Scholarly record
DEVELOPMENT OF A METHOD FOR PREDICTING THE HEAT CONSUMPTION OF BUILDINGS WITH REGARD TO THEIR INDIVIDUAL CHARACTERISTICS BASED ON THE USE OF MLP BOOSTING AND LINEAR CLASSIFIERS
Abstract
The paper analyzes the existing data mining methods to predict heat consumption in buildings of public sector facilities. The highest accuracy in solving the regression problem was obtained using the ensembles of artificial neural network models of the MLP architecture, gradient boosting models on decision-making trees and linear regression models. The ensembles of the studied methods were used in the development of the analytical subsystem in the energy resource management system of the Belgorod Region. Approbation of the developed subsystem was performed using data from 2018 and 2019. High results of experiments on real data proved the adequacy of the proposed models.
Publication Impact Profile
Publication details
References10
Seber G., Linear regression analysis Mir, 1980, P. 456
Dubrov A.M., Mhitaryan V.S., Troshin L.I., Multidimensional statistical methods, Finance and Statistics, 2003.
Antokhin P. N., Comparison of various methods of statistical forecasting of daily dynamics of surface ozone concentration, Optics of the atmosphere and the ocean, 2013, vol. 26, issue 12, pp. 1082-1089.
Shreyas R., Predicting popularity of online articles using random forest regression, 2016 Second International Conference on Cognitive Computing and Information Processing (CCIP), 2016, pp. 1-5.
Tikhonov V. V., Vorobyov S. P., Chernova E. A. Using neural networks to predict enterprise development // Executive editor, 2017. pp. 44-49.
Abdurakhmanov A. M., Methods for predicting power consumption in distribution networks (review), Electrical Engineering: Network Electronic Scientific Journal, 2016, vol. 1, pp. 3-23.
Abramovich B. N., Barabanova I. S., System for predicting power consumption using an artificial neural network, Mining science and technology, 2016, vol. 2, pp. 66-77.
Ramachandran P., Zoph B., Le Q. V., Searching for activation functions, arXiv preprint arXiv:1710.05941, 2017.
Guolin Ke, Qi Meng, Thomas Finley, Taifeng Wang, Wei Chen, Weidong Ma, Qiwei Ye, Tie-Yan Liu. LightGBM: A Highly Efficient Gradient Boosting Decision Tree, Advances in Neural Information Processing Systems, 2017, vol. 30, pp. 3149-3157.
Chen T., Guestrin C., XGBoost: A Scalable Tree Boosting System, In 22nd SIGKDD Conference on Knowledge Discovery and Data Mining, 2016.
View or Download full articleAccess options
SWS access login
Login as SWS Scientific CommitteeLogin as SWS Scientific PartnerLogin as SWS AuthorAuthors and approved SWS contributors will read and export their own linked papers after identity matching by SWS profile, email and SGEM GlobalID.
For librarian assistance: [email protected]
Purchase Instant Access
- Article can be downloaded after successful payment.
- Article may be used according to SWS library access terms.
- Article cannot be redistributed.
