@misc{10481/93404,
year = {2024},
month = {5},
url = {https://hdl.handle.net/10481/93404},
abstract = {Non-residential buildings are responsible for more than a third of global energy consumption. Estimating building energy
consumption is the first step towards identifying inefficiencies and optimizing energy management policies. This paper presents
a study of Deep Learning techniques for time series analysis applied to building energy prediction with real environments.
We collected multisource sensor data from an actual office building under normal operating conditions, pre-processed them,
and performed a comprehensive evaluation of the accuracy of feed-forward and recurrent neural networks to predict energy
consumption. The results show that memory-based architectures (LSTMs) perform better than stateless ones (MLPs) even
without data aggregation (CNNs), although the lack of ample usable data in this type of problem avoids making the most of
recent techniques such as sequence-to-sequence (Seq2Seq).},
organization = {European Union
NextGenerationEU/PRTRthrough the IA4TESproject (MIA.2021.M04.
0008)},
organization = {MICIU/AEI/10.13039/501100011033 through the SINERGYproject
(PID2021.125537NA.I00)},
organization = {byERDF/Junta deAndalucía
through the D3S project (P21.00247)},
organization = {FEDER programme 2014-
2020 (B-TIC-145-UGR18 and P18-RT-1765)},
organization = {European
Union (Energy IN TIME EeB.NMP.2013-4, No. 608981)},
organization = {Universidad de Granada/CBUA},
publisher = {Springer Nature},
keywords = {Buildings},
keywords = {Energy consumption forecasting},
keywords = {Time series},
title = {Deep learning for prediction of energy consumption: an applied use case in an office building},
doi = {10.1007/s10489-024-05451-9},
author = {Morcillo Jiménez, Roberto and Mesa, Jesús and Gómez-Romero, Juan and Vila, M. Amparo and Martín Bautista, María José},
}