Application of neural network technologies to load monitoring in aircraft structure bearing elements

Economic efficiency in the operation of aircraft local air lines (LA) can be improved by increasing the awareness of the loading and aircraft structure integrity, with the subsequent adjustment of the maintenance program in accordance with the actual operating conditions. The allowance for specific features of loading load-bearing elements at transport aircraft during particular operations provides a significant extending in the safe life (in some cases, by several times). However, at the moment, load monitoring systems in the aviation industry have not yet been implemented universally. The implementation of the well-known approaches developed for monitoring and analyzing loads requires significant changes in the programs and procedures for maintaining airworthiness, including the need to install additional sophisticated measuring equipment. We propose an alternative approach to existing methods of load monitoring without using additional measuring equipment. The main stage is formation of the relationship between the flight parameters recorded by the standard on-board recorder and the loading parameters, which are determined by computational and experimental methods as a result of processing strain gauge data. A sufficient bulk of the strain data is usually obtained during the flight tests at the certification stage. Testing of this technique is considered with reference to the example of the loads in the elements of high lift devices of aircraft flaps. The average error in the estimates of the forces in the connecting rods of the inner flaps does not exceed 6%. The results obtained make it possible to determine with the acceptable accuracy the individual accumulated damageability of an aircraft structure and to assess the residual safe life. The presented approach is a part of the methodological framework necessary for the development and implementation of modern means of analyzing the integrity of the structure, implemented on the basis of on-board monitoring systems of local airlines aircraft.

1. Shibaev V. M., Matveev V. A., Gorodnichenko V. I., et al. Forecasting the Reliability Indicators of Unmanned Aircraft / Tr. TsAGI. 2021. N 2795. P. 4 – 14 [in Russian].

2. Sukhanov V. L., Shibaev V. M., Matveev V. A., et al. The method of scoring flight safety indicators in the certification of unmanned aircraft systems / Tekhn. Vozd. Flota. 2021. Vol. XCIII. N 3 – 4(736 – 737). P. 32 – 47 [in Russian].

3. RF Pat. 2545150, MPK G01M 17/02. Method of aircraft structure inspection / Vinokurov V. I., Zykov V. N.; applicant and patent holder Vinokurov V. I., Zykov V. N. — N 2014108188/11; appl. 03.03.14; publ. 27.03.15. Byull. N 9 [in Russian].

4. RF Pat. 2599108, MPKV 64F 5/00, G01D 21/00. A method for monitoring loads and accumulated fatigue damage under aircraft operating conditions / Tsymbalyuk V. I., Orlova T. I., Frolov A. V.; applicant and patent holder Federalnoe gosudarstvennoe unitarnoe predpriyatie «Tsentralny aerogidrodinamicheskii institut imeni professora N. E. Zhukovskogo» (FGUP «TsAGI») — N 2015127185/11; appl. 07.07.15; publ. 10.10.16, Byull. N 28 [in Russian].

5. Bautin A. A. Monitoring of the elements of aviation structures using strain-gauge measurement / Zavod. Lab. Diagn. Mater. 2019. Vol. 85. N 1(I). P. 57 – 63 [in Russian]. DOI: 10.26896/1028-6861-2019-85-1-I-57-63

6. Bautin A. A., Svirsky Yu. A., Pankov A. V. Development of structural health monitoring methods through the analysis of kinetics of local stress-strain state. / Zavod. Lab. Diagn. Mater. 2019. Vol. 85. N 2. P. 42 – 47 [in Russian]. DOI: 10.26896/1028-6861-2019-85-2-42-47

7. Urnev A. S., Chernyatin A. S., Matvienko Yu. G., Razumovskii I. A. Experimental and numerical sizing of a delamination defect in layered composite materials / Zavod. Lab. Diagn. Mater. 2018. Vol. 84. N 10. P. 59 – 66 [in Russian]. DOI: 10.26896/1028-6861-2018-84-10-59-66

8. Urnev A. S., Chernyatin A. S., Matvienko Yu. G., Razumovskii I. A., Gavrikov M. Yu. Studying the destruction kinetics of a composite panel with a embedded fiber-optic sensors grid / Mashinostr. Inzh. Obrazov. 2019. N 3. P. 18 – 27 [in Russian].

9. RF Pat. 2595066, MPK V64F 5/00, G01M 7/02, G06N 3/08. A method for assessing the loading of an aircraft structure during flight strength tests using artificial neural networks / Luchinskiy M. N., Arnautov E. V., Orlov A. A., Homenko A. G., Balashova T. A.; applicant and patent holder otkrytoe aktsionernoe obshchestvo «Lyotno-issledovatelskij institut imeni M. M. Gromova» — N 2015124804/11; appl. 24.06.15; publ. 20.08.16, Byull. N 28 [in Russian].

10. Zheng J., Jiao S., Cui D. Application of Principal Component Analysis-Assisted Neural Networks for the Rotor Blade Load Prediction / International Journal of Aerospace Engineering. 2021. Vol. 2021. Article ID 5594102. P. 13. DOI: 10.1155/2021/5594102

11. Candon M., Esposito M., Fayek H., et al. Advanced multi-input system identification for next generation aircraft loads monitoring using linear regression, neural networks and deep learning / Mechanical Systems and Signal Processing. 2022. Vol. 171. N 108809. P. 25. DOI: 10.1016/j.ymssp.2022.108809

12. Hornik K., Stinchcombe M., White H. Multilayer feedforward networks are universal approximations / Neural Networks. 1989. Vol. 2. P. 359 – 366.

13. Cybenko G. Approximation by superpositions of a sigmodial function / Mathematics of Control. Signals and Systems. 1989. Vol. 2. P. 303 – 314.

14. Funahashi K. On the approximate realization of continuous mappings by neural networks / Neural Networks. 1989. Vol. 2. N 3. P. 183 – 191.

15. Bautin A., Svirskiy Y. Applying Neural Networks for Multi-site Damage Detection in Fuselage Lap Joints of Cargo Aircraft / Lecture Notes in Civil Engineeringthis. 2021. Vol. 128. P. 67 – 76. DOI: 10.1007/978-3-030-64908-1_7

16. Bushgens A. G., Gorodnichenko V. I., Desyatnik P. I., Ovsyannikov M. O., Shibaev V. M., Studnev S. R. Assessment of safe service life of spatial disorientation simulators based on the analysis of fatigue strength and safety of structural component damage / Probl. Bezopasn. Poletov. 2020. № 12. P. 3 – 11 [in Russian].