Efficient Method for Filtering Global Noises in Measuring Channels of Leak Control Systems at Nuclear Power Plants with VVER Reactors
At NPPs with VVER reactors, in accordance with GOST R 58328-2018 «Pipelines of nuclear power plants. The “leak-before-failure” concept» , the acoustic leak monitoring system (ALMS) and the humidity leak monitoring system (HLMS) are used, with each of the systems performing the leak control function locally, independently of each other.
The results of diagnostics are transferred to the In-Core Monitoring System (ICMS) for subsequent display to the operating personnel at the control room. In addition, a integrated diagnostics system (IDS) is provided, designed, among other things, to confirm the diagnosis and clarify the values of the magnitude and coordinates of the leak based on the analysis of the readings of the leak control systems and the signals of the automatic process control system (APCS). The readings of the measuring channels of the systems are made up of background noises, the sources of which are technological processes from the main equipment and auxiliary systems of the reactor plant and the signal of a leak, when it appears. The most important factor affecting the ability of leak monitoring systems to diagnose leakage is the quality of background noise filtering. The paper proposes a new effective method for filtering global noise, intended for use an integrated diagnostics system.
- GOST R 58328-2018 «Piping of nuclear power plants. “Leak before break” concept». Available at: https://files.stroyinf.ru/Data/705/70505.pdf (accessed May 05, 2020) (in Russian).
- Skomorokhov, A.O., Kudryaev A.A. Morozov S.A. Neural network models of signal filtration and diagnosis of VVER pipeline leaks. Izvestia Vysshikh Uchebnykh Zawedeniy. Yadernaya Energetika. 2010, no. 4, pp. 72-80 (in Russian).
- Goodfellow I., Bengio Y., Courville A. Deep Learning. MIT Press, 2016. 787 p.
- Nikolayenko S., Kadurin A., Arkhangelskaya E. Deep Learning. Saint Petersburg. Piter Publ., 2018, 480 p. (in Russian).
- Chollet F. Deep Learning with Python: Second edition. Manning Publications. 2017. 384 p.
- Kruschke J.K. Bayesian estimation supersedes the T test. Journal of Experimental Psychology: General. 2013, v. 142, no. 2, pp. 573-603; DOI: 10.1037/a0029146.
- Cameron D.-P. Bayesian Methods for Hackers: Probabilistic Programming and Bayesian Inference. Addison-Wesley Data and Analytics Series. 2016. 226 p.
- Barber D. Bayesian Reasoning and Machine Learning. Cambridge University Press. 2017. 666 p.
- Najim M. Modeling Estimation and Optimal Filtering in Signal Processing. Wiley. 2008. 408 р.
- Durbin J. and Koopman S.J. Time Series Analysis by State Space Methods: Second Edition. Oxford Statistical Science Series. OUP Oxford, 2012. 253 p.
- Haykin S. Adaptive Filter Theory: Fifth Edition. Pearson. 2014. 907 p.
- Grewal M.S. and Andrews A.P. Kalman Filtering: Fourth Edition. Wiley. 2015. 617 р.