Applications Of Multi-objective Genetic Algorithm To Frequencey Selective Surface Optimization

In recent years, research of intelligent algorithms is a very active area, and they have been extensively studied and used in automation, management, engineering, computers and so on by many professional scholars. The genetic algorithm(GA) is the popular one in intelligent algorithms. A simple single-objective genetic algorithm can not meet the need of the project, and it is easy to premature or converge slowly. This thesis proposes an improved algorithm, which combines the multi-objective genetic algorithm with adaptive scheme, to communicate with HFSS via VBScript. The simulation results of HFSS will be transferred to genetic algorithm as the objective function value of individuals involved in the evolution.The frequency selective surface(FSS) is often used in the military invisibility technology, and has a great contribution to solving electromagnetic compatibility problems. So it attracts attention from many scholars and researchers. The filtering characteristics of FSS has made it be promoted both in military or civilian. The development of computers increases the research of FSS, and the appearance of electromagnetic simulation software facilitates the design of FSS. The thesis chooses the electromagnetic simulation software of HFSS and optimization algorithm to design FSS.The thesis focuses on the improvement of the multi-objective genetic algorithm and its application to the design of FSS. The main work has been listed as follows.1. An adaptive multi-objective optimization algorithm aims to solve the weak ability of the traditional one in local search. The distances from the real Pareto-front to the solutions of the traditional algorithm and the improved one are compared, and their performances are judged.2. A dual-band frequency selective surface is designed by the improved multi-objective genetic algorithm combined with HFSS solver. Through the modeling, simulation and optimization, we obtain the model which meets the design requirement.3. A simple cross-shaped FSS is restructured while maintaining stability. Another notched band is added for the original structure with only a notched band. The improved multi-objective GA is used to optimize the structure to meet the design goals.