System Engineering and Productivity

System Engineering and Productivity

Developing a Deep Learning-based Power Outage Predictive Model to Improve Resilience of Power Systems

Articles in Press

Document Type : Research Paper

Authors
Reihane Montazeri Najafabadi 1
Mohammadreza Shams 2
1 M.Sc. Student, Department of Data Science, Faculty of Computer Engineering, University of Isfahan, Isfahan, Iran
2 Corresponding author: Assistant Professor, Department of Computer Engineering, Faculty of Shahreza Campus, University of Isfahan, Isfahan, Iran
10.22034/sep.2026.2076246.1420
Abstract
Unplanned power outages disrupt grid stability and increase operational costs, posing a major threat to power system efficiency. In this paper, we propose a robust outage prediction model that combines an Autoencoder-based feature extractor with a residual multi-layer perceptron (MLP) classifier. The novelty of our approach lies in its ability to maintain high predictive performance while eliminating reliance on geographic features such as latitude and longitude—commonly required by traditional models. We first train the Autoencoder on a rich, unlabeled dataset of weather and energy demand data collected over two decades (2000–2024) across Maryland, USA. The learned latent representations are then used to augment a supervised classification model trained on labeled outage data. Our final model achieves an F1-score of 81% even without location-based features, compared to 90% when using all features. This generalizability enables the deployment of predictive tools in previously unseen regions, directly enhancing grid flexibility, reliability, and system efficiency.

Highlights

  • A transfer learning framework was proposed for power outage prediction in Maryland, USA, combining an AutoEncoder-based feature extractor with a residual MLP classifier.
  • This approach eliminate dependence on geographical features such as latitude and longitude, enabling efficient generalization to unseen regions.
  • Compared to traditional models, proposed model achieved higher accuracy and overall performance across different tests with other predictive approaches.

Keywords
Power Outage
Resilience of Power Systems
Deep Learning
Transfer Learning
Autoencoder
Subjects

Copyright © Reihane Montazeri Najafabadi, Mohammadreza Shams

 

License

This article is released under the Creative Commons Attribution (CC BY 4.0) license. Anyone is free to copy, share, translate, and adapt this article for any purpose, whether commercial or non-commercial, as long as proper citation is given to the authors and original publication.

Arifeen, M., Petrovski, A., Hasan, M. J., Noman, K., Navid, W. U., & Haruna, A. (2024). Graph-Variational Convolutional Autoencoder-Based Fault Detection and Diagnosis for Photovoltaic Arrays. Machines, 12(12), 894. https://doi.org/10.3390/machines12120894
Atrigna, M., Buonanno, A., Carli, R., Cavone, G., Scarabaggio, P., Valenti, M., ... & Dotoli, M. (2023). A machine learning approach to fault prediction of power distribution grids under heatwaves. IEEE Transactions on Industry Applications, 59(4), 4835-4845. https://doi.org/10.1109/TIA.2023.3262230
Avazpour, M., Zarei, J., & Alinajad, E. (2025). Evaluation and prioritization of electricity generation technologies in Iran using a multi-criteria decision-making approach. Systems Engineering and Productivity, 5(3), 179–198 (In Persian). https://doi.org/10.22034/sep.2025.2063697.1333
Azzalini, D., Flammini, B., Emanuele, C. A., Guadagno, A., Ragaini, E., & Amigoni, F. (2025). An empirical evaluation of deep autoencoders for anomaly detection in the electricity consumption of buildings. Energy and Buildings, 327, 115069. https://doi.org/10.1016/j.enbuild.2024.115069
Campbell, R. J., & Lowry, S. (2012). Weather-related power outages and electric system resiliency. Washington, DC: Congressional Research Service, Library of Congress.
Coleman, J. (2022). Climate change made South Asian heatwave 30 times more likely. Nature. https://doi.org/10.1038/d41586-022-01444-1
Falabretti, D., Schiavo, L. L., Liotta, S., & Palazzoli, A. (2020). A novel method for evaluating the resilience of distribution networks during heat waves. International Journal of Electrical and Electronic Engineering & Telecommunications, 9(2), 73–79. https://doi.org/10.18178/ijeetc.9.2.73-79
Ghasemkhani, B., Kut, R. A., Yilmaz, R., Birant, D., Arıkök, Y. A., Güzelyol, T. E., & Kut, T. (2024). Machine learning model development to predict power outage duration (POD): A case study for electric utilities. Sensors, 24(13), 4313. https://doi.org/10.3390/s24134313
Ghosh, H. (2025). Autoencoders and transfer learning. Medium.
Huang, W., Zhang, W., Chen, Q., Feng, B., & Li, X. (2024). Prediction algorithm for power outage areas of affected customers based on CNN-LSTM. IEEE Access, 12, 15007–15015. https://doi.org/10.1109/ACCESS.2024.3355484
Namdarzadegan, M., & Bozorgi Amiri, A. (2025). Developing a hybrid sustainability–resilience model for implementing appropriate smart technologies in small and medium-sized enterprises under crisis conditions. Systems Engineering and Productivity (In Persian). https://doi.org/10.22034/sep.2025.2064025.1337
Nyame, S., Taylor, W. O., Hughes, W., Hong, M., Koukoula, M., Yang, F., ... & Cerrai, D. (2024). Transmission Failure Prediction Using AI and Structural Modeling Informed by Distribution Outages. IEEE Access, 13, 42-55. https://doi.org/10.1109/ACCESS.2024.3523415
Rahbaripour, K., Pakdelfard, M., Satarisarbanqoli, H., & Valizadeh, N. (2025). Analysis of barriers to achieving the Fourth Manufacturing Industry using the Interpretive Structural Modeling approach. Systems Engineering and Productivity (In Persian). https://doi.org/10.22034/sep.2025.2068090.1374
Razzaghy, S., Ahmadvand, A., Samadi Foroushani, M., (2024). Presenting a Dynamic Model of Iran's Electrical Energy Supply System Based on the Water-Food-Energy-Climate Change Nexus. Systems Engineering and Productivity, 3(4), 36-80. https://doi.org/10.22034/msb.2024.2022934.1183
Romitti, Y., & Sue Wing, I. (2022). Heterogeneous climate change impacts on electricity demand in world cities circa mid-century. Scientific reports, 12(1), 4280. https://doi.org/10.1038/s41598-022-07922-w
Satterlee, K. (2024). Combining big data and machine learning to predict power outages and help consumers prepare. Texas A&M University News.
Sharma, V., Hong, T., Cecchi, V., Hofmann, A., & Lee, J. Y. (2023). Forecasting weather‐related power outages using weighted logistic regression. IET Smart Grid, 6(5), 470-479. https://doi.org/10.1049/stg2.12109
Stone, B., Jr., Gronlund, C. J., Mallen, E., Hondula, D., O’Neill, M. S., Rajput, M., Grijalva, S., Lanza, K., Harlan, S., Larsen, L., et al. (2023). How blackouts during heat waves amplify mortality and morbidity risk. Environmental Science & Technology, 57(11), 8245–8255. https://doi.org/10.1021/acs.est.2c09588
Stone, B., Mallen, E., Rajput, M., Broadbent, A., Krayenhoff, E. S., Augenbroe, G., & Georgescu, M. (2021). Climate change and infrastructure risk: Indoor heat exposure during a concurrent heat wave and blackout event in Phoenix, Arizona. Urban Climate, 36, 100787. https://doi.org/10.1016/j.uclim.2021.100787
Tabassum, T., Toker, O., & Khalghani, M. R. (2024). Cyber–physical anomaly detection for inverter-based microgrid using autoencoder neural network. Applied Energy, 355, 122283. https://doi.org/10.1016/j.apenergy.2023.122283
Wang, X., Fatehi, N., Wang, C., & Nazari, M. H. (2024). Deep learning-based weather-related power outage prediction with socio-economic and power infrastructure data. IEEE Power & Energy Society General Meeting (PESGM) (pp. 1-5). IEEE. https://doi.org/10.1109/PESGM51994.2024.10688596
Zhu, S., Yao, R., Xie, Y., Qiu, F., Qiu, Y., & Wu, X. (2025). Quantifying grid resilience against extreme weather using large-scale customer power outage data. INFORMS Journal on Data Science. https://doi.org/10.1287/ijds.2023.0017
Zideh, M. J., & Solanki, S. K. (2024). Multivariate Physics-Informed Convolutional Autoencoder for Anomaly Detection in Power Distribution Systems with High Penetration of DERs. arXiv preprint arXiv:2406.02927. https://arxiv.org/abs/2406.02927
System Engineering and Productivity

Articles in Press, Accepted Manuscript
Available Online from 01 January 2026

  • Receive Date 01 November 2025
  • Revise Date 22 December 2025
  • Accept Date 01 January 2026
  • First Publish Date 01 January 2026
  • Publish Date 01 January 2026