中大機構典藏-NCU Institutional Repository-提供博碩士論文、考古題、期刊論文、研究計畫等下載:Item 987654321/97268
English  |  正體中文  |  简体中文  |  Items with full text/Total items : 83776/83776 (100%)
Visitors : 59655140      Online Users : 900
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU 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: https://ir.lib.ncu.edu.tw/handle/987654321/97268


    Title: 應用深度學習於電網互動高效建築與連結社區之用電資料分析
    Authors: 曾鎂錚;Tseng, Mei-Cheng
    Contributors: 土木工程學系
    Keywords: 建築能耗模型;節電;預測用電量發展;深度學習;Tensorflow應用;building energy consumption model;energy saving;electricity consumption prediction development;deep learning;TensorFlow application
    Date: 2025-07-21
    Issue Date: 2025-10-17 11:04:16 (UTC+8)
    Publisher: 國立中央大學
    Abstract: 隨著科技發展迅速,越來越多大型企業投入智能家電開發,具備人工智能功用的電器設備廣泛應用於住家生活,常見的像是應用物聯網(IoT)技術,採用遠端感測器連接網路方式遙控智能家電,讓使用者不必回到家也可以即時啟用遠端網路操作,既方便又省時。而這些便捷的家電設備能被應用於建築能源監控系統,透過實時監測使其更有效分析用電量使用趨勢,達到監督使用者用電情形。
    根據國際能源署所發布數據顯示,建築和建築施工部門合計佔全球最終能源消耗的三分之一以上,並佔全球總碳排放量的近40%。隨著發展中國家的城市化進程加快和氣候變化的影響,建築能耗持續增加,因此如何減緩住宅部門能源需求之成長是目前世界政府的一大難題。
    本研究將建築能耗模型結合家戶用電使用進行能耗預測分析,選擇基於Python的微框架Flask作為程式介面以處理大筆用電量數據,相較於傳統多以人力處理繁瑣資料數據,既省時又提高數據處理準確度,對於後續系統維護更為方便且快速。此外,本研究利用TensorFlow深度學習技術以預測未來短期內各項能耗結果,從模型建立考量到展示不同時間區段預測成果,期望能有效掌握家戶用電走向,並根據預測結果調整用電行為,達到更有效的節能管理。
    為了進一步提升預測的真實性與實用性,本研究引入鄰里概念,藉由自定義社區鄰居的做法,將用電模式較為相似的住戶劃分同一族群,以彼此日常家電用電行為進行比較,計算可能產生用電量,發揮同儕比較心理,提升用戶對自身用電行為的關注度,進而促使整體節能意識的提升,達到自主監督的效益。
    在全球積極推動淨零排放以及智慧能源管理的背景下,本研究的長期目標是透過智慧電表數據與深度學習技術的結合,減少家庭與商業建築的能源消耗,進而推動再生能源轉換利用,使得用電量趨於較平緩曲線,解決電力供應不及狀況。最終長期目標是減少用電能源消耗,達成永續環境,提供社會優質、舒適又安全的生活環境。

    ;With the advancement of smart technologies, AI-powered and IoT-enabled home appliances have been widely adopted, allowing real-time remote control and energy management. These smart devices can be integrated into building energy systems to monitor electricity consumption patterns and promote efficient usage.
    Given that the building sector accounts for over one-third of global energy consumption and nearly 40% of carbon emissions, reducing household energy demand is a pressing global challenge. This study proposes an energy forecasting framework that combines household electricity data with building energy models. It adoptsFlaskas a lightweight system interface and utilizes TensorFlow deep learning to predict short-term energy usage trends, improving accuracy and operational efficiency over traditional methods.
    To enhance user engagement and the practicality of predictions, a neighborhood clustering approach is introduced, enabling peer comparison and encouraging energy-saving behavior through social influence. The system supports real-time monitoring and informed energy decisions.
    This research contributes to smart energy management, facilitates renewable energy adoption, and aims to flatten demand peaks. Ultimately, it supports carbon reduction goals while promoting a sustainable and comfortable living environment.
    Appears in Collections:[Graduate Institute of Civil Engineering] Electronic Thesis & Dissertation

    Files in This Item:

    File Description SizeFormat
    index.html0KbHTML10View/Open


    All items in NCUIR are protected by copyright, with all rights reserved.

    社群 sharing

    ::: Copyright National Central University. | 國立中央大學圖書館版權所有 | 收藏本站 | 設為首頁 | 最佳瀏覽畫面: 1024*768 | 建站日期:8-24-2009 :::
    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - 隱私權政策聲明