English  |  正體中文  |  简体中文  |  Items with full text/Total items : 51931/87076 (60%)
Visitors : 8495941      Online Users : 95
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/987654321/88176


    Title: 結合經驗模態分解與類神經網路於地下水位預測之研究
    Other Titles: Prediction of groundwater level based on EMD and ANN
    Authors: 鄭鈞瑋;Cheng, Chun-Wei
    Contributors: 淡江大學水資源及環境工程學系碩士班
    黃富國;Huang, Fu-Kuo
    Keywords: 地下水位;經驗模態分解;類神經網路;自組特徵映射網路;預報模式;groundwater level;empirical mode decomposition(EMD);artificial nearul network(ANN);self-organizing map(SOM);real forecast
    Date: 2012
    Issue Date: 2013-04-13 12:04:19 (UTC+8)
    Abstract: 台灣地區山高坡陡,河道源短流急,水資源蓄積不易,且近年全球環境變化甚鉅,時有缺水情形發生。台灣地區地下水用量約佔整體水資源可利用量之三成,地下水之重要性不言可喻。因此有必要深入探討地下水位變化之規律,以掌握其脈動及特性。為達成此目標,則有賴於更具真實性與準確性之地下水位預測模式,所以本研究在既有常用預測方法之基礎上,進一步尋求一更合理可行之預測模式。
    希爾伯特-黃轉換(HHT)是近年來分析非線性非穩態性時間序列訊號非常新的時頻分析工具,透過其中之經驗模態分解(EMD),可將一複雜訊號拆解成若干具有明確物理意義且易於分析之內建模態函數(IMF),本研究嘗試結合經驗模態分解方法和類神經網路中常用之自組特徵映射網路(SOM)與倒傳遞類神經網路(BPNN),探討此複合模式用來預測地下水位變化規律之可行性,並採用較切合實際預測情境之「預報」方法,有別於一般之「後報」及「前報」,應用於雲林麥寮地區非平穩性月平均地下水位之預測,研究結果顯示較傳統模式在預測值準確性之改善與分析效率之提升上具有明顯之優勢;其中所使用單站與多站之預測模式,顯示不同測站組合之類神經網路架構對分析結果具一定之影響程度,實際預測效能最佳之組合模式須依照各站之特性作適當評估。
    Recently, the change of hydrological environment is being accelerated significantly by the impact of global warming due to climate change. It is important for the management of groundwater resources because of limited water resource. However, the use of groundwater resources efficiently relies on a more real and precise prediction model. In this study, it will seek for a reasonable and effective way to predict the changes of groundwater level by modifying the traditional method.
    HHT is a relatively new method to analyze time series data that possess intrinsic non-linear and non-stationary nature. By using EMD of HHT, complicated data will decompose into several IMFs that are easier to analyze. The “real-forecast” model combined from EMD, SOM and BPNN will further applied to predict the groundwater level. It shows that the predict result is more accurate than that from the traditional method. Among them, the mode of single station and multi station is adopted, but the combination mode with best performance needs to be appropriately evaluated in accordance with the characteristics of these stations.
    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.html0KbHTML127View/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