Паметна аналитика и анализа потрошње електричне енергије

Марија Остовић

doi:10.24867/36BE49Ostovic

Electrotechnical and Computer Engineering

Vol. 41 No. 10 (2026): Proceedings of the Faculty of Technical Sciences

Smart Analytics and Analysis of Electrical Energy Consumption

Марија Остовић

36BE49Ostovic.pdf (Serbian)

DOI:: https://doi.org/10.24867/36BE49Ostovic
Submitted: October 14, 2026
Published: 2026-10-16

Abstract

The paper explores how modern analytical methods can be used to monitor time-series data in the energy sector. The research utilizes real-world datasets alongside a controlled synthetic dataset with intentionally injected errors to test the algorithms. Linear regression is applied to forecast consumption during the peak hour at 2 PM, while an autoencoder neural network is used to detect days that deviate from typical profiles. A threshold set at the 95th percentile of the reconstruction error successfully isolated the days with the most significant shifts, and PCA visualization graphically confirmed a clear separation between normal behavior and anomalies. Combining these two approaches proved effective for rapid recognition of unusual patterns, providing a solid foundation for a better understanding of data and smarter resource management within the grid.

References

[1] C. W. Gellings, The Smart Grid: Enabling Energy Efficiency and Demand Response. Lilburn, GA, USA: Fairmont Press, 2009.
[2] M. Amin and B. F. Wollenberg, “Toward a Smart Grid: Power Delivery for the 21st Century,” IEEE Power and Energy Magazine, vol. 3, no. 5, pp. 34–41, Sep.–Oct. 2005.
[3] J. Han, M. Kamber, and J. Pei, Data Mining: Concepts and Techniques, 3rd ed. Waltham, MA, USA: Morgan Kaufmann, 2012.
[4] V. Chandola, A. Banerjee, and V. Kumar, “Anomaly Detection: A Survey,” ACM Computing Surveys, vol. 41, no. 3, Article 15, pp. 1–58, 2009.
[5] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. Cambridge, MA, USA: MIT Press, 2016.
[6] Kaggle, “Individual Household Electric Power Consumption Dataset.” Accessed: 2026. [Online]. Available: https://www.kaggle.com/datasets/uciml/electric-power-consumption-data-set

[1] [1] C. W. Gellings, The Smart Grid: Enabling Energy Efficiency and Demand Response. Lilburn, GA, USA: Fairmont Press, 2009.

[2] [2] M. Amin and B. F. Wollenberg, “Toward a Smart Grid: Power Delivery for the 21st Century,” IEEE Power and Energy Magazine, vol. 3, no. 5, pp. 34–41, Sep.–Oct. 2005.

[3] [3] J. Han, M. Kamber, and J. Pei, Data Mining: Concepts and Techniques, 3rd ed. Waltham, MA, USA: Morgan Kaufmann, 2012.

[4] [4] V. Chandola, A. Banerjee, and V. Kumar, “Anomaly Detection: A Survey,” ACM Computing Surveys, vol. 41, no. 3, Article 15, pp. 1–58, 2009.

[5] [5] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. Cambridge, MA, USA: MIT Press, 2016.

[6] [6] Kaggle, “Individual Household Electric Power Consumption Dataset.” Accessed: 2026. [Online]. Available: https://www.kaggle.com/datasets/uciml/electric-power-consumption-data-set