Izvestiya vuzov. Yadernaya Energetika

The peer-reviewed scientific and technology journal. ISSN: 0204-3327

Expert Neural Network System for Diagnosing Еlectrically Actuated Valves

9/23/2021 2021 - #03 Global safety, reliability and diagnostics of nuclear power installations

Trykov E.L. Trykova I.V. Kotsoev K.I.

DOI: https://doi.org/10.26583/npe.2021.3.06

UDC: 621.646

One of the main factors for the safe operation of a nuclear power plant (NPP) is the trouble-free operation of electrically actuated valves (EAVs). The EAVs, being an important part of critical equipment, are included both in security systems and safety-related systems. Therefore, the highest requirements are imposed on the reliability of the EAVs.

The EAVs constitute the most numerous class of NPP equipment. Depending on the design, one power unit contains from 1,500 to 3,000 units of electrically actuated valves only. As the analyses of failures in the operation of NPPs show, a significant part of them is associated with failures of electrically actuated valves of security systems and safety-related systems. The main reason for sudden failures of valves is the lack of control of their technical condition during power operation.

Violation of the EAV operation can lead to a decrease in the safety of the NPP power unit as a whole. Considering that under the conditions of an operating power unit, the EAVs are often located in rooms with an increased radiation level, it is impossible to use contact diagnostic methods. Currently, the EAVs are diagnosed by a current and voltage signal recorded from the stator windings of an electric motor (EM).

Difficulty in diagnosing EAVs arises due to the fact that during operation the valves are exposed to a significant number of factors, often random ones (changing parameters of the working environment). As a result, there is a dispersion of the parameters of the technical state of the EAVs. At the same time, the higher the variation of the technical condition parameters, the less effective the routine maintenance and repair schemes, since, in this case, there is always a factor of uncertainty in the technical condition of the object.

This paper describes an automated diagnostic system for shut-off and regulating EAVs used in NPP pipelines.


  1. Adamenkov A. K. Diagnostic Support for the Transition to Maintenance and Repair of Shut1off and Control Valves at NPP Based on their Technical Condition. Cand. Sci. (Engineering) Diss. Abstr. Volgodonsk. VITI Publ., 2009, 26 p. (in Russian).
  2. Slepov M.T., Sysoev N.P. Еlectrodrived valves diagnostics – experience of the Novovoronezh NPP. Globalnaya Yadernaya Bezopasnost’. 2014, no. 2 (11), pp. 79-85 (in Russian).
  3. MT Diagnosting of Pipeline Еlectrodrived Valves. Methodology. Put into act by the order of Rosenergoatom Concern JSC dated 27.03.2012. No. 9270-17 (in Russian).
  4. Matveev A.V., Zhidkov S.V., Adamenkov A. K.., Galivets E.Yu., Usanov D.A. An Integrated Approach to Diagnosing Еlectrodrived Valves to Applied to the Tasks of Resource Management. Armaturostroenie. 2009, no. 2 (59), pp. 53-59 (in Russian).
  5. Abidova E. A., Malik O.V., Gavrilenko D.S. System of Electrodriving Fittings Diagnostic Signals Processing. Globalnaya Yadernaya Bezopasnost’. 2015, no. 2 (15), pp. 56-62 (in Russian).
  6. Matveev A.V., Skladnikov A.F. Diagnosting of Еlectrodrived Valves Using Active Power Parameters. Armaturostroenie. 2009, no. 3 (60), pp. 67-71. Available at: https://docplayer.ru/74845967-Diagnostirovanie-armatury.html (accessed Feb. 20, 2021) (in Russian).
  7. Matveev A.V., Skladnikov A.F. Diagnosting of Еlectrodrived Valves Using Relative Electrical Parameters. Armaturostroenie. 2009, no. 3 (60), pp. 40-51. Available at: http://www.diaprom.com/publication/?id=5/ (accessed Feb. 20, 2021) (in Russian).
  8. Perslev M., Jensen M.H., Darkner S., Jennum P.J., Igel C. U-Time: A Fully Convolutional Network for Time Series Segmentation Applied to Sleep Staging. arXiv:1910.11162v1 [cs.LG] 24 Oct 2019, 19 p.
  9. Kotsoev K.I. Trykov E.L., Trykova I.V.. Application of convolutional neural network for segmentation of electric drived valves signals. Izvestiya vuzov. Yadernaya Energetika. 2021, no. 2, pp. 158-168; DOI: https://doi.org/10.26583/npe/2021.2.14 (in Russian).
  10. Abidova E.A., Hegay L.S., Chernov A.V., Draka O.E., Pugachyova O.J. Diagnosis of Valves Using Entropy Indicators. Globalnaya Yadernaya Bezopasnost’. 2016, no. 4 (21), pp. 69-77 (in Russian).
  11. Abidova E.A., Hegay L.S., Chernov A.V., Bulava V.A., Soloviev V.I., Pugachyova O.Yu. Diesel Generator Equipment Diagnostics of Nuclear Power Plant Using Deterministic and Stochastic Methods. Globalnaya Yadernaya Bezopasnost’. 2016, no. 3 (20), pp. 74-79 (in Russian).
  12. Chumak O.V. Entropies and Fractals in Data Analysis. Moscow-Izhevsk. NITs «Regulyarnaya i Khaoticheskaya Dinamika», Institut Kompyuternykh Issledovaniy Publ., 2011, 164 p. (in Russian).
  13. Hegay L.S. Software and Hardware Complex of the Diagnostic System for Electromechanical Equipment of NPP Based on Entropy Parameterization of Diagnostic Signals. Cand. Sci. (Engineering) Diss. Volgodonsk. VITI NIYaU MIFI Publ., 2017, 138 p. (in Russian).

NPP safety electrically actuated valves pipelines diagnosing neural networks segmentation automated system