Combining Principal Component Analysis and Empirical Orthogonal Function Development of Regional Groundwater Numerical Model Calibration Methodology. A Case Study of Ming-Chu Basin

• Main Authors: I-Huan Hsieh, 謝亦歡
• Other Authors: 徐年盛
• Format: Others
• Language: zh-TW
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/739emg

碩士 === 國立臺灣大學 === 土木工程學研究所 === 105 === This study is aimed to develop a regional groundwater numerical model calibration method. First, use principal component analysis (PCA) on the groundwater section to find out its temporal-spatial variable,and use it as a reference to create new assistance well. Then, applies empirical orthogonal function (EOF) with the change hydrograph of groundwater storage and simulated error hydrograph of groundwater level to quickly and accurately catch and calibrate the temporal-spatial distribuation of water recharge and hydrogeological parameters. The established method was applied on the groundwater system of Ming Chu Basin. This study is aimed to develop a groundwater numerical model calibration method. First, use principal component analysis (PCA) on the groundwater section to get its temporal-spatial variable distribution, and finding the line of eigenvalue=0, and create the new assistance well on it. After, setting the objective function is minimizing the the root mean square error (RMSE) of the simulated and observed groundwater level. The decision variables are horizontal hydraulic conductivity, vertical hydraulic conductivity and water recharge. There are three constraints of the optimization model: (1) the water recharge of groundwater system in every iteration of calibrating process must obey the mass balance; (2) the simulated groundwater level must follow the governing equation of groundwater flow; (3) the value of horizontal hydraulic conductivity and vertical hydraulic conductivity are restricted to a reasonable limits. The process of the optimization model sets the initial value of decision variables first, and inputs the variables to groundwater model. Thus, the groundwater level can be simulated and the objective function will be estimated. If the objective function doesn’t satisfy the stop condition, the simulated error hydrograph of groundwater level will be calculated and analyzed with EOF. Then, the modified decision variables is calculater according to the simulated error hydrograph of groundwater level end the result of EOF analysis. From iterations, the optimal temporal-spatial distribuation of surface water recharge and hydrogeological parameters can be obtain. This study applied the model on the calibration of the groundwater system in Ming Chu Basin. The simulated period is from January 2012 to December 2012 daily. The decision variables were selected in this study are horizontal hydraulic conductivity, vertical hydraulic conductivity of two acqufiers and rain water recharge, river water recharge and boundary water recharge of hydraulic conductivity in first acquifer. The result show that the RMSE is decreased dramatically in early iteration of the calibration and become smoothly after several iterations. The calibrated hydraulic conductivity and vertical leakence are in reasonable limits. The simulated groundwater level can reflect the approximately trendance in all acquifer and can capture the peak of the observed value in first acquifer. Hence, the established method of this study can effectively and accurately calibrate temporal-spatial distribution of surface water recharge and hydrogeological parameters.