中華民國內政部建築研究所中華民國內政部建築研究所

:::

上方連結

字級

:::

資訊與服務

:::

臺灣建築能源模擬解析用逐時標準氣象資料TMY3之建置與研究

計畫主持人	何明錦
協同主持人	黃國倉
執行單位
執行行程	102 年 03 月至 102 年 12 月
GPN
ISBN
MOIS
組別	環境控制組
執行方式	協同研究
關鍵詞	標準氣象年、建築能源模擬、TMY3
參考文獻

中文摘要：一、研究緣起
建築物的熱得除了建築物本身的外殼設計外，又與外界氣候息息相關，因此在進
行建築物能源解析前尚須掌握外界氣候的諸氣象要素，才能真正反應建築物的熱負荷
情況，並藉以計算其能源使用量。本研究以1990-2012年共23年為篩選基期，更新臺
灣建築能源解析用平均氣象年，以使氣象年具當地長期氣候之代表性。所產製符合國
際標準之TMY3標準氣象年，包括台北、新竹、台中、嘉義、台南、高雄、花蓮、台
東等地。
二、研究方法及過程
本研究採用美國再生能源實驗室(NREL)所制訂的篩選程序(Sandia Method)，作為
中選月份的篩選流程。同時以中央氣象局之原始歷史逐時氣象資料作為標準氣象年之
篩選基期，以製作TMY3。研究過程首先針對原始逐時氣象數據進行格式整理、缺漏統
計、異常資料剔除與缺漏值補建，再以權重後之各氣象要素依各月之逐日累積分佈函
數與長期氣候進行比對，以挑選出氣象分佈結構相似者為候選月。再以均方根誤差統
計量檢定各候選月對長期氣候之偏差，以偏差最小者為中選月份。
三、重要發現
以1990-2012年為基期之逐時氣象記錄資料製作符合國際Sandia Method規範之臺
灣八地(台北、新竹、台中、嘉義、台南、高雄、花蓮、台東)TMY3建築能源解析用逐
時標準氣象年，可與國際接軌。與歷屆AWY與TMY2之標準氣象年比對後，新產製
之TMY3更具長期氣候之代表性。與歷年既有之標準氣象年比較其對空調型建築耗能
變化情形：以台北既有辦公大樓為例，空調型辦公建築全年空調耗能密度在AWY,
TMY2, TMY3等標準氣象年模擬下分別為120.9, 121.9與122.3 (KWh/m2)，增加約1.9%
之耗能量。
與歷年既有之標準氣象年比較其對自然通風型住宅建築之室內熱不舒適之變化：
研究發現室內熱不舒適發生之頻率有逐漸增加之趨勢，AWY、TMY2與TMY3全年分
別有18.3%, 23.5%與26.0%之機會處於室內熱不舒適之狀態。而在熱不舒適嚴重度方
面，新產製之TMY3對AWY時期之增加率達53%。顯示今日之氣候更趨炎熱，更易
導致室內過熱情形。針對住宅建築以模擬改善策略之效益方式評估於全球氣候變遷下
之調適策略，作為未來建築能源變遷相關研究與對策之參考。
四、主要建議事項
根據研究發現，提出具體立即可行建議如下：
建議一：
建議公告本計畫所產製之臺灣TMY3標準氣象年：立即可行建議
主辦機關：內政部建築研究所
協辦機關：財團法人臺灣建築中心
有鑑於電腦動態熱負荷模擬技術日漸普及，業界已有Ecotect、eQuest等視窗化介
面之電腦模擬程式流通，然目前缺乏國家公告之標準外界氣象資料以茲利用，導致各
家業者使用來路不明之氣象資料作為模擬，使得模擬結果不具當地氣候代表性，可信
度低。透過本計畫所產至之TMY3標準氣象年，可統一各家模擬軟體之外界氣象條件，
提升模擬可信度。
建議二：
研擬反應未來氣候特徵之未來氣象年：立即可行建議
主辦機關：內政部建築研究所
本計畫以歷史氣象資料完成建築能源解析用平均氣象年之研製，有鑑於未來氣候
變遷下之建築能源變化之探討與模擬，以及早因應未來氣候下之建築節能策略與調適，
仍須仰賴足以反應未來氣候特徵之建築氣象，因此建議針對未來氣候變遷之相關研究。
英文摘要：1. Introduction
The factors that influence the heat gainof a building including building envelope
efficiency design, occupancy behavior, and, most ofall, the local climate. Therefore, before
understanding the energy consumption of a building by simulation, local meteorological data
should be firstly well prepared so as to actually reflect the energy use. The objective of this
research is to establish typical meteorological years (TMY3) for the use in building dynamic
simulation. Hourly observed weather data from 1990 to 2012 of eight locations, Taipei,
Hsingchu, Taichung, Chiayi, Tainan, Kaohsiung, Hualian, and Taitong, were prepared to
construct TMY3 by means of internationally recognized standard method.
2. Methodology
The research adopted Sandia Method developed from National Renewable Energy
Laboratory of United States (NREL) as standard method for constructing TMY3. The
processes firstly start with rearrangement,filling missing data, and omitting unreasonable
records from the raw weather data recorded from Central Weather Bureau. As TMY3 is an
hypothetical weather year that its constitutional individual months would not necessarily
comes from the same year. Therefore, in order to seek for the most representative months that
have the least deviation from the long-term weather, concepts of candidate months were
adopted. The month that have the top five least deviation are considered as candidate months.
Statistical method of root mean square error (RMSE) was used to test each candidate month’s
deviation against their long-term weather. The one that has the least RMSE was considered as
chosen months.
3. Results
The research adopted internationally recognized method in developing TMY3. As a result,
the establish TMY3s are capable of comparing toother world’s TMY3 data. By comparing the
newly constructed TMY3 with Taiwan’s historical AWY and TMY2, the TMY3s are more
representative against their corresponding local weather. For an office building located in
Taipei as an example, the annual cooling energy consumption were 120.9, 121.9, and 122.3
(KWh/m2), when simulated with the meteorological years AWY, TMY2, and TMY3,
accordingly. It is observed that there is 1.9% increase in cooling energy use.
Furthermore, in the study of indoor thermal comfort change of a natural ventilated house,
it is found that the frequency of discomforthours is gradually increased. There are 18.3%,
23.5%, and 26.0% opportunity to fall behind the thermal comfort range, respectively. While
considering the discomfort severity, there are 53% increase as simulated with TMY3 in
comparison with the case simulated with AWY. It indicates that although the indoor
overheating hours does not have a significant increase through these year period, the degree
of thermal discomfort is severely increased. The research also studied the efficiency of some
proposed adaptive strategies that could be adopted for residential buildings.
This project comes to the immediate and long-term strategies.
For immediate strategies:
1. The developed TMY3s are recommended to be announced publically for utilization.
2. It is suggested that future weather dataare mandatory fundamentals for studying the
climatic impact on building energy use and the efficiency of various adaptation
strategies. Further research on the typical meteorological data that reflects the climate
change within a century future are highly recommended.

相關檔案

臺灣建築能源模擬解析用逐時標準氣象資料TMY3之建置與研究.pdf
pdf

中華民國內政部建築研究所中華民國內政部建築研究所

研究成果

臺灣建築能源模擬解析用逐時標準氣象資料TMY3之建置與研究