Design of a bifocal metalens with tunable intensity based on deep-learning-forward genetic algorithm
Fang Wang, Xuewen Shu
Abstract
Abstract Metalenses, which control the amplitude, phase, and polarization state of incident waves based on metasurface to achieve focusing and imaging, have many important applications in various optical systems. We design a bifocal metalens that can independently control the focusing of right-handed circularly polarized light and left-handed circularly polarized light. Due to the demand for enormous simulations, traditional design methods are extremely time-consuming. Here, we propose a deep-learning-forward genetic algorithm to efficiently design the metalens parameters. The numerical simulation results of the metalens are in good agreement with the theoretical results. Meanwhile, it is flexible to change intensity ratio of the two foci through altering incident light ellipticity without redesigning the light intensity profile. This work provides a novel approach to multifunctional metasurface device realization.