灌溉水門自動化之監測與調控架構建構

Development of the supervising and controlling framework for irrigation canal automation

洪嘉鴻(Chia-Hung Hung)

指導教授 ： 余化龍

國立臺灣大學/生物資源暨農學院/生物環境系統工程學研究所/碩士(2020年)

https://doi.org/10.6342/NTU202000345

摘要

臺灣的水文狀況特殊，實際可用水量遠低於世界平均，雖然臺灣擁有非常多可以儲存水資源的水庫，但旱災的發生在近年來卻無法避免。因此水資源調度的課題一直以來都被政府與學術界頻繁討論，以目前實務上的操作，旱災發生時通常會減少農業灌溉用水量，將省下的水資源調度至工業及民生用水。過往研究針對灌溉用水的開源節流做了許多探討，為的就是降低不同用水部門間的需水衝突，但卻較少探討現行灌溉水粗放的輸配方式所導致的浪費，且因為灌溉渠道的物理限制，使現行水資源調度的研究難以真正地在水利會調配灌溉水時實現。
本研究由輸配水方式切入水資源調度的課題，若能改變現行的供水方式，就能夠節省部分一直以來被浪費的灌溉水，以達農業與其他需水部門的雙贏局面。本研究引入在世界各地已經成熟發展的渠道自動化系統(Canal automation system)架構，在臺灣現有的灌溉系統中，建立一個灌溉水門自動化監測與調控的架構，以減少輸配水造成的水資源損失，亦即以智慧灌溉水管理為目標，提出一套合理且概括性的建構流程，以供後續相關部門及或研究單位參考。
本文以軟體面探究智慧灌溉水管理的建置與執行，以概括性的思維探討智慧灌溉水管理建置初期會面臨的主要問題，分為兩大目標：(1)渠道水位感測器最佳布建、(2)渠道控制模式建構與設計，利用水文物理模型模擬渠道系統，探討水位監測資應用於渠道閘門開關的控制，使不同取水點水位能夠保持在目標的高度。
本研究以石門水利會灌區作為研究區域，利用HEC-RAS模式模擬一維的渠道水位變化。應用資訊理論(Information theory)建立智慧灌溉水管理中必要的水位感測器布建演算法，以客觀的方式選擇布建感測器的位置；再利用監測資料發展系集卡曼濾波(Ensemble Kalman filter)方法逼近真實的灌區系統，並搭配適用於高度不確定性模式的模糊控制(Fuzzy control)，進行不斷更新的灌溉水門調控策略，研究成功建立以HEC-RAS模式為基礎的渠道自動化架構，並在架構中的兩個重要主題上有全面性的探討，使相關單位更容易想像並精進智慧灌溉水管理系統。

關鍵字

智慧灌溉水管理 ； 渠道自動化 ； 感測器布建 ； 資訊理論 ； 灌溉水門調控 ； 系集卡曼濾波 ； 模糊控制

並列摘要

Taiwan has a relatively unusual hydrology state, in which the actual accessible water resource is well below the world average. As a result, allocating water resource is a constant issue in Taiwan. Although Taiwan has many reservoirs that can store water resources, the occurrence of drought has been unavoidable in recent years. Therefore, the topic of allocation of water resource among different sectors has been frequently discussed. Based on the current practice, the amount of agricultural irrigation water is usually reduced when a drought occurs, and the saved water resources are reallocated to industrial and municipal demand. In order to reduce the water demand conflicts among different sectors, much research has tried to reduce usage of irrigation water or find new source, but less on the waste caused by the current extensive water distribution methods. Because of the physical constraints of irrigation canals, it is more difficult for the current research on water resources scheduling to be realized when the Irrigation Associations allocate irrigation water.
This research discussed water resource scheduling considering water delivery and distribution. While a portion of irrigation water is currently being wasted during the process, such wastage could potentially be saved if current water supply method improves, which may lead to a win-win situation for agriculture and other water demand sectors. This study introduces the Canal Automation System that has been applied throughout the world, in order to develop a supervising and controlling framework for irrigation canal automation among the existing irrigation system, which will reduce the wastage caused by the distribution. In other words, for this research, a feasible and general structure for intelligent irrigation water management has been proposed for future studies and planning’s consideration.
This study explores the construction and implementation of intelligent irrigation water management, discussing the main problems that intelligent irrigation water management will face in the early stages in a general approach. The two main objectives are: (1) Optimal canal water level sensor deployment, and (2) Canal control model formulation and design. By modeling the irrigation canal with hydrophysical model, and applying the water level monitoring data to simulate the control of the canal gate switch, the recipients’ (farmers) water level can be maintained at the target height.
In this study, the Shimen Irrigation Association irrigation area was explored, and the HEC-RAS model was used to simulate the one-dimensional channel water level change. Information theory was applied to objectively deploying the sensors and to develop the sensor deployment algorithm that is required for the intelligent irrigation water management. The monitoring data is later used for the Ensemble Kalman filter method to approximate the real irrigation system, combined with Fuzzy Control method that suits the high uncertainty nature of this problem, which will provide a constantly updating irrigation gate control strategy. It also has a comprehensive discussion on two important topics in the framework, making it easier for the relevant units to imagine and refine the smart irrigation water management system. In conclusion, a canal automation structure based on HEC-RAS model has been proposed, and comprehensive discussion on the two aforementioned essential issue will also be beneficial for future improvement and refinement of intelligent irrigation water management system.

並列關鍵字

Intelligent irrigation water management ； Canal automation ； Sensor deployment ； Information theory ； Irrigation gate regulation ； Ensemble Kalman filter ； Fuzzy control

參考文獻

Abu-El Magad, M., Kamel, E., & Kamel, K. (1997). Economic assessment of an irrigation canal automation and control project. Paper presented at the Intelligent Control, 1997. Proceedings of the 1997 IEEE International Symposium on.

Alfonso, L., He, L., Lobbrecht, A., & Price, R. (2012). Information theory applied to evaluate the discharge monitoring network of the Magdalena River. Journal of Hydroinformatics, 15(1), 211-228.

Alfonso, L., Lobbrecht, A., & Price, R. (2010a). Information theory–based approach for location of monitoring water level gauges in polders. Water resources research, 46(3).

Alfonso, L., Lobbrecht, A., & Price, R. (2010b). Optimization of water level monitoring network in polder systems using information theory. Water resources research, 46(12).

Alfonso, L., Ridolfi, E., Gaytan-Aguilar, S., Napolitano, F., & Russo, F. (2014). Ensemble entropy for monitoring network design. Entropy, 16(3), 1365-1375.