模糊控制於地表地下聯合營運之應用

Abstract

模糊控制於地表地下聯合營運之應用 學生：陳柏成 指導教授：張良正 博士 國立交通大學土木工程研究所 摘要 地表地下聯合營運可有效的整合水資源，使水資源得到更大的使用效益，然地表水與地下水性質不同，地表水反應快速，地下水反應遲緩，且地下水系統常為非線性，使聯合營運在應用水資源時不易有效的調配。有鑑於此，本研究嘗試以模糊推論系統結合地表地下聯合營運模式，聯合調配地表水與地下水營運系統的供水量。 本研究的地表地下聯合營運模式整體架構共分為上、下層兩部分，上層部分應用模糊推論系統，將需求水量分配給地表水與地下水兩系統，決定各自系統的供應水量；下層部分則以地表水調配模式與地下水模式構成，地表水模式以線性規劃分配各水庫的放水量，地下水模式則採用類神經網路模擬抽取地下水後之地下水位變化。 本研究進ㄧ步探討本研究所發展的地表地下聯合營運供水效能，並將本模式與較簡單的「地表水先供，地下水後供之營運方式」作比較。研究結果顯示本模式相較於「地表水先供，地下水後供之營運方式」，年缺水指數(SI)可減少26.23%。除了降低缺水指數(SI)，本模式可根據地表水系統與地下水系統的狀態，提供多樣的水量調配方式，使水資源得到合適的運用。在計算量方面，其計算時間約為以遺傳演算法(Genetic Algorithm, GA)為基礎的地表地下聯合營運模式1/760倍。

Applying Fuzzy Control on the Conjunctive Use of Surface and Subsurface Water Student：Bo-Cheng Chen Advisor：Dr. Liang-Cheng Chang Department of Civil Engineering National Chiao Tung University Abstract The conjunctive use of surface and subsurface water can integrate water resources efficiently, it would gain more efficiency when using water resources. However, the characters of surface water and subsurface water are different, and subsurface system often is nonlinear, it causes more difficult while conjunctive use of water was implemented. Therefore, this study had attempted to integrate a surface water allocation model and a groundwater model by using fuzzy inference system and to regulate the water supply for both the surface and subsurface water. The framework of this study can separate into two parts. The first part applied fuzzy inference system to allocate the demand of surface and subsurface water system. The second one is composed of a surface water allocation model and a groundwater model. The surface water allocation model is a linear programming model, and the subsurface model simulates the hydraulic heads after pumping by using artificial neural network (ANN). Moreover, this study discussed the efficiency of the conjunctive use of surface and subsurface water above. Also, it compared this model with the simpler one which is operated with 「surface supply first, then subsurface supply」. Compare with the 「surface supply first, then subsurface supply」 operation strategy, the result shown that this model can decrease the shortage index (SI) 26.23%. This study also affords the more flexible and effective strategy for the conjunctive use of surface and subsurface water. On the other hand, the computing time of this system is 1/760 times of the traditional optimization method.