Application of the fuzzy data theory in durability estimation

Aim. Problems associated with the study of material fatigue, while being relevant in terms of engineering practice, have a significant degree of uncertainty. The fatigue curve is censored (which indicates the presence of items that have passed the planned load cycle and destroyed by the end of the tests), while the load block made for calculating durability can be designed in a number of ways with a sufficient share of subjective decisions. The load block is intended for calculating durability and defining test plans. It is to fully reflect the entire expected operational history. Both factors are considered in the paper as elements of fuzzy logic. The author examines the creation of a scientifically substantiated load block that would take into account the possible operating modes in a right proportion and taking into account the variability. That is due to the fact that fatigue damage accumulates over the entire life of a machine and is to be scientifically evaluated for an adequate probabilistic assessment.

Methods. As the modes of operation of a certain part are not precisely defined (and cannot be defined by virtue of the logic of random use of machines), projections of random fuzzy distributions are considered. A finite set of operating modes in a reasonable proportion was successfully scientifically substantiated. Using the example of load analysis of a critical part of rolling stock, distributions were constructed and the possible distribution of a part’s life was estimated. The output of the developed method will allow assessing the operational risks and predict the required number of spare parts. By taking into account the censored sample elements in the process of fatigue curve construction, the estimation of the fatigue curve parameters can be made more consistent.

Conclusions. The use of fuzzy sets may prove to be very useful when examining fatigue curves and estimating durability variation. Examples are given of applying the proposed method.

1. . Orlov A.I. [Artificial Intelligence: Non-numerical Statistics]. Moscow: IPR Media; 2022. Available at: https://www.iprbookshop.ru/117028.html.

2. . Orlov A.I. Change of paradigms in applied statistics. Industrial laboratory. Diagnostics of materials 2021;87(7):67. (in Russ.)

3. . Kogaev V.P. [Time-variable strength calculation]. Moscow: Mashinostroenie; 1993. (in Russ.)

4. . Gadolina I.V., Petrova I.M. Cluster Analysis of the Loading Time-Series with the Aim of Consistent Durability Estimation. In: Hu Z., Wang B., Petoukhov S., He M., editors. Advances in Artificial Systems for Power Engineering II. AIPE 2021. Lecture Notes on Data Engineering and Communications Technologies; 119. Springer: Cham. DOI: https://doi.org/10.1007/978-3-030-97064-2_2.

5. . Gadolina I.V., Kozlov A.D., Monakhova A.A., et al. Optimal Decision for Digital Signal Processing in the Durability Assessment Problem. RSUH Bulletin. Series “Information Science. Information Security. Mathematics” 2019;1:78-93. DOI: 10.28995/2686-679X-2019-1-78-93.

6. . Burger M., Dreßler K., Speckert M. Load assumption process for durability design using new data sources and data analytics. International Journal of Fatigue 2021;145:7-8. DOI: https://doi.org/10.1016/j.ijfatigue.2020.106116.

7. . Dressler K., Speckert M., Müller R. et al. Customer loads correlation in truck engineering. In: Proceedings of the World Automotive Congress 2008. MünchenVolume: Springer automotive Media / GWV Fachverlage GmbH; F200812198.

8. . Bellec E. at al. Modelling and identification of fatigue load spectra: Application in the automotive industry. International Journal of Fatigue 2021;149. DOI: 10.1016/j.ijfatigue.2021.106222.

9. . Stadele M., Mundl B., Rap A., Großkopf R. et al. Derivation and categorization of road quality in view of operation loads to consider customer requirements for customer-focused sales. In: Proceedings of the Fourth International Conference on Material and Component Performance under Variable Amplitude Loading (VAL4). Berlin: DVM; 2020. Pp. 175-183.

10. . Aghabozorgi S., Shirkhorshidi A.S., Wah T.Y. Time-series clustering – A decade review. Information Systems 2015;53:16-38. DOI: http://dx.doi.org/10.1016/j.is.2015.04.007.

11. . Plyushchenko I.V., Shakhmatov D.G., Rodin I.A. Algorithm of combining chromatography mass spectrometry-untargeted profiling and multivariate analysis for identification of marker-substances in samples of complex composition. Industrial laboratory. Diagnostics of materials 2020;86(7):12-19. (in Russ.)

12. . Meyer D., Dimitriadou E., Hornik K., et al. E1071: Misc Functions of the Department of Statistics. Probability Theory Group, TU Wien; 2017. Available at: https://CRAN.R-project.org/package=e1071. R package version 1.6-8.

13. . Tucker L., Bussa S. The SAE Cumulative Fatigue Damage Test Program. In: Wetzel R.M., editor. Fatigue under Complex Loading; 1977.

14. . Petrova I.M., Gadolina I.V. [Creating a generalized loading spectrum for various operating loads]. In: Luganova Yu.A., editor. In: [Process Equipment for the Mining and Oil Sectors. Proceedings of the XVI International Science and Technology Conference as part of the Urals Mining Ten-Day Event];2018:318-321. (in Russ.)

15. . R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing; Vienna (Austria). (accessed 30.10.2018). Available at: http://www.R-project.org/>.

16. . Orlov A.I., Lutsenko E.V. [Systemic fuzzy interval mathematics. A monograph]. Krasnodar: KubSAU; 2014. (in Russ.)

17. . Terletskyi D.A., Provotar A.I. Fuzzy ObjectOriented Dynamic Networks. Cybernetics and Systems Analysis 2015;51(1):34-40. (in Russ.) DOI: https://doi.org/10.1007/s10559-015-9694-0.

18. . Savkin A.N., Andronik A.V, Sedov A.A. [Computer simulation and analysis of the strength of structures exposed to varying loading: A monograph]. Volgograd: VSTU; 2017. Available at: https://elibrary.ru/download/elibrary_29743916_86611459.pdf. (in Russ.)