Metodología de Muestreo basada en Entropía de Shannon para la Caracterización de la Demanda Energética en Regiones de Colombia
Sampling Methodology Based on Shannon Entropy for the Characterization of Energy Demand in Regions of Colombia
DOI:
https://doi.org/10.15446/sicel.v12.121277Palabras clave:
planificación energética, Entropía de Shannon, caracterización de la demanda, muestreo energético (es)Shannon Entropy, demand characterization, energy sampling, energy planning (en)
Descargas
Este trabajo propone una metodología de muestreo basada en entropía de Shannon para caracterizar eficientemente la demanda energética en regiones de Colombia con infraestructura eléctrica de desarrollo moderado. Ante la ausencia de tecnologías de medición avanzada (AMI) y el uso de registros convencionales de consumo, se diseña una estrategia de muestreo que permita representar adecuadamente la demanda sin requerir sensores automatizados ni monitoreo complejo. El objetivo es obtener una muestra de usuarios que represente adecuadamente a la población para facilitar la planificación y gestión del sistema eléctrico. Se utilizaron datos históricos de consumo mensual, ubicación geográfica, tipo de cliente y población. Se dividió el estudio en cuatro subregiones (Norte, Sur, Centro y Oriente) y se establecieron criterios para el tamaño de muestra, el balance entre categorías de usuarios y la selección representativa de usuarios. La muestra fue evaluada según su eficiencia logística y representatividad frente a las distribuciones originales de consumo y categorías auxiliares. Se concluye que el enfoque por entropía genera una muestra densa y estratégicamente ubicada, con excelente ajuste a la distribución de consumo, siendo una opción robusta y práctica para el operador de red.
This work proposes a sampling methodology based on Shannon entropy to efficiently characterize energy demand in regions of Colombia with moderately developed electricity infrastructure. In the absence of advanced metering technologies (AMI) and the use of conventional consumption records, a sampling strategy is designed to adequately represent the demand without requiring automated sensors or complex monitoring. The objective is to obtain a sample of users that adequately represents the population to facilitate planning and management of the electricity system. Historical data on monthly consumption, geographic location, customer type and population were used. The study was divided into four sub-regions (North, South, Central and East) and criteria were established for the sample size, the balance between user categories and the representative selection of users. The sample was evaluated according to its logistic efficiency and representativeness vis-à-vis the original distributions of consumption and auxiliary categories. It is concluded that the entropy approach generates a dense and strategically located sample, with excellent fit to the consumption distribution, being a robust and practical option for the network operator.
Descargas
Citas
REFERENCIAS
[1] Carrillo Romero y A. Perdomo Arias, Caracterización y análisis del consumo energético en zonas rurales para los municipios de Arauca. Villavicencio, Colombia: Universidad de los Llanos, 2017.
[2] H. Cardot and A. De Moliner, "Conditional bias robust estimation of the total of curve data by sampling in a finite population: an illustration on electricity load curves," arXiv preprint arXiv:1806.09949, 2018.
[3] H. Imberg, X. Yang, C. Flannagan, y J. Bärgman, "Active Sampling: A Machine-Learning-Assisted Framework for Finite Population Inference with Optimal Subsamples", Technometrics, vol. 67, no. 1, pp. 46–57, sep. 2024. doi: 10.1080/00401706.2024.2374554.
[4] J. Reina García y M. Peña Varón, Diseño metodológico para la selección de sitios de muestreo en una red de monitoreo de micro–contaminantes en ríos de Valle: caso de estudio río Cauca, 2019.
[5] A. Viloria Rodriguez, Comparación de metodologías utilizadas para abordar el problema de datos faltantes en estudios longitudinales. Medellín, Colombia: Universidad Nacional de Colombia, 2024.
[6] H. Rafiee, A. A. Abin and S. S. Majd, "Cluster Sampling: A Cluster-Driven Sampling Strategy for Deep Metric Learning," 2024 14th International Conference on Computer and Knowledge Engineering (ICCKE), Mashhad, Iran, Islamic Republic of, 2024, pp. 460-465, doi: 10.1109/ICCKE65377.2024.10874603.
[7] S. Chen, J. Zheng and J. Li, "The Impact of Sample Size after Sampling on the Accuracy of Machine Learning Models," 2024 International Conference on Computers, Information Processing and Advanced Education (CIPAE), Ottawa, ON, Canada, 2024, pp. 61-66, doi: 10.1109/CIPAE64326.2024.00017.
[8] Q. -X. Zhu, Q. -Q. Zhao, Y. Xu and Y. -L. He, "Novel virtual sample generation using Gibbs Sampling integrated with GRNN for handling small data in soft sensing," 2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS), Xiangtan, China, 2023, pp. 89-94, doi: 10.1109/DDCLS58216.2023.10166679.
[9] L. G. Swan and V. I. Ugursal, “Modeling of end-use energy consumption in the residential sector: A review of modeling techniques,” Oct. 2009. doi: 10.1016/j.rser.2008.09.033.
[10] I. Gustavo et al., “Metodología para la Determinación de Curvas de Carga y Consumo Eléctrico Residencial por Uso.”
[11] İ. Y. Yarbaşı and A. K. Çelik, “The determinants of household electricity demand in Turkey: An implementation of the Heckman Sample Selection model,” Energy, vol. 283, Nov. 2023, doi: 10.1016/j.energy.2023.128431.
[12] L. Provincia, S. Elena, C. Pavón, and J. Barzola-Monteses, “Estimación de la demanda energética mensual mediante encuesta aplicada en.” [Online].Available: https://www.researchgate.net/publication/309286132
[13] X. Zhu and B. Mather, "Data-Driven Load Diversity and Variability Modeling for Quasi-Static Time-Series Simulation on Distribution Feeders," 2019 IEEE Power & Energy Society General Meeting (PESGM), Atlanta, GA, USA, 2019, pp. 1-5, doi: 10.1109/PESGM40551.2019.8973929.
[14] X. Sun, X. Li, S. Datta, X. Ke, Q. Huang, R. Huang y Z. J. Hou, "Smart Sampling for Reduced and Representative Power System Scenario Selection", IEEE Open Access Journal of Power and Energy, vol. 8, pp. 241–251, 2021. doi: 10.1109/OAJPE.2021.3093278.
[15] Asghari, P., Zakariazadeh, A., Siano, P.: Selecting and prioritizing the electricity customers for participating in demand response programs. IET Gener. Transm. Distrib. 16, 2086–2096, 2022. doi: 10.1049/gtd2.12417
[16] J. Henze, S. Kutzner y B. Sick, "Sampling Strategies for Representative Time Series in Load Flow Calculations", en Data Analytics for Renewable Energy Integration. Technologies, Systems and Society (DARE 2018), W. Woon, Z. Aung, A. C. Feliú y S. Madnick, Eds., vol. 11325, Lecture Notes in Computer Science. Cham: Springer, 2018, pp. 33–47. doi: 10.1007/978-3-030-04303-2_3.
[17] X. Zhu and B. Mather, "Data-Driven Distribution System Load Modeling for Quasi-Static Time-Series Simulation," in IEEE Transactions on Smart Grid, vol. 11, no. 2, pp. 1556-1565, March 2020, doi: 10.1109/TSG.2019.2940084.
[18] C. E. Shannon, "A Mathematical Theory of Communication," The Bell System Technical Journal, vol. 27, pp. 379–423, 623–656, Jul., Oct. 1948.
[19] S. K. Ahmed, "How to choose a sampling technique and determine sample size for research: A simplified guide for researchers", Oral Oncology Reports, vol. 12, p. 100662, 2024. doi: 10.1016/j.oor.2024.100662.
[20] R. Wang, X. Zhao, H. Qiu, X. Cheng, y X. Liu, “Uncovering urban water consumption patterns through time series clustering and entropy analysis,” Water Research, vol. 262, p. 122085, 2024. doi: 10.1016/j.watres.2024.122085
[21] T. J. Stohlgren, S. Kumar, D. T. Barnett, y P. H. Evangelista, “Using Maximum Entropy Modeling for Optimal Selection of Sampling Sites for Monitoring Networks,” Diversity, vol. 3, no. 2, pp. 252–261, 2011. doi: 10.3390/d3020252
[22] P. Ngae, H. Kouichi, P. Kumar, A.-A. Feiz, y A. Chpoun, “Optimization of an urban monitoring network for emergency response applications: An approach for characterizing the source of hazardous releases”, Quarterly Journal of the Royal Meteorological Society, 16 ene. 2019. doi: 10.1002/qj.3471
[23] M. S. Khorshidi, M. R. Nikoo y M. Sadegh, "Optimal and objective placement of sensors in water distribution systems using information theory", Water Research, vol. 143, pp. 218–228, 2018. doi: 10.1016/j.watres.2018.06.050
[24] H. Tian, Z. Lang, C. Cao y B. Wang, "Optimizing Sensor Placement for Enhanced Source Term Estimation in Chemical Plants", Processes, vol. 13, no. 3, art. 825, 2025. doi: 10.3390/pr13030825.
[25] D. Hock, M. Kappes y B. Ghita, "Entropy-Based Metrics for Occupancy Detection Using Energy Demand", Entropy, vol. 22, no. 7, art. 731, 2020. doi: 10.3390/e2207073
Licencia
Derechos de autor 2026 Oscar Bustos, Julián David Osorio, Javier Rosero-Garcia, CRISTIAN CAMILO MARIN CANO, Luis Alirio Bolaños, Pedro Jose Toro, Óscar Andrés Ocampo

Esta obra está bajo una licencia internacional Creative Commons Atribución 4.0.