Numerical and experimental generated data during project https://doi.org/10.1038/s41598-024-72478-w

The dataset was generated using a technique for fast antenna design, which leveraged a machine learning framework with an infill criterion employing predicted enhancement of the merit function, utilizing a particle swarm optimizer as the primary search engine, and employing kriging for constructing the underlying surrogate model. The model operated within a reduced-dimensionality domain, guided by directions corresponding to maximum antenna response variability identified through fast global sensitivity analysis, tailored explicitly for domain determination. Operating within the reduced domain enabled building reliable surrogates at a significantly lower computational cost. To address the accuracy loss resulting from dimensionality reduction, the global optimization phase was supplemented by local sensitivity-based parameter adjustment.