隨著國內工商產業蓬勃發展，為了維持室內空間舒適或者製造業製程
環境溫溼度需求，空調系統為不可或缺的設施，尤其近年的氣候變遷影響，
人們對空調系統的依賴，更是無法須臾或離。在整個空調系統中，空調冰
水機為最重要與耗電最高的一項設備，它可謂是空調系統的心臟，因此維
持冰水機的穩定運行，已經成為大樓物業或企業設施運行管理者一項重要
的工作。
然而，隨著空調冰水機隨著運轉時數的積累、操作不當或者是維護保
養不良，容易造成冰水機的零件老化與故障機率增高，同時亦會造成空調
設備的效能降低，產生能源額外的浪費，對企業本身與環境都是一種傷害。
所以，精準的故障預測，實為達到故障預防、即時故障處理與避免因冰水
機故障而造成空調系統的中斷重要的一項研究。
本研究將影響冰水機運轉週數因素分為三個構面: 包括機型、操作型
態、安裝環境。透過冰水機維修紀錄資料收集並以存活期間分析方法驗證
研究假說，結果顯示操作模式及安裝環境對運轉週期有著顯著的影響，此
外亦發現其運轉週數有著可預測的可能。根據本研究的結果建議冰水機廠
商可以建置並運用其維修服務的系統資料，以作為更進一步的冰水機各式
零組件的生存壽命分析判斷，進行更精準的冰水機維修預測，建立更符合
客戶需求的維護保養策略，提供更完善的維護創新服務模式，達成客戶與
冰水機製造商及其技術維修服務單位的三贏局面。

With the vigorous development of the domestic industrial and commercial
industry, the air conditioning system become an indispensable facility to
maintain not only the indoor space comfort but also the temperature and
humidity demand of the manufacturing process environment. Especially due to
the impact of climate change in recent years, people is inescapable to rely on
the air conditioning system. The air-conditioning chiller is the most important
and the most power-consuming equipment in the entire air-conditioning system.
It could be the heart of the air-conditioning system. Therefore, maintaining the
chiller in stable operation has become a very import work to the building
property or enterprise facility operation manager.
However, with the accumulation of operating hours of air conditioners,
improper operation or poor maintenance, it usually causes the aging of the parts
of the chillers and the increased probability of failure. At the same time, it will
also reduce the efficiency of air conditioning equipment and generate additional
energy. The waste is harmful to both the enterprise itself and the environment.
Therefore, accurate fault prediction is an important research to achieve fault
prevention, intime trouble shooting, and avoid the interruption of the air
conditioning system due to the failure of the chiller.
In this study, the factors that affect the operation weeks of the chiller are
divided into three aspects: including the chiller model, operation mode, and
installation environment. Through the collection of chillers maintenance
records and verification of the research hypothesis by analysis of survival time,
the results show that the operating mode and installation environment have a
significant impact on the operating weeks. In addition, it is also found that the
operating week is predictable. According to the results of this study, it is
recommended that the chiller manufacturers can build and use the system data
of their maintenance services as a further analysis and judgment of the life span
of various chillers components, and make more accurate chiller maintenance
predictions. Establish a maintenance strategy that better meets the needs of
customers, provide a more comprehensive and innovative maintenance service
model, and achieve a win-win situation for customers and chiller manufacturers
and their technical maintenance service business units.

(美商)台灣開利股份有限公司. 19XR 19XRV Centrifugal Chiller 教育訓練資料. ((美
商)台灣開利股份有限公司).
(美商)台灣開利股份有限公司. 23XRV 變頻螺旋式空調冰水主機 教育訓練資料.
((美商)台灣開利股份有限公司).
台灣日立江森自控股份有限公司. (2018). 渦卷式壓縮機. In. 台灣日立江森自控
股份有限公司網頁: 台灣日立江森自控股份有限公司.
呂金翰. (2009). 冰水主機自我故障診斷之研究. 國立台北科技大學,
林正祥, 林世俊, 陈建谋, & 萧宽民. (2010). 物业管理大楼冷水主机设备的维护
质量及成本分析. 暖通空調, 40(4), 115-118.
張佑民. (2006). 故障診斷專家系統於中央空調應用分析.
郭姿伶. (2010). 影響空調維修服務品質因素之研究: Fuzzy AHP 之應用. 撰者,
陳俊宏. (2007). 離心式冰水機組自動化故障偵測與診斷策略之研究. 國立台北
科技大學,
經濟部. (2019). 智慧化能源管理應用於中央空調系統. 經濟部高效率節能產品
或低碳技術電子報.
劉中哲. (2011). 我國冰水機市場現況. Retrieved from
賴建雄. (2016). 離心式冰水主機遠端監測與故障診斷. 國立台北科技大學,
謝宜廷. (2008). 冰水主機自我診斷之 Linux 嵌入式系統開發研究. 臺北科技大
學優質電力供電產業研發碩士專班學位論文, 1-62.
鍾朝恭, & 呂守陞. (2012). 水力發電系統壽命分佈及故障診斷分析. 農業工程學
報.
Du, Z., & Jin, X. (2007). Detection and diagnosis for sensor fault in HVAC systems.
Energy Conversion and Management, 48(3), 693-702.
Greene, F. L., Stewart, A. K., & Norton, H. J. (2002). A new TNM staging strategy for
node-positive (stage III) colon cancer: an analysis of 50,042 patients. Annals
of surgery, 236(4), 416.
Katipamula, S., & Brambley, M. R. (2005a). Methods for fault detection, diagnostics,
and prognostics for building systems—a review, part I. . Hvac&R Research,
11(1), 3-25.
Katipamula, S., & Brambley, M. R. (2005b). Methods for fault detection, diagnostics,
and prognostics for building systems—A review, part II. Hvac&R Research,
11(2), 169-187.
Wang, S., & Cui, J. (2006). A robust fault detection and diagnosis strategy for
centrifugal chillers. Hvac&R Research, 12(3), 407-428.