张文海,李磊. 2019. 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, (0):-, doi:10.11676/qxxb2019.014
人工智能在冰雹识别及临近预报中的初步应用
An initial application of artificial intelligence on the detection and nowcasting of hail weather
投稿时间:2018-03-24  修订日期:2018-07-03
DOI:10.11676/qxxb2019.014
中文关键词:  冰雹识别,临近预报,人工智能,机器学习,贝叶斯分类
英文关键词:Hail detection, Nowcasting, Artificial Intelligence, Machine Learning, Bayes Classifier
基金项目:国家自然科学基金
作者单位E-mail
张文海 深圳市强风暴科学研究院 17350048@qq.com 
李磊 深圳市国家气候观象台 chonp@163.com 
摘要点击次数: 253
全文下载次数: 278
中文摘要:
      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。在建立算法时,以雷达回波反射率的垂直和水平切片数据为基础训练集,将冰雹云的雷达反射率切片数据作为正样本,将其他雷达反射率切片数据作为负样本,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。在进行训练时,以广东省2008-2013和2015-2016年8年的数据作为训练集,在检验时,使用了2014年广东省12次冰雹过程的数据。对比检验的结果令人鼓舞,人工智能法比传统的概念模型法识别击中率高了9%。研究结果表明人工智能对冰雹这类非线性强天气过程识别的强大能力。
英文摘要:
      Based on the 3-dimensional mosaic reflectivity data from 10 S-band Doppler radars in Guangdong province, an artificial intelligence (AI) algorithm for automatic hail detection and nowcasting is developed in the light of the machine learning (ML) technology. The training data set used to develop the algorithm of include vertical and horizontal slice data of the mosaic radar reflectivity, in which the slice data of hail clouds are taken as positive samples, the other sections of the others as negative samples. The Bayes classifier method is used during the training process of ML to establish the capability of AI on recognize the characteristics of the hail cloud. The data during the period of 2008-2013 and 2015-2016 were taken as training set, while the observed data during the 12 hail weather processes in 2014 were used to validate the capability of AI. The result of comparative validation is encouraging, and the AI method is 9 percentage points higher than the traditional Conceptual Model method on identifying hit rate. The current study preliminarily shows the strong capability of AI on identify the nonlinear strong weather processes.
查看全文   查看/发表评论  下载PDF阅读器
分享按钮