English  |  正體中文  |  简体中文  |  Items with full text/Total items : 49287/83828 (59%)
Visitors : 7153324      Online Users : 51
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version
    Please use this identifier to cite or link to this item: http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/77485


    Title: 模糊群集方法之研究及其於氣候變遷水文時序辨識之應用
    Other Titles: A study on the fuzzy clustering method for detecting climate change in hydrologic time series
    Authors: 游謦竹;Yu, Chimg-Chu
    Contributors: 淡江大學水資源及環境工程學系碩士班
    虞國興;王鵬瑞;Yu, Gwo-Hsing;Wang, Peng-Jui
    Keywords: 氣候變遷;模糊C均值法;分群指標門檻值迴歸式;Climate Change;Fuzzy C-Means;Clustering indices threshold value regression equation
    Date: 2012
    Issue Date: 2012-06-21 06:49:06 (UTC+8)
    Abstract: 本研究之目的為發展一套客觀辨識氣候變遷時序之方法,以作為探究台灣近年發生之極端降雨事件究竟是短暫偶發之單一事件,或是儼然已為氣候變遷導致之長期降雨型態改變等相關問題之辨識與判定方法。
    研究中,採用模糊C均值法(Fuzzy C-Means, FCM)於氣候變遷時序資料,欲以模糊C均值法之隸屬度,辨識時序資料之分群,並以分群指標作為分群優劣之依據,若分群顯著,則表示時序資料有變異之情形,即氣候變遷之現象。研究中,引入統計虛無假設檢定理論,以建立隸屬度分群指標判定門檻值,並以時序資料個數及變異係數建立分群指標門檻值迴歸式,俾作為相關氣候變遷時序資料辨識之用。
    由研究結果顯示,利用FCM進行氣候變遷辨識時,以VPE指標之辨識結果較佳;同時,當前、後段資料年份長度接近時才有較佳之辨識度與檢定結果。在氣候變遷之年份辨識方面,當資料長度增加時,判定之年份會接近真實之變遷年份,惟當資料年份繼續增加,由於前段資料之年份相較於後段資料長度相當長,辨識度會受到影響,而產生辨識上之偏差。
    綜合上述之研究結論獲知,經由繁衍及實際雨量站等相關實際案例資料分析與適用性測試後證實,本研究發展之方法確實可作為辨識時序是否發生氣候變遷之客觀評判基準。本研究所獲致之若干成果,希冀可進一步作為未來氣候變遷後續研究之參考與應用。
    This study is to develop a method of time series analysis that could determine whether the recent occurrences of extreme rainfalls in Taiwan are part of a long-term shift in precipitation patterns due to climate change, or random events without long-term implications.
    In this study, Fuzzy C-Means (FCM) method is applied to the analysis of climatic time series data. The concept of fuzzy membership allowed one to break up the time series into distinguishable data clusters. The degree of data clustering was quantified using three clustering indices, which were obtained through regression equations from the number of time series data and the coefficient of variation. High degree of data clustering would signify a temporal shift in time series induced by climate change. A null hypothesis framework for validating a temporal shift in time series was constructed, and a threshold value for the clustering indices were determined for the hypothesis validation.
    Comparing three clustering indices used, the FCM method perform best in identifying temporal shifts when the validity partition entropy (VPE) index was used. Also, when time series data was divided into two parts, a greater accuracy in identifying the years of climate change was achieved if the two parts were of more or less equal time spans. The accuracy of the identification could increase with the time span of the data series, yet a large difference between the time spans of the two component data series could lead to greater errors in the identification.
    In conclusion, based on the results of applicability testing against simulation and real precipitation data, the method proposed in this study can provide a basis for the identification of climate change in time series data. Hopefully, future research and application could be built on the results of this study.
    Appears in Collections:[水資源及環境工程學系暨研究所] 學位論文

    Files in This Item:

    File SizeFormat
    index.html0KbHTML138View/Open

    All items in 機構典藏 are protected by copyright, with all rights reserved.


    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - Feedback