Automated Machine Learning for Optimized Load Forecasting and Economic Impact in the Greek Wholesale Energy Market
Source abstract
Abstract
This study investigates the use of automated machine learning to forecast the demand of electrical loads. A stochastic optimization algorithm minimizes the cost and risk of the traded asset across different markets using a generic framework for trading activities of load portfolios. Assuming an always overbought condition in the Day-Ahead as well as in the Futures Market, the excess energy returns without revenue to the market, and the results are compared with a standard contract in Greece, which stands as the lowest as far as the billing price is concerned. The analysis achieved a mean absolute percentage error (MAPE) of 12.89% as the best fitted model and without using any kind of pre-processing methods.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Energy Load and Power Forecasting · Market Dynamics and Volatility · Energy Efficiency and Management
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Chemical Process and Energy Resources Institute, Centre for Research and Technology Hellas, Egialias 52, 15125 Athens, Greece
References (43)
Show source-provided references
1. Short-term load forecasting using neural networks and global climate models: An application to a large-scale electrical power system10.1016/j.apenergy.2023.121439 ↗
2. Maragkos, N., Tzelepi, M., Passalis, N., Adamakos, A., and Tefas, A. (2022, January 26–29). Electric load demand forecasting on Greek Energy Market using lightweight neural networks. Proceedings of the 2022 IEEE 14th Image, Video, and Multidimensional Signal Processing Workshop (IVMSP), Nafplio, Greece.10.1109/IVMSP54334.2022.9816189 ↗
3. Cortes, C., Lawrence, N., Lee, D., Sugiyama, M., and Garnett, R. (2015). Efficient and Robust Automated Machine Learning. Advances in Neural Information Processing Systems, Curran Associates, Inc.. Available online: https://proceedings.neurips.cc/paper_files/paper/2015/file/11d0e6287202fced83f79975ec59a3a6-Paper.pdf.
4. Hafeez, S., and Kathirisetty, N. (2022, January 10–12). Effects and Comparison of different Data pre-processing techniques and ML and deep learning models for sentiment analysis: SVM, KNN, PCA with SVM and CNN. Proceedings of the 2022 First International Conference on Artificial Intelligence Trends and Pattern Recognition (ICAITPR), Hyderabad, India.10.1109/ICAITPR51569.2022.9844192 ↗
5. Machine learning algorithms for predicting electrical load demand: An evaluation and comparison10.1007/s12046-023-02354-2 ↗
6. Westergaard, G., Erden, U., Mateo, O.A., Lampo, S.M., Akinci, T.C., and Topsakal, O. (2024). Time Series Forecasting Utilizing Automated Machine Learning (AutoML): A Comparative Analysis Study on Diverse Datasets. Information, 15.10.3390/info15010039 ↗
7. Incorporating Black-Litterman views in portfolio construction when stock returns are a mixture of normals10.1016/j.omega.2018.11.017 ↗
8. Portfolio Selection
9. Hyndman, R.J., and Athanasopoulos, G. Forecasting: Principles and Practice, OTexts. [3rd ed.]. Available online: https://otexts.com/fpp3/.
10. Load Forecasting via Deep Neural Networks10.1016/j.procs.2017.11.374 ↗
11. Neural Architecture Search: A Survey
12. L’Heureux, A., Grolinger, K., and Capretz, M.A.M. (2022). Transformer-Based Model for Electrical Load Forecasting. Energies, 15.10.3390/en15144993 ↗
13. Thompson, N.C., Greenewald, K., Lee, K., and Manso, G.F. (2020). The Computational Limits of Deep Learning. arXiv.
14. An intelligent hybrid short-term load forecasting model optimized by switching delayed PSO of micro-grids10.1063/1.5021728 ↗
15. Sequential Model-Based Optimization for General Algorithm Configuration10.1007/978-3-642-25566-3_40 ↗
16. A Comparison of Automated Machine Learning Tools for Predicting Energy Building Consumption in Smart Cities
17. Golmohamadi, H., Keypour, R., Hassanpour, A., and Davoudi, M. (2015, January 4–6). Optimization of green energy portfolio in retail market using stochastic programming. Proceedings of the 2015 North American Power Symposium (NAPS), Charlotte, NC, USA.10.1109/NAPS.2015.7335233 ↗
18. Fotopoulou, M., Rakopoulos, D., and Blanas, O. (2021). Day Ahead Optimal Dispatch Schedule in a Smart Grid Containing Distributed Energy Resources and Electric Vehicles. Sensors, 21.10.3390/s21217295 ↗
19. Risk-aware two-stage stochastic programming for electricity procurement of a large consumer with storage system and demand response10.1016/j.est.2022.104478 ↗
20. Deng, S., Zhu, Y., Huang, X., Duan, S., and Fu, Z. (2022). High-Frequency Direction Forecasting of the Futures Market Using a Machine-Learning-Based Method. Future Internet, 14.10.3390/fi14060180 ↗
21. Bidding Strategy for Microgrid in Day-Ahead Market Based on Hybrid Stochastic/Robust Optimization10.1109/TSG.2015.2476669 ↗
22. Local Energy Trading Behavior Modeling With Deep Reinforcement Learning10.1109/ACCESS.2018.2876652 ↗
23. (2024, September 25). Household Data. Available online: https://data.open-power-system-data.org/household_data/.
24. LeDell, E., and Poirier, S. (2024, August 12). H2O AutoML: Scalable Automatic Machine Learning. Available online: https://api.semanticscholar.org/CorpusID:221338558.
25. Wang, C., Wu, Q., Weimer, M., and Zhu, E. (2021, January 5–9). FLAML: A Fast and Lightweight AutoML Library. Proceedings of the Fourth Conference on Machine Learning and Systems (MLSys 2021), Virtual. Available online: https://www.microsoft.com/en-us/research/publication/flaml-a-fast-and-lightweight-automl-library/.
26. Wang, C., Chen, X., Wu, C., and Wang, H. (2022). AutoTS: Automatic Time Series Forecasting Model Design Based on Two-Stage Pruning. arXiv, Available online: http://arxiv.org/abs/2203.14169.10.2139/ssrn.4614633 ↗
27. Optimization of diabetes prediction methods based on combinatorial balancing algorithm10.1038/s41387-024-00324-z ↗
28. Prado, F.F., and Digiampietri, L.A. (2020, January 3–6). A systematic review of automated feature engineering solutions in machine learning problems. Proceedings of the SBSI ‘20: Proceedings of the XVI Brazilian Symposium on Information Systems, São Bernardo do Campo, Brazil.10.1145/3411564.3411610 ↗
29. Yang, L., and Shami, A. (2020). On Hyperparameter Optimization of Machine Learning Algorithms: Theory and Practice. arXiv.10.1016/j.neucom.2020.07.061 ↗
30. Automated machine learning with dynamic ensemble selection10.1007/s10489-023-04770-7 ↗
31. Another look at measures of forecast accuracy10.1016/j.ijforecast.2006.03.001 ↗
32. Sparse and robust mean–variance portfolio optimization problems10.1016/j.physa.2019.04.151 ↗
33. Dorsman, A., Westerman, W., Karan, M.B., and Arslan, Ö. (2011). The Electricity Market, Day-Ahead Market and Futures Market. Financial Aspects in Energy, Springer.10.1007/978-3-642-19709-3 ↗
34. Demand response performance and uncertainty: A systematic literature review10.1016/j.esr.2022.100857 ↗
35. Risk assessment of microgrid aggregators considering demand response and uncertain renewable energy sources10.1007/s40565-019-0513-x ↗
36. Futures hedging in electricity retailing10.1007/s10479-022-04969-w ↗
37. Hedging in Financial Markets10.2143/AST.28.1.519076 ↗
38. RETRACTED: Robust bidding strategy of interconnected multi-carrier systems in the electricity markets under the uncertainty in electricity load
39. The M4 Competition: 100,000 time series and 61 forecasting methods10.1016/j.ijforecast.2019.04.014 ↗
40. Comparative Analysis an Accounting Technique of Analysis and Interpretation of Financial Statement10.55248/gengpi.5.0524.1220 ↗
41. Vakhrushev, A., Ryzhkov, A., Savchenko, M., Simakov, D., Damdinov, R., and Tuzhilin, A. (2021). LightAutoML: AutoML Solution for a Large Financial Services Ecosystem. arXiv.
42. Zhang, H., Li, Z., Xue, Y., Chang, X., Su, J., Wang, P., Guo, Q., and Sun, H. (IEEE Trans. Consum. Electron., 2024). A Stochastic Bi-level Optimal Allocation Approach of Intelligent Buildings Considering Energy Storage Sharing Services, IEEE Trans. Consum. Electron., early access.10.1109/TCE.2024.3412803 ↗
43. Alghamdi, H., Hafeez, G., Ali, S., Ullah, S., Khan, M.I., Murawwat, S., and Hua, L.-G. (2023). An Integrated Model of Deep Learning and Heuristic Algorithm for Load Forecasting in Smart Grid. Mathematics, 11.10.3390/math11214561 ↗