Sizing and Characterization of Load Curves of Distribution Transformers Using Clustering and Predictive Machine Learning Models

The efficient sizing and characterization of the load curves of distribution transformers are crucial challenges for electric utilities, especially given the increasing variability of demand, driven by emerging loads such as electric vehicles. This study applies clustering techniques and predictive...

Full description

Saved in:
Bibliographic Details
Main Authors: Pedro Torres-Bermeo, Kevin López-Eugenio, Carolina Del-Valle-Soto, Guillermo Palacios-Navarro, José Varela-Aldás
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Energies
Subjects:
Online Access:https://www.mdpi.com/1996-1073/18/7/1832
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850212279717462016
author Pedro Torres-Bermeo
Kevin López-Eugenio
Carolina Del-Valle-Soto
Guillermo Palacios-Navarro
José Varela-Aldás
author_facet Pedro Torres-Bermeo
Kevin López-Eugenio
Carolina Del-Valle-Soto
Guillermo Palacios-Navarro
José Varela-Aldás
author_sort Pedro Torres-Bermeo
collection DOAJ
description The efficient sizing and characterization of the load curves of distribution transformers are crucial challenges for electric utilities, especially given the increasing variability of demand, driven by emerging loads such as electric vehicles. This study applies clustering techniques and predictive models to analyze and predict the behavior of transformer demand, optimize utilization factors, and improve infrastructure planning. Three clustering algorithms were evaluated, K-shape, DBSCAN, and DTW with K-means, to determine which one best characterizes the load curves of transformers. The results show that DTW with K-means provides the best segmentation, with a cross-correlation similarity of 0.9552 and a temporal consistency index of 0.9642. For predictive modeling, supervised algorithms were tested, where Random Forest achieved the highest accuracy in predicting the corresponding load curve type for each transformer (0.78), and the SVR model provided the best performance in predicting the maximum load, explaining 90% of the load variability (R<sup>2</sup> = 0.90). The models were applied to 16,696 transformers in the Ecuadorian electrical sector, validating the load prediction with an accuracy of 98.55%. Additionally, the optimized assignment of the transformers’ nominal power reduced installed capacity by 39.27%, increasing the transformers’ utilization factor from 31.79% to 52.35%. These findings highlight the value of data-driven approaches for optimizing electrical distribution systems.
format Article
id doaj-art-022db72b51af46de9b5d42541f58affd
institution OA Journals
issn 1996-1073
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Energies
spelling doaj-art-022db72b51af46de9b5d42541f58affd2025-08-20T02:09:22ZengMDPI AGEnergies1996-10732025-04-01187183210.3390/en18071832Sizing and Characterization of Load Curves of Distribution Transformers Using Clustering and Predictive Machine Learning ModelsPedro Torres-Bermeo0Kevin López-Eugenio1Carolina Del-Valle-Soto2Guillermo Palacios-Navarro3José Varela-Aldás4Centro de Investigación MIST, Facultad de Ingenierías, Maestría en Big Data y Ciencia de Datos, Universidad Tecnológica Indoamérica, Ambato 180103, EcuadorCentro de Investigación MIST, Facultad de Ingenierías, Maestría en Big Data y Ciencia de Datos, Universidad Tecnológica Indoamérica, Ambato 180103, EcuadorFacultad de Ingeniería, Universidad Panamericana, Álvaro del Portillo 49, Zapopan 45010, MexicoDepartment of Electronic Engineering and Communications, University of Zaragoza, 44003 Teruel, SpainCentro de Investigación MIST, Facultad de Ingenierías, Maestría en Big Data y Ciencia de Datos, Universidad Tecnológica Indoamérica, Ambato 180103, EcuadorThe efficient sizing and characterization of the load curves of distribution transformers are crucial challenges for electric utilities, especially given the increasing variability of demand, driven by emerging loads such as electric vehicles. This study applies clustering techniques and predictive models to analyze and predict the behavior of transformer demand, optimize utilization factors, and improve infrastructure planning. Three clustering algorithms were evaluated, K-shape, DBSCAN, and DTW with K-means, to determine which one best characterizes the load curves of transformers. The results show that DTW with K-means provides the best segmentation, with a cross-correlation similarity of 0.9552 and a temporal consistency index of 0.9642. For predictive modeling, supervised algorithms were tested, where Random Forest achieved the highest accuracy in predicting the corresponding load curve type for each transformer (0.78), and the SVR model provided the best performance in predicting the maximum load, explaining 90% of the load variability (R<sup>2</sup> = 0.90). The models were applied to 16,696 transformers in the Ecuadorian electrical sector, validating the load prediction with an accuracy of 98.55%. Additionally, the optimized assignment of the transformers’ nominal power reduced installed capacity by 39.27%, increasing the transformers’ utilization factor from 31.79% to 52.35%. These findings highlight the value of data-driven approaches for optimizing electrical distribution systems.https://www.mdpi.com/1996-1073/18/7/1832machine learningclusteringtransformer load characterizationloadabilitypredictive modelingDTW with K-means
spellingShingle Pedro Torres-Bermeo
Kevin López-Eugenio
Carolina Del-Valle-Soto
Guillermo Palacios-Navarro
José Varela-Aldás
Sizing and Characterization of Load Curves of Distribution Transformers Using Clustering and Predictive Machine Learning Models
Energies
machine learning
clustering
transformer load characterization
loadability
predictive modeling
DTW with K-means
title Sizing and Characterization of Load Curves of Distribution Transformers Using Clustering and Predictive Machine Learning Models
title_full Sizing and Characterization of Load Curves of Distribution Transformers Using Clustering and Predictive Machine Learning Models
title_fullStr Sizing and Characterization of Load Curves of Distribution Transformers Using Clustering and Predictive Machine Learning Models
title_full_unstemmed Sizing and Characterization of Load Curves of Distribution Transformers Using Clustering and Predictive Machine Learning Models
title_short Sizing and Characterization of Load Curves of Distribution Transformers Using Clustering and Predictive Machine Learning Models
title_sort sizing and characterization of load curves of distribution transformers using clustering and predictive machine learning models
topic machine learning
clustering
transformer load characterization
loadability
predictive modeling
DTW with K-means
url https://www.mdpi.com/1996-1073/18/7/1832
work_keys_str_mv AT pedrotorresbermeo sizingandcharacterizationofloadcurvesofdistributiontransformersusingclusteringandpredictivemachinelearningmodels
AT kevinlopezeugenio sizingandcharacterizationofloadcurvesofdistributiontransformersusingclusteringandpredictivemachinelearningmodels
AT carolinadelvallesoto sizingandcharacterizationofloadcurvesofdistributiontransformersusingclusteringandpredictivemachinelearningmodels
AT guillermopalaciosnavarro sizingandcharacterizationofloadcurvesofdistributiontransformersusingclusteringandpredictivemachinelearningmodels
AT josevarelaaldas sizingandcharacterizationofloadcurvesofdistributiontransformersusingclusteringandpredictivemachinelearningmodels