天空云量预报及支持向量机和神经网络方法比较研究
THE STUDY ON FORECAST OF CLOUD AMOUNT WITH SVM AND ANN METHODS
-
摘要: 使用支持向量机和人工神经网络两种方法,分别建立了天空云量的预报模型。利用2001年5月1日~2004年12月31日的武汉市地面、高空观测值及欧洲中心的24小时预报场等资料,通过按不同比例随机抽取样本进行交叉验证的方法,分析了SVM和ANN模型的预报能力和鲁棒性;然后再用全部样本资料建立预报模型,来预报2005年1月1日~5月31日武汉市天空云量。交叉验证和实例预报的结果显示:虽然SVM和ANN模型都表现了较好的预报能力,但SVM的预报能力高于ANN方法,且在计算速度上有ANN无法比拟的优势。Abstract: The cloud amount forecast is carried out based on SVM and ANN methods,and comparisons are made between the two methods. Using the surface and high-level observations in Wuhan and numerical forecast field from ECMWF during May 1st 2001 to December 31st 2004,cross-validations are performed with random samples in different proportion. It shows that both models are able to forecast cloud amount with robustness. And then the models are rebuilt with the whole sample data to forecast the cloud amount in Wuhan from January 1st to May 31st 2005.Cross-validations and real time forecast reveal that SVM is better than ANN in the ability of forecasting cloud amount,especially in computing speed.
-
Key words:
- Support Vector Machine(SVM) /
- Artificial Neural Network(ANN) /
- Model /
- Cloud amount /
- Forecast
-
[1] 李学桥,马莉.神经网络·工程应用[M].重庆:重庆大学出版社,1996:37-41. [2] 薛宇峰,宋丽红,罗泽举.湛江降水量的径向基神经网络预测模型[J].热带气象学报,2006,22(1):91-95. [3] 胡江林,张礼平,宇如聪.神经网络模型预报湖北汛期降水量的应用研究[J].气象学报,2001,59(6):776-783. [4] 王艳姣,张鹰,邓自旺,等.RBF神经网络的汛期旱涝预报方法研究[J].热带气象学报,2005,21(1):102-108. [5] 胡江林,涂松柏,冯光柳.基于人工神经网络的暴雨预报方法探讨[J].热带气象学报,2003,19(4):422-428. [6] JIN Long,LUO Ying,GUO Guang,et al.Study on mixed model of neural network for farmland flood/drought prediction[J].Acta Meteoro Sinica,1997,11(3):364-373. [7] VAPNIK V N.The Nature of Statistical Learning Theory[M].New York:Springer Verlag,2000. [8] CRISTIANINI N,SHAWA-TAYLOR J.An Introduction of Support Vector Machines and Other Kernel-based Learning Methods[M].Cambridge:Cambridge University Press,2000. [9] SCHOLKOPF B,BURGES CH-J C,SMOLA A J,et al.Advances in Kernel Methods-Support Vector Learning[M].Cambridge:MIT Press,1999. [10] 祁亨年.支持向量机及其应用研究综述[J].计算机工程,2004,30(10):6-9. [11] 冯汉中,陈永义.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[J].应用气象学报,2004,15(3):355-365. [12] 张礼平,陈永义,周筱兰.支持向量机(SVM)及其在场预测中的应用[J].热带气象学报,2006,22(3):278-282. [13] 胡江林,李劲.湖北省天空云量的特征分析及其预报[J].湖北气象,2000(2):15-17. [14] 陈永义,俞小鼎,高学浩,等.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[J].应用气象学报,2004,15(3):345-354.
点击查看大图
计量
- 文章访问数: 923
- HTML全文浏览量: 0
- PDF下载量: 976
- 被引次数: 0