Publications
2022
Chen, Yimin; Wen, Jin; Pradhan, Ojas; Lo, L. James; Wu, Teresa
Using discrete Bayesian networks for diagnosing and isolating cross-level faults in HVAC systems Journal Article
In: Applied Energy, vol. 327, pp. 120050, 2022, ISSN: 0306-2619.
Abstract | Links | BibTeX | Tags: Cross-level fault, Discrete Bayesian Network, HVAC system, Pattern Matching, Root cause fault diagnosis
@article{CHEN2022120050,
title = {Using discrete Bayesian networks for diagnosing and isolating cross-level faults in HVAC systems},
author = {Yimin Chen and Jin Wen and Ojas Pradhan and L. James Lo and Teresa Wu},
url = {https://www.sciencedirect.com/science/article/pii/S0306261922013071},
doi = {https://doi.org/10.1016/j.apenergy.2022.120050},
issn = {0306-2619},
year = {2022},
date = {2022-01-01},
journal = {Applied Energy},
volume = {327},
pages = {120050},
abstract = {Fault detection and diagnosis (FDD) technologies are critical to ensure satisfactory building performance, such as reducing energy wastes and negative impacts on occupant comfort and productivity. Existing FDD technologies mainly focus on component-level FDD solutions, which could lead to mis-diagnosis of cross-level faults in heating, ventilating, and air-conditioning (HVAC) systems. Cross-level faults are those faults that occur in one component or subsystem, but cause operational abnormalities in other components or subsystems, and result in a building level performance degradation. How to effectively diagnose the root cause of a cross-level fault is the focus of this study. This paper presents a novel discrete Bayesian Network (DisBN)-based method for diagnosing cross-level faults in an HVAC system commonly used in commercial buildings. A two-level DisBN structure model is developed in this study. The parameters used in the DisBN model are obtained either from expert knowledge or through machine-learning strategies from normal system operation data. Meanwhile, the probability parameters are discretized to incorporate the uncertainties associated with typical expert knowledge. Thus, the developed DisBN method addresses the challenges many other BN based FDD methods face, i.e., the lack of fault data for BN parameter training. The developed DisBN represents causal relationships between a fault and its cross-level system impacts (i.e., fault symptoms or fault indicators) by considering how fault impacts propagate across different levels in an HVAC system. A weather and schedule information-based Pattern Matching (WPM) method is employed to automatically create WPM baseline data sets for each incoming real time snapshot data from the building systems. Consequently, BN inference and real-time diagnostics are achieved by comparing incoming snapshot data and the WPM baseline data set. The proposed method is evaluated using experimental fault data collected in a campus building. Fault diagnosis results demonstrate that the WPM-DisBN method is effective at locating the root causes of cross-level faults in an HVAC system.},
keywords = {Cross-level fault, Discrete Bayesian Network, HVAC system, Pattern Matching, Root cause fault diagnosis},
pubstate = {published},
tppubtype = {article}
}
Fault detection and diagnosis (FDD) technologies are critical to ensure satisfactory building performance, such as reducing energy wastes and negative impacts on occupant comfort and productivity. Existing FDD technologies mainly focus on component-level FDD solutions, which could lead to mis-diagnosis of cross-level faults in heating, ventilating, and air-conditioning (HVAC) systems. Cross-level faults are those faults that occur in one component or subsystem, but cause operational abnormalities in other components or subsystems, and result in a building level performance degradation. How to effectively diagnose the root cause of a cross-level fault is the focus of this study. This paper presents a novel discrete Bayesian Network (DisBN)-based method for diagnosing cross-level faults in an HVAC system commonly used in commercial buildings. A two-level DisBN structure model is developed in this study. The parameters used in the DisBN model are obtained either from expert knowledge or through machine-learning strategies from normal system operation data. Meanwhile, the probability parameters are discretized to incorporate the uncertainties associated with typical expert knowledge. Thus, the developed DisBN method addresses the challenges many other BN based FDD methods face, i.e., the lack of fault data for BN parameter training. The developed DisBN represents causal relationships between a fault and its cross-level system impacts (i.e., fault symptoms or fault indicators) by considering how fault impacts propagate across different levels in an HVAC system. A weather and schedule information-based Pattern Matching (WPM) method is employed to automatically create WPM baseline data sets for each incoming real time snapshot data from the building systems. Consequently, BN inference and real-time diagnostics are achieved by comparing incoming snapshot data and the WPM baseline data set. The proposed method is evaluated using experimental fault data collected in a campus building. Fault diagnosis results demonstrate that the WPM-DisBN method is effective at locating the root causes of cross-level faults in an HVAC system.
2014
Li, Shun; Wen, Jin
Application of pattern matching method for detecting faults in air handling unit system Journal Article
In: Automation in Construction, vol. 43, pp. 49-58, 2014, ISSN: 0926-5805.
Abstract | Links | BibTeX | Tags: AHU, BAS, Fault detection and diagnostics, HVAC, Pattern Matching, Principle Component Analysis
@article{LI201449,
title = {Application of pattern matching method for detecting faults in air handling unit system},
author = {Shun Li and Jin Wen},
url = {https://www.sciencedirect.com/science/article/pii/S0926580514000545},
doi = {https://doi.org/10.1016/j.autcon.2014.03.002},
issn = {0926-5805},
year = {2014},
date = {2014-01-01},
journal = {Automation in Construction},
volume = {43},
pages = {49-58},
abstract = {This paper presents a hybrid air handling unit (AHU) fault detection strategy based on Principal Component Analysis (PCA) method and Pattern Matching method. The basic idea of the pattern matching method is to locate periods of operation from a historical data set whose operational conditions are similar to the target operating condition. The proposed Pattern Matching-PCA method uses two similarity factors, PCA similarity factors and Distance similarity factors, to characterize the degree of similarity between historical data window and current snapshot data. PCA model is then built using the historical AHU operation dataset that are identified to be similar to current snapshot operation data. The method is validated by operational data of an AHU system in real building. The results show that the sensibility of PCA models is enhanced by preprocessing the training data with the Pattern Matching method.},
keywords = {AHU, BAS, Fault detection and diagnostics, HVAC, Pattern Matching, Principle Component Analysis},
pubstate = {published},
tppubtype = {article}
}
This paper presents a hybrid air handling unit (AHU) fault detection strategy based on Principal Component Analysis (PCA) method and Pattern Matching method. The basic idea of the pattern matching method is to locate periods of operation from a historical data set whose operational conditions are similar to the target operating condition. The proposed Pattern Matching-PCA method uses two similarity factors, PCA similarity factors and Distance similarity factors, to characterize the degree of similarity between historical data window and current snapshot data. PCA model is then built using the historical AHU operation dataset that are identified to be similar to current snapshot operation data. The method is validated by operational data of an AHU system in real building. The results show that the sensibility of PCA models is enhanced by preprocessing the training data with the Pattern Matching method.