The Numerical Modeling of Confined Aquifer Pumping Test for a Partial Penetration Well

碩士 === 國立中正大學 === 應用地球物理研究所 === 93 === A single well pumping test is usually performed with a partially penetrating well. Under this situation, the drawdown in the well is highly affected by penetration ratio and the position of screen in aquifer. By using finite element method, this study simulat...

Full description

Bibliographic Details
Main Authors: Bobin, 黃保彬
Other Authors: 呂學諭
Format: Others
Language:zh-TW
Published: 2005
Online Access:http://ndltd.ncl.edu.tw/handle/15857729015087605313
Description
Summary:碩士 === 國立中正大學 === 應用地球物理研究所 === 93 === A single well pumping test is usually performed with a partially penetrating well. Under this situation, the drawdown in the well is highly affected by penetration ratio and the position of screen in aquifer. By using finite element method, this study simulated models with a confined aquifer with 10 m thick and 2000 m wide to describe the relationship between pre-mentioned factors and drawdown in the well but not in the aquifer as the analytical solution. The test model shows that drawdown for the first tens second is non linear in the semi-log plot of drawdown versus log time due to well storage. Therefore, we choose the drawdown value at 1000 second to represent the model drawdown which guarantees a linear relationship in the same plot for all cases. The first model simulated drawdowns under variable penetration ratios. It comes up a regression equation as: .In addition, the models with different screen positions were also simulated. It is interesting that the relationship between drawdown and screen position does not follow a smooth trend and goes up and down. Therefore, a simple regression equation does not exist. To smooth out the curve, a simplified formula is derived as: Despite of penetration ratio and screen position, the drawdown is also variable with pumping rate, hydraulic conductivity and specific storage and they are designated as 10-3m3/s, 10-4m/s and 10-5 m-1, respectively, in this study. It is demonstrated that these parameters show linear relationship with drawdown at 1000 second. Therefore, it is plausible to combined them into a parameter to maintain the linear relationship. Under this situation, the derived regression equations in this study can be applied to other hydrogeological parameters.