資料探勘(Data Mining)為美國麻省理工學院(MIT Review)與ZDNET News所推崇為21世紀的前瞻趨勢與創新技術之一，目前已有具體資料探勘模型應用於零售物流、金融服務、健康保險、營建製造等產業。然而，隨著「節能科技」在國際潮流的推動，台灣在冷凍設備耗能資料的處理未能跟上智慧節能電表的安裝速度。
龐大的耗能監測資料雖受妥善儲存，所累積而看似無用的時間序列資料與自動紀錄之高維度屬性卻無從採擷蘊涵的用電行為與耗能推測。再者，冷凍設備屬耗電量較大之設備且長時間用電，故應有較大之節能空間，故本研究以冷凍設備為研究對象，針對不同型態(液態、氣態)之冷媒洩漏狀況下之耗能相關屬性資料進行智慧探勘，並分析不同冷媒填充量對其冷凍能力及性能之影響。
本研究建立遠端監測系統量測與記錄傳統控制系統之設備運轉效能，量測設備異常造成冷凍設備之訊號、性能與耗電量變化，預測性能係數，研究結果發現針對液態冷媒洩漏試驗部份，類神經模型最能準確預測設備性能係數，MAPE值為8.257%；而針對氣態冷媒洩漏試驗，則是廣義線性回歸最能預測C.O.P值，MAPE值為10.665%，且針對3.24M2冷凍庫而言，當冷媒填充量為3kg時，設備運轉性能及冷凍能力皆為最佳狀態。
後續研究期望可建立設備異常與相關性之特徵資料庫，透過雲端監測系統分析設備異常訊號及異常特徵分析，建立異常反應機制點與處理作業流程，並量測排除異常對耗能之影響，研究成果亦期能將系統化建構與資料分析流程技術轉移至節能科技產業以作為知識管理的一環。

This study presents six data mining techniques for the prediction of coefficient of performance (C.O.P) of refrigerant (R404A). These techniques include artificial neural networks (ANNs), support vector machines (SVMs), classification and regression tree (CART), multiple regression (MR), generalized linear regression (GLR), chi-squared automatic interaction detector (CHAID). The purpose of this study is to predict C.O.P for the refrigeration equipment under different amount of refrigerant. After getting the value of C.O.P, we can judge the abnormal situation of equipment which might cause the overdose leakage of refrigerant. Analytical results from cross-fold validation are compared and the best model is investigated. The study shows that the data mining techniques can effectively and efficiently determine the C.O.P for particular amount, temperature and pressure of refrigerants. In the liquid leakage phase, ANNs predicts C.O.P the best, which MAPE is 8.257%; while in the vapor leakage phase, GLR got the most accurate value in MAPE, which is 10.665%. The models built in this study can help to judge the performance of the refrigeration equipment.

[1] S.C. Lo, C.C. Lin. Applying back propagation neural network and sequential pattern mining to construct corporation crisis prediction model - A case of Taiwan's Electronic Industry. 2009. pp. 409-14.
[2] A. Ahmed, N.E. Korres, J. Ploennigs, H. Elhadi, K. Menzel. Mining building performance data for energy-efficient operation. Advanced Engineering Informatics. (2010).
[3] 李涵茵. 能源報導-低電價、深省思. in: 經濟部能源局, (Ed.).2011. pp. 35-8.
[4] 湯政仁. 能源報導-智慧電表國內發展現況. in: 經濟部能源局, (Ed.).2010. pp. 26-9.
[5] J.S. Chou, Y. Tai, L.J. Chang. Predicting the development cost of TFT-LCD manufacturing equipment with artificial intelligence models. International Journal of Production Economics. 128 (2010) 339-50.
[6] J. Zhang, C. Cheng. Day-ahead electricity price forecasting using artificial intelligence. Inst. of Elec. and Elec. Eng. Computer Society, Vancouver, BC, Canada, 2008.
[7] E.U. Kucuksille, R. Selbaş, A. Şencan. Data mining techniques for thermophysical properties of refrigerants. Energy Conversion and Management. 50 (2009) 399-412.
[8] A.T. Şahin. Performance analysis of single-stage refrigeration system with internal heat exchanger using neural network and neuro-fuzzy. Renewable Energy. 36 (2011) 2747-52.
[9] C. Liu, H. Yi. Research on power system load forecasting model based on data mining technology. 2010 International Conference on Intelligent System Design and Engineering Application2011. pp. 240-3.
[10] J.W. Taylor, R. Buizza. Using weather ensemble predictions in electricity demand forecasting. International Journal of Forecasting. 19 (2003) 57-70.
[11] S. Saab, E. Badr, G. Nasr. Univariate modeling and forecasting of energy consumption: The case of electricity in Lebanon. Energy. 26 (2001) 1-14.
[12] C.S. Chen, Y.M. Tzeng, J.C. Hwang. The application of artificial neural networks to substation load forecasting. Electric Power Systems Research. 38 (1996) 153-60.
[13] J.H. Park, Y.M. Park, K.Y. Lee. Composite modeling for adaptive short-term load forecasting. IEEE Transactions on Power Systems. 6 (1991) 450-7.
[14] B. Ye, C. Guo, Y. Cao. Short-term load forecasting using a new fuzzy modeling strategy. The 5th World Congress on Intelligent control and Automation, Hangxhou, P.R. China, 2004. pp. 5045-9.
[15] T. Al-Saba, I. El-Amin. Artificial neural networks as applied to long-term demand forecasting. Artificial Intelligence in Engineering. 13 (1999) 189-97.
[16] P. Mandal, T. Senjyu, T. Funabashi. Neural networks approach to forecast several hour ahead electricity prices and loads in deregulated market. Energy Conversion and Management. 47 (2006) 2128-42.
[17] L.C. Ying, M.C. Pan. Using adaptive network based fuzzy inference system to forecast regional electricity loads. Energy Conversion and Management. 49 (2008) 205-11.
[18] A. Şencan. Modeling of thermodynamic properties of refrigerant/absorbent couples using data mining process. Energy Conversion and Management. 48 (2007) 470-80.
[19] A. Şencan, S.L. Kose, R. Selbaş. Prediction of thermophysical properties of mixed refrigerants using artificial neural network. Energy Conversion and Management. 52 (2011) 958-74.
[20] H. Bechtler, M.W. Browne, P.K. Bansal, V. Kecman. New approach to dynamic modelling of vapour-compression liquid chillers: Artificial neural networks. Applied Thermal Engineering. 21 (2001) 941-53.
[21] D.J. Swider. A comparison of empirically based steady-state models for vapor-compression liquid chillers. Applied Thermal Engineering. 23 (2003) 539-56.
[22] E. Arcaklioglu, A. Erişen, R. Yilmaz. Artificial neural network analysis of heat pumps using refrigerant mixtures. Energy Conversion and Management. 45 (2004) 1917-29.
[23] E.U. Kuuksille, R. Selba, A. Encan. Prediction of thermodynamic properties of refrigerants using data mining. Energy Conversion and Management. 52 (2011) 836-48.
[24] G.K.F. Tso, K.K.W. Yau. Predicting electricity energy consumption: A comparison of regression analysis, decision tree and neural networks. Energy. 32 (2007) 1761-8.
[25] H.T. Pao. Comparing linear and nonlinear forecasts for Taiwan's electricity consumption. Energy. 31 (2006) 1793-805.
[26] J. Zhang, C. Cheng. Day-ahead electricity price forecasting using artificial intelligence. (2008).
[27] D. Liu, Y. Yuan, S. Liao. Artificial neural network vs. nonlinear regression for gold content estimation in pyrometallurgy. Expert Systems with Applications. 36 (2009) 10397-400.
[28] T.Y. Li, Y. Chen, M. Yang, X.W. Mu. Data mining model based on improved fuzzy k-means algorithm and neural network algorithm. Dalian Haishi Daxue Xuebao/Journal of Dalian Maritime University. 34 (2008).
[29] S. Haykin. Neural networks: a comprehensive foundation, 2nd Edition. Prentice Hall. (1999).
[30] A. Yezioro, B. Dong, F. Leite. An applied artificial intelligence approach towards assessing building performance simulation tools. Energy and Buildings. 40 (2008) 612-20.
[31] B. Ashuri, J. Lu. Time series analysis of ENR construction cost index. Journal of Construction Engineering and Management. 136 (2010) 1227-37.
[32] V.N. Vapnik. Statistical Learning Theory. John Wiley and Sons, New York, 1998.
[33] L.J. Scheetz, J. Zhang, J. Kolassa. Classification tree modeling to identify severe and moderate vehicular injuries in young and middle-aged adults. Artificial Intelligence in Medicine. 45 (2009) 1-10.
[34] B.D. Pitt. Applications of data mining techniques to electric load profiling. Electrical and electronic engineering2000.
[35] L. Breiman, J.H. Friedman, R.A. Olshen, C.J. Stone. Classification and Regression Trees. Chapman & Hall/CRC, New York, 1984.
[36] B. Samanta, G.L. Bird, M. Kuijpers, R.A. Zimmerman, G.P. Jarvik, G. Wernovsky, et al. Prediction of periventricular leukomalacia. Part I: Selection of hemodynamic features using logistic regression and decision tree algorithms. Artificial Intelligence in Medicine. 46 (2009) 201-15.
[37] SPSS. Clementine 12.0 Algorithms Guide SPSS.
[38] W.-K. Kao, H.-M. Chen, J.-S. Chou. Aseismic ability estimation of school building using predictive data mining models. Expert Systems with Applications. 38 (2011) 10252-63.
[39] G. Shmueli, N.R. Patel, P.C. Bruce. Data Mining for Business Intelligence. John Wiley & Sons, Inc., New Jersey, 2007.
[40] S.L. Salzberg. On comparing classifiers: Pitfalls to avoid and a recommended approach. Data Mining and Knowledge Discovery. 1 (1997) 317-28.
[41] T. Orenstein, Z. Kohavi, I. Pomeranz. An optimal algorithm for cycle breaking in directed graphs. Journal of Electronic Testing. 7 (1995) 71-81.