地下水系統場置性抽水型態之辨識 及其抽水量之估算

Local Pumping Source Identification and Pumping Estimation of Groundwater System

王仁甫(Ren-Fu Wang)

指導教授 ： 徐年盛

國立臺灣大學/工學院/土木工程學研究所/碩士(2011年)

https://doi.org/10.6342/NTU.2011.01183

摘要

本研究之目的在於建立一套方法，結合訊號分析技術與地下水數值模式，以供辨識個別場置性抽水型態及估算其抽水量，並將建立之方法應用於濁水溪沖積扇溪州地區以彰顯其可用性。本研究使用獨立成份分析法，由觀測井所觀測到的地下水位變化分離出未知的抽水型態，並計算不同抽水型態所造成的水位變化，再藉由模式的模擬，在觀測井的有效觀測範圍內佈設水井抽水，檢定該場置性抽水量以擬合水位振幅。
本研究將方法應用於濁水溪沖積扇上之溪州鄉抽水定量，結果顯示，溪州(1)觀測井觀測到農業型抽水造成7.96公分洩降，抽水量為每天4457噸；節費型A抽水造成3.09公分洩降，抽水量每天1730噸；節費型B抽水造成5.11公分洩降，抽水量為每天2861噸；溪州(2)觀測井觀測到農業型抽水造成8.9公分洩降，抽水量為每天8188噸；節費型A抽水造成4.02公分洩降，抽水量為每天3698噸；節費型B抽水造成3.23公分洩降，抽水量為每天2971噸。溪州(3)觀測井觀測到農業型抽水造成9.79公分洩降，抽水量為每天9006噸；節費型A抽水造成1.92公分洩降，抽水量每天1766噸；節費型B抽水造成3.14公分洩降，抽水量為每天2888噸。

關鍵字

地下水 ； 獨立成分分析 ； 抽水量 ； 濁水溪沖積扇 ； 數值模式

並列摘要

The purpose of this study is to establish a method for Combining with signal analysis techniques and numerical groundwater model for the identification of local pumping source s and estimate the amount of its pumping .The method of this study is applied to pumping estimation at Hsichou of Jhuoshuei river alluvial fan.This study use ICA to identify the pumping source from observed groundwater level and estimate the amount of pumping source. By model simulations, set pumping wells in the effective diameter of local observation wells and estimate the local pumping to fit the level of pumping amplitude.
The method of this study is applied to pumping estimation at Hsichou of Jhuoshuei river alluvial fan .The amount of local pumping at Hsichou(1) shows 4457 ton/day of agricultural pumping and with 7.96cm drawdown, 1730 ton/day of charge-saving pumping A and with 3.09cm drawdown, 2861 ton/day of charge-saving pumping B and with 5.11cm drawdown；The amount of local pumping at Hsichou(2) shows 3698 ton/day of agricultural pumping and with 4.02cm drawdown, 1730 ton/day of charge-saving pumping A and with 3.09cm drawdown, 2971 ton/day of charge-saving pumping B and with 3.23cm drawdown；The amount of local pumping at Hsichou(3) shows 9006 ton/day of agricultural pumping and with 9.79cm drawdown, 1766 ton/day of charge-saving pumping A and with 1.92cm drawdown, 2888 ton/day of charge-saving pumping B and with 3.14cm drawdown.

並列關鍵字

Groundwater ； ICA ； Pumping ； Modflow ； numerical model

參考文獻

石麗娟，「獨立成份分析法與類神經網路於股價時間序列分析與預測之研究」，國立成功大學資訊管理研究所碩士論文，2007。

趙昶凱，「新穎獨立成份應用於雜訊語音辨識」，國立成功大學資訊工程研究所碩士論文，2007。

Van Hateren, J. H. and Ruderman, D. L., “Independent Component Analysis of Natural Image Sequences Yields Spatio-temporal Filters Similar to Simple Cells in Primary Visual Cortex,” Proc. Royal Society, pp. 2315-2320, 1998.

Hyvärinen,A., “Independent Component Analysis in the Presence of Gaussian Noise by Maximizing Joint Likelihood,” Neurocomputing, Vol. 22, pp. 49-67, 1998.

Hyvärinen, A., “Fast and Robust Fixed-point Algorithms for Independent Component Analysis,” IEEE Trans. on Neural Networks, Vol. 10, pp. 626–634, 1999.