Increasing the Safety of Operation of Industrial Technical Systems from the Composite Materials by Predicting Their Life on the Basis of New Methods of Non-Destructive Control and Deep Neural Networks


The problem is considered related to increase of the operational safety of industrial facilities made of composite materials by means of an a priori assessment of the maximum service life.

Two tasks are being solved: development of the new methods and means of non-destructive testing allowing to identify the defects that appear in the process of testing products with various loads and in the process of their operation; development of the new methods and means for assessing service life of the products based on the results of non-destructive testing.

The first problem is being solved by the development of optical-thermographic non-destructive testing, including the technologies of ultrasonic thermotomography and electric force thermography, which determine the state of the object by dynamic temperature fields and optical control technology based on the fiber-optic sensors that measure the amount of material internal deformation under a force effect on the structure.

Solution to the second problem is based on the use of neural network analysis (artificial neural networks) for assessment and prediction of the service life using the results of non-destructive testing with preliminary training of the neural network.

An estimate was obtained by the experimental studies related to the error in determining the products service life, which is 12.6 %.

The implementation of the proposed approach will allow to create the new technologies for predicting the service life of elements and structures made of composite materials using the results of non-destructive testing, which will provide an additional opportunity for developing practical recommendations on the confirmation or extension of the service life and improvement of safety for structures operation.

  1. Ostrik A.V., Nikolaev D.N., Cheprunov A.A. Explosive Technologies for Strength Tests of Thin-Walled Composite Constructions on Action of Side Non-Stationary Loadings Having Various Physical Nature. Konstruktsii iz kompozitsionnykh materialov = Composite Materials Constructions. 2019. № 3. pp. 55–63. (In Russ.).
  2. Storodubtseva T.N., Tomilin A.I. Composite structural materials, methods for studying their properties. Sovremennye naukoemkie tekhnologii = Modern high technologies. 2012. № 11. pp. 53–54. (In Russ.).
  3. Zershchikov K.Yu. Investigation of Stress Relaxation in Sealing Units with Polymer or Composite Sealings Used in Pipeline Valves. Konstruktsii iz kompozitsionnykh materialov = Composite Materials Constructions. 2018. № 1. pp. 50–55. (In Russ.).
  4. Kaledin V.O., Vyachkina E.A., Vyachkin E.S., Budadin O.N., Kozelskaya S.O. Applying Ultrasonic Thermotomography and Electric-Loading Thermography for Thermal Characterization of Small-Sized Defects in Complex-Shaped Spatial Composite Structures. Russian Journal of Nondestructive Testing. 2020. Vol. 56. № 1. pp. 58–69.
  5. Kozelskaya S.O. Computer System of Ultrasonic Thermotomography for Detecting Small-sized Defects in Products from Polymer Composite Materials. Promyshlennye ASU i kontrollery = Industrial Automatic Control Systems and Controllers. 2020. № 7. pp. 9–16. (In Russ.). DOI: 10.25791/asu.7.2020.1197
  6. Budadin O.N., Kulkov A.A., Kozelskaya S.O., Kaledin V.O., Vyachkin E.S. Method of Electric Power Thermography of Spatial Objects and Device for its Implementation. Patent RF № 2690033. Applied: September 14, 2018. Published: May 30, 2019. (In Russ.).
  7. Larin A.A., Fedotov M.Yu., Bukharov S.V., Reznichenko V.I. New Applications Systems of Fiber-Optical Sensors. Prikladnaya fotonika = Applied Photonics. 2017. № 4. pp. 311–324. (In Russ.).
  8. Nazirov M.F., Zhukov Yu.A., Yakovitskaya S.Yu. Strain Measurement of Carbon Fiber Reinforced Plastic (CFRP) Samples with Embedded Optical Fiber Sensors.
  9. Voprosy oboronnoy tekhniki. Seriya 16: Tekhnicheskie sredstva protivodeystviya terrorizmu = Military Enginery. Counter-terrorism technical devices. Issue 16. 2015. № 9–10 (87–88). pp. 95–101. (In Russ.).
  10. Morozova T.Yu., Bekarevich A.A., Budadin O.N. The New Approach to Identification of Product Defects. Kontrol. Diagnostika = Testing. Diagnostics. 2014. № 8. pp. 42–48. (In Russ.). DOI: 10.14489/td.2014.08.pp.042-048
  11. Bekarevich A.A., Budadin O.N., Morozova T.Yu., Toporov V.I. The method for adaptive forecast of the operation residual resource of the complex objects and the device for its implementation. Patent RF № 2533321. Applied: June 28, 2013. Published: November 20, 2014. Bulletin № 32. (In Russ.).
  12. Liu W., Wang Z., Liu X., Zeng N., Liu Y., Alsaadi F.E. A survey of deep neural network architectures and their applications. Neurocomputing. 2017. Vol. 234. pp. 11–26. DOI: 10.1016/j.neucom.2016.12.038
  13. Akimov D.A., Dyatchenkova A.Yu., Sachkov V.E. Self-Diagnostics of Technical Nodes of the Automobot in the Conditions of Incomplete Information is Based on Abduction. Sovremennaya nauka: aktualnye problemy teorii i praktiki. Seriya: Estestvennye i tekhnicheskie nauki = Modern Science: actual problems of theory and practic. Series «Natural & Technical Sciences». 2018. № 2. pp. 18–24. (In Russ.).
  14. Cox P.T., Pietrzykowski T. General diagnosis by abductive inference. Proc. IEEE Sympos. Logic Programming. San Francisco, 1987. pp. 183–189.
  15. Kotelnikov V.V., Akimov D.A. Training of convolutional neural networks for predicting and assessing the level of criticality of structures defects. Novshestva v oblasti tekhnicheskikh nauk: sb. nauch. tr. po itogam Mezhdunar. nauch.-prakt. konferentsii. (Innovations in the field of technical sciences: collection of research papers on the results of the International scientific-practical conference). Tyumen, 2016. pp. 73–77. (In Russ.).
  16. Sozykin A.V. An Overview of Methods for Deep Learning in Neural Networks. Vestnik Yuzhno-Uralskogo gosudarstvennogo universiteta. Seriya. Vychislitelnaya matematika i informatika = Bulletin of the South Ural State University. Series «Computational Mathematics and Software Engineering». 2017. Vol. 6. № 3. pp. 28–59. (In Russ.). DOI: 10.14529/cmse170303
DOI: 10.24000/0409-2961-2021-4-7-12
Year: 2021
Issue num: April
Keywords : non-destructive testing operational safety service life artificial neural network composite materials data processing
  • Kozelskaya S.O.
    Kozelskaya S.O.
    Cand. Sci. (Eng.), Senior Research Assistant, Joint Stock Company «Central Research Institute of Special Engineering», Khotkovo, Russia