In this paper, an evolutionary optimization procedure is presented to generate band-pass metasurface-based filters in terahertz regime. As a measure of novelty, pass-band, transition, and out-band characteristics are investigated separately, all of which result in different metasurfaces for filtering applications. The presented approach is defined based on random hill climbing algorithm, regarding the established link between Matlab and HFSS software. A metasurface-based filter with specific properties is considered as the main problem to be solved by the optimization method. Moreover, the fuzzy theory, mean square method, and weighting coefficient procedure are considered to define an efficient fitness function evaluation approach. Also, a step-by-step procedure is used to generate desired structures with a great note of efficiency. The final generated structure has magnificent characteristics including sharp transitions together with transmittance around 0.68 and less than 0.04 at pass-band and out-band regions, respectively. Also, the generated metasurface benefits from wide bandwidth (65%) and great compactness compared to other previous works.

在本文中，提出了一种进化优化程序，以在太赫兹状态下生成基于带通超表面的滤波器。作为新颖性的一种度量，通带，过渡和带外特性将分别进行研究，所有这些都导致在滤波应用中使用不同的超颖表面。考虑到Matlab与HFSS软件之间已建立的联系，提出的方法是基于随机爬山算法定义的。具有特定属性的基于超表面的滤波器被认为是通过优化方法要解决的主要问题。此外，考虑了模糊理论，均方方法和加权系数程序来定义一种有效的适应度函数评估方法。同样，使用分步过程来生成具有很高效率的所需结构。最终产生的结构具有宏伟的特性，包括急剧的跃迁以及在通带和带外区域的透射率分别约为0.68和小于0.04。而且，与其他先前的工作相比，生成的超表面受益于宽带宽（65％）和紧凑性。