EEMD應用於暴雨基流分離

Baseflow separation using EEMD

劉宏毅(Hong-Yi Liu)

指導教授 ： 陳彥璋

國立臺北科技大學/工程學院/土木與防災研究所/碩士(2014年)

https://doi.org/10.6841/NTUT.2014.00779

摘要

水文歷線代表直接逕流及基流兩部分總和，直接逕流大部分是由超滲降雨形成，為造成洪水之主因；基流為河川在乾旱情況下所排出之流量。河川流域的水文特性與各類水文災害之間的關係十分密切。因此，了解台灣河川水文特性以及影響台灣水文特性的因素，是了解台灣各類水文災害的關鍵。台灣因位處亞熱帶以及世界最大陸地與最大海洋之間，季風氣候極為顯著，夏季高溫多雨，降雨豐沛；冬季則十分乾旱少雨，使得台灣河川洪枯流量差距極大。
不同以往使用經驗模式做基流分離之方法，這些方法或過於主觀、或不具物理概念或與實際水文狀況不符。本研究主要為暴雨期間，比較傳統的基流分離方法與使用EEMD做基流分離之間的差異，將台灣北部北勢溪-思源橋及黸魚堀溪-大林橋兩河川流量資料，為解決模態混雜(mode mixing)的問題，使用了整體經驗模式解構法(Ensemble Empirical Model Decomposition(EEMD))做為基流分離之工具。若未來可以有效地利用EEMD進行基流分離，將可以客觀且具物理意義的方式做基流分離分析，無論在學術理論上或工程實務上都可已有很大的幫助。

關鍵字

基流分離 ； 暴雨 ； EEMD

並列摘要

Hydrograph representative the sum of two parts-the direct runoff and the baseflow. The direct runoff is formed by a super majority of rainfall infiltration is the primary cause of floods. The baseflow in the river flow is discharged from the drought conditions. The relationship between the hydrological characteristics of river basin and various types of hydrological disasters are very close. Therefore, understanding the hydrological characteristics of the Taiwan’s rivers, and the factors that affect the hydrologic characteristics of Taiwan, is the key to understanding all kinds of hydrological disasters of Taiwan. Taiwan is located at subtropical climate, and between the world's largest land and largest ocean, the monsoon climate is extremely significant. It’s hot and rainy in summer, and abundant rainfall. However it’s very drought in winter, that makes the gap of Taiwan rivers is enormous between flow and dry season.
Different from the past by using empirical mode as the method of baseflow separation, these methods are either too subjective, or non-physical concepts, or inconsistent with the actual hydrological conditions. This study compare the difference between traditional method of base flow separation method, and by using EEMD (Ensemble Empirical Model Decomposition) as the base flow separation in the rainstorm season. We study the flow data of two rivers in northern Taiwan- the Peishih River – Siyuan bridge and Daiyujue River – Dalin bridg. In order to solve the problem of mode mixing, We use the Ensemble Empirical Model Decomposition (EEMD) as a tool for the separation of base flow. If it can be effectively utilized EEMD as the base flow separation in the future, it will be able to do the base flow separation analysis in objectively possessed and physical sense way. It can be a great help both in academic theory or in engineering practices.

並列關鍵字

Baseflow separation ； storm ； EEMD

參考文獻

21. 林明億，感潮河段水位構成要素分析，碩士論文，國立台北科技大學，土木與防災研究所，台北市，2011。

25. 陳慰平，由河川流量資料與流量歷線推估地下水補注量，碩士論文，國立成功大學，資源工程研究所，台南市，1999。

27. 陳岱源，修正型水桶模式應用於基流分離之研究，碩士論文，國立成功大學，水利暨海洋工程研究所，台南市，1995。

28. 楊宗翰，地震引致河川基流量變化分析~以集集大地震為例，碩士論文，國立成功大學，水利暨海洋工程研究所，台南市，1993。

1. Birtles, A.B. ,"Identification and separation of major base flow components from a stream hydrograph", Water Resources Research, Volume 14, Issue 5, 1978, pp.791-803.