Full-Text Search:
Home|About CNKI|User Service|中文
Add to Favorite Get Latest Update

Quantity of Rainfall of a Slope Based on BP Neural Network

LIU Yang1a,LI Hai-ying1b,WANG Lian-yuan1b,LIU Xiao-duan2,GE Xiao-yuan2(1a.Computer Teaching and Research Center;1b.College of Physics,Jilin University,Changchun 130012,China;2.National Research Center for Geoanalysis,Beijing 100037,China)  
In order to predict rainfall of a slope,based on nonlinear characteristics of rainfall runoff of slope,using method of three-layer back-propagation network model,we study the rainfall runoff of slope of the experimental zones in miyun reservoir.The structure of the model has five input variables including slope, length of slope,rainfall intensity,rainfall last time,soil roughness,and the output variable is quantity of rainfall runoff of slope.The network model was trained by using observed data of the zones,after a hundred times the network trends convergence.The error of trained sample books is 2.040 96×10-10,smaller than predetermined precision,the average opposite error is 1.67%.The model provides new method to predict quantity of rainfall runoff of slope.
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©CNKI All Rights Reserved