Full Text PDF
ESTD Year: 2017 | Impact Factor (2026): 8.7
DOI Prefix: 10.47001/IRJIET
ESTD Year: 2017 | Impact Factor (2026): 8.7
DOI Prefix: 10.47001/IRJIET
Vol 10 No 6 (2026): Volume 10, Issue 6, June 2026 | Pages: 227-241
International Research Journal of Innovations in Engineering and Technology
OPEN ACCESS | Research Article | Published Date: 24-06-2026
The study addresses the challenge of accurate long-term load forecasting in power distribution systems, focusing on uncertainties in transformer load behavior across multiple substations. Conventional models, including high-order polynomial regressions and standard artificial neural networks, often exhibit overfitting, large forecast errors, and unrealistic growth predictions. To overcome these limitations, the study implements a Neuronal Auditory Machine Intelligence (NeuroAMI) framework, integrating sequential time-series learning with advanced probabilistic forecasting. The Online Sequential Extreme Learning Machine (OS-ELM) and Linear Regression (LR) models are employed for comparative analysis, while forecast accuracy is evaluated using Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), Continuous Ranked Probability Score (CRPS), and skill scores. Results indicate that NeuroAMI consistently outperforms baseline methods, achieving a low RMSE of 440.49 kW, a minimal MAPE of 4.05%, and the highest accuracy at 95.95%. CRPS analysis further confirms improved calibration and tighter predictive intervals, reflecting reduced uncertainty in transformer load projections. Forecast comparisons over a ten-year horizon demonstrate that NeuroAMI closely tracks true load trends for feeders with varying growth rates, in contrast to the unrealistic exponential predictions of polynomial models. The findings highlight the potential of NeuroAMI for enhancing operational planning, capacity expansion decisions, and reliability assessments in electricity distribution. This study aligns with policies promoting data-driven, predictive management of power systems, offering a robust, scalable framework adaptable to varying load profiles and substations.
NeuroAMI, OS-ELM, Load Forecasting, RMSE, Probabilistic Forecasting.
B. E. Nemine, C. O. Ahiakwo, S. L. Braide, & H. N. Amadi. (2026). Comparative Evaluation of NeuroAMI, OS ELM, Linear Regression, and Reinforcement Learning. International Research Journal of Innovations in Engineering and Technology - IRJIET, 10(6), 227-241. Article DOI https://doi.org/10.47001/IRJIET/2026.106029
This work is licensed under Creative common Attribution Non Commercial 4.0 Internation Licence
Z.A. Bashir and M. E. El-Hawary, “Applying wavelets to short-term load forecasting using PSO-based neural networks,” IEEE Trans. Power Syst., vol. 24, no. 1, pp. 20–27, 2009.
J. Bedi and D. Toshniwal, “Deep learning framework to forecast electricity demand,” Appl. Energy, vol. 238, pp. 1312–1326, 2019, doi: 10.1016/j.apenergy.2019.01.113.
C. J. Bennett, R. A. Stewart, and J. W. Lu, “Forecasting low voltage distribution network demand profiles using a pattern recognition based expert system,” Energy, vol. 67, pp. 200–212, 2014.
C. J. Bennett, R. A. Stewart, and J. W. Lu, “Development of a three-phase battery energy storage scheduling and operation system for low voltage distribution networks,” Appl. Energy, vol. 146, pp. 122–134, 2015.
J. A. Bennett, C. N. Trevisan, J. F. DeCarolis, C. Ortiz-García, M. Pérez-Lugo, B. T. Etienne, and A. F. Clarens, “Extending energy system modelling to include extreme weather risks and application to hurricane events in Puerto Rico,” Nat. Energy, vol. 6, no. 3, pp. 240–249, 2021.
P. M. Bento, J. A. Pombo, M. R. Calado, and S. J. Mariano, “Stacking ensemble methodology using deep learning and ARIMA models for short-term load forecasting,” Energies, vol. 14, no. 21, Art. no. 7378, 2021.
A.Bersani, A. Borghetti, C. Bossi, L. De Biase, O. Lamquet, S. Massucco, and F. Silvestro, “Management of low voltage grids with high penetration of distributed generation: Concepts, implementations and experiments (Cigré 2006 Session Paper),” Cigré, 2006.
C.Bersani, H. Dagdougui, A. Ouammi, and R. Sacile, “Stochastic constrained linear quadratic control in a network of smart microgrids,” IET Renew. Power Gener., vol. 14, no. 7, pp. 1193–1200, 2020.
C. Bersani, M. Fossa, A. Priarone, R. Sacile, and E. Zero, “Model predictive control versus traditional relay control in a high energy efficiency greenhouse,” Energies, vol. 14, no. 11, Art. no. 3353, 2021.
C. E. Borges, O. Kamara-Esteban, T. Castillo-Calzadilla, C. M. Andonegui, and A. Alonso-Vicario, “Enhancing the missing data imputation of primary substation load demand records,” Sustain. Energy, Grids Netw., vol. 23, Art. no. 100369, 2020.
A.Carvalho, A. Chouchene, T. Lima, and F. Charrua-Santos, “Cognitive manufacturing in Industry 4.0 toward cognitive load reduction: A conceptual framework,” Appl. Syst. Innov., vol. 3, no. 4, Art. no. 55, 2020, doi: 10.3390/asi3040055.
J.W. Chan and C. K. Yeo, “A transformer-based approach to electricity load forecasting,” The Electricity Journal, vol. 37, Art. no. 107370, 2024, doi: 10.1016/j.tej.2024.107370.
L. R. Chandran, G. A. Babu, M. G. Nair, and K. Ilango, “A review on status monitoring techniques of transformer and a case study on loss of life calculation of distribution transformers,” Mater. Today: Proc., vol. 46, pp. 4659–4666, 2021.
S. J. Chapman, Electric machinery fundamentals, 3rd ed. New York: McGraw-Hill, 2003.
K. Chen, K. Chen, Q. Wang, Z. He, J. Hu, and J. He, “Short-term load forecasting with deep residual networks,” IEEE Trans. Smart Grid, vol. 10, no. 4, pp. 3943–3952, 2018.
L. Chen, H. Yu, L. Tong, X. Huai, P. Jin, Y. Huang, and C. Dou, “Research on load forecasting method of distribution transformer based on deep learning,” in Proc. 2020 7th IEEE Int. Conf. Cyber Secur. Cloud Comput. (CSCloud)/2020 6th IEEE Int. Conf. Edge Comput. Scalable Cloud (EdgeCom), Aug. 2020, pp. 228–233.
N. Ding, C. Benoit, G. Foggia, Y. Besanger, and F. Wurtz, “Neural network-based model design for short-term load forecast in distribution systems,” IEEE Trans. Power Syst., vol. 31, no. 1, pp. 72–81, 2015.
M. Dong, A. B. Nassif, and B. Li, “A data-driven residential transformer overloading risk assessment method,” IEEE Trans. Power Del., vol. 34, no. 1, pp. 387–396, 2018.
G. Duan and W. Han, “Heavy overload prediction method of distribution transformer based on GBDT,” Int. J. Pattern Recognit. Artif. Intell., vol. 36, no. 09, Art. no. 2259014, 2022.
L. Fauziah, D. Devianto, and M. Maiyastri, “Medium-term electricity load forecasting in Kuantan Bay area using Cheng's fuzzy time series method,” UNAND Math. J., vol. 8, no. 2, pp. 84–92, 2019.