A systematic feature extraction and selection framework for data-driven whole-building automated fault detection and diagnostics in commercial buildings

Zhang, Liang; Frank, Stephen; Kim, Janghyun; Jin, Xin; Leach, Matt

doi:10.1016/j.buildenv.2020.107338

Title: A systematic feature extraction and selection framework for data-driven whole-building automated fault detection and diagnostics in commercial buildings

Journal Article · Tue Sep 29 00:00:00 EDT 2020 · Building and Environment

DOI:https://doi.org/10.1016/j.buildenv.2020.107338· OSTI ID:1710174

Zhang, Liang ^[1]; Frank, Stephen ^[1];

^[1]; Jin, Xin ^[1]; Leach, Matt ^[1]

National Renewable Energy Lab. (NREL), Golden, CO (United States)

In data-driven automated fault detection and diagnostics (AFDD) modeling for building energy systems, feature engineering is a critical process of extracting information from high-dimensional and noisy sensor measurement and turning it into informative and representative inputs or features for data-driven modeling. However, few studies specifically discuss the feature engineering, especially the interactions between feature extraction and feature selection in whole-building AFDD. We developed a systematic feature extraction and selection framework for whole-building AFDD. In this framework, features are aggressively extracted from raw sensor data using statistical feature extraction techniques with various window sizes and statistics. With many features extracted, a hybrid feature selection algorithm that combines the filter and wrapper method then selects the best feature set. The framework considers diversity in the duration of fault behavior among fault types in whole-building AFDD, thus achieving high model generalization. We implemented our developed framework in a virtual testbed calibrated with measured data from Oak Ridge National Laboratory's Flexible Research Platform designed to mimic the operation of a typical small commercial building. The AFDD model is trained by the simulation data generated from the virtual testbed. The results show that (1) the developed framework improves the generalization of the AFDD model by 10.7% compared with literature-reported feature extraction and selection methods and (2) features with diverse window sizes and statistics are selected, providing insight into physical systems beyond the current understanding of buildings and faults and improving the detection and diagnostics of multiple fault types.

View Accepted Manuscript (DOE)

View Accepted Manuscript (Publisher)

Cite

Export

Save

Research Organization:: National Renewable Energy Lab. (NREL), Golden, CO (United States)

Sponsoring Organization:: USDOE Office of Energy Efficiency and Renewable Energy (EERE), Energy Efficiency Office. Building Technologies Office

Grant/Contract Number:: AC36-08GO28308

OSTI ID:: 1710174

Alternate ID(s):: OSTI ID: 1781483

Report Number(s):: NREL-JA-5500-75756; MainId:6783; UUID:b33ba875-472e-ea11-9c2f-ac162d87dfe5; MainAdminID:18752

Journal Information:: Building and Environment, Vol. 186; ISSN 0360-1323

Publisher:: ElsevierCopyright Statement

Country of Publication:: United States

Language:: English

References (18)

Optimal Sensor Configuration and Feature Selection for AHU Fault Detection and Diagnosis Li, Dan; Zhou, Yuxun; Hu, Guoqiang IEEE Transactions on Industrial Informatics, Vol. 13, Issue 3 https://doi.org/10.1109/TII.2016.2644669	journal	June 2017
A review of fault detection and diagnostics methods for building systems Kim, Woohyun; Katipamula, Srinivas Science and Technology for the Built Environment, Vol. 24, Issue 1 https://doi.org/10.1080/23744731.2017.1318008	journal	April 2017
Using intelligent data analysis to detect abnormal energy consumption in buildings Seem, John E. Energy and Buildings, Vol. 39, Issue 1 https://doi.org/10.1016/j.enbuild.2006.03.033	journal	January 2007
Review Article: Methods for Fault Detection, Diagnostics, and Prognostics for Building Systems—A Review, Part I Katipamula, Srinivas; Brambley, Michael HVAC&R Research, Vol. 11, Issue 1 https://doi.org/10.1080/10789669.2005.10391123	journal	January 2005
A systematic feature selection procedure for short-term data-driven building energy forecasting model development Zhang, Liang; Wen, Jin Energy and Buildings, Vol. 183 https://doi.org/10.1016/j.enbuild.2018.11.010	journal	January 2019
Representing Small Commercial Building Faults in EnergyPlus, Part II: Model Validation Kim, Janghyun; Frank, Stephen; Im, Piljae Buildings, Vol. 9, Issue 12 https://doi.org/10.3390/buildings9120239	journal	November 2019
Building fault detection and diagnostics: Achieved savings, and methods to evaluate algorithm performance Lin, Guanjing; Kramer, Hannah; Granderson, Jessica Building and Environment, Vol. 168 https://doi.org/10.1016/j.buildenv.2019.106505	journal	January 2020
Data-Driven Modeling, Fault Diagnosis and Optimal Sensor Selection for HVAC Chillers Namburu, Setu Madhavi; Azam, Mohammad S.; Luo, Jianhui IEEE Transactions on Automation Science and Engineering, Vol. 4, Issue 3 https://doi.org/10.1109/TASE.2006.888053	journal	July 2007
Fault detection and diagnosis for building cooling system with a tree-structured learning method Li, Dan; Zhou, Yuxun; Hu, Guoqiang Energy and Buildings, Vol. 127 https://doi.org/10.1016/j.enbuild.2016.06.017	journal	September 2016
Review of automated fault detection and diagnostic tools in air handling units Bruton, Ken; Raftery, Paul; Kennedy, Barry Energy Efficiency, Vol. 7, Issue 2 https://doi.org/10.1007/s12053-013-9238-2	journal	November 2013
Robust model-based fault diagnosis for air handling units Mulumba, Timothy; Afshari, Afshin; Yan, Ke Energy and Buildings, Vol. 86 https://doi.org/10.1016/j.enbuild.2014.10.069	journal	January 2015
Data mining in building automation system for improving building operational performance Xiao, Fu; Fan, Cheng Energy and Buildings, Vol. 75 https://doi.org/10.1016/j.enbuild.2014.02.005	journal	June 2014
Important sensors for chiller fault detection and diagnosis (FDD) from the perspective of feature selection and machine learning Han, H.; Gu, B.; Wang, T. International Journal of Refrigeration, Vol. 34, Issue 2 https://doi.org/10.1016/j.ijrefrig.2010.08.011	journal	March 2011
A review on time series data mining Fu, Tak-chung Engineering Applications of Artificial Intelligence, Vol. 24, Issue 1 https://doi.org/10.1016/j.engappai.2010.09.007	journal	February 2011
Study on a hybrid SVM model for chiller FDD applications Han, H.; Gu, B.; Kang, J. Applied Thermal Engineering, Vol. 31, Issue 4 https://doi.org/10.1016/j.applthermaleng.2010.10.021	journal	March 2011
Automated daily pattern filtering of measured building performance data Miller, Clayton; Nagy, Zoltán; Schlueter, Arno Automation in Construction, Vol. 49 https://doi.org/10.1016/j.autcon.2014.09.004	journal	January 2015
Representing Small Commercial Building Faults in EnergyPlus, Part I: Model Development Kim, Janghyun; Frank, Stephen; Braun, James E. Buildings, Vol. 9, Issue 11 https://doi.org/10.3390/buildings9110233	journal	November 2019
ARX model based fault detection and diagnosis for chillers using support vector machines Yan, Ke; Shen, Wen; Mulumba, Timothy Energy and Buildings, Vol. 81 https://doi.org/10.1016/j.enbuild.2014.05.049	journal	October 2014

Similar Records

Voltron Compatible Whole Building Root-Fault Detection and Diagnosis

Technical Report · Sat Dec 01 00:00:00 EST 2018 · OSTI ID:1710174

Wen, Jin; Chen, Yimin

Sensor impact evaluation and verification for fault detection and diagnostics in building energy systems: A review

Journal Article · Sun Aug 01 00:00:00 EDT 2021 · Advances in Applied Energy · OSTI ID:1710174

Zhang, Liang; Leach, Matt; Bae, Yeonjin; +7 more

Data-driven building energy modeling with feature selection and active learning for data predictive control

Journal Article · Thu Sep 09 00:00:00 EDT 2021 · Energy and Buildings · OSTI ID:1710174

Zhang, Liang

Related Subjects

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATION
whole building
AFDD
statistical feature extraction
feature selection
commercial buildings
automated fault detection and diagnostics

Title: A systematic feature extraction and selection framework for data-driven whole-building automated fault detection and diagnostics in commercial buildings

Citation Formats

References (18)

Similar Records

Related Subjects