隨著科技的進步，用電需求與日俱增，供應商為解決大量的用電需求，將節電政策的重心放在需求面管理，以有效的掌握用電。諸如推行契約需量、時間電價等政策，希望藉此控制使用者在尖峰時段的用電量。但此一構想之落實不單從供應商著手，使用者的配合也非常重要。然而在實際情況下，使用者對於自身用電習慣並不瞭解，因此需要透過用電視覺化提高使用者的用電意識，同時藉由用電警示提早告知使用者用電即將超約，以利使用者從中控管尖峰用電。因此本研究設計出一套結合用電視覺化與用電預測的系統，系統能於網頁上呈現即時用電資訊與需量預測趨勢，讓使用者能依照面板上的狀態得知用電是否即將超過契約需量。在預測模型中，利用LSTM模型加入用電資料的特性實現短期需量預測，同時模型也會隨著時間的推進而更新以符合現況。而基於真實資料所進行的實驗結果顯示，系統能在尖峰時段有效的預測用電。且網頁上能即時更新用電狀況與預測趨勢。達到即時通知使用者，讓使用者能防止超過契約容量，同時增加使用者對於自身用電的意識。

As technology advances and the demand for electricity increases, suppliers are focusing on demand side management to effectively control electricity consumption in order to solve the large demand. For example, they have implemented policies such as contract demand and time-of-use tariffs, hoping to control electricity consumption of users during peak hours. However, in reality, users do not understand their own electricity consumption habits. Therefore, we need to raise awareness of users through power visualization, and at the same time, inform users in advance that they are about to exceed their electricity consumption contracts through power alerts, so that they control their peak electricity consumption at an early stage. We design a system that combines electricity visualization and electricity consumption prediction. The system presents real time electricity consumption information and demand prediction trend on the web page, so that users can know whether electricity consumption is about to exceed the contracted demand according to the status on the panel. In the prediction model, the LSTM model uses the characteristics of electricity consumption data to achieve short-term load forecasting, and the model updates over time to match the current situation. Experimental results based on real data show that the system is able to predict power consumption effectively during peak hours. The webpage has been updated with the electricity consumption status and forecast trend in real time. This allows users to be notified immediately to prevent exceeding the contracted capacity and to increase their awareness of their own electricity consumption.

[1] Logenthiran, Thillainathan, Dipti Srinivasan, and Tan Zong Shun. "Demand side management in smart grid using heuristic optimization." IEEE transactions on smart grid 3.3 (2012): 1244-1252.
[2] Strbac, Goran. "Demand side management: Benefits and challenges." Energy policy 36.12 (2008): 4419-4426.
[3] Logenthiran, Thillainathan, Dipti Srinivasan, and Tan Zong Shun. "Demand side management in smart grid using heuristic optimization." IEEE transactions on smart grid 3.3 (2012): 1244-1252.
[4] Goyal, Govind Rai, and Shelly Vadhera. "Challenges of implementing demand side management in developing countries." Journal of Power Technologies 100.1 (2020): 43.
[5] Singh, Swati, and Surabhi Chandra. "Energy Efficiency and Demand Side Management." Int. Adv. Res. J. Sci. Eng. Technol.(IARJSET) 2 (2015).
[6] 財團法人台灣綠色生產力基金會, 經 濟 部 能 源 局 "電能管理需量控制器 Q&A節能技術手冊." (2006).
[7] Torres, José F., et al. "Deep learning for time series forecasting: a survey." Big Data 9.1 (2021): 3-21.
[8] Gardner Jr, Everette S. "Exponential smoothing: The state of the art." Journal of forecasting 4.1 (1985): 1-28.
[9] Deb, Chirag, et al. "A review on time series forecasting techniques for building energy consumption." Renewable and Sustainable Energy Reviews 74 (2017): 902-924.
[10] Box, George EP, et al. Time series analysis: forecasting and control. John Wiley & Sons, 2015.
[11] Chabouni, Naima, Yacine Belarbi, and Wassim Benhassine. "Electricity load dynamics, temperature and seasonality Nexus in Algeria." Energy 200 (2020): 117513.
[12] Pardo, Angel, Vicente Meneu, and Enric Valor. "Temperature and seasonality influences on Spanish electricity load." Energy Economics 24.1 (2002): 55-70.
[13] Taylor, James W. "Triple seasonal methods for short-term electricity demand forecasting." European Journal of Operational Research 204.1 (2010): 139-152.
[14] Fekri, Mohammad Navid, et al. "Deep learning for load forecasting with smart meter data: Online Adaptive Recurrent Neural Network." Applied Energy 282 (2021): 116177.
[15] Liang, Fan, et al. "Towards online deep learning-based energy forecasting." 2019 28th International Conference on Computer Communication and Networks (ICCCN). IEEE, 2019.
[16] Vexler, Julian, and Stefan Kramer. "Integrating LSTMs with online density estimation for the probabilistic forecast of energy consumption." International Conference on Discovery Science. Springer, Cham, 2019.
[17] Jagait, Rashpinder Kaur, et al. "Load forecasting under concept drift: Online ensemble learning with recurrent neural network and ARIMA." IEEE Access 9 (2021): 98992-99008.
[18] Von Krannichfeldt, Leandro, Yi Wang, and Gabriela Hug. "Online ensemble learning for load forecasting." IEEE Transactions on Power Systems 36.1 (2020): 545-548.
[19] Murugesan, Latha Karthigaa, Rashina Hoda, and Zoran Salcic. "Visualization of Electricity Consumption: Software Prototype through Literature Survey." Target 24 (2014): 29.
[20] Chandramitasari, Widyaning, Bobby Kurniawan, and Shigeru Fujimura. "Building deep neural network model for short term electricity consumption forecasting." 2018 International Symposium on Advanced Intelligent Informatics (SAIN). IEEE, 2018.
[21] Zhang, Ning, et al. "Comparison of three short-term load forecast models in Southern California." Energy 189 (2019): 116358.
[22] Hong, Tao, Pu Wang, and H. Lee Willis. "A naïve multiple linear regression benchmark for a load forecasting." 2011 IEEE Power and Energy Society General Meeting. IEEE, 2011.
[23] Fard, Abdollah Kavousi, and Mohammad-Reza Akbari-Zadeh. "A hybrid method based on wavelet, ANN and ARIMA model for short-term load forecasting." Journal of Experimental & Theoretical Artificial Intelligence 26.2 (2014): 167-182.
[24] Waseem, Muhammad, Zhenzhi Lin, and Li Yang. "Data-driven load forecasting of air conditioners for demand response using levenberg–marquardt algorithm-based ANN." Big Data and Cognitive Computing 3.3 (2019): 36.
[25] Hochreiter, Sepp, et al. "Gradient flow in recurrent nets: the difficulty of learning long-term dependencies." (2001).
[26] Chen, Zexi, et al. "Load forecasting based on LSTM neural network and applicable to loads of “replacement of coal with electricity”." Journal of Electrical Engineering & Technology 16.5 (2021): 2333-2342.
[27] Brownlee, Jason. Long short-term memory networks with python: develop sequence prediction models with deep learning. Machine Learning Mastery, 2017.
[28] Li, Youru, et al. "EA-LSTM: Evolutionary attention-based LSTM for time series prediction." Knowledge-Based Systems 181 (2019): 104785.
[29] Ding, Yukai, et al. "Interpretable spatio-temporal attention LSTM model for flood forecasting." Neurocomputing 403 (2020): 348-359.
[30] Lin, Jun, et al. "Short-term load forecasting based on LSTM networks considering attention mechanism." International Journal of Electrical Power & Energy Systems 137 (2022): 107818.
[31] Zang, Haixiang, et al. "Residential load forecasting based on LSTM fusing self-attention mechanism with pooling." Energy 229 (2021): 120682.
[32] Hu, Yuntong, and Fuyuan Xiao. "Network self attention for forecasting time series." Applied Soft Computing (2022): 109092.
[33] Monigatti, Paul, Mark Apperley, and Bill Rogers. "Power and energy visualization for the micro-management of household electricity consumption." Proceedings of the International Conference on Advanced Visual Interfaces. 2010.