An Evolutionary Algorithm and Its Application in Antenna Design

An Evolutionary algorithms (EA) is usually implemented to design an antenna straight without general performance test of the EA before the antenna design. This paper first designs an evolutionary algorithm (EA) and tests it by using typical benchmark problems and gets an efficient algorithm. Then the tested algorithm is used to design antenna. The EA designed in this paper is simple and easy to implement. But its performance is not poor. The difference of the proposed algorithm from others is that it combines three operators to produce offsprings: affine crossover, uniform crossover and uniform mutation. Since the three operators are independent to some extent, population diversity could be well maintained, then it could enhance capacity in global search. The performance of the proposed algorithm is better or not worse than that of referred state-of-the-art algorithm as a whole for the optimization problems with only inequality constraints. Finally, the new algorithm was used to design a single-arm antenna. The evolved antenna meets the requirements.