Abstract
This paper discusses data-driven fault diagnosis of the power plant reheater tube leakage based on their operating data. From the temperature sensors, fault data and normal data are measured. Mahalanobis distance (MD) analysis was performed to quantitatively analyze whether the distribution of fault data differed from that of the normal data. Then, sequential probability ratio test (SPRT) was performed to determine the time to anomalies (TTAs). To verify detected TTAs, power-generation data was used. This paper demonstrated the feasibility of the proposed approach to detect reheater tube leakage prior to the failure.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Abbreviations
- A :
-
Upper boundary
- α :
-
The probability of accepting H1 when H0 is true
- B :
-
Lower boundary
- β :
-
The probability of accepting H0 when H1 is true
- H 0 :
-
Null hypothesis
- H 1 :
-
Alternative hypothesis
- L n :
-
Likelihood ratio
- n :
-
Number of variables
- s :
-
Scale factor
- Σ:
-
Covariance matrix of training data
- σ :
-
Standard deviation
- x :
-
Column vector of the test sample data
- \(\bar x\) :
-
Mean vector of the training data
- y k :
-
Sample data
References
R. J. Reichl, M. Schmidthaler and F. Schneider, Power outage cost evaluation: Reasoning, methods and an application, Journal of Scientific Research & Reports, 2 (2013) 249–276.
D. Kwon, M. R. Hodkiewicz, J. Fan, T. Shibutani and M. G. Pecht, IoT-based based prognostics and systems health management for industrial applications, IEEE Access, 4 (2016) 3659–3670.
S. S. Hamid, D. N. Jamal and M. Shajahan, Automatic detection and analysis of boiler tube leakage system, International Journal of Computer Applications, 84 (16) (2013) 19–23.
Q. Yang, Q. Yang and W. J. Yan, PSO based LS-SVM approach for fault prediction of primary air fan, 2015 Chinese Automation Congress (CAC) (2015).
K. Rostek, L. Morytko and A. Jankowska, Early detection and prediction of leaks in fluidized-bed boilers using artificial neural networks, Energy, 89 (2015) 914–935.
C. Leys, O. Kleina, Y. Dominicy and C. Leyc, Detecting multivariate outliers: Use a robust variant of the Mahalanobis distance, Journal of Experimental Social Psychology, 74 (2018) 150–156.
J. Yu, J. Jang, J. Yoo and S. Kim, Fault detection method for steam boiler tube using mahalanobis distance, Journal of Korean Institute of Intelligent Systems, 26 (3) (2016) 246–252.
M. Davarifar, A. Rabhi, A. Hajjaji and Z. Daneshifar, Real-time diagnosis of PV system by using the sequential probability ratio test (SPRT), 16th International Power Electronics and Motion Control Conference and Exposition (2014) 21–24.
C. G. Kenny and W. Lu, Early detection of signal and process anomalies in enterprise computing systems, Proceedings of the 2002 International Conference on Machine Learning and Applications (2002) 24–27.
Acknowledgments
This research was supported in part by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2017R1D1A1B03028604), and in part by National Research Foundation of Korea (NRF) grant funded by the Korea government the Ministry of Science, ICT & Future Planning (No. 2019R1A2C1090228).
Author information
Authors and Affiliations
Corresponding author
Additional information
Hongjun Choi was born in Pocheon city, Republic of Korea in 1995. He received the B.S. degree in Mechanical Engineering from Konkuk University, Seoul, Republic of Korea in 2019. He is currently working toward the M.S. degree in Mechanical Design and Production Engineering at Konkuk University, Seoul city, Republic of Korea. His research interests include PHM, vibration and noise of motor.
Chang-Wan Kim was born in Pohang city, Republic of Korea, in 1969. He received the B.S. degree in Mechanical Engineering from Hanyang University, Seoul, Republic of Korea, in 1987. He received the M.S. degree in Mechanical Engineering from Pohang University of Science and Technology, Pohang, Republic of Korea, in 1993. He received the M.S. degree in Computational Applied Mathematics, and Ph.D. degree in Aerospace and Engineering Mechanics from University of Texas at Austion, Texas, USA, in 1997 and 1999, respectively. He is currently a Professor of Department of Mechanical Engineering, Konkuk University, Seoul city, Korea. His research interests include vibration and noise analysis, multi-body dynamics, finite element analysis, and multi-physics analysis of system.
Daeil Kwon received the Bachelor’s degree in Mechanical Engineering from POSTECH, South Korea, and the Ph.D. degree in Mechanical Engineering from the University of Maryland, College Park, MD, USA. He was a Senior Reliability Engineer with Intel Corporation, Chandler, AZ, USA, where he developed use condition-based reliability models and methodologies for assessing package and system reliability performance. He is currently an Associate Professor with Sungkyunkwan University, Seoul, South Korea. His research interests are focused on prognostics and health management of electronics, reliability modeling, and use condition characterization.
Rights and permissions
About this article
Cite this article
Choi, H., Kim, CW. & Kwon, D. Data-driven fault diagnosis based on coal-fired power plant operating data. J Mech Sci Technol 34, 3931–3936 (2020). https://doi.org/10.1007/s12206-020-2202-0
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12206-020-2202-0