隨著氣候變遷所造成的全球暖化日趨嚴重，各國已陸續宣示將於2050年達成淨零排放目標，台灣亦於2022年3月公布2050淨零排放路徑藍圖，其中一項關鍵規劃即為電力能源去碳化。依據我國電力能源去碳化的規劃目標，2050年總電力供應的60～70% 將來自再生能源，特別是太陽能與風力發電等變動性再生能源。由於變動性再生能源的發電量會受到氣候和地理條件的影響，使其供應具有間歇性，如何確保電力系統的穩定性，小時別的電力負載預測就顯得非常重要。此外，若能進一步進行區域負載預測，有助於不同區域間的經濟調度，減少由於供需差異引起的電力系統不穩定性。而相較於單點的確定性預測，機率性預測模型可進一步提供不同風險水準的預測區間，有助於電力系統的機組排程；此外，季節為在短期負載預測中一個相當重要的影響因素。因此，本研究使用機器學習演算法結合氣候與時序資料建構短期區域電力負載機率預測模型，並進一步考慮季節因素對預測誤差的影響。
研究結果顯示，本研究所建構的機率性預測模型 (分位數梯度提升迴歸) 不論在判定係數、平均絕對誤差百分比及均方根誤差等評估指標的表現皆優於確定性預測模型，此顯示本研究建構之機率性預測模型具備較佳之預測能力。模型在各區域之表現方面，機率性預測模型在北部地區之預測準確度最高，其次為中部、南部與東部。
將各區域根據季節區分後，研究結果顯示機率性預測模型在特定季節的預測準確度的確有所提高；其中，北部地區在春季、秋季與冬季的預測準確度加以提昇；中部地區在春季的預測結果亦更加精準；南部地區特別在秋季時，模型的預測準確度表現最佳；東部區域則在春季、夏季與冬季的預測準確度表現出明顯的改善。
最後，依據研究結果，建議政府應增加智慧電網的投資以提升電力負載預測準確性，並建議台電設立區域電力負載預測平台以協助大型用電者即時調整用電行為與進行電力規劃。此外，亦建議將機率性預測的方法融入到電力負載預測中，可以使電力營運者在面對風險或不確定因素時，能有更多的輔助決策之資訊。

In March 2022, Taiwan announced a net-zero CO2 emission roadmap based largely on the decarbonization of electricity, wherein 60-70% of the grid electricity in 2050 will be generated using renewable energy sources, such as solar and wind power. Note however that the intermittency of these energy sources will require hourly load forecasting to facilitate pre-emptive dispatch decisions and thereby ensure grid stability while avoiding the costs associated with responding to unexpected situations. This paper presents a probabilistic machine learning model for the short-term prediction of regional power loads based on climatic and temporal data, while taking into account the effect of seasonal variables on forecasting accuracy. In simulations, the proposed probabilistic Quantile Gradient Boosting Regression model outperformed deterministic forecasting models in terms of the coefficient of determination, mean absolute percentage error, and root mean square error. Based on our analysis, we recommend that governments invest in smart grids and power distributors establish regional platforms for the hourly forecasting of electricity loads.

1. 王長春 (2006)，應用各種預測理論預測台灣電力系統區域負載，國立臺北科技大學博士論文，臺灣博碩士論文知識加值系統，https://hdl.handle.net/11296/g4dxzu。
2. 交通部中央氣象局臺灣的溫度和雨量特徵全書，https://www.cwb.gov.tw/V8/C/K/Encyclopedia/climate/climate2_all.html
3. 李宜馨、陳彥銘、李明峯 (2018)，動態迴歸模型於短期區域電力負載預測之研究臺灣能源期刊，5(4)，395-406。
4. 徐正威 (2008)，考慮氣象因素之分區負載預測，大同大學碩士論文，臺灣博碩士論文知識加值系統，https://hdl.handle.net/11296/w7v22g。
5. 國家發展委員會 (2022)，都市及區域發展統計彙編。
6. 國家發展委員會 (2022)，臺灣 2050 淨零排放路徑及策略總說明 (全文)。
7. 國家發展委員會 (2022)，臺灣 2050 淨零排放路徑及策略總說明 (簡報)。
8. 許哲強 (2002)，台灣區域電力負載預測分析系統之建立與應用研究，國立成功大學博士論文，臺灣博碩士論文知識加值系統，https://hdl.handle.net/11296/fr84g2。
9. 鄭伊秀、陳翔雄 (2021)，電業營運的基礎-探討國外電力負載預測做法，臺灣經濟研究月刊，44(12)，99-106。
10. Ahmad, T., Zhang, H., & Yan, B. (2020). A review on renewable energy and electricity requirement forecasting models for smart grid and buildings. Sustainable Cities and Society, 55, 102052.
11. Alfares, H. K., & Nazeeruddin, M. (2002). Electric load forecasting: literature survey and classification of methods. International Journal of Systems Science, 33(1), 23-34.
12. Breiman, L. (2001). Random forests. Machine Learning, 45, 5-32.
13. Cheng, Y. Y., Chan, P. P., & Qiu, Z. W. (2012, July). Random forest based ensemble system for short term load forecasting. In 2012 International Conference on Machine Learning and Cybernetics (Vol. 1, pp. 52-56). IEEE.
14. Dudek, G. (2015). Short-term load forecasting using random forests. In Intelligent Systems' 2014: Proceedings of the 7th IEEE International Conference Intelligent Systems IS’2014, September 24‐26, 2014, Warsaw, Poland, Volume 2: Tools, Architectures, Systems, Applications (pp. 821-828). Springer International Publishing.
15. ESENER, İ. I., YÜKSEL, T., & Kurban, M. (2015). Short-term load forecasting without meteorological data using AI-based structures. Turkish Journal of Electrical Engineering and Computer Sciences, 23(2), 370-380.
16. Fahad, M. U., & Arbab, N. (2014). Factor affecting short term load forecasting. Journal of Clean Energy Technologies, 2(4), 305-309.
17. Friedman, J. H. (2001). Greedy function approximation: a gradient boosting machine. Annals of Statistics, 1189-1232.
18. Gneiting, T., & Katzfuss, M. (2014). Probabilistic forecasting. Annual Review of Statistics and Its Application, 1, 125-151.
19. Gross, G., & Galiana, F. D. (1987). Short-term load forecasting. Proceedings of the IEEE, 75(12), 1558-1573.
20. Guo, W., Che, L., Shahidehpour, M., & Wan, X. (2021). Machine-Learning based methods in short-term load forecasting. The Electricity Journal, 34(1), 106884.
21. Hahn, H., Meyer-Nieberg, S., & Pickl, S. (2009). Electric load forecasting methods: Tools for decision making. European Journal of Operational Research, 199(3), 902-907.
22. He, Y., Liu, R., Li, H., Wang, S., & Lu, X. (2017). Short-term power load probability density forecasting method using kernel-based support vector quantile regression and Copula theory. Applied Energy, 185, 254-266.
23. Hong, T., & Fan, S. (2016). Probabilistic electric load forecasting: A tutorial review. International Journal of Forecasting, 32(3), 914-938.
24. Hong, T., Pinson, P., Fan, S., Zareipour, H., Troccoli, A., & Hyndman, R. J. (2016). Probabilistic energy forecasting: Global energy forecasting competition 2014 and beyond. International Journal of Forecasting, 32(3), 896-913.
25. Hong, T., Pinson, P., Wang, Y., Weron, R., Yang, D., & Zareipour, H. (2020). Energy forecasting: A review and outlook. IEEE Open Access Journal of Power and Energy, 7, 376-388.
26. James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning (Vol. 112, p. 18). New York: springer.
27. Juban, R., Ohlsson, H., Maasoumy, M., Poirier, L., & Kolter, J. Z. (2016). A multiple quantile regression approach to the wind, solar, and price tracks of GEFCom2014. International Journal of Forecasting, 32(3), 1094-1102.
28. Khatoon, S., & Singh, A. K. (2014, December). Effects of various factors on electric load forecasting: An overview. In 2014 6th IEEE Power India International Conference (PIICON) (pp. 1-5). IEEE.
29. Koenker, R., & Hallock, K. F. (2001). Quantile regression. Journal of Economic Perspectives, 15(4), 143-156.
30. Kohavi, R. (1995, August). A study of cross-validation and bootstrap for accuracy estimation and model selection. In Ijcai (Vol. 14, No. 2, pp. 1137-1145).
31. Kvålseth, T. O. (1985). Cautionary note about R 2. The American Statistician, 39(4), 279-285.
32. Liu, B., Nowotarski, J., Hong, T., & Weron, R. (2015). Probabilistic load forecasting via quantile regression averaging on sister forecasts. IEEE Transactions on Smart Grid, 8(2), 730-737.
33. Meinshausen, N., & Ridgeway, G. (2006). Quantile regression forests. Journal of Machine Learning Research, 7(6).
34. Minhas, D. M., Khalid, R. R., & Frey, G. (2017, August). Activation of electrical loads under electricity price uncertainty. In 2017 IEEE International Conference on Smart Energy Grid Engineering (SEGE) (pp. 373-378). IEEE.
35. Moghram, I., & Rahman, S. (1989). Analysis and evaluation of five short-term load forecasting techniques. IEEE Transactions on Power Systems, 4(4), 1484-1491.
36. Montgomery, D. C., Peck, E. A., & Vining, G. G. (2021). Introduction to Linear Regression Analysis. John Wiley & Sons.
37. Nowotarski, J., & Weron, R. (2015). Computing electricity spot price prediction intervals using quantile regression and forecast averaging. Computational Statistics, 30, 791-803.
38. Papadopoulos, S., & Karakatsanis, I. (2015, February). Short-term electricity load forecasting using time series and ensemble learning methods. In 2015 IEEE Power and Energy Conference at Illinois (PECI) (pp. 1-6). IEEE.
39. Pinson, P. (2013). Wind energy: Forecasting challenges for its operational management.
40. Protasiewicz, J., & Szczepaniak, P. S. (2012). Neural models of demands for electricity-prediction and risk assessment. Electrical Review, 88(6), 272-279.
41. Quan, H., Srinivasan, D., & Khosravi, A. (2014). Uncertainty handling using neural network-based prediction intervals for electrical load forecasting. Energy, 73, 916-925.
42. Scikit-learn developers. (n.d.). DecisionTreeRegressor: sklearn.tree.DecisionTreeRegressor. Scikit-learn. Retrieved June 27, 2023, from https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeRegressor.html
43. Taylor, J. W., & Buizza, R. (2002). Neural network load forecasting with weather ensemble predictions. IEEE Transactions on Power systems, 17(3), 626-632.
44. Taylor, J. W., & McSharry, P. E. (2007). Short-term load forecasting methods: An evaluation based on european data. IEEE Transactions on Power Systems, 22(4), 2213-2219.
45. Verbois, H., Rusydi, A., & Thiery, A. (2018). Probabilistic forecasting of day-ahead solar irradiance using quantile gradient boosting. Solar Energy, 173, 313-327.
46. Wang, S., Wang, S., & Wang, D. (2019). Combined probability density model for medium term load forecasting based on quantile regression and kernel density estimation. Energy Procedia, 158, 6446-6451.
47. Wang, Y., Gan, D., Zhang, N., Xie, L., & Kang, C. (2019). Feature selection for probabilistic load forecasting via sparse penalized quantile regression. Journal of Modern Power Systems and Clean Energy, 7(5), 1200-1209.
48. Xie, J., & Hong, T. (2016). GEFCom2014 probabilistic electric load forecasting: An integrated solution with forecast combination and residual simulation. International Journal of Forecasting, 32(3), 1012-1016.
49. Yang, Y., Li, S., Li, W., & Qu, M. (2018). Power load probability density forecasting using Gaussian process quantile regression. Applied Energy, 213, 499-509.
50. Zhang, Y., Wang, J., & Wang, X. (2014). Review on probabilistic forecasting of wind power generation. Renewable and Sustainable Energy Reviews, 32, 255-270.
51. Zor, K., Timur, O., & Teke, A. (2017, June). A state-of-the-art review of artificial intelligence techniques for short-term electric load forecasting. In 2017 6th International Youth Conference on Energy (IYCE) (pp. 1-7). IEEE.