• Yu. V. Parfenenko Sumy State University, Sumy, Ukraine, Ukraine
  • V. V. Shendryk Sumy State University, Sumy, Ukraine, Ukraine
  • O. S. Galichenko Sumy State University, Sumy, Ukraine, Ukraine



heat consumption, management, modeling, information systems, prediction, neural networks, energy saving.


The method of improvement the process of decision support to improve management of heat supplies’ modes through the development
of prediction heat consumption model of the social and public sector building is proposed. The object of the study is the process of choosing
the most optimal architecture of the neural network to solve goals of forecasting of heat consumption of the building of social and public
sector. The subject of the study is models prediction heat consumption of buildings of social and public sector using artificial neural networks.
The purpose of this study is to improve the forecasting reliability of heat energy demand of social and public sector buildings. Models of shortterm prediction of heat energy demand of social and public sector buildings using artificial neural networks that take into account the influence of weather conditions, fluctuations in demand for thermal energy depending on the type of day of the week and the previous values of heat energy demand are proposed. Models are based on such architectures of neural network’s as a nonlinear network input-output, a nonlinear autoregressive network, a nonlinear autoregressive neural network with external inputs. The proposed models were implemented in the programming environment Matlab, to train their the Levenberg-Marquard algorithm was used. Experiments on the study of the accuracy of the developed models, which showed that the greatest accuracy of prediction can be achieved by using a model type NARX, were carried. Using the proposed model within the information system «HeatCAM» for the prediction of building’s heat consumption during the heating season
can increase the effectiveness of decision support in the management of heating modes, which reduces heat consumption.


Системные исследования в энергетике: Ретроспектива научных направлений СЭИ-ИСЭМ / отв. ред. Н. И. Воропай. – Новосибирск : Наука, 2010. – 686 с. 2. Вороновский Г. К. Проблемы и перспективы использования искусственных нейронных сетей в энергетике. Ч. 2. Управление / Г. К. Вороновский, К. В. Махотило, С. А. Сергеев // Пробл. заг. енергетики. – 2007. – № 16. – С. 54–67. 3. Chramcov B. Heat Demand Forecasting for Concrete District Heating System / B. Chramcov, Ja. Baláte, M. Princ // International Journal of Mathematical Models and Methods in Applied Sciences. – 2010. – Vol. 4, No. 4. – P. 231–239. 4. Bacher P. Short-term heat load forecasting for single family houses / P. Bacher, H. Madsen // Energy and Buildings. – 2013. – Vol. 65. – P.101–112. 5. Kontu K. Forecasting district heating consumption based on customer measurements [Electronic resource] / K. Kontu // International association EUROHEAT & POWER. – Access mode: Kontu.pdf. 6. Grosswindhager S. Online Short-Term Forecast of System heat Load in District Heating Networks / S. Grosswindhager, A. Voigt, M. Kozek // ISF 2011 : the 31st International Symposium on Forecasting, Prague, Czech Republic, 26–29 June 2011 :conference proceedings. – Lancaster : International Institute of Forecasters, 2011. – P. 1–8. 7. Dostál P. Prediction of the heat supply daily diagram via artificial neural network / P. Dostál, B. Chramcov, Ja.Baláte // East West Fuzzy Colloquium : proceedings. Zittau, Germany, 2002. – P. 178–183. 8. Buhari M. Short-Term Load Forecasting Using Artificial Neural Network / M. Buhari, S. S. Adamu // IMECS 2012 : International MultiConference of Engineers and Computer Scientist, Hong Kong, 14–16 Mar 2012 : proceedings. – Hong Kong, 2012. 9. Grzenda M. Consumer-oriented heat consumption prediction / M. Grzenda // Control and Cybernetics. – 2012. – Vol. 41, No. 1. – P. 213–240.



How to Cite

Parfenenko, Y. V., Shendryk, V. V., & Galichenko, O. S. (2014). PREDICTION THE HEAT CONSUMPTION OF SOCIAL AND PUBLIC SECTOR BUILDINGS USING NEURAL NETWORKS. Radio Electronics, Computer Science, Control, (2).



Neuroinformatics and intelligent systems