Analysis of the Integration of Drift Detection Methods in Learning Algorithms for Electrical Consumption Forecasting in Smart Buildings

Sustainability

Luis Hernández-Callejo¹  Oscar Duque-Pérez²  Angel Zorita-Lamadrid²  Victor Alonso-Gómez³  Luis Gonzalez-Morales⁴  Hugo J. Bello⁵  Adalberto Ospino-Castro⁶  Alvaro Jaramillo-Duque⁷  Martín Solís⁸  Felix Santos García⁹  Deyslen Mariano-Hernández^1,10 

[1] ADIRE-ITAP, Departamento Ingeniería Agrícola y Forestal, Universidad de Valladolid, 42004 Soria, Spain;ADIRE-ITAP, Departamento de Ingeniería Eléctrica, Universidad de Valladolid, 47002 Valladolid, Spain;Departamento de Física, Universidad de Valladolid, 47011 Valladolid, Spain;Departamento de Ingeniería Eléctrica, Electrónica y Telecomunicaciones–DEET, Facultad de Ingeniería, Universidad de Cuenca, Cuenca 010107, Ecuador;Departamento de Matemática Aplicada, Universidad de Valladolid, 47002 Valladolid, Spain;Facultad de Ingeniería, Universidad de la Costa, Barranquilla 080002, Colombia;GIMEL, Departamento de Ingeniería Eléctrica, Universidad de Antioquia, Medellín 050010, Colombia;Tecnológico de Costa Rica, Cartago 30101, Costa Rica;Área de Ciencias Básicas y Ambientales, Instituto Tecnológico de Santo Domingo, Santo Domingo 10602, Dominican Republic;Área de Ingeniería, Instituto Tecnológico de Santo Domingo, Santo Domingo 10602, Dominican Republic;

关键词: drift detection;    electrical consumption forecasting;    energy forecasting;    machine learning;    smart buildings;   

DOI  :  10.3390/su14105857

来源: DOAJ

【 摘 要 】

Buildings are currently among the largest consumers of electrical energy with considerable increases in CO₂ emissions in recent years. Although there have been notable advances in energy efficiency, buildings still have great untapped savings potential. Within demand-side management, some tools have helped improve electricity consumption, such as energy forecast models. However, because most forecasting models are not focused on updating based on the changing nature of buildings, they do not help exploit the savings potential of buildings. Considering the aforementioned, the objective of this article is to analyze the integration of methods that can help forecasting models to better adapt to the changes that occur in the behavior of buildings, ensuring that these can be used as tools to enhance savings in buildings. For this study, active and passive change detection methods were considered to be integrators in the decision tree and deep learning models. The results show that constant retraining for the decision tree models, integrating change detection methods, helped them to better adapt to changes in the whole building’s electrical consumption. However, for deep learning models, this was not the case, as constant retraining with small volumes of data only worsened their performance. These results may lead to the option of using tree decision models in buildings where electricity consumption is constantly changing.

【 授权许可】

Unknown