Evaluation of neural networks for residential load forecasting and the impact of systematic feature identification

被引：1
作者：
Vanting N.B. [1 ]
Ma Z. [1 ]
Jørgensen B.N. [1 ]
机构：
[1] SDU Center for Energy Informatics, The Maersk Mc-Kinney Moller Institute, University of Southern Denmark, Odense
来源：
Energy Informatics | 2022年 / 5卷 / Suppl 4期
关键词：
Artificial neural network; Ecosystem; Feature identification; Feature selection; Recurrent neural network; Residential electricity consumption; Short-term load forecasting;
D O I：
10.1186/s42162-022-00224-5
中图分类号：
学科分类号：
摘要：
Energy systems face challenges due to climate change, distributed energy resources, and political agenda, especially distribution system operators (DSOs) responsible for ensuring grid stability. Accurate predictions of the electricity load can help DSOs better plan and maintain their grids. The study aims to test a systematic data identification and selection process to forecast the electricity load of Danish residential areas. The five-ecosystem CSTEP framework maps relevant independent variables on the cultural, societal, technological, economic, and political dimensions. Based on the literature, a recurrent neural network (RNN), long-short-term memory network (LSTM), gated recurrent unit (GRU), and feed-forward network (FFN) are evaluated and compared. The models are trained and tested using different data inputs and forecasting horizons to assess the impact of the systematic approach and the practical flexibility of the models. The findings show that the models achieve equal performances of around 0.96 adjusted R2 score and 4–5% absolute percentage error for the 1-h predictions. Forecasting 24 h gave an adjusted R2 of around 0.91 and increased the error slightly to 6–7% absolute percentage error. The impact of the systematic identification approach depended on the type of neural network, with the FFN showing the highest increase in error when removing the supporting variables. The GRU and LSTM did not rely on the identified variables, showing minimal changes in performance with or without them. The systematic approach to data identification can help researchers better understand the data inputs and their impact on the target variable. The results indicate that a focus on curating data inputs affects the performance more than choosing a specific type of neural network architecture. © 2022, The Author(s).
引用
收藏
相关论文
共 59 条
  • [51] Energy and Air Emission Accounts, (2022)
  • [52] Tanoto Y., Ongsakul W., Marpaung C.O.P., Levenberg-Marquardt recurrent networks for long-term electricity peak load forecasting, Telkomnika, 9, 2, pp. 257-266, (2011)
  • [53] Vanting N.B., Ma Z., Jorgensen B.N., A scoping review of deep neural networks for electric load forecasting, Energy Inform, 4, 2, (2021)
  • [54] Vonk B.M.J., Nguyen P.H., Grand M.O.W., Slootweg J.G., Kling W.L., Improving short-term load forecasting for a local energy storage system, 2012 47Th International Universities Power Engineering Conference, (2012)
  • [55] Xu B., Sun Y., Wang H., Yi S., Short-term electricity consumption forecasting method for residential users based on cluster classification and backpropagation neural network, 11Th International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC 2019, pp. 55-59, (2019)
  • [56] Xypolytou E., Meisel M., Sauter T., Short-term electricity consumption forecast with artificial neural networks—a case study of office buildings, 2017 IEEE Manchester Powertech, (2017)
  • [57] Yan X., Chen H., Zhang X., Tan C., Energy storage sizing for office buildings based on short-term load forecasting, 2012 IEEE 6Th International Conference on Information and Automation for Sustainability, ICIAFS 2012, pp. 290-295, (2012)
  • [58] Yong B., Shen Z., Wei Y., Shen J., Zhou Q., Short-term electricity demand forecasting based on multiple LSTMs, 10Th International Conference on Brain Inspired Cognitive Systems, BICS 2019, pp. 192-200, (2020)
  • [59] Zhu J., Yang Z., Guo Y., Zhang J., Yang H., Short-term load forecasting for electric vehicle charging stations based on deep learning approaches, Appl Sci, (2019)
123456