請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/70929
標題: | 應用SWAT模式結合最大熵法模擬灌溉配水過程–以石門灌區為例 Application of SWAT Model Coupled with Maximum Entropy Method to Simulate Irrigation Water Allocation - A Case Study of Shimen Irrigation Area |
作者: | Chia-Ying Lee 李佳穎 |
指導教授: | 余化龍 |
關鍵字: | 灌溉配水,不確定性,SWAT,最大熵法, irrigation water allocation,uncertainty,SWAT,maximum entropy method, |
出版年 : | 2018 |
學位: | 碩士 |
摘要: | 台灣由於其氣候條件,造成整體可利用水資源相當不足,根據2016年水利署統計資料,灌溉用水佔了65.23%的總用水比例,因此如何適當地進行灌溉配水對水資源的調度有很大的幫助,由於灌溉配水過程充滿不確定性,需要考量推估過程中的所有不確定性,並以序率(stochastic)的架構去探討灌溉配水量。
本研究以SWAT模式(Soil and Water Assessment Tool)進行石門灌區水文過程的模擬,並可推估出灌區的需水量。配水端則採用最大熵法(maximum entropy method),使用最大熵法分析灌溉配水過程的優勢在於能在有限的資訊下做推估,且可以考慮現有操作下之不確定性,其所採取的原則就是要盡可能保留推估過程的所有不確定性,使推估結果盡可能保持客觀,其限制式以區間表示不確定性,並嘗試放鬆限制式的不確定性範圍後進行模式收斂,達到隨需而供的灌溉配水架構。 在此不確定性的配水架構下,灌溉系統中的影響參數皆可表示為機率密度函數(probability density function),例如:支渠配水量、地面水取水量、田間灌溉用水量和灌溉系統中之流量,若有更多可靠的資料,滿足更多的限制式,可以使配水量不確定性的範圍縮小,使推估的結果更精確。此序率的配水架構並可結合風險分析領域,提供決策者農業水資源調度的參考,期待本研究對台灣的灌溉配水方式有所幫助。 Due to the climatic conditions in Taiwan, the overall available water resources are quite inadequate. According to the statistics of the Taiwan’s Water Resources Agency in 2016, irrigation water accounts for 65.23% of the total water consumption. Therefore, how to properly implement irrigation water allocation will greatly help the water resources management. Because the process of irrigation water allocation is full of uncertainties, it is necessary to incorporate all the uncertainties into the estimation process, and to analyse the amount of irrigation water in a stochastic framework. In this study, the SWAT model (Soil and Water Assessment Tool) was used to simulate the hydrological process in the Shimen irrigation area so that the irrigation water demand was estimated. The approach adopted to allocate irrigation water is maximum entropy method. The most important advantage of using the maximum entropy method to analyze the irrigatioin water allocation is that the random variables can be estimated under limited information, and the uncertainty under existing irrigation systens can be considered. The principle of the maximum entropy method is to keep all the uncertainties of the estimation process in order to keep the estimation results as objective as possible. The constraints of estimation process were expressed as intervals to indicate uncertainties. After attempting to relax the limited range of uncertainty, the model converges and the uncertainty-based framework of irrigation water allocation is available on water demand. Under this uncertainty-based framework, the influence parameters in the irrigation system can be expressed as probability density functions, such as irrigation water allocation in branch canals, surface water withdrawals, irrigation water demand and flow rates in the irrigation system. Also, If there are more reliable data and more constraints are satisfied, the range of uncertainty in the irrigation water allocation can be reduced so that the estimated results would be more accurate. Furthermore, the framework can be further combined with risk analysis to provide some useful information for the decision-makers to manage the agricultural water resources. It is expected that this study would be beneficial to the allocation methods of irrigation water in Taiwan. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/70929 |
DOI: | 10.6342/NTU201802443 |
全文授權: | 有償授權 |
顯示於系所單位: | 生物環境系統工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-107-1.pdf 目前未授權公開取用 | 8.69 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。