透過您的圖書館登入
IP:18.117.176.186
  • 學位論文

應用自適應神經模糊系統於變頻氣冷式冰水主機之性能分析

Application of Adaptive Neuro-Fuzzy Inference System to Performance Analysis for Air-cooled Scroll Chiller with Variable Speed

指導教授 : 張永宗
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


近年來能源的短缺日漸嚴重,導致電費價格不斷上漲,而台灣位處於亞熱帶地區,每當夏季來臨時,人們更依賴空調設備,使空調設備用電增加,而空調設備的核心裝置莫過於冰水主機。為了改善空調設備耗電量,本研究應用變頻器於氣冷式冰水主機的壓縮機與冷凝器散熱風扇,透過變頻器的影響使冰水主機的運轉頻率下降,實驗結果發現,頻率的下降雖能降低總耗電量,但其冷凍能力也下降,導致主機性能係數也隨之變化,故本研究提出自適應神經模糊系統建立變頻前冰水主機之性能模型,利用此模型探討變頻前後之性能改善程度,此方法解決了傳統模糊系統以專家的經驗定義模糊規則庫的困難度。 本論文考慮冰水主機變頻前之外氣濕球溫度、冰水出水溫度、冷凍能力等自變數,利用自適應神經模糊系統建立變頻前主機性能模型,將變頻後之自變數代入此模型並獲得變頻前之性能係數,依據變頻前性能係數與變頻後性能係數,分析其變頻前後之性能改善程度,而本研究將針對變頻40Hz、45Hz、50Hz、55Hz,分別比較其性能改善程度。由實驗結果顯示,頻率為40Hz時,其改善程度高於其他頻率,約提升48.42%。

並列摘要


In recent years energy shortages lead to rising of electricity prices. Taiwan is located in located in subtropical regions during summer people more dependent on air-conditioning equipment cause increasing of electricity. However the core unit of air-conditioning equipment is chiller. To reduce the power consumption of air-conditioning in this study. We apply an inverter on compressor and condenser cooling fan adjust both inverter to reduce operation frequency. The experimental results showed that the decrease of the frequency would reduce the total power consumption, its cooling capacity would be reduced too it might cause the change of the coefficient of performance. This study proposed an application of adaptive neuro-fuzzy system to establish performance model at constant frequency and explore the improvement of C.O.P using this model. This method solved the difficulty of fuzzy systems in order to define the experience of expert fuzzy rule base. This study consider outside air wet bulb temperature chiller water supply temperature and cooling capacity of chiller during fixed-frequency. We used adaptive neuro-fuzzy system to establish performance model at constant frequency, we put the independent variables of frequency conversion into the model and would get the coefficient of performance before frequency conversion. According to the coefficient of performance before and after frequency conversion we can analyse the improvement of performance. This research focus on frequency 40Hz,45Hz,50Hz,55Hz,and compare the improvement of performance respectively results shown that when the frequency was 40Hz,the improvement was better than other frequency, it would about 48.42%.

參考文獻


[12] 朱正宇,「應用灰色系統理論於冰水主機耗電量預測」,碩士論文,國立台北科技大學,2007。
[15] 莊虔毅,「以基因演算法求解含凹凸特性曲線之冰水主機最佳負載分配」,碩士論文,國立台北科技大學,2005。
[16] 李俊賢,「應用模擬退火法於空調系統運轉之最佳化」,碩士論文,國立台北科技大學,2008。
[3] 顧孝偉,「住宅用電量監測與解析之研究」,碩士論文,國立成功大學。
[4] K.T.Chau,K.C.Wu,C.C.Chan and W..X. Shen,"A new battery capacity indicator for nickel–metal hydride battery powered electric vehicles using adaptive neuro-fuzzy inference system",Energy Conversion and Management 44: 2059–2071, 2003.

延伸閱讀