A Machine Learning Approach for Transformer Load Allocation in Power Distribution Networks
Keywords:
K-means, load allocation, machine learning, CYMDIST, Power Distribution NetworksAbstract
Accurate transformer load allocation is essential for distribution network planning and operation, particularly in systems without advanced metering infrastructure. This paper proposes a bottom-up data-driven methodology that integrates clustering of measured load curves, supervised classification, and energy-based scaling to estimate transformer-level demand while preserving temporal characteristics and energy consistency. The methodology was validated using real data from an Ecuadorian distribution feeder with measurements available at the feeder head and at an intermediate recloser. Power flow results were compared with the conventional load allocation methods available in CYMDIST in order to evaluate the accuracy of the proposed method against standard approaches. At the intermediate recloser, where downstream allocation accuracy is critical, the proposed method achieved the lowest error, with a mean absolute error of 32.23 kVA and a mean absolute percentage error of 6.60%, compared with the range of errors obtained with the CYMDIST methods, which reached up to 186.77 kVA and 33.05%, respectively. These results suggest that the proposed methodology provides a practical approach for improving load allocation in minimally monitored distribution networks.
Downloads
References
E. B´aez, R. Duarte, E. Buzarquis, and P. Gardel, “Load curves estimation of distribution transformers using rna,” in 2018 IEEE Biennial Congress of Argentina (ARGENCON), 2018, pp. 1–6. [Online]. Available: https://doi.org/10.1109/ARGENCON.2018.8646235
G. Rouwhorst, E. M. S. Duque, P. H. Nguyen, and H. Slootweg, “Improving clustering-based forecasting of aggregated distribution transformer loadings with gradient boosting and feature selection,” IEEE Access, vol. 10, pp. 443–455, 2022. [Online]. Available: https://doi.org/10.1109/ACCESS.2021.3137870
P. Torres-Bermeo, J. Varela-Ald´as, K. L´opez-Eugenio, N. Velasco, and G. Palacios-Navarro, “Optimization of electric transformer operation through load estimation based on the k-means algorithm,” Energies, vol. 18, no. 14, p. 3755, 2025. [Online]. Available: https://doi.org/10.3390/en18143755
H. Ren, N. N. Schulz, V. Krishnan, and Y. Zhang, “Online static load model estimation in distribution systems,” in 2019 IEEE 28th International Symposium on Industrial Electronics (ISIE), 2019, pp. 153–158. [Online]. Available: https://doi.org/10.1109/ISIE.2019.8781530
Y. Tan, C. Liao, Y. Li, Y. Cao, M. Shahidehpour, and C. Chen, “A linear power flow model for balanced distribution network with droop-controlled dstatcom and voltage controlled dg,” International Journal of Electrical Power & Energy Systems, vol. 117, p. 105665, 2020. [Online]. Available: https://doi.org/10.1016/j.ijepes.2019.105665
J. Jardini, C. Tahan, S. Ahn, and E. Ferrari, “Distribution transformer loading evaluation based on load profiles measurements,” IEEE Transactions on Power Delivery, vol. 12, no. 4, pp. 1766–1770, 1997. [Online]. Available: https://doi.org/10.1109/61.634203
P. F. Austnes, C. Garc´ıa-Pareja, F. Nobile, and M. Paolone, “Probabilistic load forecasting of distribution power systems based on empirical copulas,” Sustainable Energy, Grids and Networks, vol. 42, p. 101708, 2025. [Online]. Available: https://doi.org/10.1016/j.segan.2025.101708
W. Yan, G. Xu, X. Zhang, and R. Zhao, “A forward/backward robust state estimation algorithm for radial and simple loop distribution systems,” International Journal of Electrical Power & Energy Systems, vol. 163, p. 110366, 2024. [Online]. Available: https://doi.org/10.1016/j.ijepes.2024.110366
A. Primadianto and C.-N. Lu, “A review on distribution system state estimation,” IEEE Transactions on Power Systems, vol. 32, no. 5, pp. 3875–3883, 2017. [Online]. Available: https://doi.org/10.1109/TPWRS.2016.2632156
J. Wu and Y. Yu, “Cbr-based load estimation for distribution networks,” in MELECON 2006 - 2006 IEEE Mediterranean Electrotechnical Conference, 2006, pp. 952–955. [Online]. Available: https://doi.org/10.1109/MELCON.2006.1653256
A. Bragantini, A. Sumper, V. Gadelha, and S. Galceran-Arellano, “Neural network-based three-phase state estimation for unobservable low voltage grids,” International Journal of Electrical Power & Energy Systems, vol. 167, p. 110568, 2025. [Online]. Available: https://doi.org/10.1016/j.ijepes.2025.110568
M. Fotopoulou, S. Petridis, I. Karachalios, and D. Rakopoulos, “A review on distribution system state estimation algorithms,” Applied Sciences, vol. 12, no. 21, 2022. [Online]. Available: https://doi.org/10.3390/app122111073
E. Mulliez, M. A. Sanz-Bobi, A. Sanchez, P. Mazidi, A. Gonzalez, and R. Bachiller, “Life characterization of power distribution transformers using clustering techniques,” PHM Society European Conference, vol. 4, no. 1, 2018. [Online]. Available: https://doi.org/10.36001/phme.2018.v4i1.129
A. Sepp¨al¨a, Load research and load estimation in electricity distribution, 1996. [Online]. Available: https://publications.vtt.fi/pdf/publications/1996/P289.pdf
M. Mahmoodi, S. M. N. R.A., L. Blackhall, and P. Scott, “A comparison on power flow models for optimal power flow studies in integrated medium-low voltage unbalanced distribution systems,” Sustainable Energy, Grids and Networks, vol. 38, p. 101339, 2024. [Online]. Available: https://doi.org/10.1016/j.segan.2024.101339
B. Taheri, M. Sedighizadeh, M. R. Nasiri, and A. S. Fini, “A non-intrusive load monitoring algorithm based on real-time feature extraction and deep learning model,” International Journal of Electrical Power & Energy Systems, vol. 168, p. 110637, 2025. [Online]. Available: https://doi.org/10.1016/j.ijepes.2025.110637
W. Kersting and W. Phillips, “Load allocation based upon automatic meter readings,” in 2008 IEEE/PES Transmission and Distribution Conference and Exposition. IEEE, 2008, pp. 1–7. [Online]. Available: https://doi.org/10.1109/TDC.2008.4517057
EATON, “Cymdist – distribution system analysis software package,” 2024, accessed: January 2025. [Online]. Available: https://www.eaton.com/us/en-us/products/utility-grid-solutions/software-modules/distribution-system-analysis-package-CYMDIST.html
