Search academic texts

Information × Registration Number 0226U001445, (0124U000450) , R & D reports Title Algorithms, models, and tools of artificial intelligence for two-level modeling of complex materials behavior for dual-use technologies. popup.stage_title Створення наборів даних та регресійних моделей, що описують напружений, граничний та тепловий стан композитних, полікристалічних та мезопористих матеріалів. Head Vodka Oleksii О., Кандидат технічних наук Registration Date 27-01-2026 Organization National Technical University "Kharkiv Polytechnic Institute" popup.description1 The main objective of this project is to develop methods, algorithms, and mathematical models, as well as create information technology and software for the synthesis and analysis of microstructures of heterogeneous (polycrystalline, composite, mesoporous (aerogels), etc.) materials. Additionally, the project involves creating regression models of various types to accelerate computations. Improving the accuracy and/or speed of modeling the mechanical behavior of such materials will elevate the solution to synthesizing the internal structure of the material for specific operating conditions. This is crucial for designing structures, enabling better performance in terms of strength, durability, thermal characteristics, and weight reduction for the bodies and airframes of aerial and aquatic drones, monocrystalline blades of the AL-31F engine, etc. popup.description2 The research project is devoted to the development of innovative approaches to the application of artificial intelligence methods for materials modeling in advanced dual-use engineering technologies. The study focuses on the development of algorithms and software tools for the generation and analysis of microstructures of composite, polycrystalline, and mesoporous materials, as well as on investigating the possibilities of their parallel processing and scalability. Approaches are developed for constructing regression models of the deformation, limit, and thermal states of complex materials under various loading conditions. The main objective of the project is to create multilevel algorithms for modeling the physical and mechanical behavior of materials, enabling efficient synthesis of their internal structure using physics-informed neural networks and symbolic regression methods. This provides a significant acceleration in the search for optimal material parameters compared to traditional numerical approaches and expands the range of technological solutions for the application of composite, polycrystalline, and mesoporous materials, including aerogels. The second stage of the research was dedicated to the development and practical validation of algorithmic and software tools for three-dimensional modeling of material microstructures and the analysis of their mechanical properties. Within this stage, the MatViz3D software was improved, parallel algorithms for microstructure generation were implemented, and data preparation tools for finite element analysis were developed. A series of numerical experiments was conducted for composite, polycrystalline, and mesoporous materials, resulting in the formation of open synthetic datasets suitable for further scientific research. Product Description popup.authors Kseniia Y. Potopalska Yuliia A. Viazovychenko Mariia I. Shapovalova Nataliia O. Fomenko Valeriia O. Rykova Vladyslav V. Zhykhariev Valeriia I. Hritskova Oleh S. Semenenko popup.nrat_date 2026-01-27 Close