Hierarchical Utility Data Forecasting Using Section-Wise Short-Term Demand Prediction with SFFO-Tuned Residual Networks
Received: 6 March 2026 | Revised: 18 April 2026 | Accepted: 26 April 2026 | Online: 2 June 2026
Corresponding author: Kanchan Ashish Khedikar
Abstract
Accurate short-term power demand forecasting is crucial for load planning and energy efficiency in distribution networks, particularly at fine-grained levels such as sections and transformers. This study presents a new Section-Wise Predictive Analysis framework based on a Sparrow FireFly Optimization-tuned Residual Deep Neural Network (SFFO-R-DNN) to forecast monthly kWh consumption using consumer-level data from the Sankh subdivision of Maharashtra State Electricity Distribution Company Limited (MSEDCL). Consumer ID, DTC code, section name, tariff, calendar details, and lag consumption are some of the features included in the more than 200,000 records of the dataset (2021–2023). While residual blocks represent non-linear dependencies, hierarchical embeddings incorporate behaviors at the spatial and tariff levels. Compared to default setups, the SFFO algorithm reduces validation RMSE by 3 kWh by automatically selecting appropriate hyperparameters. The suggested model outperforms ARIMA, DNN, LSTM, and GRU baselines on the test data, with an RMSE of 24.7 kWh, a MAE of 15.2 kWh, a MAPE of 6.8%, and an R2 of 0.963. Performance is consistent across tariffs, seasons, and demand bands, and section-wise MAPE stays within 6.1% to 7.3%. These results demonstrate the spatial robustness and operational reliability of the model. In rural and semi-urban energy networks, the proposed SFFO-R-DNN can assist with decentralized planning, detect overloads at the DTC level, and implement tailored DSM programs.
Keywords:
Maharashtra State Electricity Distribution Company Limited, residual deep neural network, sparrow firefly optimization, consumer-level data, power demandReferences
Y. Qian, Y. Wang, and J. Shao, "Enhancing power utilization analysis: detecting aberrant patterns of electricity consumption," Electrical Engineering, vol. 106, no. 5, pp. 5639–5654, Oct. 2024.
W. Liao, R. Zhu, T. Ishizaki, Y. Li, Y. Jia, and Z. Yang, "Can Gas Consumption Data Improve the Performance of Electricity Theft Detection?," IEEE Transactions on Industrial Informatics, vol. 20, no. 6, pp. 8453–8465, June 2024.
A. I. Kawoosa et al., "Improving Electricity Theft Detection Using Electricity Information Collection System and Customers’ Consumption Patterns," Energy Exploration & Exploitation, vol. 42, no. 5, pp. 1684–1714, Sept. 2024.
J. L. Rojas-Renteria, T. D. Espinoza-Huerta, F. S. Tovar-Pacheco, J. L. Gonzalez-Perez, and R. Lozano-Dorantes, "An Electrical Energy Consumption Monitoring and Forecasting System," Engineering, Technology & Applied Science Research, vol. 6, no. 5, pp. 1130–1132, Oct. 2016.
N. Zhang and L. Zhu, "Abnormal Electricity Consumption Detection Method for Smart Grid Using Fusion Matrix Completion and Improved Clustering Algorithm," IEEE Access, vol. 12, pp. 76634–76647, 2024.
S. K. Filipova-Petrakieva and V. Dochev, "Short-Term Forecasting of Hourly Electricity Power Demand: Reggresion and Cluster Methods for Short-Term Prognosis," Engineering, Technology & Applied Science Research, vol. 12, no. 2, pp. 8374–8381, Apr. 2022.
J. Wang and X. Li, "Abnormal Electricity Detection of Users Based on Improved Canopy-Kmeans and Isolation Forest Algorithms," IEEE Access, vol. 12, pp. 99110–99121, 2024.
X. Wang, H. Wang, B. Bhandari, and L. Cheng, "AI-Empowered Methods for Smart Energy Consumption: A Review of Load Forecasting, Anomaly Detection and Demand Response," International Journal of Precision Engineering and Manufacturing-Green Technology, vol. 11, no. 3, pp. 963–993, May 2024.
M. Al-Rajab and S. Loucif, "Sustainable EnergySense: a predictive machine learning framework for optimizing residential electricity consumption," Discover Sustainability, vol. 5, no. 1, Apr. 2024, Art. no. 55.
J. Einolander, A. Kiviaho, and R. Lahdelma, "Detecting changes in price-sensitivity of household electricity consumption: The impact of the global energy crisis on implicit demand response behavior of Finnish detached households," Energy and Buildings, vol. 306, Mar. 2024, Art. no. 113941.
S. F. Luna-Romero, X. Serrano-Guerrero, M. A. de Souza, and G. Escrivá-Escrivà, "Enhancing anomaly detection in electrical consumption profiles through computational intelligence," Energy Reports, vol. 11, pp. 951–962, June 2024.
B. Nemade, K. Kishor Maharana, V. Kulkarni, C. Srivardhankumar, and M. Shelar, "Revolutionizing smart grid security: a holistic cyber defence strategy," Frontiers in Artificial Intelligence, vol. 7, Dec. 2024.
A. Binbusayyis and M. Sha, "Energy consumption prediction using modified deep CNN-Bi LSTM with attention mechanism," Heliyon, vol. 11, no. 1, Jan. 2025.
X. Dong et al., "Building electricity load forecasting based on spatiotemporal correlation and electricity consumption behavior information," Applied Energy, vol. 377, Jan. 2025, Art. no. 124580.
B. N. Subitha and K. K. S. Harish, "Prediction of Electricity Consumption in Residential Areas using Temporal Fusion Transformer and Convolutional Neural Network," Journal of Machine and Computing, pp. 209–219, Jan. 2025.
P. C. Hsu, L. Gao, and Y. Hwang, "Comparative study of LSTM and ANN models for power consumption prediction of variable refrigerant flow (VRF) systems in buildings," International Journal of Refrigeration, vol. 169, pp. 55–68, Jan. 2025.
L. Zec, J. Mikulović, and M. Žarković, "Application of artificial neural network to power consumption forecasting for the Sarajevo region," Electrical Engineering, vol. 107, no. 3, pp. 3561–3572, Mar. 2025.
Downloads
How to Cite
License
Copyright (c) 2026 Kanchan Ashish Khedikar, Piyush Kumar Pareek
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
