Vector evaluated GA-ICT for optimum design of arbitrarily arranged wire grid model antenna

GA-ICT is a new method that combines a genetic algorithm (GA) and improved circuit theory (ICT) to determine the optimum antenna design without having to provide an initial basic structure. It is considered as a kind of multi-objective optimization to satisfy all design conditions such as directivity, beam width, and side lobe level. In this case, it may become difficult to obtain all design requirements at the same time due to nonlinear characteristics since the evaluation function of this method is constructed based on the weights derived from the characteristics. To overcome this problem, we introduce a vector evaluation method to GA-ICT that generates different individual sets in parallel calculations. The evaluation functions have different weights to generate different chromosomes that satisfy other design conditions in the first parallel GA operation and the sets are merged in the next GA operation. By applying this vector evaluated GA-ICT to the sector antenna downsizing problem, the multi-sector monopole Yagi-Uda array antenna (MSMPYA) was downsized by 70%.