Investigation of building load prediction models based on integration of mechanism methods and data-driven models

Xiaojie Lin, Ning Zhang, Liuliu Du-Ikonen, Xiaolei Yuan and Wei Zhong

Energy, 2025, vol. 324, issue C

Abstract: In the district energy systems, the quality of data often proves inferior, resulting in that the historical building data may be partially or entirely absent. The traditional data-driven models may generate poor fitting results in such scenarios, while mechanism models typically involve a time-consuming simulation process, especially for large-space building load calculation. This paper proposes building load prediction models based on the integration of mechanism methods and data-driven models to deal with the problem for different building types and in different degrees of data quantity. The mechanism methods are performed based on the specific building in the case, and the base structure of data-driven models is not limited by this method. Two cases with different building types and load types are selected for the experiment. This paper investigates the building load prediction capabilities of different models, including different base structures and whether mechanism methods are integrated, in different training data sampling scenarios. Based on the experiment results, the proposed models achieve generalization and robustness in different cases and scenarios.

Keywords: Deep learning; Load prediction; Data-driven model; District energy systems (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544225015750
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:324:y:2025:i:c:s0360544225015750

DOI: 10.1016/j.energy.2025.135933

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 ().