违反了 PRIMARY KEY 约束 'PK_t_counter'。不能在对象 'dbo.t_counter' 中插入重复键。 语句已终止。 利用人工神经网络模型预测西北太平洋热带气旋生成频数-An Artificial Neural Network Model to Predict the Frequency of Tropical Cyclones Formed over the Western North Pacific
doi:  10.3878/j.issn.1006-9585.2018.18110
利用人工神经网络模型预测西北太平洋热带气旋生成频数

An Artificial Neural Network Model to Predict the Frequency of Tropical Cyclones Formed over the Western North Pacific
摘要点击 93  全文点击 46  投稿时间:2018-08-11  修订日期:2018-10-29
查看HTML全文   下载PDF   查看/发表评论  下载PDF阅读器
基金:  国家自然科学基金项目 41775063和41475074,国家重点研发计划项目2017YFC1501901
中文关键词:  人工神经网络,热带气旋,西北太平洋,频数
英文关键词:  Artificial Neural Network,tropical cyclone,western North Pacific,frequency
     
作者中文名作者英文名单位
海滢haiying成都信息工程大学大气科学学院
陈光华chenguanghua中国科学院大气物理研究所季风系统研究中心;中国科学院大气物理研究所季风系统研究中心
引用:海滢,陈光华..利用人工神经网络模型预测西北太平洋热带气旋生成频数[J].气候与环境研究
Citation:haiying,chenguanghua..An Artificial Neural Network Model to Predict the Frequency of Tropical Cyclones Formed over the Western North Pacific[J].Climatic and Environmental Research(in Chinese)
中文摘要:
      通过对60年(1950-2009年)北半球夏秋季节(6-10月)热带气旋(TC)频数与春季(3-5月)大尺度环境变量的相关分析,挑选出8个相关性较高的前期预报因子建立人工神经网络(ANN)模型,对2010-2017年8年夏秋季TC频数进行预测,并将预测结果与传统多元线性回归(MLR)方法所得结果进行对比分析。结果表明,ANN模型对60年历史数据的拟合精度高,相关系数高达0.99,平均绝对误差(MAE)低至0.77。在8年预测中,ANN模型相关系数为0.8,MAE值为1.97;而MLR模型相关系数仅为0.46,MAE值为3.3。ANN模型在历史数据拟合和预测中的表现都明显优于MLR模型,未来可考虑应用于实际的业务预测中。
Abstract:
      This study uses the Artificial Neural Network (ANN) method and the Multiple Linear Regression (MLR) method to predict the numbers of the tropical cyclone(TC) formed over the western North Pacific from June to October. The correlations between the frequency of TCs and the large scale environmental variables during the boreal spring (March-May) have been analyzed in 1950-2009, and then the eight highly correlated predictors are selected to predict TC frequency in 2010-2017. The comparison between the ANN and MLR models shows that the ANN model exhibits a better performance. Specifically, the correlation coefficient(R) reaches 0.99 and mean absolute error (MAE) is 0.77 during the historical data simulation. During the prediction period, the R values of ANN and MLR models are 0.8 and 0.46, respectively. The MAE of ANN and MLR models are 1.97 and 3.3, respectively, which further confirms that the performance in the ANN model takes significant advantage over that in the MLR model in both simulation and prediction, and has a good potential for application in the operational forecast.
主办单位:中国科学院大气物理研究所/中国气象学会 单位地址:北京市9804信箱
联系电话: 010-82995048,010-82995413传真:010-82995048 邮编:100029 Email:qhhj@mail.iap.ac.cn
本系统由北京勤云科技发展有限公司设计
京ICP备09060247号