Nuclear Engineering and Technology

Korean Nuclear Society (한국원자력학회)
권호 표지
DOI QR코드

DOI QR Code

Quantification of predicted uncertainty for a data-based model

  • Chai, Jangbom (Dept. of Mechanical Engineering, Ajou University) ;
  • Kim, Taeyun (Dept. of Mechanical Engineering, Ajou University)
  • Received : 2019.12.12
  • Accepted : 2020.08.04
  • Published : 2021.03.25
Download PDF
⟨ Previous Next ⟩

Abstract

A data-based model, such as an AAKR model is widely used for monitoring the drifts of sensors in nuclear power plants. However, since a training dataset and a test dataset for a data-based model cannot be constructed with the data from all the possible states, the model uncertainty cannot be good enough to represent the uncertainty of estimations. In fact, the errors of estimation grow much bigger if the incoming data come from inexperienced states. To overcome this limitation of the model uncertainty, a new measure of uncertainty for a data-based model is developed and the predicted uncertainty is introduced. The predicted uncertainty is defined in every estimation according to the incoming data. In this paper, the AAKR model is used as a data-based model. The predicted uncertainty is similar in magnitude to the model uncertainty when the estimation is made for the incoming data from the experienced states but it goes bigger otherwise. The characteristics of the predicted model uncertainty are studied and the usefulness is demonstrated with the pressure signals measured in the flow-loop system. It is expected that the predicted uncertainty can quite reduce the false alarm by using the variable threshold instead of the fixed threshold.

Keywords

Acknowledgement

This work was supported by the Nuclear Safety Research Program through the Korea Foundation Of Nuclear Safety(KoFONS) using the financial resource granted by the Nuclear Safety and Security Commission(NSSC) of the Republic of Korea (No. 1805007).

References

  1. H. Hashemian, On-line monitoring applications in nuclear power plants, Prog. Nucl. Energy 53 (2) (2011) 167-181. https://doi.org/10.1016/j.pnucene.2010.08.003
  2. J.B. Coble, R.M. Meyer, P. Ramuhalli, L.J. Bond, H. Hashemian, B. Shumaker, D. Cummins, A Review of Sensor Calibration Monitoring for Calibration Interval Extension in Nuclear Power Plants. No. PNNL-21687. Pacific Northwest National Lab(PNNL), 2012. Richland, WA USA.
  3. J.W. Hines, R. Seibert, Technical Review of On-Line Monitoring Techniques for Performance Assessment (NUREG/CR-6895) Vol. 1. State-Of-The-Art, US Nuclear Regulatory Committee, 2006.
  4. C.K. Williams, C.E. Rasmussen, Gaussian Processes for Machine Learning, MIT Press, 2006.
  5. S. Lee, J. Chai, An enhanced prediction model for the on-line monitoring method using the Gaussian process regression, Journal of Mechanical Science and Technology. 33 (2019) 2249-2257, in re-view. https://doi.org/10.1007/s12206-019-0426-7
  6. Yi Liu, Qing-Yang Wu, Junghui Chen, Active selection of informative data for sequential quality enhancement of soft sensor models with latent variables, Ind. Eng. Chem. Res. 56 (16) (2017) 4804-4817. https://doi.org/10.1021/acs.iecr.6b04620
  7. J.W. Hines, R. Seibert, Technical Review of On-Line Monitoring Techniques for Performance Assessment (NUREG/CR-6895) Vol. 2, in: Theoretical Issues, US Nuclear Regulatory Committee, 2008. May.
  8. F. Di Maio, P. Baraldi, E. Zeo, R. Seraoui, Fault detection in nuclear power plants components by a combination of statistical methods, IEEE Trans. Reliab. 62 (4) (2013) 833-845. https://doi.org/10.1109/TR.2013.2285033
  9. N. Sairam, S. Mandal, Thermocouple Sensor Fault Detection Using AutoAssociative Kernel Regression and Generalized Likelihood Ratio Test., Computer, Electrical & Communication Engineering (ICCECE), 2016 International Conference on, IEEE, 2016, pp. 1-6.