Research on thermal load prediction of district heating station based on transfer learning

Chendong Wang, Jianjuan Yuan, Ke Huang, Ji Zhang, Lihong Zheng, Zhihua Zhou and Yufeng Zhang

Energy, 2022, vol. 239, issue PE

Abstract: Precise prediction of thermal load plays a critical role in China to fulfill the demand of energy saving, carbon emission reduction and environmental protection, and to realize the "3060″ target. This study proposed layer transfer model and merged transfer model for thermal load prediction of the district heating station. Experiment schemes were elaborated to simulate cross-year and cross-site scenarios, and practical data was collected serving the experiments. The prediction accuracy can be maintained without degradation in cross-year scenario, specifically, the coefficient of variation of the root mean squared error fluctuated between −1.09% and +0.45% compared to previous heating season when proposed two models were used. In the cross-site scenario, proposed models can achieve good prediction performance when the training data is insufficient. The coefficient of variation of the root mean squared error of the new model with insufficient training data was reduced by 7.62% on average when merged transfer model was used, which is equivalent to an overall reduction of 41.67%. Furthermore, proposed models can be applied to further optimize the prediction performance, even if beyond the scenarios discussed in this study.

Keywords: District heating; District heating station; Heating energy consumption prediction; Machine learning; Transfer learning (search for similar items in EconPapers)
Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (5)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544221025573
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:239:y:2022:i:pe:s0360544221025573

DOI: 10.1016/j.energy.2021.122309

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 ().