近年來，由於科技的不斷進步，導致能源的需求以及消耗顯著增加。根據臺灣經濟部能源局的統計，在2018年工業能耗佔臺灣總能耗的31.01％以及55.93%的電力使用量，使其成為臺灣最大的能源使用者。因此，許多企業致力於開發節能方法以節約能源，並且成為對生態友善的企業。
對於半導體製造業而言，冰水主機系統的主要功能是為空調系統供應冷卻水與冰水循環，並保持潔淨室及實驗室的環境溫度，以便設備和工廠能夠正常運行。冰水主機系統的能源損耗約佔整個工廠用電量的25%到30％。因此，如果可以提高冰水主機系統的整體性能，則可以節省能源和電力成本。本研究分析了當前冰水主機系統的各種運行參數，並使用系統歷史運行數據和偏最小平方迴歸法（PLS）建立了系統能耗預測模型。通過調整可控參數，例如冷卻水泵的頻率和冰水泵的頻率等，來預測系統性能指標：COP（性能係數）、冷凍噸、瓦特和負載率。最後，針對每個可操作方案的各項指標進行了預測，並通過應用我們的最佳化流程，可以為設備操作員找到的最佳操作設置以達到節省能源之目的。

In recent years, advances in technology have led to a significant increase in energy consumption. According to statistical from the Bureau of Energy, Ministry of Economic Affairs, industry accounts for 31.01% of Taiwan's total energy consumption and 55.93% of total electricity usage in 2018, making it the largest energy user in Taiwan. Thus, many companies are committed to developing energy conservation methods to save energy, and becoming environment friendly companies.
For the semiconductor manufacturing industry, the main function of the chiller system is to supply chilled-water for the air conditioning system, and to maintain the ambient temperature so that the equipment and factories can operate normally. The energy consumption of the chiller system accounts for about 70% of the electricity consumption of the whole plant. Therefore, if the overall performance of the chiller system can be improved, it will save energy and electricity costs. This study analyzes the various operational parameters of the current chiller system, and uses historical operational data and Partial Least Squares Regression (PLSR) to establish a system energy consumption prediction model. By adjusting the controllable parameters, such as the frequency of the cooling water pumps and the frequency of the chilled-water pumps, etc., the system performance indicators, COP (Coefficient of Performance), refrigeration tons, and kilowatt and load rate per unit, are predicted. Finally, the performance of each feasible solution is simulated, and by applying our optimization framework, we can find the best operational settings for the equipment operators to achieve the goal of energy saving.

1. ASHRAE, B. A. I., & Design, E. (2011). ASHRAE Handbook‐HVAC Applications(SI). American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc., Atlanta, USA.
2. Guideline, A. S. H. R. A. E. (2002). Guideline 14-2002, Measurement of energy and demand savings. American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc., Atlanta, USA.
3. Bourdouxhe, J., Grodent, M., and Lebrun, J. (1997). HVAC1 Toolkit: algorithms and subroutines for primary HVAC system energy calculations. American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc., Atlanta, USA.
4. Energy Statistics Handbook 2017. (2017). In Bureau of Energy, Ministry of Economic Affairs. Retrieved 10/1, 2019, from https://www.moeaboe.gov.tw/
5. Energy Statistics Handbook 2018. (2018). In Bureau of Energy, Ministry of Economic Affairs. Retrieved 10/1, 2019, from https://www.moeaboe.gov.tw/
6. Gordon, J. M., & Ng, K. C. (1995). Predictive and diagnostic aspects of a universal thermodynamic model for chillers. International Journal of Heat and Mass Transfer, 38(5), 807-818.
7. Hydeman, M., & Gillespie, K. L. (2002). Tools and techniques to calibrate electric chiller component models. ASHRAE transactions, 108(1), 733-741.
8. Lee, W. S., & Lin, L. C. (2009). Optimal chiller loading by particle swarm algorithm for reducing energy consumption. Applied Thermal Engineering, 29(8-9), 1730-1734.
9. Lee, T. S. (2004). Thermodynamic modeling and experimental validation of screw liquid chillers. Ashrae Transactions, 110(1), 206-216.
10. Namburu, S. M., Luo, J., Azam, M., Choi, K., & Pattipati, K. R. (2005). Fault detection, diagnosis, and data-driven modeling in HVAC chillers. Signal Processing, Sensor Fusion, and Target Recognition XIV, 5809, 143-154.
11. Nelder, J. A., & Mead, R. (1965). A simplex method for function minimization. The computer journal, 7(4), 308-313.
12. Reddy, T. A., & Andersen, K. K. (2002). An evaluation of classical steady-state off-line linear parameter estimation methods applied to chiller performance data.
Hvac&R Research, 8(1), 101-124.
13. Sezgan, O., Smith, B., & Moezzi, M. (1996). Building performance evaluation and tracking. Berkeley, Lawrence Berkeley National Laboratories.
14. Sreedharan, P., & Haves, P. (2001). Comparison of chiller models for use in model-based fault detection.
15. Wold, S., Sjöström, M., & Eriksson, L. (2001). PLS-regression: a basic tool of chemometrics. Chemometrics and intelligent laboratory systems, 58(2), 109-130.
16. Yik, F. W., & Lam, V. K. (1998). Chiller models for plant design studies. Building Services Engineering Research and Technology, 19(4), 233-241.
17. 胡石政、蔡尤溪(2000)。高科技產業耗能耗電研究。89年度電力科技產業學術
合作研究計畫。
18. 楊景超(2006)。12吋晶圓廠冰水系統節能實務。國立交通大學工學院精密與自
動化工程學程碩士論文。