Short-Term Power Load Forecasting Using an Improved Model Integrating GCN and Transformer
Source abstract
Abstract
Improving the accuracy of power load forecasting is an important link in the optimization of power systems. Most of the existing studies in the short-term load forecasting task at present have the problem of insufficient extraction of multi-scale features. Therefore, in order to improve prediction accuracy, this study designs a short-term power load forecasting model integrating multi-scale GCN and the improved Transformer, as well as the prediction method based on this model. First, multi-feature power load data were collected. Second, the random forest algorithm was used to preprocess the data. Next, multi-scale GCN was utilized to model the multi-scale spatio-temporal features in the power load data. The data processed by the multi-scale GCN were input into the improved Transformer module based on MLLA to extract long-term temporal dependencies. Subsequently, comparative experiments and ablation experiments were conducted on three public power datasets. The experimental results show that, compared to the comparative model, for the ETTh1 dataset, the RMSE index of this model decreased by up to 0.314, the MAE decreased by up to 0.304, and the R2 index result improved by up to 9.45%. For the ETTm1 dataset, the RMSE index of this model decreased by up to 0.266, the MAE decreased by up to 0.231, and the R2 index result improved by up to 3.3%. For the Australian dataset, the RMSE index of this model decreased by up to 494.366, the MAE decreased by up to 493.127, and the R2 index result improved by up to 54%, verifying the superiority and effectiveness of the proposed model.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Energy Load and Power Forecasting · Geoscience and Mining Technology · Smart Grid and Power Systems
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Information and Communication Engineering, Hainan University, Haikou 570228, China
School of Computing and Artificial Intelligence, Hainan College of Software Technology, Qionghai 571499, China
School of Tourism, Hainan Normal University, Haikou 570228, China
Funding
“Rising Star in the South China Sea” project of Hainan Province · Education Department of Hainan Province
References (44)
Show source-provided references
1. Mustapha, M., Mustafa, M., Khalid, S., Abubakar, I., and Shareef, H. (2015, January 19–20). Classification of electricity load forecasting based on the factors influencing the load consumption and methods used: An-overview. Proceedings of the 2015 IEEE Conference on Energy Conversion (CENCON 2015), Johor Bahru, Malaysia.10.1109/CENCON.2015.7409585 ↗
2. Klyuev, R., Morgoev, I., Morgoeva, A., Gavrina, O.A., Martyushev, N.V., Efremenkov, E.A., and Mengxu, Q. (2022). Methods of forecasting electric energy consumption: A literature review. Energies, 15.10.3390/en15238919 ↗
3. Short-term load forecasting in smart meters with sliding window-based ARIMA algorithms10.1007/s40595-018-0119-7 ↗
4. Short-term load forecasting method based on fuzzy time series, seasonality and long memory process10.1016/j.ijar.2017.01.006 ↗
5. A novel multivariate combined power load forecasting system based on feature selection and multi-objective intelligent optimization10.1016/j.eswa.2023.122970 ↗
6. Short-term photovoltaic power forecasting with feature extraction and attention mechanisms10.1016/j.renene.2024.120437 ↗
7. A hybrid model for deep learning short-term power load forecasting based on feature extraction statistics techniques10.1016/j.eswa.2023.122012 ↗
8. Recurrent neural networks for short-term load forecasting10.1109/59.651623 ↗
9. Application of bidirectional recurrent neural network combined with deep belief network in short-term load forecasting10.1109/ACCESS.2019.2950957 ↗
10. Ultra-short-term industrial power demand forecasting using LSTM based hybrid ensemble learning10.1109/TPWRS.2019.2963109 ↗
11. A hybrid intelligent model based on recurrent neural networks and excitable dynamics for price prediction in deregulated electricity market10.1016/j.engappai.2012.12.012 ↗
12. Zhang, J., Li, H., Cheng, P., and Yan, J. (2024). Interpretable Wind Power Short-Term Power Prediction Model Using Deep Graph Attention Network. Energies, 17.10.3390/en17020384 ↗
13. A short-term wind power prediction method via self-adaptive adjacency matrix and spatiotemporal graph neural networks10.1016/j.compeleceng.2024.109715 ↗
14. Powerformer: A temporal-based transformer model for wind power forecasting10.1016/j.egyr.2023.12.030 ↗
15. A novel model for ultra-short term wind power prediction based on Vision Transformer10.1016/j.energy.2024.130854 ↗
16. SmartFormer: Graph-based transformer model for energy load forecasting
17. A short-term load forecasting model of multi-scale CNN-LSTM hybrid neural network considering the real-time electricity price10.1016/j.egyr.2020.11.078 ↗
18. Multi-temporal-spatial-scale temporal convolution network for short-term load forecasting of power systems10.1016/j.apenergy.2020.116328 ↗
19. Electricity demand loads modeling using AutoRegressive Moving Average (ARMA) models10.1016/j.energy.2008.05.008 ↗
20. Short-Term Wind Power Generation Forecasting: Direct Versus Indirect Arima-Based Approaches10.1080/15435075.2011.546755 ↗
21. Short-Term Load Forecasting With Exponentially Weighted Methods10.1109/TPWRS.2011.2161780 ↗
22. An ensemble approach for short-term load forecasting by extreme learning machine10.1016/j.apenergy.2016.02.114 ↗
23. A Strategy for Short-Term Load Forecasting by Support Vector Regression Machines10.1109/TPWRS.2013.2269803 ↗
24. Comprehensive learning particle swarm optimization based memetic algorithm for model selection in short-term load forecasting using support vector regression10.1016/j.asoc.2014.09.007 ↗
25. Abumohsen, M., Owda, A., and Owda, M. (2023). Electrical load forecasting using LSTM, GRU, and RNN algorithms. Energies, 16.10.3390/en16052283 ↗
26. L’Heureux, A., Grolinger, K., and Capretz, M. (2022). Transformer-based model for electrical load forecasting. Energies, 15.10.3390/en15144993 ↗
27. Short-Term Load Forecasting Based on Improved TCN and DenseNet10.1109/ACCESS.2022.3218374 ↗
28. LSTM enhanced by dual-attention-based encoder-decoder for daily peak load forecasting10.1016/j.epsr.2022.107860 ↗
29. Short-term load forecasting based on LSTM networks considering attention mechanism10.1016/j.ijepes.2021.107818 ↗
30. Short-term multi-energy load forecasting for integrated energy systems based on CNN-BiGRU optimized by attention mechanism10.1016/j.apenergy.2022.118801 ↗
31. Cai, W., Liang, Y., Liu, X., Feng, J., and Wu, Y. (2024, January 20–27). MSGNet: Learning multi-scale inter-series correlations for multivariate time series forecasting. Proceedings of the 38th AAAI Conference on Artificial Intelligence (AAAI2024), Vancouver, BC, Canada.10.1609/aaai.v38i10.28991 ↗
32. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, Ł., and Polosukhin, I. (2017, January 9–12). Attention is All You Need. Proceedings of the 31st Proceedings of Advances in Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA.
33. Han, D., Wang, Z., Xia, Z., Han, Y., Pu, Y., Ge, C., Song, J., Song, S., Zheng, B., and Huang, G. (2024, January 4–9). Demystify Mamba in Vision: A Linear Attention Perspective. Proceedings of the 38th Proceedings of Advances in Neural Information Processing Systems (NIPS 2024), Vancouver, BC, Canada.
34. Gu, A., and Dao, T. (2024, January 21–27). Mamba: Linear-Time Sequence Modeling with Selective State Spaces. Proceedings of the 2024 International Conference on Machine Learning (ICML2024), Vienna, Austria.
35. Devlin, J., Chang, M., Lee, K., and Toutanova, K. (2018, January 1–4). Bert: Pre-training of deep bidirectional transformers for language understanding. Proceedings of the Empirical Methods in Natural Language Processing (EMNLP2018), Brussels, Belgium.
36. Zhou, H., Zhang, S., Peng, J., Zhang, S., Li, J., Xiong, H., and Zhang, W. (2021, January 2–6). Informer: Beyond efficient transformer for long sequence timeseries forecasting. Proceedings of the 35th AAAI Conference on Artificial Intelligence (AAAI2021), Vancouver, BC, Canada.10.1609/aaai.v35i12.17325 ↗
37. Kitaev, N., Kaiser, L., and Levskaya, A. (2020, January 26–30). Reformer: The efficient transformer. Proceedings of the 8th International Conference on Learning Representations (ICLR 2020), Addis Ababa, Ethiopia.
38. Forecasting tourist arrivals via random forest and long short-term memory10.1007/s12559-020-09747-z ↗
39. Airborne Particle Pollution Predictive Model Using Gated Recurrent Unit (GRU) Deep Neural Networks10.1007/s12145-020-00462-9 ↗
40. Bai, S., Kolter, Z., and Koltun, V. (2018). An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling. arXiv.
41. Jung, S., Moon, J., Park, S., and Hwang, E. (2021). An Attention-Based Multi Layer GRU Model for Multi-Step-Ahead Short-Term Load Forecasting. Sensors, 21.10.3390/s21051639 ↗
42. Short-term Load Forecasting Based on TCN-Attention Neural Network
43. MrCAN: Multi-relations aware convolutional attention network for multivariate time series forecasting10.1016/j.ins.2023.119277 ↗
44. Kingma, D., and Ba, J. (2015, January 7–9). Adam: A Method for Stochastic Optimization. Proceedings of the 3rd International Conference of Learning Representations (ICLR 2015), San Diego, CA, USA.