasting.Acta Meteor Sinica,73(3):471-480 (in Chinese)
陈永义,俞小鼎,高学浩等.2004.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,15(3):345-354.Chen Y Y,Yu X D,Gao X H,et al.2004.A new method for non-linear classify and non-linear regression Ⅰ:Introduction to support vector machine.J Appl Meteor Sci,15(3):345-354 (in Chinese)
冯汉中,陈永义.2004.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,15(3):355-365.Feng H Z,Chen Y Y.2004.A new method for non-linear classify and non-linear regression Ⅱ:Application of support vector machine to weather forecast.J Appl Meteor Sci,15(3):355-365 (in Chinese)
胡家美,刘峰,黄奕铭等.2008.冰雹和低空风切变天气的多普勒雷达回波特征.广东气象,30(1):24-28.Hu J M,Liu F,Huang Y M,et al.2008.Signatures of hail and low level shear from Doppler radar products.Guangdong Qixiang,30(1):24-28 (in Chinese)
兰红平,孙向明,梁碧玲等.2009.雷暴云团自动识别和边界相关追踪技术研究.气象,35(7):101-111.Lan H P,Sun X M,Liang B L,et al.2009.An automatic tracking and recognition algorithm for thunderstorm cloud-cluster (TRACER).Meteor Mon,35(7):101-111 (in Chinese)
苏德斌,马建立,张蔷等.2011.X波段双线偏振雷达冰雹识别初步研究.气象,37(10):1228-1232.Su D B,Ma J L,Zhang Q,et al.2011.Preliminary research on method of hail detection with X band dual linear polarization radar.Meteor Mon,37(10):1228-1232 (in Chinese)
王瑾,李明元,汪华.2010.基于多雷达三维插值格点强冰雹诊断因子的强冰雹识别方法.高原气象,29(6):1533-1545.Wang J,Li M Y,Wang H.2010.A new hail detection algorithm of hail diagnostic parameters based on 3D mosaicing data of multi-radar.Plateau Meteor,29(6):1533-1545 (in Chinese)
吴书君,王凤娇,刘昭武等.2006.两种垂直累积液态含水量产品的应用对比分析.气象,32(10):34-40.Wu S J,Wang F J,Liu Z W,et al.2006.Applied comparison of two products of vertically integrated liquid water content.Meteor Mon,32(10):34-40 (in Chinese)
肖艳姣,刘黎平.2006.新一代天气雷达网资料的三维格点化及拼图方法研究.气象学报,64(5):647-657.Xiao Y J,Liu L P.2006.Study of methods for interpolating data from weather radar network to 3-D grid and mosaics.Acta Meteor Sinica,64(5):647-657 (in Chinese)
肖艳姣,马中元,李中华.2009.改进的雷达回波顶高、垂直积分液态水含量及其密度算法.暴雨灾害,28(3):210-214.Xiao Y J,Ma Z Y,Li Z H.2009.Advanced algorithms of ET,VIL and VIL density for the CINRAD.Torr Rain Dis,28(3):210-214 (in Chinese)
俞小鼎,姚秀萍,熊廷南等.2006.多普勒天气雷达原理与业务应用.北京:气象出版社,314pp.Yu X D,Yao X P,Xiong T N,et al.2006.The Principle and Operational Application of Doppler Weather Radar.Beijing:China Meteorological Press,314pp (in Chinese)
俞小鼎,周小刚,王秀明.2012.雷暴与强对流临近天气预报技术进展.气象学报,70(3):311-337.Yu X D,Zhou X G,Wang X M.2012.The advances in the nowcasting techniques on thunderstorms and severe convection.Acta Meteor Sinica,70(3):311-337 (in Chinese)
俞小鼎.2014.关于冰雹的融化层高度.气象,40(6):649-654.Yu X D.2014.A note on the melting level of hail.Meteor Mon,40(6):649-654 (in Chinese)
曾晓梅.1999.国外人工智能技术在天气预报中的应用综述.气象科技,(1):4-10.Zeng X M.1999.A review on the application of artificial intelligence in weather forecast abroad.Meteor Sci Technol,(1):4-10 (in Chinese)
张承福.1994.人工神经网络在天气预报中的应用研究.气象,20(6):43-47.Zhang C F.1994.On the application of artificial neural network for weather forecast.Meteor Mon,20(6):43-47 (in Chinese)
周德平,杨洋,王吉宏等.2007.冰雹云雷达识别方法及防雹作业经验.气象科技,35(2):258-263.Zhou D P,Yang Y,Wang J H,et al.2007.Hail cloud identification method by radar echoes and experiences from hail suppression operation in Liaoning province.Meteor Sci Technol,35(2):258-263 (in Chinese)
朱军,胡文波.2015.贝叶斯机器学习前沿进展综述.计算机研究与发展,52(1):16-26.Zhu J,Hu W B.2015.Recent advances in Bayesian machine learning.J Comp Res Dev,52(1):16-26 (in Chinese)
Moninger W R,Davis J,Dyer R,et al.1987.Summary of the first conference on the artificial intelligence research in environmental science (AIRIES).Bull Amer Meteor Soc,68:793-800
Sánchez J L,López L,García-Ortega E,et al.2013.Nowcasting of kinetic energy of hail precipitation using radar.Atmos Res,123:48-60
Silver D,Huang A,Maddison C J,et al.2016.Mastering the game of Go with deep neural networks and tree search.Nature,529(7587):484-489
Skripniková K,■ezá■ová D.2014.Radar-based hail detection.Atmos Res,144:175-185
Smyth T J,Blackman T M,Illingworth A J.1999.Observations of oblate hail using dual polarization radar and implications for hail-detection schemes.Quart J Roy Meteor Soc,125(555):993-1016
Witt A,Eilts M D,Stumpf G J,et al.1998.An enhanced hail detection algorithm for the WSR-88D.Wea Forcasting,13(2):286-303
陈永义,俞小鼎,高学浩等.2004.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,15(3):345-354.Chen Y Y,Yu X D,Gao X H,et al.2004.A new method for non-linear classify and non-linear regression Ⅰ:Introduction to support vector machine.J Appl Meteor Sci,15(3):345-354 (in Chinese)
冯汉中,陈永义.2004.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,15(3):355-365.Feng H Z,Chen Y Y.2004.A new method for non-linear classify and non-linear regression Ⅱ:Application of support vector machine to weather forecast.J Appl Meteor Sci,15(3):355-365 (in Chinese)
胡家美,刘峰,黄奕铭等.2008.冰雹和低空风切变天气的多普勒雷达回波特征.广东气象,30(1):24-28.Hu J M,Liu F,Huang Y M,et al.2008.Signatures of hail and low level shear from Doppler radar products.Guangdong Qixiang,30(1):24-28 (in Chinese)
兰红平,孙向明,梁碧玲等.2009.雷暴云团自动识别和边界相关追踪技术研究.气象,35(7):101-111.Lan H P,Sun X M,Liang B L,et al.2009.An automatic tracking and recognition algorithm for thunderstorm cloud-cluster (TRACER).Meteor Mon,35(7):101-111 (in Chinese)
苏德斌,马建立,张蔷等.2011.X波段双线偏振雷达冰雹识别初步研究.气象,37(10):1228-1232.Su D B,Ma J L,Zhang Q,et al.2011.Preliminary research on method of hail detection with X band dual linear polarization radar.Meteor Mon,37(10):1228-1232 (in Chinese)
王瑾,李明元,汪华.2010.基于多雷达三维插值格点强冰雹诊断因子的强冰雹识别方法.高原气象,29(6):1533-1545.Wang J,Li M Y,Wang H.2010.A new hail detection algorithm of hail diagnostic parameters based on 3D mosaicing data of multi-radar.Plateau Meteor,29(6):1533-1545 (in Chinese)
吴书君,王凤娇,刘昭武等.2006.两种垂直累积液态含水量产品的应用对比分析.气象,32(10):34-40.Wu S J,Wang F J,Liu Z W,et al.2006.Applied comparison of two products of vertically integrated liquid water content.Meteor Mon,32(10):34-40 (in Chinese)
肖艳姣,刘黎平.2006.新一代天气雷达网资料的三维格点化及拼图方法研究.气象学报,64(5):647-657.Xiao Y J,Liu L P.2006.Study of methods for interpolating data from weather radar network to 3-D grid and mosaics.Acta Meteor Sinica,64(5):647-657 (in Chinese)
肖艳姣,马中元,李中华.2009.改进的雷达回波顶高、垂直积分液态水含量及其密度算法.暴雨灾害,28(3):210-214.Xiao Y J,Ma Z Y,Li Z H.2009.Advanced algorithms of ET,VIL and VIL density for the CINRAD.Torr Rain Dis,28(3):210-214 (in Chinese)
俞小鼎,姚秀萍,熊廷南等.2006.多普勒天气雷达原理与业务应用.北京:气象出版社,314pp.Yu X D,Yao X P,Xiong T N,et al.2006.The Principle and Operational Application of Doppler Weather Radar.Beijing:China Meteorological Press,314pp (in Chinese)
俞小鼎,周小刚,王秀明.2012.雷暴与强对流临近天气预报技术进展.气象学报,70(3):311-337.Yu X D,Zhou X G,Wang X M.2012.The advances in the nowcasting techniques on thunderstorms and severe convection.Acta Meteor Sinica,70(3):311-337 (in Chinese)
俞小鼎.2014.关于冰雹的融化层高度.气象,40(6):649-654.Yu X D.2014.A note on the melting level of hail.Meteor Mon,40(6):649-654 (in Chinese)
曾晓梅.1999.国外人工智能技术在天气预报中的应用综述.气象科技,(1):4-10.Zeng X M.1999.A review on the application of artificial intelligence in weather forecast abroad.Meteor Sci Technol,(1):4-10 (in Chinese)
张承福.1994.人工神经网络在天气预报中的应用研究.气象,20(6):43-47.Zhang C F.1994.On the application of artificial neural network for weather forecast.Meteor Mon,20(6):43-47 (in Chinese)
周德平,杨洋,王吉宏等.2007.冰雹云雷达识别方法及防雹作业经验.气象科技,35(2):258-263.Zhou D P,Yang Y,Wang J H,et al.2007.Hail cloud identification method by radar echoes and experiences from hail suppression operation in Liaoning province.Meteor Sci Technol,35(2):258-263 (in Chinese)
朱军,胡文波.2015.贝叶斯机器学习前沿进展综述.计算机研究与发展,52(1):16-26.Zhu J,Hu W B.2015.Recent advances in Bayesian machine learning.J Comp Res Dev,52(1):16-26 (in Chinese)
Moninger W R,Davis J,Dyer R,et al.1987.Summary of the first conference on the artificial intelligence research in environmental science (AIRIES).Bull Amer Meteor Soc,68:793-800
Sánchez J L,López L,García-Ortega E,et al.2013.Nowcasting of kinetic energy of hail precipitation using radar.Atmos Res,123:48-60
Silver D,Huang A,Maddison C J,et al.2016.Mastering the game of Go with deep neural networks and tree search.Nature,529(7587):484-489
Skripniková K,■ezá■ová D.2014.Radar-based hail detection.Atmos Res,144:175-185
Smyth T J,Blackman T M,Illingworth A J.1999.Observations of oblate hail using dual polarization radar and implications for hail-detection schemes.Quart J Roy Meteor Soc,125(555):993-1016
Witt A,Eilts M D,Stumpf G J,et al.1998.An enhanced hail detection algorithm for the WSR-88D.Wea Forcasting,13(2):286-303