A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Fault Detection and Diagnosis of Chillers Under Transient Conditions
https://orcid.org/http://orcid.org/0000-0002-7956-1921
https://orcid.org/http://orcid.org/0000-0003-1705-1093
https://orcid.org/http://orcid.org/0000-0002-7423-0621
Chillers are widely used to provide heating, cooling, and control the indoor environment. A significant portion of energy in buildings is consumed by HVAC systems. Therefore, controlling and commissioning chillers are of great importance to have an energy-efficient system. Faults commonly occur in the system due to the lack of maintenance, poor component performance, installation errors, and improper control of chillers. They are among the primary reasons for the waste of energy, reducing the lifespan of equipment, thermal discomfort, and growth in CO2 emission. Identifying and diagnosing faults of chillers in the early stages can reduce adverse consequences of faults to a great extent. Various Automatic Fault Detection and Diagnosis (AFDD) methods have been developed subject to the system's operational condition to address the mentioned challenges. Generally, chillers experience both steady and transient states during their operational conditions. According to the literature, most of the algorithms have been developed for the steady state condition of the system, while chillers, especially in commercial buildings, work under transient conditions most of the time. Accordingly, this study focuses on dynamic fault detection and diagnosis of chillers to investigate the feasibility and challenges of AFDD within the transient state. A dataset published by ASHRAE RP-1043 that considers seven typical faults and one normal condition for a 90-ton chiller is classified using Support Vector Machine (SVM) classification method. Faulty and normal behaviors of the system were successfully detected and isolated through the proposed algorithm which were trained on dynamic data.
Fault Detection and Diagnosis of Chillers Under Transient Conditions
https://orcid.org/http://orcid.org/0000-0002-7956-1921
https://orcid.org/http://orcid.org/0000-0003-1705-1093
https://orcid.org/http://orcid.org/0000-0002-7423-0621
Chillers are widely used to provide heating, cooling, and control the indoor environment. A significant portion of energy in buildings is consumed by HVAC systems. Therefore, controlling and commissioning chillers are of great importance to have an energy-efficient system. Faults commonly occur in the system due to the lack of maintenance, poor component performance, installation errors, and improper control of chillers. They are among the primary reasons for the waste of energy, reducing the lifespan of equipment, thermal discomfort, and growth in CO2 emission. Identifying and diagnosing faults of chillers in the early stages can reduce adverse consequences of faults to a great extent. Various Automatic Fault Detection and Diagnosis (AFDD) methods have been developed subject to the system's operational condition to address the mentioned challenges. Generally, chillers experience both steady and transient states during their operational conditions. According to the literature, most of the algorithms have been developed for the steady state condition of the system, while chillers, especially in commercial buildings, work under transient conditions most of the time. Accordingly, this study focuses on dynamic fault detection and diagnosis of chillers to investigate the feasibility and challenges of AFDD within the transient state. A dataset published by ASHRAE RP-1043 that considers seven typical faults and one normal condition for a 90-ton chiller is classified using Support Vector Machine (SVM) classification method. Faulty and normal behaviors of the system were successfully detected and isolated through the proposed algorithm which were trained on dynamic data.
Fault Detection and Diagnosis of Chillers Under Transient Conditions
https://orcid.org/http://orcid.org/0000-0002-7956-1921
https://orcid.org/http://orcid.org/0000-0003-1705-1093
https://orcid.org/http://orcid.org/0000-0002-7423-0621
Chillers are widely used to provide heating, cooling, and control the indoor environment. A significant portion of energy in buildings is consumed by HVAC systems. Therefore, controlling and commissioning chillers are of great importance to have an energy-efficient system. Faults commonly occur in the system due to the lack of maintenance, poor component performance, installation errors, and improper control of chillers. They are among the primary reasons for the waste of energy, reducing the lifespan of equipment, thermal discomfort, and growth in CO2 emission. Identifying and diagnosing faults of chillers in the early stages can reduce adverse consequences of faults to a great extent. Various Automatic Fault Detection and Diagnosis (AFDD) methods have been developed subject to the system's operational condition to address the mentioned challenges. Generally, chillers experience both steady and transient states during their operational conditions. According to the literature, most of the algorithms have been developed for the steady state condition of the system, while chillers, especially in commercial buildings, work under transient conditions most of the time. Accordingly, this study focuses on dynamic fault detection and diagnosis of chillers to investigate the feasibility and challenges of AFDD within the transient state. A dataset published by ASHRAE RP-1043 that considers seven typical faults and one normal condition for a 90-ton chiller is classified using Support Vector Machine (SVM) classification method. Faulty and normal behaviors of the system were successfully detected and isolated through the proposed algorithm which were trained on dynamic data.
Fault Detection and Diagnosis of Chillers Under Transient Conditions
Lecture Notes in Civil Engineering
Desjardins, Serge (editor) / Poitras, Gérard J. (editor) / Nik-Bakht, Mazdak (editor) / Bezyan, Yashar (author) / Nik-Bakht, Mazdak (author) / Nasiri, Fuzhan (author)
Canadian Society of Civil Engineering Annual Conference ; 2023 ; Moncton, NB, Canada
Proceedings of the Canadian Society for Civil Engineering Annual Conference 2023, Volume 3 ; Chapter: 26 ; 369-381
2024-10-16
13 pages
Article/Chapter (Book)
Electronic Resource
English
Performance Monitoring, Fault Detection, and Diagnosis of Reciprocating Chillers
| British Library Online Contents | 1996
Generative adversarial network for fault detection diagnosis of chillers
| Elsevier | 2020
Performance Monitoring, Fault Detection, and Diagnosis of Reciprocating Chillers
| British Library Conference Proceedings | 1996
A Robust Fault Detection and Diagnosis Strategy for Centrifugal Chillers
| Taylor & Francis Verlag | 2006
Decoupling features for fault detection and diagnosis on centrifugal chillers (1486-RP)
| Taylor & Francis Verlag | 2011