請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/73772
標題: | 智慧綠建築高績效節電管理模式探討 Discussion on the high-performance power-saving management model of smart green buildings |
作者: | Jun-Mao Liao 廖俊茂 |
指導教授: | 張陸滿(Luh-Maan Chang) |
關鍵字: | 智慧綠建築評估系統,機器學習,隨機森林,微智慧電網,耗電密度, Smart Green Building Evaluation System,Machine Learning,Random Forest,Mini Smart Grid,Energy Use Intensity, |
出版年 : | 2021 |
學位: | 博士 |
摘要: | 面對氣候變遷能源短缺而公共建築具有耗電密度高以及管理相對集中的特點,但取得最高等級智慧綠建築標章的研發公共建築,是否於運轉期間可發揮節能最大績效值得探討研究。本研究的目的係針對智慧綠建築標章的研發公共建築,探討於運轉階段是否仍有節能績效並建置節電管理模式,俾以提升整體節能之績效。 研究中係以中台灣產業創新園區為例,該園區於施工階段取得鑽石級智慧及綠建築候選證書,完工後取得鑽石級智慧及綠建築標章,於運轉階段,運用量測Measure、管理Manage、節電Reduce系統,整合微智慧電網系統及建築資訊模型(Building Information Modeling)之空間管理資訊,彙整出各類空間及設施系統之詳細耗電資料,並以時間序列及影響耗電因子進行統計分析,運用線性回歸及機器學習、深度學習技術,預報未來三個月空調耗電,進而訂定各空間及設施系統的耗電密度基線,作為運轉調控,建構一套高績效節電管理模式,以發揮智慧綠建築較佳節能效益。 本研究除深入統計分析研發大樓各類空間及設施詳細耗電資訊外,並展示節電管理的建模架構流程,此流程架構將可供建築師、業主、物業管理單位,於智慧綠建築規劃設計及進入運轉階段時之參考使用。 Public buildings often have common features of high power consumption density and relatively centralized management. To deal with global climate change and energy shortages, public sector lauched many incentive programs to mitigate carbon dioxide and energy saving of buildings and those distinguished buildings with Smart Green Building Emblems. Nevertheless, whether or not these public buildings have better operational performance in energy saving is worthy of exploring. Therefore, the purpose of this research is to verify if these emblemed public buildings are saving energy in the operation phase, and to construct a generic power-saving management mode for improving the overall energy-saving performance. In this research, the Central Taiwan Industrial Innovation Campus (CTIC) was used to exemplify the operational performance and the generic mode. The Park obtained a highest level of smart and green building with diamond emblem after the finish of construction. During operation, systems of measurement, management and power reduction were used to integrate space management information through the mechanisms of the micro-smart grid and Building Information Modeling. The systems facilitated the collection of detailed power consumption data of various spaces and their corresponding facilities. Then, statistical analysis on time series and influential power consumption factors were performed. Moreover, linear regression, machine learning, and deep learning technologies were utilized to forecast the next three-month of the air-conditioning power consumption. Meanwhile, the baseline of power consumption density was set for each space and each facility to regulate the operation and to construct a set of high-performance power-conservation management mode for improving energy efficiency of the smart green public buildings. In addition to statistically analyze the detailed power consumption information. Various spaces and facilities in the R D spaces of CTIC were scrutized. This research results in a generic framework. Architects, design engineers, facility owners, and property managers could base on the framework to plan, to design and to forecast the power consumption of their smart green buildings in the future. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/73772 |
DOI: | 10.6342/NTU202100249 |
全文授權: | 有償授權 |
顯示於系所單位: | 土木工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
U0001-2901202111431300.pdf 目前未授權公開取用 | 14.98 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。