journal article·2025-07-10Open access

A Novel Energy Control Digital Twin System with a Resource-Aware Optimal Forecasting Model Selection Scheme

Jin-Woo Kwon, Anwar Rubab, Won-Tae Kim

Applied Sciences· Volume 15· Issue 14

Abstract

As global energy demand intensifies across industrial, commercial, and residential domains, efficient and accurate energy management and control become crucial. Energy Digital Twins (EDTs), leveraging sensor measurement data and precise time-series forecasting models, offer promising monitoring, prediction, and optimization solutions for such services. Edge computing enables EDTs to deliver real-time management services placed closer to users. However, the existing energy management methodologies may fail to consider the limited resources of edge environments, which may cause service delays and reduced accuracy in management services. To solve this problem, we propose a novel energy control digital twin system with a resource-aware optimal forecasting mode selection scheme. The system dynamically selects optimal forecasting models by integrating statistical features of the input time series with available resources. It employs a two-stage approach: first, it identifies promising models through similarity detection in past time series; second, this initial recommendation is refined by considering the available computing resources to pinpoint the optimal forecasting model. This mechanism enhances adaptability and responsiveness in resource-constrained environments. Utilizing real-world LPG consumption data from 887 sensors, the proposed system achieves forecasting accuracy comparable to previous methods while reducing latency by up to 19 times in low-resource settings.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Energy Efficiency and Management · Digital Transformation in Industry · Forecasting Techniques and Applications

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Future Convergence and Engineering, Computer Science Department, Korea University of Technology and Education, Cheonan-si 31253, Republic of Korea

Computer Science and Engineering, Computer Science Department, Korea University of Technology and Education, Cheonan-si 31253, Republic of Korea

Funding

Technology Innovation Program (Development of SDF-Based AI Autonomous Manufacturing Core Technology to Advance the Automobile Industry) funded by the Ministry of Trade, Industry, and Energy (MOTIE), South Korea · Star Professor Research Program of Korea University of Technology and Education

References (22)

Show source-provided references
1. Global Energy-Efficiency Transition Tendencies: Development Phenomenon or Not?10.1016/j.esr.2024.101524 ↗
2. Paving the Way towards a Sustainable Future or Lagging Behind? An Ex-Post Analysis of the IEA’s World Energy Outlook10.1016/j.rser.2025.115371 ↗
3. Ba, L., Tangour, F., El Abbassi, I., and Absi, R. (2025). Analysis of Digital Twin Applications in Energy Efficiency: A Systematic Review. Sustainability, 17.10.3390/su17083560 ↗
4. Advancements in Digital Twin Technology and Machine Learning for Energy Systems: A Comprehensive Review of Applications in Smart Grids, Renewable Energy, and Electric Vehicle Optimisation
5. Han, F., Du, F., Jiao, S., and Zou, K. (2024). Predictive Analysis of a Building’s Power Consumption Based on Digital Twin Platforms. Energies, 17.10.3390/en17153692 ↗
6. Short-Term Photovoltaic Power Prediction Based on a Digital Twin Model10.3390/jmse12071219 ↗
7. An Edge Computing-Oriented Net Power Forecasting for PV-Assisted Charging Station: Model Complexity and Forecasting Accuracy Trade-Off10.1016/j.apenergy.2021.118456 ↗
8. Makridakis, S., Spiliotis, E., and Assimakopoulos, V. (2018). Statistical and Machine Learning forecasting methods: Concerns and ways forward. PLoS ONE, 13.10.1371/journal.pone.0194889 ↗
9. A review of ARIMA vs. machine learning approaches for time series forecasting in data driven networks10.3390/fi15080255 ↗
10. Yemets, K., Izonin, I., and Dronyuk, I. (2025). Time Series Forecasting Model Based on the Adapted Transformer Neural Network and FFT-Based Features Extraction. Sensors, 25.10.3390/s25030652 ↗
11. A comprehensive survey of deep learning for time series forecasting: Architectural diversity and open challenges10.1007/s10462-025-11223-9 ↗
12. Comprehensive Review of Edge Computing for Power Systems: State of the Art, Architecture, and Applications10.3390/app15084592 ↗
13. Deep learning with edge computing: A review10.1109/JPROC.2019.2921977 ↗
14. Rule-based forecasting: Development and validation of an expert systems approach to combining time series extrapolations10.1287/mnsc.38.10.1394 ↗
15. Meta-learning approaches to selecting time series models10.1016/j.neucom.2004.03.008 ↗
16. Shchur, O., Turkmen, A.C., Erickson, N., Shen, H., Shirkov, A., Hu, T., and Wang, B. (2023, January 12–15). AutoGluon–TimeSeries: AutoML for probabilistic time series forecasting. Proceedings of the International Conference on Automated Machine Learning, Berlin, Germany.
17. Jin, H., Song, Q., and Hu, X. (2019, January 4–8). Auto-keras: An efficient neural architecture search system. Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery Data Mining, Anchorage, AK, USA.10.1145/3292500.3330648 ↗
18. Talkhi, N., Akhavan Fatemi, N., Jabbari Nooghabi, M., Soltani, E., and Jabbari Nooghabi, A. (2024). Using meta-learning to recommend an appropriate time-series forecasting model. BMC Public Health, 24.10.1186/s12889-023-17627-y ↗
19. Fischer, R., and Saadallah, A. (2024, January 25–29). AutoXPCR: Automated multi -objective model selection for time series forecasting. Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Barcelona, Spain.10.1145/3637528.3672057 ↗
20. SimpleTS: An efficient and universal model selection framework for time series forecasting10.14778/3611540.3611561 ↗
21. A sensor data-driven decision support system for liquefied petroleum gas suppliers10.3390/app11083474 ↗
22. Gluonts: Probabilistic and neural time series modeling in python