Prediction of hourly heating energy use for HVAC using feedforward neural networks
2017
Аутори
Sretenović, AleksandraJovanović, Radiša
Novaković, Vojislav M.
Nord, Nataša M.
Živković, Branislav
Конференцијски прилог (Објављена верзија)
Метаподаци
Приказ свих података о документуАпстракт
In this paper, feedforward neural network, one of the most widely used
artificial intelligence methods, was proposed for the prediction of hourly
heating energy use of one university campus. Two different approaches were
presented: network that provides one output (heating energy use for selected
hour) and network with 24 outputs (daily profile of heating energy use). The
proposed models were trained and tested using real measured hourly energy
use and meteorological data. It has been shown that both models can be used
for the prediction with satisfying accuracy. This kind of prediction can be
used for calculating accurate energy bills, which is very useful considering
that the significant part of the campus is being leased. Estimating energy use
for different weather conditions can help in energy planning.
Кључне речи:
neural networks / building energy use predictionИзвор:
International Scientific Conference on Information Technology and Data Related Research, SINTEZA, 2017, 297-301Колекције
Институција/група
Mašinski fakultetTY - CONF AU - Sretenović, Aleksandra AU - Jovanović, Radiša AU - Novaković, Vojislav M. AU - Nord, Nataša M. AU - Živković, Branislav PY - 2017 UR - https://machinery.mas.bg.ac.rs/handle/123456789/4713 AB - In this paper, feedforward neural network, one of the most widely used artificial intelligence methods, was proposed for the prediction of hourly heating energy use of one university campus. Two different approaches were presented: network that provides one output (heating energy use for selected hour) and network with 24 outputs (daily profile of heating energy use). The proposed models were trained and tested using real measured hourly energy use and meteorological data. It has been shown that both models can be used for the prediction with satisfying accuracy. This kind of prediction can be used for calculating accurate energy bills, which is very useful considering that the significant part of the campus is being leased. Estimating energy use for different weather conditions can help in energy planning. C3 - International Scientific Conference on Information Technology and Data Related Research, SINTEZA T1 - Prediction of hourly heating energy use for HVAC using feedforward neural networks EP - 301 SP - 297 DO - 10.15308/Sinteza-2017-297-301 ER -
@conference{ author = "Sretenović, Aleksandra and Jovanović, Radiša and Novaković, Vojislav M. and Nord, Nataša M. and Živković, Branislav", year = "2017", abstract = "In this paper, feedforward neural network, one of the most widely used artificial intelligence methods, was proposed for the prediction of hourly heating energy use of one university campus. Two different approaches were presented: network that provides one output (heating energy use for selected hour) and network with 24 outputs (daily profile of heating energy use). The proposed models were trained and tested using real measured hourly energy use and meteorological data. It has been shown that both models can be used for the prediction with satisfying accuracy. This kind of prediction can be used for calculating accurate energy bills, which is very useful considering that the significant part of the campus is being leased. Estimating energy use for different weather conditions can help in energy planning.", journal = "International Scientific Conference on Information Technology and Data Related Research, SINTEZA", title = "Prediction of hourly heating energy use for HVAC using feedforward neural networks", pages = "301-297", doi = "10.15308/Sinteza-2017-297-301" }
Sretenović, A., Jovanović, R., Novaković, V. M., Nord, N. M.,& Živković, B.. (2017). Prediction of hourly heating energy use for HVAC using feedforward neural networks. in International Scientific Conference on Information Technology and Data Related Research, SINTEZA, 297-301. https://doi.org/10.15308/Sinteza-2017-297-301
Sretenović A, Jovanović R, Novaković VM, Nord NM, Živković B. Prediction of hourly heating energy use for HVAC using feedforward neural networks. in International Scientific Conference on Information Technology and Data Related Research, SINTEZA. 2017;:297-301. doi:10.15308/Sinteza-2017-297-301 .
Sretenović, Aleksandra, Jovanović, Radiša, Novaković, Vojislav M., Nord, Nataša M., Živković, Branislav, "Prediction of hourly heating energy use for HVAC using feedforward neural networks" in International Scientific Conference on Information Technology and Data Related Research, SINTEZA (2017):297-301, https://doi.org/10.15308/Sinteza-2017-297-301 . .