Building Electricity Load Forecasting Considering Climate Change Impacts: A Multi-Factor Deep Learning Approach

Yuqing Wang; Qiaoqiao Li; Yan Xu

doi:10.53941/aieng.2025.100003

Abstract

Accurate electricity load forecasting is crucial for efficient power system operation and planning, especially in urban environments where building energy dynamics play a significant role. This study focuses on improving building load forecasting by accounting for the differences between indoor and outdoor climate conditions, which significantly influence energy consumption patterns in buildings. In this paper, a comprehensive comparative study was conducted to assess the performance of several deep neural network models, with multi-factors incorporated to enhance forecast performance. The input features considered in this paper include the input sequence structures, lagged correlation-based temperature conditions, calendar information and categorical Building-Adjusted Internal Temperature (BAIT) index. The proposed approach innovatively integrates the disparities between indoor and outdoor climatic conditions to better capture the complexities of energy demand influenced by building characteristics and external weather conditions. The proposed method was tested on Singapore EMA load data from 2019 to 2022. The experimental outcomes demonstrate that the BAIT-enhanced models exhibit strong performance across various metrics, which yields superior predictions.

References

1.
Hong, T.; Fan, S. Probabilistic electric load forecasting: A tutorial review. Int. J. Forecast. 2016, 32, 914–938.
2.
Pomponi, F.; Moncaster, A. Circular economy for the built environment: A research framework. J. Clean. Prod. 2017, 143, 710–718. https://doi.org/10.1016/j.jclepro.2016.12.055.
3.
Zhao, H.; Magoulès, F. A review on the prediction of building energy consumption. Renew. Sustain. Energy Rev. 2012, 16, 3586–3592. https://doi.org/10.1016/j.rser.2012.02.049.
4.
Hochreiter, S.; Schmidhuber, J. Long short-term memory. Neural Comput. 1997, 9, 1735–1780.
5.
Kong, W.; Dong, Z.Y.; Jia, Y.; et al. Short-term residential load forecasting based on LSTM recurrent neural network. IEEE Trans. Smart Grid 2019, 10, 841–851.
6.
Cho, K.; van Merriënboer, B.; Bahdanau, D.; et al. On the properties of neural machine translation: Encoder-decoder approaches. arXiv 2014, arXiv:1409.1259.
7.
Zheng, Z.; Zhu, J.; Lu, Y. Load forecasting using a GRU-based RNN model. In Proceedings of the International Conference on Energy, Electrical and Power Engineering, Munich, Germany, 17–18 July 2017.
8.
Li, S.; Jin, X.; Xuan, Y.; et al. Enhancing Transformer Efficiency with Channel-wise Tokenization for Time Series Forecasting. In Advances in Neural Information Processing Systems; NeurIPS: La Jolla, CA, USA, 2023; Volume 36.
9.
Dao, T.; Fu, D.; Ermon, S.; et al. FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness. In Advances in Neural Information Processing Systems NeurIPS: La Jolla, CA, USA, 2022; Volume 35.
10.
Liu, Y.; Gu, J.; Zeng, X.; et al. Flowformer: A Continuous-Time Transformer for Time Series Forecasting. In Proceedings of the 39th International Conference on Machine Learning, PMLR, Baltimore, Maryland, USA, 17–23 July 2022; Volume 162, pp. 14329–14343.
11.
Zhou, H.; Zhang, S.; Peng, J.; et al. Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting. In Proceedings of the AAAI Conference on Artificial Intelligence, Virtually, 2–9 February 2021; Volume 35, pp. 11106–11115.
12.
Kitaev, N.; Kaiser, Ł.; Levskaya, A. Reformer: The Efficient Transformer. In Proceedings of the International Conference on Learning Representations, Addis Ababa, Ethiopia, 26–30 April 2020.
13.
Chen, Y.; Zhang, D. Theory-guided deep-learning for electrical load forecasting (TgDLF) via ensemble long short-term memory, Adv. Appl. Energy 2021, 1, 100004. https://doi.org/10.1016/j.adapen.2020.100004.
14.
Staffell, I.; Pfenninger, S.; Johnson, N. A global model of hourly space heating and cooling demand at multiple spatial scales. Nat. Energy 2023, 8, 1328–1344. https://doi.org/10.1038/s41560-023-01341-5.
15.
Yang, W.; Sparrow, S.N.; Wallom, D.C. A comparative climate-resilient energy design: Wildfire Resilient Load Forecasting Model using multi-factor deep learning methods. Appl. Energy 2024, 368, 123365. https://doi.org/10.1016/j.apenergy.2024.123365.
16.
Schuster, M.; Paliwal, K.K. Bidirectional recurrent neural networks. IEEE Trans. Signal Process. 1997, 45, 2673–2681.

Scilight Press

Author Information

Abstract

Keywords

References

About Scilight

Journals

Publishing Policies

Contact Us