DocumentCode :
495779
Title :
A Retrospective Temporal Integration Method for Richards´ Equation
Author :
Xiangdong, Chen ; Jun, Xia ; Aizhong, Ye ; Yongyong, Zhang
Author_Institution :
Key Lab. of Water Cycle & Related Land Surface Processes, Chinese Acad. of Sci., Beijing, China
Volume :
2
fYear :
2009
fDate :
March 31 2009-April 2 2009
Firstpage :
591
Lastpage :
595
Abstract :
Numerical simulation is the effective method of solving Richardspsila equation for whose highly-nonlinear character. Most methods of the numerical solutions improved the special discrete methods, or designed on the basis of physical conservation. Those methods can be solved when one time initial value is given. Combining the self-memory principle, the retrospective time integration method for Richardspsila equation is proposed. It is a new kind of time integration scheme which could include multi-time historical data, absorbs the character of stochastic method which forecast by making use of multi-time historical data before the initial time. This scheme is applied to simulate the fixed head vertical infiltration. The results of 1-order retrospective scheme are calculated with given memory coefficients. The stable region of the memory coefficients in 1-order retrospective scheme is also given. The results show that the retrospective scheme can get higher accuracy than the implicit scheme.
Keywords :
finite difference methods; forecasting theory; integration; moisture; partial differential equations; soil; stochastic processes; 1-order retrospective scheme; Richards´ equation; groundwater; numerical simulation; partial differential equations; retrospective temporal integration method; surface water; time integration scheme; Computer science; Difference equations; Differential equations; Finite element methods; Geography; Laboratories; Land surface; Partial differential equations; Soil moisture; Water resources;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Science and Information Engineering, 2009 WRI World Congress on
Conference_Location :
Los Angeles, CA
Print_ISBN :
978-0-7695-3507-4
Type :
conf
DOI :
10.1109/CSIE.2009.591
Filename :
5171407
Link To Document :
بازگشت