|

Developing an Algorithm for Determining Malfunctions in the Measuring Channels of the Acoustic Leak Control System

Authors: Kotsoev K.I., Trykov E.L., Kudryaev A.A., Perevezentsev V.V. Published: 11.09.2021
Published in issue: #3(136)/2021  
DOI: 10.18698/0236-3933-2021-3-100-112

 
Category: Instrument Engineering, Metrology, Information-Measuring Instruments and Systems | Chapter: Acoustic  
Keywords: acoustic leak monitoring systems, acoustic sensors, diagnostics, neural network, classification

Nowadays Russian and foreign NPP operate systems for monitoring coolant leakage at the primary circuit based on measuring the dispersion of generated acoustic signals (acoustic waves) propagating over the metal surface. In the acoustic leak monitoring systems provision are made to self-diagnosis of measuring channels, as well as an adaptive algorithm is applied that allows automatic readjusting to the use of neighboring measuring channels instead of those that have failed. At the same time, there may be such malfunctions in the system technical means that do not automatically diagnose the malfunction of the measuring channels, which may lead to the failure of the system function to determine the magnitude and coordinate of the leak of the primary circuit coolant. That is why, the task of developing algorithms for determining the malfunction of the measuring channels of the acoustic leak monitoring system, implemented using software without making changes to the technical means of the system, is urgent. An algorithm is proposed for determining the malfunction of the measuring channels of the acoustic leak control system using a test signal of increased duration. An analysis of the applicability of the algorithm was performed on a representative sample of signals from measuring channels of the acoustic leak monitoring system of campaign 2018--2019 Novovoronezh NPP-2 рower Unit no. 1. The proposed algorithm has been implemented in the testing mode at power Unit no. 1 of Novovoronezh NPP-2 since the start of a new campaign in July 2019

References

[1] Arkadov G.V., Pavelko V.I., Finkel’ B.M. Sistemy diagnostirovaniya VVER [VVER diagnostics sytems]. Moscow, Energoatomizdat Publ., 2010.

[2] Shvetsov D.M., Trykov E.L., Leskin S.T., et al. Analysis of acoustic signals of leak for increases in sensitivity of control due to creation of effective diagnostic features. Izvestiya vysshikh uchebnykh zavedeniy. Yadernaya energetika, 2018, no. 1, pp. 112--121 (in Russ.). DOI: https://doi.org/10.26583/NPE.2018.1.11

[3] Katser Yu.D., Kozitsin V.O., Maksimov I.V. NPP equipment fault detection methods. Izvestiya vysshikh uchebnykh zavedeniy. Yadernaya energetika, 2019, no. 4, pp. 5--27 (in Russ.). DOI: https://doi.org/10.26583/npe.2019.4.01

[4] Trykov E.L., Trykova I.V., Kotsoev K.I. Detection of anomalies in the reactor equipment operation using neural network algorithms. Izvestiya vysshikh uchebnykh zavedeniy. Yadernaya energetika, 2020, no. 3, pp. 136--147 (in Russ.). DOI: https://doi.org/10.26583/npe.2020.3.14

[5] Leskin S.T. Development of algorithms for abnormality detection in NPP equipment by analyzing data of operating technological control. Izvestiya vysshikh uchebnykh zavedeniy. Yadernaya energetika, 1997, no. 4, pp. 4--12 (in Russ.).

[6] Chiang L.H., Russell E.L., Braatz R.D. Fault detection and diagnosis in industrial systems. London, Springer Verlag, 2001.

[7] Muller A.C., Guido S.S. Introduction to machine learning with Python. O’Reilly Media, 2016.

[8] Michie D., Spiegelhalter D.J., Taylor C.C. Machine learning, neural and statistical classification. Overseas Press, 2009.

[9] Cai B., Huang L., Xie M. Bayesian networks in fault diagnosis. IEEE Trans. Industr. Inform., 2017, vol. 13, no. 5, pp. 2227--2240. DOI: https://doi.org/10.1109/TII.2017.2695583

[10] Buyankin V.M., Zakharov V.G. Neuroidentification for control electrodrive artificial neural network fuzzy. Vestnik MADI, 2009, no. 4, pp. 22--25 (in Russ.).

[11] Ardakani M.H., Shokry A., Escudero G., et al. Unsupervised automatic updating of classification models of fault diagnosis for novelty detection. Comput. Aided Chem. Eng., 2018, vol. 43, pp. 1123--1128.DOI: https://doi.org/10.1016/B978-0-444-64235-6.50196-0

[12] Van Ryzin J., ed. Classification and clustering. Academic Press, 1977.

[13] Nilsson N.J. Introduction to machine learning. Stanford University, 1998.

[14] Gayer A.V., Sheshkus A.V., Chernyshova Yu.S. Augmentation on the fly for the neural networks learning. Trudy Instituta sistemnogo analiza RAN [Proceedings of the Institute for Systems Analysis, Russian Academy of Sciences], 2018, vol. 68, no. S1, pp. 150--157 (in Russ.). DOI: https://doi.org/10.14357/20790279180517

[15] Chiang L.H., Russell E.L., Braatz R.D. Fault detection and diagnosis in industrial systems. London, Springer Verlag, 2000.