Design and Simulation of Mushroom Nanorods for Plasmonic Applications Using FDTD Methods and Deep Learning

This paper introduces a dataset generated through Finite-Difference Time-Domain (FDTD)simulations, focusing on the electromagnetic behavior of Mushroom Nanorods. The dataset captures a rangeof numerical parameters and field images across different configurations, including materials like Gold,Al₂O₃, and SiO₂. These simulations are intended to advance research in photonics, optics, andelectromagnetic wave dynamics. The dataset also serves as a key resource for machine learning models,specifically reinforcement learning, aimed at improving the prediction accuracy of electromagnetic fieldbehavior influenced by material properties and structural variations. In comparison to other existing datasetsin the field, our dataset offers a more comprehensive and parameterized set of images and simulation data,which allows for a deeper understanding of field behavior in nanophotonic systems. This work not onlycontributes to the development of more efficient photonic devices but also highlights the potential of usingsimulation-driven data for optimizing nanostructure designs in various research areas, includingnanophotonics and electromagnetic modelling.