Colombia
Bello Acosta, José Antonio
2021-10-08T20:03:05Z
2021-10-08T20:03:05Z
2021
http://hdl.handle.net/20.500.12010/22157
Electric energy consumption forecasting is a relevant issue to design and implement public policies related to energy generation and distribution at urban scale. This problem has been addressed from different standpoints, including traditional time series analysis and machine learning techniques, focused on the prediction of future energy demand according to metered data. This work proposes a hybrid model, based on computer simulation results from the City Energy Analyst toolbox and metered environmental and energy consumption data from a representative set of buildings located in Singapore. Such model is intended to provide reliable energy demand forecasts by using regression models. Three different regression methods were evaluated: a traditional SARIMA model and both NARX and Recurrent neural network architectures. The results obtained using this approach point out that RNN models provide accurate forecasts for 1 and 24 hours, outperforming NARX based models, while the SARIMA is, in general, unable to represent the electrical demand time series patterns.
72 páginas
application/pdf
eng
Universidad de Bogotá Jorge Tadeo Lozano
reponame:Expeditio Repositorio Institucional UJTL
instname:Universidad de Bogotá Jorge Tadeo Lozano
Inteligencia artificial
Neural network model for building electric energy consumption forecasting in densely populated tropical areas
Trabajo de grado de maestría
Inteligencia artificial -- Tesis y disertaciones académicas
Inteligencia computacional -- Tesis y disertaciones académicas
Sistemas adaptativos -- Tesis y disertaciones académicas
info:eu-repo/semantics/openAccess
info:eu-repo/semantics/acceptedVersion
Abierto (Texto Completo)
Magíster en Modelado y Simulación
Maestría en Modelado y Simulación MM&S
Martın Abadi et al. “TensorFlow: A system for large-scale machine learning”. en. In: (2016), p. 21.
YuvrajAgarwal et al. “Occupancy-driven energy management for smart building automation”. In: Proceedings of the 2nd ACM Workshop on Embedded Sensing Systems for Energy-Efficiency in Building. BuildSys ’10. Zurich, Switzerland: Association for Computing Machinery, Nov. 2010, pp. 1–6. isbn: 978-14503-0458-0. doi: 10.1145/1878431.1878433. url: https://doi.org/10. 1145/1878431.1878433 (visited on 07/22/2020).
Muhammad Waseem Ahmad, Monjur Mourshed, and Yacine Rezgui. “Trees vs Neurons: Comparison between random forest and ANN for high-resolution prediction of building energy consumption”. en. In: Energy and Buildings 147 (July 2017), pp. 77–89. issn: 0378-7788. doi: 10.1016/j.enbuild.2017. 04.038. url: http://www.sciencedirect.com/science/article/pii/ S0378778816313937 (visited on 12/08/2020).
Muhammad Waseem Ahmad et al. “Random Forests and Artificial Neural Network for Predicting Daylight Illuminance and Energy Consumption”. en. In: (2017), p. 7.
HAkbari, MPomerantz, andHTaha.“Coolsurfaces and shade trees to reduce energy use and improve air quality in urban areas”. en. In: Solar Energy. Urban Environment 70.3 (Jan. 2001), pp. 295–310. issn: 0038-092X. doi: 10.1016/S0038-092X(00)00089-X. url: http://www.sciencedirect.com/ science/article/pii/S0038092X0000089X (visited on 06/02/2020).
K. P. Amber, M. W. Aslam, and S. K. Hussain. “Electricity consumption forecasting models for administration buildings of the UK higher education sector”. en. In: Energy and Buildings 90 (Mar. 2015), pp. 127–136. issn: 0378-7788. doi: 10.1016/j.enbuild.2015.01.008. url: http://www. sciencedirect.com/science/article/pii/S0378778815000110 (visited on 07/22/2020).
K. P. Amber et al. “Intelligent techniques for forecasting electricity consumption of buildings”. en. In: Energy 157 (Aug. 2018), pp. 886–893. issn: 0360-5442. doi: 10.1016/j.energy.2018.05.155. url: http://www. sciencedirect.com/science/article/pii/S036054421830999X (visited on 06/02/2020).
A. Angelova, A. Krizhevsky, and V. Vanhoucke. “Pedestrian detection with a Large-Field-Of-View deep network”. In: 2015 IEEE International Conference on Robotics and Automation (ICRA). May 2015, pp. 704–711. doi: 10.1109/ ICRA.2015.7139256
Maher Azaza, Douglas Eriksson, and Fredrik Wallin. “A study on the viability of an on-site combined heat-and power supply system with and without electricity storage for office building”. en. In: Energy Conversion and Management 213 (June 2020), p. 112807. issn: 0196-8904. doi: 10.1016/j. enconman.2020.112807. url: http://www.sciencedirect.com/science/ article/pii/S0196890420303459 (visited on 06/02/2020).
Jean-Marie Bahu et al. “Towards a 3D Spatial Urban Energy Modelling Approach”. en. In: International Journal of 3-D Information Modeling (IJ3DIM) (July 2014). ISSN: 2156-1710 Issue: 3 Journal Abbreviation: IJ3DIM Pages: 1-16 Publisher: IGI Global Volume: 3. doi: 10.4018/ij3dim.2014070101. url: www.igi-global.com/article/towards-a-3d-spatial-urbanenergy-modelling-approach/122864 (visited on 12/09/2020).
WorldBank. Electric power consumption (kWh per capita) | Data. url: https: //data.worldbank.org/indicator/EG.USE.ELEC.KH.PC?end=2018& start=1960&view=chart (visited on 01/15/2020).
S. Bhavya and Anitha S. Pillai. “Prediction Models in Healthcare Using Deep Learning”. en. In: Proceedings of the 11th International Conference on Soft Computing and Pattern Recognition (SoCPaR 2019). Ed. by Ajith Abraham et al. Advances in Intelligent Systems and Computing. Cham: Springer International Publishing, 2021, pp. 195–204. isbn: 978-3-030-49345-5. doi: 10. 1007/978-3-030-49345-5_21.
Mathieu Bourdeau et al. “Modeling and forecasting building energy consumption: A review of data-driven techniques”. en. In: Sustainable Cities and Society 48 (July 2019), p. 101533. issn: 2210-6707. doi: 10.1016/j.scs.2019. 101533. url: http://www.sciencedirect.com/science/article/pii/ S2210670718323862 (visited on 06/02/2020).
G. E. P. Box and D. R. Cox. “An Analysis of Transformations”. In: Journal of the Royal Statistical Society. Series B (Methodological) 26.2 (1964). Publisher: [Royal Statistical Society, Wiley], pp. 211–252. issn: 0035-9246. url: https: //www.jstor.org/stable/2984418 (visited on 12/10/2020).
George EdwardPelhamBoxandGwilymJenkins. Time Series Analysis, Forecasting and Control. Holden-Day, Incorporated, 1976. isbn: 978-0-8162-11043.
BP. Statistical Review of World Energy | Energy economics | Home. en. url: https://www.bp.com/en/global/corporate/energy-economics/ statistical-review-of-world-energy.html (visited on 06/02/2020).
James E. Braun and Nitin Chaturvedi. “An Inverse Gray-Box Model for Transient Building Load Prediction”. In: HVAC&R Research 8.1 (Jan. 2002). Publisher: Taylor & Francis eprint: https://www.tandfonline.com/doi/pdf/10.1080/10789669.2002.10391290, pp. 73–99. issn: 1078-9669. doi: 10.1080/10789669.2002.10391290. url: https://www.tandfonline.com/doi/abs/10.1080/10789669.2002.
Lars Buitinck et al. “API design for machine learning software: experiences from the scikit-learn project”. In: arXiv:1309.0238 [cs] (Sept. 2013). arXiv: 1309.0238. url: http://arxiv.org/abs/1309.0238 (visited on 02/01/2019).
Cao, Xiaodong, Xilei, Dai, and Liu, Junjie. “Building energy-consumption status worldwide and the state-of-the-art technologies for zero-energy buildings during the past decade”. en. In: Energy and Buildings 128 (Sept. 2016), pp. 198–213. issn: 0378-7788. doi: 10.1016/j.enbuild.2016.06.089. url: https://ezproxy.ucentral.edu.co:2154/science/article/pii/ S0378778816305783 (visited on 01/15/2020).
KyunghyunChoetal. “Learning Phrase Representations using RNN EncoderDecoder for Statistical Machine Translation”. In: arXiv:1406.1078 [cs, stat] (Sept. 2014). arXiv: 1406.1078. url: http://arxiv.org/abs/1406.1078 (visited on 11/20/2020).
#SistemasDeEnergía
1 recurso en línea (archivo
de texto)
El pronóstico del consumo de energía eléctrica es un asunto de relevancia para diseñar e implementar políticas públicas relacionadas con la generación y distribución de energía a escala urbana. Este problema ha sido abordado desde distintos frentes, incluyendo el análisis estadístico de series temporales y los modelos basados en aprendizaje de máquina, enfocándose en la predicción de la futura demanda de energía según datos tomados mediante medición directa. Este trabajo propone un modelo híbrido, que combina resultados de simulaciones obtenidas mediante el City Energy Analyst (CEA) y variables ambientales medidas junto con los datos de consumo de energía para un conjunto representativo de edificios localizados en la ciudad de Singapur. Este propuesta pretende obtener pronósticos fiables de consumo de electricidad usando modelos de regresión. Los resultados obtenidos usando esta aproximación, muestran que la configuración de redes neuronales recurrentes (RNN) proveen un pronóstico acertado para distintas ventanas de tiempo, mejorando los resultados obtenidos al usar un modelo no lineal autoregresivo con variables exógenas (NARX), mientras que los modelos autoregresivos integrados de media móvil no son capaces de representar apropiadamente los patrones de la serie de tiempo de la demanda de electricidad.
Facultad de Ciencias Naturales e Ingeniería
info:eu-repo/semantics/masterThesis
http://purl.org/coar/resource_type/c_bdcc