TY - GEN
T1 - Tandem Connectionist Anomaly Detection
T2 - 2018 IEEE International Conference on Prognostics and Health Management, ICPHM 2018
AU - Hasegawa, Takanori
AU - Ogata, Jun
AU - Murakawa, Masahiro
AU - Ogawa, Tetsuji
N1 - Funding Information:
port our work. This work was partly supported by the New Energy and Industrial Technology Development Organization (NEDO).
Funding Information:
We thank the National Renewable Energy Laboratory (NREL) for providing the "wind turbine gearbox condition monitoring vibration analysis benchmarking datasets" to support our work. This work was partly supported by the New Energy and Industrial Technology Development Organization (NEDO).
Publisher Copyright:
© 2018 IEEE.
PY - 2018/8/27
Y1 - 2018/8/27
N2 - An effective use of faulty-state data is proposed to achieve robust, accurate data-driven anomaly (fault) detection for rotating machine. Although using faulty data in the training process generally can improve the performance of anomaly detection system, it is rare to obtain enough samples to train failures or defects on a target machine. We therefore utilize the existing data from non-target (different-type) machines for feature representation learning to improve anomaly detection for the target machine. Specifically, deep neural networks (DNNs) that are trained to discriminate the normal and faulty states of the non-target machines are used to extract features. The extracted features are then taken as inputs to an anomaly detector based on Gaussian mixture models (GMMs). This architecture is called DNN/GMM tandem connectionist anomaly detection. Experimental comparisons using vibration signals from actual wind turbine components demonstrated that the developed tandem connectionist system yielded significant improvements over existing systems, and that the representation learning performed robustly with respect to differences in machine types.
AB - An effective use of faulty-state data is proposed to achieve robust, accurate data-driven anomaly (fault) detection for rotating machine. Although using faulty data in the training process generally can improve the performance of anomaly detection system, it is rare to obtain enough samples to train failures or defects on a target machine. We therefore utilize the existing data from non-target (different-type) machines for feature representation learning to improve anomaly detection for the target machine. Specifically, deep neural networks (DNNs) that are trained to discriminate the normal and faulty states of the non-target machines are used to extract features. The extracted features are then taken as inputs to an anomaly detector based on Gaussian mixture models (GMMs). This architecture is called DNN/GMM tandem connectionist anomaly detection. Experimental comparisons using vibration signals from actual wind turbine components demonstrated that the developed tandem connectionist system yielded significant improvements over existing systems, and that the representation learning performed robustly with respect to differences in machine types.
KW - Anomaly detection
KW - Condition monitoring
KW - Data-driven method
KW - Machine learning
KW - Representation learning
KW - Vibration signals
KW - Wind turbine
UR - http://www.scopus.com/inward/record.url?scp=85062833817&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85062833817&partnerID=8YFLogxK
U2 - 10.1109/ICPHM.2018.8448450
DO - 10.1109/ICPHM.2018.8448450
M3 - Conference contribution
AN - SCOPUS:85062833817
T3 - 2018 IEEE International Conference on Prognostics and Health Management, ICPHM 2018
BT - 2018 IEEE International Conference on Prognostics and Health Management, ICPHM 2018
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 11 June 2018 through 13 June 2018
ER -