English  |  正體中文  |  简体中文  |  全文筆數/總筆數 : 18278/19583 (93%)
造訪人次 : 1053318      線上人數 : 522
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
搜尋範圍 查詢小技巧:
  • 您可在西文檢索詞彙前後加上"雙引號",以獲取較精準的檢索結果
  • 若欲以作者姓名搜尋,建議至進階搜尋限定作者欄位,可獲得較完整資料
  • 進階搜尋
    請使用永久網址來引用或連結此文件: http://nhuir.nhu.edu.tw/handle/987654321/22511


    題名: 應用『資料探勘技術』探討配電系統高事故停電饋線的問題—以台灣電力公司嘉義區處為例
    其他題名: Data Mining to Distribution Power System--Taiwan Power Company Chia-Yi Branch As an Example
    作者: 姜朝原
    Chiang, Chao-yuan
    貢獻者: 資訊管理學研究所
    吳光閔
    Guang-ming Wu
    關鍵詞: 事故停電;饋線;資料探勘;關聯法則
    Feeder;Data Mining;Association Rules;Outage
    日期: 2006
    上傳時間: 2015-08-04 14:50:22 (UTC+8)
    摘要:   當配電系統發生事故停電時,除了造成設備的損壞之外,將增加維護成本的支出及電費收入的減少等損失。為提供穩定的電力品質、增加供電可靠度及降低配電線路維護成本,來減少事故停電次數,以提昇用戶的滿意度。因此,充分了解配電系統發生事故停電饋線之特質與其事故原因之關聯性是非常重要的課題,最後找出真正引起事故停電的原因,事先做好各種防範措施,並針對高事故停電饋線擬定改善策略方針予以重點管理。   本研究採用之研究方法係應用目前企業管理上炙手可熱的「資料探勘(Data Mining)」技術及管理上常用之統計分析方法,從個案-台電公司嘉義區營業處之配電事故停電記錄資料庫中,經由資料探勘與一般統計分析,來找出高事故停電饋線之潛在特質。資料探勘與一般統計分析結果如下:1.「饋線一」在晴天的架空低壓線路上,損壞部位以熔絲鏈開關為主,事故原因為品質不良最多且停電總時間為61~120分。2.「饋線二」在雨天的架空高壓分歧線路上,設備無損壞,事故原因以雷害最多且停電總時間為61~120分。3.「饋線三」在雨天的架空高壓分歧線路上,設備無損壞,事故原因以雷害最多且停電總時間為61~120分。   經資料探勘技術與一般統計分析驗證後,結果顯示兩者相互契合,並能找出發生事故停電的真正原因,最後本研究針對個別饋線所發生之事故原因提出相關的防範對策及建議,期望能將結果提供給配電系統維護人員擬定維護計畫、高階管理者擬定事故停電防範對策之參酌。
      When an outage ocuurs in the distribution power system,not only causing the damage of the equipment but also increasing the losses of the expenditure of safeguarding the cost and electric rate income.In order to improve user's satisfaction which is applying for steady power quality、increasing supply power reliability and reducing distribution line mainteinace cost,That is to decrease the frequency of an outage occurs.So,It is very important subjects that fully understood the speciality of the feeder and its cause of accident happens to an outage in distribution power system is related.Finally,Find out the reason that really causes an outage and to do various kinds of precautionary measures well in advanced.In accordance with an outage to high accident feeder draft improve tactics policy management in focus.   The research approach department adopted in this research uses enterprise's managerial hot 「Data Mining」technology and managerial commonly used statistics analytical method at present.From the case - Database of the record that the distribution an outage occurs in Taiwan Power Company Chia-Yi Branch,via Data Mining and general statistical analysis to find out the potential speciality of the feeder that the high an outage occurs.Data Mining and general statistics analysis result are as follows:1.「the first feeder 」On the over head low-voltage line when it is a sunny day,the damage position of the fuse cut switch is as the core which cause of accident is bad for quality and lose electrical power at total time is 61~120 minutes.2.「the second feeder」On the over head high-voltage branch line when it is a rainy day,the equipment has not been damaged which is struck by lightning and lose electrical power at total time is 61~120 minutes.3.「the third feeder」On the over head high-voltage branch line when it is a rainy day,the equipment has not been damaged which is struck by lightning and lose electrical power at total time is 61~120 minutes.   After Mata Mining technology and general statistical analysis are verified that the result shows both agree with each other and can find out real reason why an outage occurs.When an outage occurs to the specific feeder in this research puts forward relevant strick precaution countermeasure and suggestion finally.Expect to be able to offer the result to the systematic attenbant of the distribution to draft the maintenance plan, high-order administrator draft an outage occurs and takes precautions against the considering in light of actual conditions of the countermeasure.
    顯示於類別:[資訊管理學系] 博碩士論文

    文件中的檔案:

    檔案 描述 大小格式瀏覽次數
    094NHU05396023-001.pdf1994KbAdobe PDF0檢視/開啟
    index.html0KbHTML580檢視/開啟


    在NHUIR中所有的資料項目都受到原著作權保護.

    TAIR相關文章

    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - 回饋