Detecting and Diagnosing Incipient Building Faults Using Uncertainty Information from Deep Neural Networks

Jin, Baihong; Li, Dan; Srinivasan, Seshadhri; Ng, See-Kiong; Poolla, Kameshwar; Sangiovanni-Vincentelli, Alberto

doi:10.1109/icphm.2019.8819438

Title: Detecting and Diagnosing Incipient Building Faults Using Uncertainty Information from Deep Neural Networks

Journal Article · Sat Jun 01 00:00:00 EDT 2019 · Prognostics and System Health Management Conference

DOI:https://doi.org/10.1109/icphm.2019.8819438· OSTI ID:1572819

Jin, Baihong ^[1]; Li, Dan ^[2]; Srinivasan, Seshadhri ^[1]; Ng, See-Kiong ^[2]; Poolla, Kameshwar ^[1]; Sangiovanni-Vincentelli, Alberto ^[1]

Univ. of California, Berkeley, CA (United States)
National Univ. of Singapore (Singapore)

Early detection of incipient faults is of significant importance to reducing maintenance costs, saving energy, and enhancing occupant comfort in buildings. Popular supervised learning models such as deep neural networks are considered promising due to their ability to directly learn from labeled fault data; however, it is known that the performance of supervised learning approaches highly relies on the availability and quality of labeled training data. In Fault Detection and Diagnosis (FDD) applications, the lack of labeled incipient fault data has posed a major challenge to applying these supervised learning techniques to commercial buildings. To address this concern, this paper proposes using Monte Carlo dropout (MC-dropout) to enhance the supervised learning pipeline, so that the resulting neural network is able to detect and diagnose unseen incipient fault examples. We also examine the proposed MC-dropout method on the RP-1043 dataset to demonstrate its effectiveness in indicating the most likely incipient fault types.

View Accepted Manuscript (DOE)

Cite

Export

Save

Research Organization:: Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)

Sponsoring Organization:: USDOE Office of Science (SC)

Grant/Contract Number:: AC02-05CH11231

OSTI ID:: 1572819

Journal Information:: Prognostics and System Health Management Conference, Vol. 2019; Conference: International Conference on Prognostics and Health Management (ICPHM), San Francisco, CA, (United States), 17-20 Jun 2019; ISSN 2166-563X

Publisher:: IEEECopyright Statement

Country of Publication:: United States

Language:: English

References (29)

From Model, Signal to Knowledge: A Data-Driven Perspective of Fault Detection and Diagnosis Dai, Xuewu; Gao, Zhiwei IEEE Transactions on Industrial Informatics, Vol. 9, Issue 4 https://doi.org/10.1109/tii.2013.2243743	journal	November 2013
A Survey of Fault Diagnosis and Fault-Tolerant Techniques—Part I: Fault Diagnosis With Model-Based and Signal-Based Approaches Gao, Zhiwei; Cecati, Carlo; Ding, Steven X. IEEE Transactions on Industrial Electronics, Vol. 62, Issue 6 https://doi.org/10.1109/tie.2015.2417501	journal	June 2015
A robust pattern recognition-based fault detection and diagnosis (FDD) method for chillers Zhao, Yang; Xiao, Fu; Wen, Jin HVAC&R Research, Vol. 20, Issue 7 https://doi.org/10.1080/10789669.2014.938006	journal	October 2014
An intelligent chiller fault detection and diagnosis methodology using Bayesian belief network Zhao, Yang; Xiao, Fu; Wang, Shengwei Energy and Buildings, Vol. 57 https://doi.org/10.1016/j.enbuild.2012.11.007	journal	February 2013
Bayesian network based FDD strategy for variable air volume terminals Xiao, Fu; Zhao, Yang; Wen, Jin Automation in Construction, Vol. 41 https://doi.org/10.1016/j.autcon.2013.10.019	journal	May 2014
Mining Building Energy Management System Data Using Fuzzy Anomaly Detection and Linguistic Descriptions Wijayasekara, Dumidu; Linda, Ondrej; Manic, Milos IEEE Transactions on Industrial Informatics, Vol. 10, Issue 3 https://doi.org/10.1109/TII.2014.2328291	journal	August 2014
Incipient fault diagnosis of chemical processes via artificial neural networks Watanabe, Kajiro; Matsuura, Ichiro; Abe, Masahiro AIChE Journal, Vol. 35, Issue 11 https://doi.org/10.1002/aic.690351106	journal	November 1989
Pattern recognition-based chillers fault detection method using Support Vector Data Description (SVDD) Zhao, Yang; Wang, Shengwei; Xiao, Fu Applied Energy, Vol. 112 https://doi.org/10.1016/j.apenergy.2012.12.043	journal	December 2013
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
An incipient fault detection approach via detrending and denoising He, Zhangming; Shardt, Yuri A. W.; Wang, Dayi Control Engineering Practice, Vol. 74 https://doi.org/10.1016/j.conengprac.2018.02.005	journal	May 2018
Detection of anomalous events from unlabeled sensor data in smart building environments Jaikumar, Padmini; Gacic, Aca; Andrews, Burton 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) https://doi.org/10.1109/ICASSP.2011.5946934	conference	May 2011
Design Automation for Smart Building Systems Jia, Ruoxi; Jin, Baihong; Jin, Ming Proceedings of the IEEE, Vol. 106, Issue 9 https://doi.org/10.1109/JPROC.2018.2856932	journal	September 2018
A fault detection and diagnosis strategy of VAV air-conditioning systems for improved energy and control performances Qin, Jianying; Wang, Shengwei Energy and Buildings, Vol. 37, Issue 10 https://doi.org/10.1016/j.enbuild.2004.12.011	journal	October 2005
Variable-Structure Multiple-Model Approach to Fault Detection, Identification, and Estimation Ru, J.; Rong Li, X. IEEE Transactions on Control Systems Technology, Vol. 16, Issue 5 https://doi.org/10.1109/TCST.2007.916318	journal	September 2008
Deep learning LeCun, Yann; Bengio, Yoshua; Hinton, Geoffrey Nature, Vol. 521, Issue 7553 https://doi.org/10.1038/nature14539	journal	May 2015
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
Fault detection and diagnosis for buildings and HVAC systems using combined neural networks and subtractive clustering analysis Du, Zhimin; Fan, Bo; Jin, Xinqiao Building and Environment, Vol. 73 https://doi.org/10.1016/j.buildenv.2013.11.021	journal	March 2014
Anomaly Detection Using Autoencoders with Nonlinear Dimensionality Reduction Sakurada, Mayu; Yairi, Takehisa Proceedings of the MLSDA 2014 2nd Workshop on Machine Learning for Sensory Data Analysis - MLSDA'14 https://doi.org/10.1145/2689746.2689747	conference	January 2014
A hybrid FDD strategy for local system of AHU based on artificial neural network and wavelet analysis Fan, Bo; Du, Zhimin; Jin, Xinqiao Building and Environment, Vol. 45, Issue 12 https://doi.org/10.1016/j.buildenv.2010.05.031	journal	December 2010
Building Energy Doctors: An SPC and Kalman Filter-Based Method for System-Level Fault Detection in HVAC Systems Sun, Biao; Luh, Peter B.; Jia, Qing-Shan IEEE Transactions on Automation Science and Engineering, Vol. 11, Issue 1 https://doi.org/10.1109/TASE.2012.2226155	journal	January 2014
Multiple faults diagnosis for sensors in air handling unit using Fisher discriminant analysis Du, Zhimin; Jin, Xinqiao Energy Conversion and Management, Vol. 49, Issue 12 https://doi.org/10.1016/j.enconman.2008.06.032	journal	December 2008
PCA-SVM-Based Automated Fault Detection and Diagnosis (AFDD) for Vapor-Compression Refrigeration Systems Han, Hua; Cao, Zhikun; Gu, Bo HVAC&R Research, Vol. 16, Issue 3 https://doi.org/10.1080/10789669.2010.10390906	journal	May 2010
Using SVM based method for equipment fault detection in a thermal power plant Chen, Kai-Ying; Chen, Long-Sheng; Chen, Mu-Chen Computers in Industry, Vol. 62, Issue 1 https://doi.org/10.1016/j.compind.2010.05.013	journal	January 2011
A model-based fault detection and diagnostic methodology based on PCA method and wavelet transform Li, Shun; Wen, Jin Energy and Buildings, Vol. 68 https://doi.org/10.1016/j.enbuild.2013.08.044	journal	January 2014
Model-based Fault Detection and Diagnosis of HVAC systems using Support Vector Machine method Liang, J.; Du, R. International Journal of Refrigeration, Vol. 30, Issue 6 https://doi.org/10.1016/j.ijrefrig.2006.12.012	journal	September 2007
Fault diagnosis for sensors in air handling unit based on neural network pre-processed by wavelet and fractal Zhu, Yonghua; Jin, Xinqiao; Du, Zhimin Energy and Buildings, Vol. 44 https://doi.org/10.1016/j.enbuild.2011.09.043	journal	January 2012
The Sensitivity of Chiller Performance to Common Faults Comstock, M. C.; Braun, J. E.; Groll, E. A. HVAC&R Research, Vol. 7, Issue 3 https://doi.org/10.1080/10789669.2001.10391274	journal	July 2001
Development of room temperature and relative humidity linear parametric models for an open office using BMS data Mustafaraj, G.; Chen, J.; Lowry, G. Energy and Buildings, Vol. 42, Issue 3 https://doi.org/10.1016/j.enbuild.2009.10.001	journal	March 2010
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

Cited By (1)

An Encoder-Decoder Based Approach for Anomaly Detection with Application in Additive Manufacturing Tan, Yingshui; Jin, Baihong; Nettekoven, Alexander 2019 18th IEEE International Conference On Machine Learning And Applications (ICMLA) https://doi.org/10.1109/icmla.2019.00171	conference	December 2019

Similar Records

Using discrete Bayesian networks for diagnosing and isolating cross-level faults in HVAC systems

Journal Article · Mon Oct 10 00:00:00 EDT 2022 · Applied Energy · OSTI ID:1572819

Chen, Yimin; Wen, Jin; Pradhan, Ojas; +2 more

A review of computing-based automated fault detection and diagnosis of heating, ventilation and air conditioning systems

Journal Article · Wed Apr 06 00:00:00 EDT 2022 · Renewable and Sustainable Energy Reviews · OSTI ID:1572819

Chen, Jianli; Zhang, Liang; Li, Yanfei; +3 more

Incipient fault diagnosis of chemical processes via artificial neural networks

Journal Article · Wed Nov 01 00:00:00 EST 1989 · A.I.Ch.E. Journal (American Institute of Chemical Engineers); (USA) · OSTI ID:1572819

Watanabe, K; Matsuura, I

Related Subjects

42 ENGINEERING
97 MATHEMATICS AND COMPUTING
buildings (structures)
condition monitoring
fault diagnosis
learning (artificial intelligence)
Monte Carlo methods
neural nets
structural engineering computing

Title: Detecting and Diagnosing Incipient Building Faults Using Uncertainty Information from Deep Neural Networks

Citation Formats

References (29)

Cited By (1)

Similar Records

Related Subjects