A review of data-driven fault detection and diagnostics for building HVAC systems

Zhelun Chen; Zheng O'Neill; Jin Wen; Ojas Pradhan; Tao Yang; Xing Lu; Guanjing Lin; Shohei Miyata; Seungjae Lee; Chou Shen; Roberto Chiosa; Marco Savino Piscitelli; Alfonso Capozzoli; Franz Hengel; Alexander Kührer; Marco Pritoni; Wei Liu; John Clauß; Yimin Chen; Terry Herr

doi:10.1016/j.apenergy.2023.121030

A review of data-driven fault detection and diagnostics for building HVAC systems

Zhelun Chen
, Zheng O'Neill
, Jin Wen
, Ojas Pradhan
, Tao Yang
, Xing Lu
, Guanjing Lin
, Shohei Miyata
, Seungjae Lee
, Chou Shen
, Roberto Chiosa
, Marco Savino Piscitelli
, Alfonso Capozzoli
, Franz Hengel
, Alexander Kührer
, Marco Pritoni
, Wei Liu
, John Clauß
, Yimin Chen
, Terry Herr

Research output: Contribution to journal › Article › peer-review

219 Scopus citations

Abstract

With the wide adoption of building automation system, and the advancement of data, sensing, and machine learning techniques, data-driven fault detection and diagnostics (FDD) for building heating, ventilation, and air conditioning systems has gained increasing attention. In this paper, data-driven FDD is defined as those that are built or trained from data via machine learning or multivariate statistical analysis methods. Following this definition, this paper reviews and summarizes the literature on data-driven FDD from three aspects: process, systems studied (including the systems being investigated, the faults being identified, and the associated data sources), and evaluation metrics. A data-driven FDD process is further divided into the following steps: data collection, data cleansing, data preprocessing, baseline establishment, fault detection, fault diagnostics, and potential fault prognostics. Literature reported data-driven methods used in each step of an FDD process are firstly discussed. Applications of data-driven FDD in various HVAC systems/components and commonly used data source for FDD development are reviewed secondly, followed by a summary of typical metrics for evaluating FDD methods. Finally, this literature review concludes that despite the promising performance reported in the literature, data-driven FDD methods still face many challenges, such as real-building deployment, performance evaluation and benchmarking, scalability and transferability, interpretability, cyber security and data privacy, user experience, etc. Addressing these challenges is critical for a broad real-building adoption of data-driven FDD.

Original language	English
Article number	121030
Journal	Applied Energy
Volume	339
DOIs	https://doi.org/10.1016/j.apenergy.2023.121030
State	Published - Jun 1 2023
Externally published	Yes

Funding

This work has been performed within the framework of the International Energy Agency - Energy in Buildings and Communities Program (IEA-EBC) Annex 81 “Data-Driven Smart Buildings”. Drexel University and Texas A&M University’s efforts were partly supported by: (1) NSF project # 2050509 “PFI-RP: Data-Driven Services for High Performance and Sustainable Buildings.” (2) the Building Technologies Office at the U.S. Department of Energy through the Emerging Technologies program under award number DE-EE0009150 and DE-EE0009153 , and (3) the Building America program at the U.S. Department of Energy under award number DE-EE0008694 . Lawrence Berkeley National Laboratory’s efforts were partly supported by the U.S. Department of Energy under contract number DE-AC0205CH11231 . The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. This work has been performed within the framework of the International Energy Agency - Energy in Buildings and Communities Program (IEA-EBC) Annex 81 “Data-Driven Smart Buildings”. Drexel University and Texas A&M University's efforts were partly supported by: (1) NSF project # 2050509 “PFI-RP: Data-Driven Services for High Performance and Sustainable Buildings.” (2) the Building Technologies Office at the U.S. Department of Energy through the Emerging Technologies program under award number DE-EE0009150 and DE-EE0009153, and (3) the Building America program at the U.S. Department of Energy under award number DE-EE0008694. Lawrence Berkeley National Laboratory's efforts were partly supported by the U.S. Department of Energy under contract number DE-AC0205CH11231. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

Keywords

Anomaly detection
Building HVAC
Data imputation
Data-driven
Fault detection
Fault diagnostics
Fault prognostics
Feature selection
Machine learning

Access to Document

10.1016/j.apenergy.2023.121030

Cite this

Chen, Z., O'Neill, Z., Wen, J., Pradhan, O., Yang, T., Lu, X., Lin, G., Miyata, S., Lee, S., Shen, C., Chiosa, R., Piscitelli, M. S., Capozzoli, A., Hengel, F., Kührer, A., Pritoni, M., Liu, W., Clauß, J., Chen, Y., & Herr, T. (2023). A review of data-driven fault detection and diagnostics for building HVAC systems. Applied Energy, 339, Article 121030. https://doi.org/10.1016/j.apenergy.2023.121030

@article{33fa680865e44b09958e0d52d20c577c,

title = "A review of data-driven fault detection and diagnostics for building HVAC systems",

abstract = "With the wide adoption of building automation system, and the advancement of data, sensing, and machine learning techniques, data-driven fault detection and diagnostics (FDD) for building heating, ventilation, and air conditioning systems has gained increasing attention. In this paper, data-driven FDD is defined as those that are built or trained from data via machine learning or multivariate statistical analysis methods. Following this definition, this paper reviews and summarizes the literature on data-driven FDD from three aspects: process, systems studied (including the systems being investigated, the faults being identified, and the associated data sources), and evaluation metrics. A data-driven FDD process is further divided into the following steps: data collection, data cleansing, data preprocessing, baseline establishment, fault detection, fault diagnostics, and potential fault prognostics. Literature reported data-driven methods used in each step of an FDD process are firstly discussed. Applications of data-driven FDD in various HVAC systems/components and commonly used data source for FDD development are reviewed secondly, followed by a summary of typical metrics for evaluating FDD methods. Finally, this literature review concludes that despite the promising performance reported in the literature, data-driven FDD methods still face many challenges, such as real-building deployment, performance evaluation and benchmarking, scalability and transferability, interpretability, cyber security and data privacy, user experience, etc. Addressing these challenges is critical for a broad real-building adoption of data-driven FDD.",

keywords = "Anomaly detection, Building HVAC, Data imputation, Data-driven, Fault detection, Fault diagnostics, Fault prognostics, Feature selection, Machine learning",

author = "Zhelun Chen and Zheng O'Neill and Jin Wen and Ojas Pradhan and Tao Yang and Xing Lu and Guanjing Lin and Shohei Miyata and Seungjae Lee and Chou Shen and Roberto Chiosa and Piscitelli, \{Marco Savino\} and Alfonso Capozzoli and Franz Hengel and Alexander K{\"u}hrer and Marco Pritoni and Wei Liu and John Clau{\ss} and Yimin Chen and Terry Herr",

note = "Publisher Copyright: {\textcopyright} 2023 Elsevier Ltd",

year = "2023",

month = jun,

day = "1",

doi = "10.1016/j.apenergy.2023.121030",

language = "English",

volume = "339",

journal = "Applied Energy",

issn = "0306-2619",

publisher = "Elsevier",

}

Chen, Z, O'Neill, Z, Wen, J, Pradhan, O, Yang, T, Lu, X, Lin, G, Miyata, S, Lee, S, Shen, C, Chiosa, R, Piscitelli, MS, Capozzoli, A, Hengel, F, Kührer, A, Pritoni, M, Liu, W, Clauß, J, Chen, Y & Herr, T 2023, 'A review of data-driven fault detection and diagnostics for building HVAC systems', Applied Energy, vol. 339, 121030. https://doi.org/10.1016/j.apenergy.2023.121030

TY - JOUR

T1 - A review of data-driven fault detection and diagnostics for building HVAC systems

AU - Chen, Zhelun

AU - O'Neill, Zheng

AU - Wen, Jin

AU - Pradhan, Ojas

AU - Yang, Tao

AU - Lu, Xing

AU - Lin, Guanjing

AU - Miyata, Shohei

AU - Lee, Seungjae

AU - Shen, Chou

AU - Chiosa, Roberto

AU - Piscitelli, Marco Savino

AU - Capozzoli, Alfonso

AU - Hengel, Franz

AU - Kührer, Alexander

AU - Pritoni, Marco

AU - Liu, Wei

AU - Clauß, John

AU - Chen, Yimin

AU - Herr, Terry

PY - 2023/6/1

Y1 - 2023/6/1

N2 - With the wide adoption of building automation system, and the advancement of data, sensing, and machine learning techniques, data-driven fault detection and diagnostics (FDD) for building heating, ventilation, and air conditioning systems has gained increasing attention. In this paper, data-driven FDD is defined as those that are built or trained from data via machine learning or multivariate statistical analysis methods. Following this definition, this paper reviews and summarizes the literature on data-driven FDD from three aspects: process, systems studied (including the systems being investigated, the faults being identified, and the associated data sources), and evaluation metrics. A data-driven FDD process is further divided into the following steps: data collection, data cleansing, data preprocessing, baseline establishment, fault detection, fault diagnostics, and potential fault prognostics. Literature reported data-driven methods used in each step of an FDD process are firstly discussed. Applications of data-driven FDD in various HVAC systems/components and commonly used data source for FDD development are reviewed secondly, followed by a summary of typical metrics for evaluating FDD methods. Finally, this literature review concludes that despite the promising performance reported in the literature, data-driven FDD methods still face many challenges, such as real-building deployment, performance evaluation and benchmarking, scalability and transferability, interpretability, cyber security and data privacy, user experience, etc. Addressing these challenges is critical for a broad real-building adoption of data-driven FDD.

AB - With the wide adoption of building automation system, and the advancement of data, sensing, and machine learning techniques, data-driven fault detection and diagnostics (FDD) for building heating, ventilation, and air conditioning systems has gained increasing attention. In this paper, data-driven FDD is defined as those that are built or trained from data via machine learning or multivariate statistical analysis methods. Following this definition, this paper reviews and summarizes the literature on data-driven FDD from three aspects: process, systems studied (including the systems being investigated, the faults being identified, and the associated data sources), and evaluation metrics. A data-driven FDD process is further divided into the following steps: data collection, data cleansing, data preprocessing, baseline establishment, fault detection, fault diagnostics, and potential fault prognostics. Literature reported data-driven methods used in each step of an FDD process are firstly discussed. Applications of data-driven FDD in various HVAC systems/components and commonly used data source for FDD development are reviewed secondly, followed by a summary of typical metrics for evaluating FDD methods. Finally, this literature review concludes that despite the promising performance reported in the literature, data-driven FDD methods still face many challenges, such as real-building deployment, performance evaluation and benchmarking, scalability and transferability, interpretability, cyber security and data privacy, user experience, etc. Addressing these challenges is critical for a broad real-building adoption of data-driven FDD.

KW - Anomaly detection

KW - Building HVAC

KW - Data imputation

KW - Data-driven

KW - Fault detection

KW - Fault diagnostics

KW - Fault prognostics

KW - Feature selection

KW - Machine learning

UR - https://www.scopus.com/pages/publications/85151031548

U2 - 10.1016/j.apenergy.2023.121030

DO - 10.1016/j.apenergy.2023.121030

M3 - Article

AN - SCOPUS:85151031548

SN - 0306-2619

VL - 339

JO - Applied Energy

JF - Applied Energy

M1 - 121030

ER -

A review of data-driven fault detection and diagnostics for building HVAC systems

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this