Energyplus冰水主機能耗模型探討

Investigation on chiller performance curves in Energyplus

10.29911/JEHVACE_NEW.202105_(127).0007

謝佳興(Chia-Hsing Hsieh)

建築能源模擬 ； 冰水主機性能曲線 ； 迴歸模型 ； EnergyPlus ； Chiller performance curves ； Regression models

冷凍空調&能源科技

127期（2021 / 05 / 10）

71 - 76

繁體中文

本研究旨在探討EnergyPlus預設冰水主機性能曲線、EnergyPlus資料庫冰水主機性能曲線與實際量測迴歸冰水主機性能曲線，三者間模擬耗電量與實際耗電量之差異。共選取了3個實際場域，量測4台不同冰水主機的運轉數據，並以均方根誤差的變異係數（Coefficient of Variation of the Root Mean Squared Error, CVRMSE）比較不同性能曲線之間的預測誤差。結果顯示實際量測迴歸曲線有著最低的CVRMSE值，但若實務上難以量測實際運轉數據，使用資料庫中最接近的性能曲線來模擬是較適合的作法。

This study aims to investigate the difference between default chiller performance curves in EnergyPlus, chiller performance curves database in EnergyPlus datasets, and the actual regression chiller performance curves. Three different buildings were selected and the operating data of four different chillers were measured. The coefficient of variation of the root mean square error was used to compare the prediction error between the chiller performance curves. The results show that the actual regression curve has the lowest CVRMSE value, but if it is difficult to measure the actual operating data in practice, it is more appropriate to use the most similar performance curve in EnergyPlus datasets to simulate.

1. Hydeman, M.,Gillespie, K.L.(2002).Tools and Techniques to Calibrate Electric Chiller Component Models.ASHRAE Transactions
2. International Energy Agency(2013).Transition to sustainable buildings: strategies and opportunities to 2050.