4.7 Article

System-level fouling detection of district heating substations using virtual-sensor-assisted building automation system

Journal

ENERGY
Volume 227, Issue -, Pages -

Publisher

PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.energy.2021.120515

Keywords

Fault detection; District heating substations; Virtual sensors; Heat exchanger fouling; Sensors; Building automation system

Funding

  1. Basic Science Research Program through the National Research Foundation of Korea (NRF) - Ministry of Education [2019R1F1A1060834]
  2. National Research Foundation of Korea [2019R1F1A1060834] Funding Source: Korea Institute of Science & Technology Information (KISTI), National Science & Technology Information Service (NTIS)

Ask authors/readers for more resources

An automated fouling detection method for district heating substations (DHS) is proposed, which significantly improves detection performance for light fouling conditions with the assistance of virtual sensors. Virtual measurements are fed into the input layer of the detection model to reinforce the physical relationship of the trained model and address the issue of sensor absences.
The operational fouling of heat exchangers in a district heating substation (DHS) has a negative effect on the heating efficiency in the primary district heating side and on the heating charge and thermal comfort in the terminal building side. It is effective for early detection of heat exchanger fouling when using the existing building automation system installed in a building. Therefore, an automated fouling detection method is proposed for DHSs. Practical challenges, including sensor absences and limited operation datasets, are considered for the proposed method to be effective for application in a real DHS at the building side. The assistance of virtual sensors is suggested to address the issue of sensor absences. The virtual measurements are fed into the input layer of the fouling detection model to reinforce the physical relationship of the trained model. Based on the physical relationship in the model, the virtual-sensor assisted fouling detection method can significantly improve fouling detection performance for light fouling conditions. The detection is also successful for new DHS operation patterns that are not included in the training datasets. Compared with the application without a virtual sensor, the virtual-sensor assisted application showed an approximately 61% improvement in accurate detection with a 17-min fast alarm. (c) 2021 Elsevier Ltd. All rights reserved.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.7
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available