Nuclear Engineering and Technology

Korean Nuclear Society (한국원자력학회)
권호 표지
DOI QR코드

DOI QR Code

Maintenance-based prognostics of nuclear plant equipment for long-term operation

  • Received : 2017.06.05
  • Accepted : 2017.06.06
  • Published : 2017.08.25
Download PDF
⟨ Previous Next ⟩

Abstract

While industry understands the importance of keeping equipment operational and well maintained, the importance of tracking maintenance information in reliability models is often overlooked. Prognostic models can be used to predict the failure times of critical equipment, but more often than not, these models assume that all maintenance actions are the same or do not consider maintenance at all. This study investigates the influence of integrating maintenance information on prognostic model prediction accuracy. By incorporating maintenance information to develop maintenance-dependent prognostic models, prediction accuracy was improved by more than 40% compared with traditional maintenance-independent models. This study acts as a proof of concept, showing the importance of utilizing maintenance information in modern prognostics for industrial equipment.

Keywords

References

  1. J. Coble, J.W. Hines, Prognostic algorithm categorization with application to the PHM challenge problem, in: Prognostics and Health Management Conference PHM, Denver, CO, USA, 2008.
  2. M. Sharp, J. Coble, A. Nam, J.W. Hines, B. Upadhyaya, Lifecycle prognostics: transitioning between information types, Proc. Inst. Mech. Eng. O J. Risk Reliab. 229 (4) (2014) 279-290.
  3. A. Nam, B. Jeffries, J.W. Hines, B. Upadhyaya, A Bayesian Statistical Updating Method: Error Reduction in Remaining Useful Life Estimation, NPIC-HMIT, University of Tennessee, 2015.
  4. A. Jardine, A. Tsang, Maintenance, Replacement, and Reliability: Theory and Applications, second ed., CRC Press, Florida, USA, 2013.
  5. J. Moubray, Reliability-Centered Maintenance, Industrial Press Inc, Great Britain, 1997.
  6. S. Martorell, A. Sanchez, V. Serradell, Age-dependent reliability model considering effects of maintenance and working conditions, Reliab. Eng. Syst. Saf. 64 (1999) 19-31. https://doi.org/10.1016/S0951-8320(98)00050-7
  7. Z. Welz, A. Nam, M. Sharp, J.W. Hines, B. Upadhyaya, Improved heat exchanger lifecycle prognostic methods for enhanced light water reactor sustainability, Int. J. Prognost. Health Manag. 6 (2015) 1-13.
  8. F. Incropera, D. DeWitt, T. Bergman, A. Lavine, Fundamentals of Heat and Mass Transfer, sixth ed., Wiley, Jefferson City, 2007.
  9. D. Specht, A general regression neural network, IEEE Trans. Neural Netw. 2 (1991) 568-576. https://doi.org/10.1109/72.97934
  10. W. Weibull, A statistical distribution function of wide applicability, J. Appl. Mech. 18 (1951) 293-297.
  11. C. Lu, W. Meeker, Using degradation measures to estimate a time-to-failure distribution, Technometrics 35 (1993) 161-174. https://doi.org/10.1080/00401706.1993.10485038
  12. J. Coble, J.W. Hines, Applying the general path model to estimation of remaining useful life, Int. J. Prognost. Health Manag. 2 (2011) 71-82.
  13. A. Saxena, J. Celaya, B. Saha, S. Saha, K. Goebel, Metrics for offline evaluation of prognostic performance, Int. J. Prognost. Health Manag. 1 (2010) 4-23.
  14. M. Sharp, Simple metrics for evaluating and conveying prognostic model performance to users with varied backgrounds, in: Annual Conference of the Prognostics and Health Management Society, 2013.