首页 | 本学科首页   官方微博 | 高级检索  
     检索      

不同径流尺度的小波神经网络预测
引用本文:汪丽娜,陈晓宏,李艳.不同径流尺度的小波神经网络预测[J].华南师范大学学报(自然科学版),2013,45(2):108-110.
作者姓名:汪丽娜  陈晓宏  李艳
作者单位:1.1.华南师范大学地理科学学院
基金项目:国家自然科学基金重点项目(50839005);国家自然科学基金项目(41001019);广东省自然科学基金博士启动项目(S2011040005992);广东高校优秀青年创新人才培养计划(育苗工程)项目(LYM11049);教育部高等学校博士学科点专项科研基金新教师类资助课题(20114407120006)
摘    要:本文探讨了小波神经网络在径流预测中的应用。该方法结合了神经网络和小波转化的优点,弥补了人工神经网络在预测应用中的不足,较好的追踪预测径流时间序列中的突变性,强转折性,奇异性点等。并且本研究通过与人工神经网络预测方法比较研究,得出小波神经网络能取得更高精度的不同尺度的径流预测值。

关 键 词:径流预测
收稿时间:2011-10-15

Forecasting Approach for Different Scales of Runoff Based on the Wavelet Analysis and Neural Network
WANG Lina,CHEN Xiaohong,LI Yan.Forecasting Approach for Different Scales of Runoff Based on the Wavelet Analysis and Neural Network[J].Journal of South China Normal University(Natural Science Edition),2013,45(2):108-110.
Authors:WANG Lina  CHEN Xiaohong  LI Yan
Institution:1.School of Geography Science,South China Normal University,Guangzhou 510631,China; 2.Center for Water Resources and Environment,Sun Yat-sen University,Guangzhou 510275,China; 3.Key Laboratory of Guangdong Provincial Water and Water Security in South China; 4.Department of Natural Resources and Environment,Guangdong University of Business,Guangzhou 510320,China)
Abstract:This pape discussed the application of the wavelet analysis and neural network method in runoff predicted. This method unified the merits of which are the neural networks and the wavelet transform . The wavelet analysis and neural network method made up the defective of Artifical Neural Network in forecast application. And this method can well trace forecast the special points which have some characteristic, sudden change ,strong sinuous and irregularity, in runoff time series. Comparison was made between the wavelet analysis and neural network method with artificial neural network. The comparison result indicated the wavelet analysis and neural network method obtained the high accuracy runoff predicted value.
Keywords:
本文献已被 CNKI 等数据库收录!
点击此处可从《华南师范大学学报(自然科学版)》浏览原始摘要信息
点击此处可从《华南师范大学学报(自然科学版)》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号