A.R. Bestugin1, E.A. Gushchina2, E.A. Frolova3

1-3 Saint-Petersburg State University of Aerospace Instrumentation (Saint-Petersburg, Russia)

1 fresguap@mail.ru; 2guschina_guap@mail.ru; 3frolovaelena@mail.ru

Modern methods of determining the optimal composition of polymer composite materials are based on multi-stage experimental studies that require significant time, financial and labor resources. The traditional approach involves a large number of laboratory tests to select components and their ratios, which leads to high costs and increases the environmental load due to energy consumption, chemical reagents and waste generation. In this regard, the actual task is the development of alternative methods that allow to reduce the volume of experimental studies without losing the accuracy of predicting the properties of materials.

The aim of the work is to develop a methodology for automated selection of polymer composite compositions, which allows to significantly reduce the number of tests and minimize costs.

In the course of the research, a model using machine learning and statistical analysis methods was developed to determine the composition of polymer composites based on a set of experimental data. The proposed approach demonstrates high efficiency, reducing the number of required tests by 19 times compared to traditional methods. The application of the technique provides significant economic benefits, reducing the cost of research from 17-73 million rubles to 0.9-4 million rubles depending on the complexity of the composition of composite material.

An important achievement of the work is a significant reduction of environmental load - CO₂ emissions during the development process are reduced from 605 kg to 32 kg per research cycle. For practical implementation of the method a specialized software interface was created, which provides convenient input of initial data, automated selection of optimal compositions and clear visualization of the obtained results.

The developed methodology makes it possible to significantly accelerate the creation of new polymer composites with specified properties, which is especially in demand in the aircraft, automotive and medical industries. Reduction of R&D costs makes the technology affordable for small and medium-sized enterprises, and minimization of environmental impact corresponds to the principles of sustainable development. The developed software product can be easily integrated into existing computer-aided design systems, ensuring quick implementation into production processes.

Bestugin A.R., Gushchina E.A., Frolova E.A. Methodology of diagnostics of polymer composite materials composition. Radiotekhnika. 2025. V. 89. № 5. P. 166−172. DOI: https://doi.org/10.18127/j00338486-202505-18 (In Russian)

1. Semin A.Ju., Skopinova A.V., Pleshanov A.N. Analiz konstruktivnyh reshenij, primenjaemyh pri proektirovanii bortovyh registratorov kosmicheskih sistem global'nogo monitoringa. Radiotehnika. 2023. T. 87. № 4. S. 32-35. DOI: https://doi.org/10.18127/j00338486-202304-05 (in Russian).
2. Mason R.D., Lind D.A. Statistical techniques in business and economics. Irwin/McGraw Hill. 1999. 791 p.
3. Zhao J., Ramos N. M. M., Simoes M. L., et al. Application of clustering technique for definition of generic objects in a material database. Journal of Building Physics. 2015. V. 39. № 2. P. 124-146.
4. Jiang Z., Zhang Z., Friedrich K. Prediction on wear properties of polymer composites with artificial neural networks. Composites Science and Technology. 2007. V. 67. № 2. P. 168-176.
5. Hamidi Y.K., Berrado A., Altan M.C. Machine learning applications in polymer composites. AIP Conference Proceedings. AIP Publishing. 2020. V. 2205. № 1.
6. Tao L., Chen G., Li Y. Machine learning discovery of high-temperature polymers. Patterns. 2021. V. 2. № 4. P. 100225.
7. GOST 11262-2017 (ISO 527-2:2012) Plastmassy. Metod ispytanija na rastjazhenie. M.: Standartinform. 2018. 20 s. (in Russian).
8. Komarovskaja A.S. Vozdejstvie predprijatija po pererabotke othodov plastika na atmosfernyj vozduh. Materialy XVI Mezhdunar. nauch.-praktich. konf. studentov, aspirantov i molodyh uchenyh «Masherovskie chtenija». Vitebsk. 2022. T. 1 (in Russian).
9. Ustojchivoe razvitie [jelektronnyj resurs] URL: https://www.un.org/sustainabledevelopment/sustainable-consumption-production/ data obrashhenija 14.04.2025 (in Russian).
10. Gushhina E.A. Proektirovanie bazy dannyh polimernyh kompozitnyh materialov. Vestnik Kazanskogo gos. tehnich. un-ta im. A.N. Tupoleva. 2024. № 2. S. 46-49 (in Russian).
11. GOST 14236-81 Plenki polimernye. Metod ispytanija na rastjazhenie. M.: Izdatel'stvo standartov. 1981. 10 s. (in Russian).
12. GOST 33693-2015 (ISO 20753:2008) Plastmassy. Obrazcy dlja ispytanija. M.: Standartinform. 2016. 16 s. (in Russian).
13. GOST 12423-2013 (ISO 291:2008) Plastmassy. Uslovija kondicionirovanija i ispytanija obrazcov (prob). M.: Standartinform, 2014. 8 s. (in Russian).