Journal of Platform Technology

ICT Platform Society (ICT플랫폼학회)
권호 표지
DOI QR코드

DOI QR Code

A Comprehensive Survey of Short-Term Load Forecasting: Methods, Challenges, and Emerging Trends

  • Jihoon Moon (Duksung Women's Univ., Dept. of Data Science)
  • Received : 2025.05.11
  • Accepted : 2025.06.19
  • Published : 2025.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

Short-term load forecasting (STLF) is the process of predicting electricity usage from a few minutes to a few days in advance. STLF is essential for platform-based energy management systems (EMSs) because it enables accurate demand prediction, real-time scheduling, and adaptive control. Due to the growing complexity of electricity usage patterns and the abundance of high-resolution sensor data, forecasting models must efficiently handle nonlinear, multivariate, and nonstationary time series on a large scale. This paper thoroughly reviews STLF techniques ranging from traditional statistical models, such as the autoregressive integrated moving average (ARIMA) model, to advanced deep learning approaches including long short-term memory (LSTM) models, gated recurrent unit (GRU) models, convolutional neural networks (CNNs), temporal convolutional networks (TCNs), and Transformerbased architectures. It also covers hybrid and ensemble strategies, performance enhancement methods, benchmark datasets, and evaluation metrics. It emphasizes the practical integration of STLF into intelligent energy platforms and highlights industrial use cases and emerging trends, such as explainable artificial intelligence (XAI), model compression, and federated learning (FL). The goal of this survey is to provide researchers and practitioners with a framework for selecting and implementing STLF solutions that are accurate, scalable, interpretable, and platform-ready.

Keywords

References

  1. E. A. Feinberg and D. Genethliou, "Load forecasting," in Applied Mathematics for Restructured Electric Power Systems, Springer, Boston, MA, 2005, pp. 269–285.
  2. C. Kuster, Y. Rezgui, and M. Mourshed, "Electrical load forecasting models: A critical systematic review," Sustain. Cities Soc., Vol. 35, pp. 257–270, 2017. https://doi.org/10.1016/j.scs.2017.08.009
  3. Q. Dong, R. Huang, C. Cui, D. Towey, L. Zhou, J. Tian, and J. Wang, "Short-term electricity-load forecasting by deep learning: a comprehensive survey," Eng. Appl. Artif. Intell., Vol. 154, 110980, 2025.
  4. J.-F. Chen, W.-M. Wang, and C.-M. Huang, "Analysis of an adaptive time-series autoregressive moving-average (ARMA) model for short-term load forecasting," Electr. Power Syst. Res., Vol. 34, No. 3, pp. 187–196, 1995. https://doi.org/10.1016/0378-7796(95)00977-1
  5. S.-J. Huang and K.-R. Shih, "Short-term load forecasting via ARMA model identification including non-Gaussian process considerations," IEEE Trans. Power Syst., Vol. 18, No. 2, pp. 673–679, May 2003. https://doi.org/10.1109/TPWRS.2003.811010
  6. M. T. Hagan and S. M. Behr, "The time series approach to short term load forecasting," IEEE Trans. Power Syst., Vol. 2, No. 3, pp. 785–791, Aug. 1987. https://doi.org/10.1109/TPWRS.1987.4335210
  7. C.-M. Huang, C.-J. Huang, and M.-L. Wang, "A particle swarm optimization to identifying the ARMAX model for short-term load forecasting," IEEE Trans. Power Syst., Vol. 20, No. 2, pp. 1126–1133, May 2005. https://doi.org/10.1109/TPWRS.2005.846106
  8. H.-T. Yang, C.-M. Huang, and C.-L. Huang, "Identification of ARMAX model for short term load forecasting: an evolutionary programming approach," in Proc. of the Power Industry Computer Applications Conf. (PICA), IEEE, 1995, pp. 1–6.
  9. G. R. Newsham and B. J. Birt, "Building-level occupancy data to improve ARIMA-based electricity use forecasts," in Proc. of the 2nd ACM Workshop on Embedded Sensing Systems for Energy-Efficiency in Buildings (BuildSys), ACM, 2010, pp. 13–18.
  10. J. Moon, S. Jun, J. Park, Y.-H. Choi, and E. Hwang, "An electric load forecasting scheme for university campus buildings using artificial neural network and support vector regression," KIPS Trans. Comput. Commun. Syst., Vol. 5, No. 10, pp. 293–302, Oct. 2016. https://doi.org/10.3745/KTCCS.2016.5.10.293
  11. Y. Chen and H. Tan, "Short-term prediction of electric demand in building sector via hybrid support vector regression," Appl. Energy, Vol. 204, pp. 1363–1374, Oct. 2017. https://doi.org/10.1016/j.apenergy.2017.03.070
  12. K. Grolinger, A. L'Heureux, M. A. M. Capretz, and L. Seewald, "Energy forecasting for event venues: Big data and prediction accuracy," Energy Build., Vol. 112, pp. 222–233, Jan. 2016. https://doi.org/10.1016/j.enbuild.2015.12.010
  13. J. Cecati, P. Kolbusz, P. Różycki, P. Siano, and B. M. Wilamowski, "A novel RBF training algorithm for short-term electric load forecasting and comparative studies," IEEE Trans. Ind. Electron., Vol. 62, No. 10, pp. 6519–6529, Oct. 2015. https://doi.org/10.1109/TIE.2015.2424399
  14. Y. Chen, P. B. Luh, C. Guan, Y. Zhao, L. D. Michel, M. A. Coolbeth, S. J. Rourke, and R. J. Boorman, "Short-term load forecasting: Similar day-based wavelet neural networks," IEEE Trans. Power Syst., Vol. 25, No. 1, pp. 322–330, Feb. 2010. https://doi.org/10.1109/TPWRS.2009.2030426
  15. Y. Zhao, P. B. Luh, C. Bomgardner, and G. H. Beerel, "Short-term load forecasting: Multi-level wavelet neural networks with holiday corrections," in Proc. of the IEEE Power & Energy Society General Meeting, Calgary, Canada, Jul. 2009, pp. 1–7.
  16. R. Zhang, Z. Y. Dong, Y. Xu, K. Meng, and K. P. Wong, "Short-term load forecasting of Australian national electricity market by an ensemble model of extreme learning machine," IET Gener. Transm. Distrib., Vol. 7, No. 4, pp. 391–397, Apr. 2013. https://doi.org/10.1049/iet-gtd.2012.0541
  17. H. S. Hippert, C. E. Pedreira, and R. C. Souza, "Neural networks for short-term load forecasting: A review and evaluation," IEEE Trans. Power Syst., Vol. 16, No. 1, pp. 44–55, Feb. 2001. https://doi.org/10.1109/59.910780
  18. E. Ceperic, V. Ceperic, and A. Baric, "A strategy for short-term load forecasting by support vector regression machines," IEEE Trans. Power Syst., Vol. 28, No. 4, pp. 4356–4364, Nov. 2013. https://doi.org/10.1109/TPWRS.2013.2269803
  19. R. Salakhutdinov, A. Mnih, and G. Hinton, "Restricted Boltzmann machines for collaborative filtering," in Proc. of the 24th Int. Conf. on Machine Learning (ICML), Corvalis, OR, USA, Jun. 2007, pp. 791–798.
  20. M. Cai, M. Pipattanasomporn, and S. Rahman, "Day-ahead building-level load forecasts using deep learning vs. traditional time-series techniques," Appl. Energy, Vol. 236, pp. 1078–1088, Feb. 2019. https://doi.org/10.1016/j.apenergy.2018.12.042
  21. S. Nayaka, A. Yelamali, and K. Byahatti, "Electricity short term load forecasting using Elman recurrent neural network," in Proc. of the 2010 Int. Conf. on Power, Control and Embedded Systems (ICPCES), Allahabad, India, Dec. 2010, pp. 351–354.
  22. F. M. Bianchi, E. Maiorino, M. C. Kampffmeyer, A. Rizzi, and R. Jenssen, "An overview and comparative analysis of recurrent neural networks for short term load forecasting," arXiv preprint arXiv:1705.04378, May 2017.
  23. H. Zheng, J. Yuan, and L. Chen, "Short-term load forecasting using EMD-LSTM neural networks with a Xgboost algorithm for feature importance evaluation," Energies, Vol. 10, No. 8, 1168, Aug. 2017. https://doi.org/10.3390/en10081168
  24. J. Zheng, C. Xu, Z. Zhang, and X. Li, "Electric load forecasting in smart grids using long-short term memory based recurrent neural network," in Proc. of the 2017 51st Annual Conf. on Information Sciences and Systems (CISS), Baltimore, MD, USA, Mar. 2017, pp. 1–6.
  25. Y. Wang, M. Liu, Z. Bao, and S. Zhang, "Short-term load forecasting with multi-source data using gated recurrent unit neural networks," Energies, Vol. 11, No. 5, 1138, May 2018. https://doi.org/10.3390/en11051138
  26. J. Chung, C. Gulcehre, K. Cho, and Y. Bengio, "Empirical evaluation of gated recurrent neural networks on sequence modeling," arXiv preprint arXiv:1412.3555, Dec. 2014.
  27. D. L. Marino, K. Amarasinghe, and M. Manic, "Building energy load forecasting using deep neural networks," in Proc. of the 2016 IEEE 42nd Annual Conf. of the Industrial Electronics Society (IECON), Florence, Italy, Oct. 2016, pp. 7046–7051.
  28. Y. Huang, N. Wang, W. Gao, X. Guo, C. Huang, T. Hao, and J. Zhan, "LoadCNN: A low training cost deep learning model for day-ahead individual residential load forecasting," arXiv preprint arXiv:1908.00298, Aug. 2019.
  29. Y. Eren and İ. Küçükdemiral, "A comprehensive review on deep learning approaches for short-term load forecasting," Renew. Sustain. Energy Rev., Vol. 189, 114031, Jan. 2024. https://doi.org/10.1016/j.rser.2023.114031
  30. A. van den Oord, S. Dieleman, H. Zen, K. Simonyan, O. Vinyals, A. Graves, N. Kalchbrenner, A. Senior, and K. Kavukcuoglu, "WaveNet: A generative model for raw audio," arXiv preprint arXiv:1609.03499, Sep. 2016.
  31. S. Bai, J. Z. Kolter, and V. Koltun, "Convolutional sequence modeling revisited," in Proc. of the 6th Int. Conf. on Learning Representations (ICLR), Vancouver, Canada, Apr. 2018.
  32. R. Wan, S. Mei, J. Wang, M. Liu, and F. Yang, "Multivariate temporal convolutional network: A deep neural networks approach for multivariate time series forecasting," Electronics, Vol. 8, No. 8, 876, Aug. 2019. https://doi.org/10.3390/electronics8080876
  33. K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition," in Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, Jun. 2016, pp. 770–778.
  34. W. Jiang, "Deep learning based short‐term load forecasting incorporating calendar and weather information," Internet Technol. Lett., Vol. 5, No. 4, e383, 2022. https://doi.org/10.1002/itl2.383
  35. W. He, "Load forecasting via deep neural networks," in Proc. of the 5th Int. Conf. on Information Technology and Quantitative Management (ITQM), Nov. 2017, pp. 308–314.
  36. C. Tian, J. Ma, C. Zhang, and P. Zhan, "A deep neural network model for short-term load forecast based on long short-term memory network and convolutional neural network," Energies, Vol. 11, No. 12, 3493, Dec. 2018. https://doi.org/10.3390/en11123493
  37. S. Zhang, R. Chen, J. Cao, and J. Tan, "A CNN and LSTM-based multi-task learning architecture for short and medium-term electricity load forecasting," Electr. Power Syst. Res., Vol. 222, 109507, Sep. 2023. https://doi.org/10.1016/j.epsr.2023.109507
  38. Y. Jin, M. A. Acquah, M. Seo, and S. Han, "Short-term electric load prediction using transfer learning with interval estimate adjustment," Energy Build., Vol. 258, 111846, Mar. 2022. https://doi.org/10.1016/j.enbuild.2022.111846
  39. W. Lin, D. Wu, and B. Boulet, "Spatial-temporal residential short-term load forecasting via graph neural networks," IEEE Trans. Smart Grid, Vol. 12, No. 6, pp. 5373–5384, Nov. 2021. https://doi.org/10.1109/TSG.2021.3093515
  40. Q. Gao, K. Liu, K. Wu, M. You, and H. Liu, "Short-term load forecasting for typical buildings based on VMD-Informer-DMD model," in Proc. of the IEEE 2nd Industrial Electronics Society Annual On-Line Conf. (ONCON'23), Dec. 2023, pp. 1–6.
  41. X. Bu, Q. Wu, B. Zhou, and C. Li, "Hybrid short-term load forecasting using CGAN with CNN and semi-supervised regression," Appl. Energy, Vol. 338, 120920, May 2023. https://doi.org/10.1016/j.apenergy.2023.120920
  42. A. Krogh and J. A. Hertz, "A simple weight decay can improve generalization," in Advances in Neural Information Processing Systems, Vol. 4, Denver, CO, USA, 1992, pp. 950–957.
  43. N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, "Dropout: a simple way to prevent neural networks from overfitting," J. Mach. Learn. Res., Vol. 15, No. 1, pp. 1929– 1958, Jun. 2014.
  44. S. Ioffe and C. Szegedy, "Batch normalization: Accelerating deep network training by reducing internal covariate shift," arXiv preprint arXiv:1502.03167, Feb. 2015.
  45. J. L. Ba, J. R. Kiros, and G. E. Hinton, "Layer normalization," arXiv preprint arXiv:1607.06450, Jul. 2016.
  46. D. Ulyanov, A. Vedaldi, and V. Lempitsky, "Instance normalization: The missing ingredient for fast stylization," arXiv preprint arXiv:1607.08022, Jul. 2016.
  47. Y. Wu and K. He, "Group normalization," in Proc. of the European Conf. on Computer Vision (ECCV), Munich, Germany, Sep. 2018, pp. 3–19.
  48. T. Salimans and D. P. Kingma, "Weight normalization: A simple reparameterization to accelerate training of deep neural networks," in Advances in Neural Information Processing Systems, Vol. 29, Barcelona, Spain, Dec. 2016, pp. 901–909.
  49. S. Qiao, W. Shen, Z. Zhang, and A. Yuille, "Weight standardization," arXiv preprint arXiv:1903.10520, Mar. 2019.
  50. S. H. Rafi, S. R. Deeba, and E. Hossain, "A short-term load forecasting method using integrated CNN and LSTM network," IEEE Access, Vol. 9, pp. 32436–32448, 2021. https://doi.org/10.1109/ACCESS.2021.3060654
  51. S. Ungureanu, V. Topa, and A. C. Cziker, "Deep learning for short-term load forecasting— industrial consumer case study," Appl. Sci., Vol. 11, No. 21, 10126, Nov. 2021. https://doi.org/10.3390/app112110126
  52. A. Duttagupta, J. Zhao, and S. Shreejith, "Exploring lightweight federated learning for distributed load forecasting," arXiv preprint arXiv:2404.03320, Apr. 2024.
  53. A. X. Wang and J. J. Li, "A novel cloud-edge collaboration based short-term load forecasting method for smart grid," Front. Energy Res., Vol. 10, 977026, Aug. 2022. https://doi.org/10.3389/fenrg.2022.977026
  54. E. White Smith, M. Alexander, M. Pum, and H. Castro, "Explainable AI in load forecasting: Understanding GRU-LGBM predictions," ResearchGate, Jan. 2025. [Online]. Available: https://www.researchgate.net/publication/388625428_Explainable_AI_in_Load_Forecasting_Understanding_GRU-LGBM_Predictions
  55. J. Delgado Fernández, S. Potenciano Menci, C. M. Lee, A. Rieger, and G. Fridgen, "Privacypreserving federated learning for residential short-term load forecasting," Appl. Energy, Vol. 326, 119915, Nov. 2022. https://doi.org/10.1016/j.apenergy.2022.119915