建立冰水機與冷卻水塔量測驗證基準線之研究

Study on Performance Measurement and Verification Method for Chillers and Cooling Towers

林育群(Yu-Chun Lin)

指導教授 ： 柯明村

國立臺北科技大學/機電學院/能源與冷凍空調工程系碩士班/碩士(2009年)

摘要

國內空調機房節能改善驗證方法可利用經濟部能源局所公告的『空調冰水機房節能改善之節能績效量測與驗證方法』，用此方法進行經驗方程式係數係數之迴歸建立後，得到的迴歸方程式其計算之耗電量存在著一定程度的誤差，因此本研究利用粒子演算法則以進行這方面的改善，在相同的經驗方程式下，使迴歸方程式所計算出來的耗電量誤差能夠降低。研究結果顯示，本研究所撰寫的粒子演算程式應用於經驗方程式之迴歸，其求得的迴歸方程式之平均耗電量誤差小於ASHRAE Guideline-14方法者約4.6 %，而相關係數R2則從0.68提升到0.81，且在整體迴歸耗電量趨勢與原始量測數據比較方面，也具備不失準確性的吻合趨勢。此外，本研究亦嘗試以美國聯邦能源管理專案(Federal energy management program, FEMP)所訂定的『美國能源局節能績效保證契約量測與驗證冰水機組改善計畫』之範例使用之經驗公式，與G-14內採用的經驗方程式進行比較，期能獲得較準確之基準線經驗公式。研究結果顯示，該FEMP使用之經驗公式在耗電量整體趨勢上比G-14經驗模型更為吻合，其平均耗電量與原始數據之誤差為6.37 %，較G-14減少了5.6 % 誤差，而R2也提升至0.91。最後在冷卻水側模型建立方面，本研究提出三種方式可應用於冷卻水側基準線之建立，此三種方法皆有不錯的準確性，提供使用者一個參考的方法。其中以利用外氣濕球溫度與進出口水溫來建立基準線方程式之方法較為準確，而與實際案例之數據相比，其平均耗電量誤差為5.01 %、耗電量迴歸方程式之R2為0.99。

關鍵字

量測與驗證 ； 冰水機 ； 冷卻水塔 ； 粒子族群演算法

並列摘要

Today's energy-saving chilled-water plant to improve the authentication method is followed by the Ministry of Economic Affairs “chilled-water plant energy flow to improve the performance of energy-saving methods of measurement and verification” Bureau of Energy Notice, And this method of coefficient of empirical formula was established, there is a certain degree of error, this study is the use of particle algorithm to improve the experience of the same formula, let the regression coefficient calculated from the power error can reduce the degree, Particle algorithm writing its power is way more G-14 were small 4.6% average error, and the return of power from R2 from 0.68 to 0.81, and the total return of the overall power consumption also has yet accurate trend of the match. Furthermore, the use of FEMP set: “Standard Measurement & Verification Plan for Chiller Replacement Projects Conducted Under DOE’s Super ESPC Program” an example, And G-14, as defined by the experience of model comparison, there is hope that a more accurate baseline room air-conditioned ice model, the results show that the total power consumption, the overall trend than the G-14 experience model more in line, the power consumption of the average error is 6.37%, representing a way of G-14 will reduce the number of error of 5.6 percent, the return of power from R2 also upgraded to 0.91. Finally, in the cooling water side model, using three ways to complete the cooling water side of the baseline established, the main way to provide users with a reference, and the three methods have good accuracy, with the use of outside air wet-bulb temperature and imports and exports to establish baseline water temperature is more accurate formula, the average error of 5.01 percent power consumption, as well as power consumption R2 error is 0.99.

並列關鍵字

Measurement and Verification ； Chillers ； Cooling Towers ； Particle Swarm Optimization

參考文獻

[5] 柯明村、戴邦文，應用於節能績效保證合約之冰水主機，國立台北科技大學能源與冷凍空調工程系碩士論文，2006年。

[8] ASHRAE Handbook, “Supervisory Control Strategies And Optimization”, 2007.

[9] ASHRAE Transactions, “Optimized Pump Speed Control Using Pump Water Flow Station for HVAC Systems” , 2007.

[10] EVO: International Performance Measurement and Verification Protocol, “Concepts and Options for Determining Energy and Water Savings”, 2007.

[11] 李文興、龔仲寬，粒子族群演算法應用於室內溫水游泳池熱泵系統最佳化分析設計，國立台北科技大學能源與冷凍空調工程系碩士論文，2006年。