Predicting the energy consumption in buildings using the optimized support vector regression model

Wei Cai, Xiaodong Wen, Chaoen Li, Jingjing Shao and Jianguo Xu

Energy, 2023, vol. 273, issue C

Abstract: One of the most significant axes of regional, national, and worldwide energy policy is energy efficiency in building design. In particular, the energy efficiency of HVAC systems is of paramount importance. They provide energy for both residential and commercial sectors. As a result, assessing building energy usage is a critical step in optimizing building energy consumption. The impact of eight input factors on the two output variables, heating and cooling loads, for residential structures was explored in this study. For this purpose, the SVR-supervised machine learning algorithm was used. Despite its advantages, such as predictive accuracy and robustness, this method suffers from the fact that there is no specific rule for fitting its parameters. Therefore, six meta-heuristic optimization algorithms were investigated, and the strongest algorithm for optimal parameter fitting for the SVR model was presented. The correlation and error parameters analysis showed that the hybrid model SVR-AEO has the best performance in simulating residential buildings' heating and cooling loads. According to the obtained results, the value of R2 for the cooling and heating loads prediction in the training data is obtained to be 0.9975 and 0.99955, respectively.

Keywords: Support vector machine; Support vector regression; Meta-heuristic algorithms; Building energy prediction; Heating load; Cooling load (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (8)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544223005820
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:273:y:2023:i:c:s0360544223005820

DOI: 10.1016/j.energy.2023.127188

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().