Akademik Veri Yönetim Sistemi
5th International Conference on Light and Light-Based Technologies, Ankara, Türkiye, 15 - 17 Mayıs 2025, ss.68, (Özet Bildiri)
This study addresses ultra-compact and ultra-high quality factor waveguide cavities designed by artificial intelligence. All high quality factor cavitation designs naturally include a slow wave light mode that is a linear combination of another fundamental propagation modes. In this manner, initially genetic algorithm was used as the main optimization methodology and waveguides capable of achieving slow light have been considered. Later on, a regression based model was trained by a dataset consisting randomized simulation parameters of the hole radii on dielectric (or dielectric rod radii on air background), positions, and numbers with corresponding quality factors. Resonance modes were also examined with a modeling program package utilizing finite difference method in time domain and comparisons were made with frequency domain. The quality factor values above 106 were obtained for both transverse magnetic and transverse electric polarization modes. This outcome will facilitate the identification of essential subcomponents for various devices, including electromagnetic and optical logic circuits, as well as sensor structures.