08.02.2026
The Q1 scientific journal Energies has published the results of a research study conducted by Professor Yevgeniy Bodyanskiy of the Department of Artificial Intelligence in collaboration with colleagues from Ukraine, Poland, and Germany.
The scientific paper presents an innovative real-time diagnostic method for wind turbine blades, opening up new capabilities for sustaining energy infrastructure under critical conditions.
The authors proposed a fundamentally new approach — a hypersector-based method grounded in Fuzzy Learning Vector Quantization (FLVQ). Unlike traditional techniques, this method models each defect class as a “hypersector” on a multidimensional sphere, significantly improving recognition accuracy and noise immunity.
The diagnostic process utilizes images captured by Unmanned Aerial Vehicles (UAVs), enabling regular inspections without endangering human personnel or shutting down turbine operations.
This development holds strategic significance for both Ukraine and Europe. Amid targeted attacks on energy infrastructure, having advanced tools for the rapid detection and repair of damage at restored facilities is crucial. Renewable energy, particularly wind power, is becoming a key element of national and regional resilience. The technology supports the long-term, stable operation of renewable energy sources, aligning directly with Ukraine and the EU’s shared commitment to energy independence and sustainable development.
This research stands as a prime example of how fundamental AI research can deliver practical solutions to global challenges, especially in the context of hybrid threats and the necessity for rapid adaptation of critical infrastructure.
The full text of the article “Hypersector-Based Method for Real-Time Classification of Wind Turbine Blade Defects” is available at the link:
