Mostrar el registro sencillo del documento

dc.rights.licenseReconocimiento 4.0 Internacional
dc.contributor.advisorVelásquez Henao, Juan David
dc.contributor.authorMontoya Cardona, José Fernando
dc.date.accessioned2021-10-16T16:07:01Z
dc.date.available2021-10-16T16:07:01Z
dc.date.issued2021-10
dc.identifier.urihttps://repositorio.unal.edu.co/handle/unal/80568
dc.descriptionilustraciones, diagramas, tablas
dc.description.abstractLa precisión del pronóstico de la demanda horaria de energía eléctrica es fundamental para realizar una planificación adecuada de los recursos de generación, ya que las desviaciones altas en el pronóstico generan sobrecostos en la operación del sistema. En este trabajo se propone una metodología novedosa de pronóstico basada en el agrupamiento de series de tiempo y los modelos ARIMA; específicamente, el modelo realiza el agrupamiento por tipos de día; seguidamente, agrega las series pertenecientes a un mismo grupo; luego, descompone las series agregadas usando descomposición aditiva; después, se pronostican las series con modelos ARIMA donde se utilizan como variables exógenas las componentes espectrales de Fourier para considerar la estacionalidad y finalmente, se combinan los pronósticos. El modelo propuesto fue utilizado para pronosticar la demanda horaria desde el 13 de enero de 2020 hasta el 15 de marzo de 2020. Los pronósticos del modelo propuesto fueron comparados con los pronósticos del modelo del Centro Nacional de Despacho (Colombia), encontrándose mejoras de hasta un 50% en la precisión con el modelo propuesto. (Texto tomado de la fuente)
dc.description.abstractThe forecasting accuracy of the hourly electricity demand is essential for planning the resources of generation, since high deviations in the forecast generate cost overruns in the system’s operation. In this research, a novel forecasting methodology based in clustering time series and ARIMA models is proposed; specifically, the model performs the clustering by types of day, then adds the time series belonging to the same cluster; it later decomposes the aggregate series using additive decomposition, then time series are forecasted with ARIMA models where the Fourier spectral components are used as exogenous variables to consider seasonality and finally, the results of the forecast are combined. The proposed model was used to forecast hourly demand from January 13, 2020 to March 15, 2020. The results of the proposed model were compared with the model of the National Dispatch Center (Colombia), getting improvements of up to 50% of accuracy with the proposed model.
dc.format.extentxiv, 39 páginas
dc.format.mimetypeapplication/pdf
dc.language.isospa
dc.publisherUniversidad Nacional de Colombia
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.subject.ddc330 - Economía::333 - Economía de la tierra y de la energía
dc.titlePronóstico de la demanda de energía en Colombia a corto plazo basado en un modelo híbrido adaptativo.
dc.typeTrabajo de grado - Maestría
dc.type.driverinfo:eu-repo/semantics/masterThesis
dc.type.versioninfo:eu-repo/semantics/acceptedVersion
dc.publisher.programMedellín - Minas - Maestría en Ingeniería - Analítica
dc.coverage.countryColombia
dc.description.degreelevelMaestría
dc.description.degreenameMagíster en Ingeniería - Analítica
dc.description.researchareaAnalítica de Mercados de Energía
dc.identifier.instnameUniversidad Nacional de Colombia
dc.identifier.reponameRepositorio Institucional Universidad Nacional de Colombia
dc.identifier.repourlhttps://repositorio.unal.edu.co/
dc.publisher.departmentDepartamento de la Computación y la Decisión
dc.publisher.facultyFacultad de Minas
dc.publisher.placeMedellín, Colombia
dc.publisher.branchUniversidad Nacional de Colombia - Sede Medellín
dc.relation.referencesP. Ledesma, «Regulacion de frecuencia y potencia», p. 33.
dc.relation.referencesT.-Y. Kim y S.-B. Cho, «Predicting residential energy consumption using CNN-LSTM neural networks», Energy, vol. 182, pp. 72-81, 2019, doi: 10.1016/j.energy.2019.05.230.
dc.relation.referencesJ. Davidvelásquez, C. J. Franco, y H. A. García, «A non-linear model for forecasting the monthly demand for electricity in Colombia», Estud. Gerenciales, vol. 25, n.o 112, pp. 37-54, 2009, doi: 10.1016/S0123-5923(09)70079-8.
dc.relation.referencesA. S. Ahmad et al., «A review on applications of ANN and SVM for building electrical energy consumption forecasting», Renew. Sustain. Energy Rev., vol. 33, pp. 102-109, 2014, doi: 10.1016/j.rser.2014.01.069.
dc.relation.referencesR. Y. M. Li, S. Fong, y K. W. S. Chong, «Forecasting the REITs and stock indices: Group method of data handling neural network approach», Pac. Rim Prop. Res. J., vol. 23, n.o 2, pp. 123-160, 2017, doi: 10.1080/14445921.2016.1225149.
dc.relation.references«Acuerdo 1303 Por el cual se actualizan los procedimientos para la gestión integral de la demanda C.N.O». https://www.cno.org.co/content/acuerdo-1303-por-el-cual-se-actualizan-los-procedimientos-para-la-gestion-integral-de-la (accedido jun. 29, 2020)
dc.relation.referencesE. G. Tajeuna, M. Bouguessa, y S. Wang, «A network-based approach to enhance electricity load forecasting», 2019, vol. 2018-November, pp. 266-275. doi: 10.1109/ICDMW.2018.00046.
dc.relation.referencesL. N. Ferreira y L. Zhao, «A time series clustering technique based on community detection in networks», 2015, vol. 53, n.o 1, pp. 183-190. doi: 10.1016/j.procs.2015.07.293.
dc.relation.referencesM. E. J. Newman, «Fast algorithm for detecting community structure in networks», Phys. Rev. E - Stat. Phys. Plasmas Fluids Relat. Interdiscip. Top., vol. 69, n.o 6, p. 5, 2004, doi: 10.1103/PhysRevE.69.066133.
dc.relation.referencesG. Le Ray y P. Pinson, «Online adaptive clustering algorithm for load profiling», Sustain. Energy Grids Netw., vol. 17, 2019, doi: 10.1016/j.segan.2018.100181.
dc.relation.referencesL. F. S. Vilela, R. C. Leme, C. A. M. Pinheiro, y O. A. S. Carpinteiro, «Forecasting financial series using clustering methods and support vector regression», Artif. Intell. Rev., vol. 52, n.o 2, pp. 743-773, 2019, doi: 10.1007/s10462-018-9663-x.
dc.relation.referencesK. Gajowniczek y T. Zabkowski, «Simulation Study on Clustering Approaches for Short-Term Electricity Forecasting», Complexity, vol. 2018, 2018, doi: 10.1155/2018/3683969.
dc.relation.referencesP. Laurinec, M. Lóderer, M. Lucká, y V. Rozinajová, «Density-based unsupervised ensemble learning methods for time series forecasting of aggregated or clustered electricity consumption», J. Intell. Inf. Syst., vol. 53, n.o 2, pp. 219-239, 2019, doi: 10.1007/s10844-019-00550-3.
dc.relation.referencesS. Sun, J. Fu, y A. Li, «A compound wind power forecasting strategy based on clustering, two-stage decomposition, parameter optimization, and optimal combination of multiple machine learning approaches», Energies, vol. 12, n.o 18, 2019, doi: 10.3390/en12183586.
dc.relation.referencesY. Wang, Y. Shen, S. Mao, X. Chen, y H. Zou, «LASSO and LSTM integrated temporal model for short term solar intensity forecasting», IEEE Internet Things J., vol. 6, n.o 2, pp. 2933-2944, 2019, doi: 10.1109/JIOT.2018.2877510.
dc.relation.referencesP. Laurinec y M. Lucká, «Interpretable multiple data streams clustering with clipped streams representation for the improvement of electricity consumption forecasting», Data Min. Knowl. Discov., vol. 33, n.o 2, pp. 413-445, 2019, doi: 10.1007/s10618-018-0598-2.
dc.relation.referencesR. Jain, N. Jain, Y. Gupta, T. Chugh, T. Chugh, y D. J. Hemanth, «A Modified Fuzzy Logic Relation Based Approach for Electricity Consumption Forecasting in India», Int. J. Fuzzy Syst., vol. 22, n.o 2, pp. 461-475, 2020, doi: 10.1007/s40815-019-00704-z.
dc.relation.referencesS. Singh y A. Yassine, «Big data mining of energy time series for behavioral analytics and energy consumption forecasting», Energies, vol. 11, n.o 2, 2018, doi: 10.3390/en11020452.
dc.relation.referencesE. Y. Shchetinin, «Cluster-based energy consumption forecasting in smart grids», Commun. Comput. Inf. Sci., vol. 919, pp. 445-456, 2018, doi: 10.1007/978-3-319-99447-5_38.
dc.relation.referencesB. Auder, J. Cugliari, Y. Goude, y J.-M. Poggi, «Scalable clustering of individual electrical curves for profiling and bottom-up forecasting», Energies, vol. 11, n.o 7, 2018, doi: 10.3390/en11071893.
dc.relation.referencesL. Li, K. Ota, y M. Dong, «Everything is image: CNN-based short-term electrical load forecasting for smart grid», 2017, vol. 2017-November, pp. 344-351. doi: 10.1109/ISPAN-FCST-ISCC.2017.78.
dc.relation.referencesM. Ghofrani, R. Azimi, F. M. Najafabadi, y N. Myers, «A new day-ahead hourly electricity price forecasting framework», presentado en 2017 North American Power Symposium, NAPS 2017, 2017. doi: 10.1109/NAPS.2017.8107269.
dc.relation.referencesJ. Yang et al., «k-Shape clustering algorithm for building energy usage patterns analysis and forecasting model accuracy improvement», Energy Build., vol. 146, pp. 27-37, 2017, doi: 10.1016/j.enbuild.2017.03.071.
dc.relation.referencesI. P. Panapakidis, «Clustering based day-ahead and hour-ahead bus load forecasting models», Int. J. Electr. Power Energy Syst., vol. 80, pp. 171-178, 2016, doi: 10.1016/j.ijepes.2016.01.035.
dc.relation.referencesX. Liang, T. Hong, y G. Q. Shen, «Occupancy data analytics and prediction: A case study», Build. Environ., vol. 102, pp. 179-192, 2016, doi: 10.1016/j.buildenv.2016.03.027.
dc.relation.referencesY. H. Chen, W.-C. Hong, W. Shen, y N. N. Huang, «Electric load forecasting based on a least squares support vector machine with fuzzy time series and global harmony search algorithm», Energies, vol. 9, n.o 2, pp. 1-13, 2016, doi: 10.3390/en9020070.
dc.relation.referencesM. Chaouch, «Clustering-based improvement of nonparametric functional time series forecasting: Application to intra-day household-level load curves», IEEE Trans. Smart Grid, vol. 5, n.o 1, pp. 411-419, 2014, doi: 10.1109/TSG.2013.2277171.
dc.relation.referencesH. Mori y M. Takahashi, «Application of preconditioned Generalized radial Basis Function Network to prediction of photovoltaic power generation», presentado en IEEE PES Innovative Smart Grid Technologies Conference Europe, 2012. doi: 10.1109/ISGTEurope.2012.6465877.
dc.relation.referencesB. Kitchenham, «Procedures for Performing Systematic Reviews», p. 33.
dc.relation.referencesM. R. Cogollo y J. D. Velásquez, «Methodological advances in artificial neural networks for time series forecasting», IEEE Lat. Am. Trans., vol. 12, n.o 4, pp. 764-771, 2014, doi: 10.1109/TLA.2014.6868881.
dc.relation.referencesW. Zhang, G. Mu, C. Song, G. Yan, y V. Heuveline, «Extraction of Spatialoral Features of Bus Loads in Electric Grids Through Clustering in a Dynamic Model Space», IEEE Access, vol. 8, pp. 5852-5861, 2020, doi: 10.1109/ACCESS.2019.2963071.
dc.relation.referencesW. Sun y C. Zhang, «A hybrid BA-ELM model based on factor analysis and similar-day approach for short-term load forecasting», Energies, vol. 11, n.o 5, 2018, doi: 10.3390/en11051282.
dc.relation.referencesM. Ghofrani, D. Carson, y M. Ghayekhloo, «Hybrid clustering-time series-Bayesian neural network short term load forecasting method», presentado en NAPS 2016 - 48th North American Power Symposium, Proceedings, 2016. doi: 10.1109/NAPS.2016.7747865.
dc.relation.referencesP. Laurinec, M. Loderer, P. Vrablecova, M. Lucka, V. Rozinajova, y A. B. Ezzeddine, «Adaptive Time Series Forecasting of Energy Consumption Using Optimized Cluster Analysis», 2016, vol. 0, pp. 398-405. doi: 10.1109/ICDMW.2016.0063.
dc.relation.referencesA. Laouafi, M. Mordjaoui, F. Laouafi, y T. E. Boukelia, «Daily peak electricity demand forecasting based on an adaptive hybrid two-stage methodology», Int. J. Electr. Power Energy Syst., vol. 77, pp. 136-144, 2016, doi: 10.1016/j.ijepes.2015.11.046.
dc.relation.referencesR. L. Talavera-Llames, R. Pérez-Chacón, M. Martínez-Ballesteros, A. Troncoso, y F. Martínez-Álvarez, «A nearest neighbours-based algorithm for big time series data forecasting», Lect. Notes Comput. Sci. Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinforma., vol. 9648, pp. 174-185, 2016, doi: 10.1007/978-3-319-32034-2_15.
dc.relation.referencesW. Wang, W. Pedrycz, y X. Liu, «Time series long-term forecasting model based on information granules and fuzzy clustering», Eng. Appl. Artif. Intell., vol. 41, pp. 17-24, 2015, doi: 10.1016/j.engappai.2015.01.006.
dc.relation.referencesC. J. Bennett, R. A. Stewart, y J. W. Lu, «Forecasting low voltage distribution network demand profiles using a pattern recognition based expert system», Energy, vol. 67, pp. 200-212, 2014, doi: 10.1016/j.energy.2014.01.032.
dc.relation.referencesS. Humeau, T. K. Wijaya, M. Vasirani, y K. Aberer, «Electricity load forecasting for residential customers: Exploiting aggregation and correlation between households», presentado en 2013 Sustainable Internet and ICT for Sustainability, SustainIT 2013, 2013. doi: 10.1109/SustainIT.2013.6685208.
dc.relation.referencesZ. Ding, P. Yang, X. Yang, y Z. Zhang, «Wind power prediction method based on sequential time clustering support vector machine», Dianli Xitong ZidonghuaAutomation Electr. Power Syst., vol. 36, n.o 14, pp. 131-135+149, 2012, doi: 10.3969/j.issn.1000-1026.2012.14.025.
dc.relation.references«Complex Fourier Series». 2020. [En línea]. Disponible en: https://eng.libretexts.org/Bookshelves/Electrical_Engineering/Book%3A_Electrical_Engineering_(Johnson)/04%3A_Frequency_Domain/4.02%3A_Complex_Fourier_Series
dc.relation.referencesM. Charrad, N. Ghazzali, V. Boiteau, y A. Niknafs, «NbClust: An R Package for Determining the Relevant Number of Clusters in a Data Set», J. Stat. Softw., vol. 61, n.o 1, Art. n.o 1, nov. 2014, doi: 10.18637/jss.v061.i06.
dc.relation.references«pmdarima», API Reference. https://alkaline-ml.com/pmdarima/modules/classes.html
dc.rights.accessrightsinfo:eu-repo/semantics/openAccess
dc.subject.lembDemanda de energía eléctrica
dc.subject.lembEnergy consumption
dc.subject.lembConsumo de energía
dc.subject.proposalFourier
dc.subject.proposalDemanda de energía
dc.subject.proposalPrecisión del pronóstico
dc.subject.proposalAgrupamiento
dc.subject.proposalSeries de tiempo
dc.subject.proposalK-means
dc.subject.proposalARIMA
dc.subject.proposalFourier
dc.subject.proposalEdemand
dc.subject.proposalForecasting Accuracy
dc.subject.proposalClustering
dc.subject.proposalTime series
dc.subject.proposalEnergy demand
dc.title.translatedThe forecasting energy demand in Colombia in the short term based on an adaptive hybrid model.
dc.type.coarhttp://purl.org/coar/resource_type/c_bdcc
dc.type.coarversionhttp://purl.org/coar/version/c_ab4af688f83e57aa
dc.type.contentText
dc.type.redcolhttp://purl.org/redcol/resource_type/TM
oaire.accessrightshttp://purl.org/coar/access_right/c_abf2
dcterms.audience.professionaldevelopmentInvestigadores


Archivos en el documento

Thumbnail

Este documento aparece en la(s) siguiente(s) colección(ones)

Mostrar el registro sencillo del documento

Reconocimiento 4.0 InternacionalEsta obra está bajo licencia internacional Creative Commons Reconocimiento-NoComercial 4.0.Este documento ha sido depositado por parte de el(los) autor(es) bajo la siguiente constancia de depósito