Optimization And Design Of Reconfigurable Antennas Based On Multi-Objective Intelligent Optimization Algorithm

In practical applications, most problems of science and engineering are multi-objective optimization problems. Because various objective functions may be not compromised or conflict each other, no single solution can make all the required objectives achieve optimal at the same time. For these problems, a non-dominated(Pareto) optimal solution set can be usually obtained.As one of the most adaptable antenna structures for reconfiguration, the reconfigurable pixel antenna generally consists of many electrically small metallic patches which are interconnected by RF-switches. By controlling the states of switches, the reconfigurable pixel antenna can synthesize a rich variety of antenna shapes flexibly, and is easier to realize reconfigurable performance. However, due to the large mumber of switches, the design of reconfigurable pixel antennas is more complex, so it has to rely on an efficient search method to excavate the potential reconfigurable capability.This thesis is mainly about the study on multi-objective intelligent optimization algorithms and reconfigurable pixel antennas. The main works are listed as follows:1. A self-adaptive elitist non-dominated sorting genetic algorithm(self-adaptive NSGA-II) is proposed. Several benchmark functions with different characteristics are used to test the performance of the algorithm and compared with performances of the conventional elitist non-dominated sorting genetic algorithm(NSGA-II) and the multi-objective particle swarm optimization(MOPSO). Then, convergence metric and spacing metric are used to evaluate the results, which prove the efficiency of self-adaptive NSGA-II.2. Using the proposed self-adaptive NSGA-II, a reconfigurable pixel antenna with radiation pattern reconfigurability is optimized and compared with the optimized result of microgenetic algorithm. Through the comparison, it is proved that the multi-objecive intelligent optimization algorithm has more advantages than the single objective intelligent optimization algorithm for the design and optimization of antennas.3. In a reconfigurable pixel antenna, the effects of switches with different distances from RF-port are different. In order to equalize the effects of these switches to antenna reconfiguration, and reduce the number of switch and antenna complexity, we proposed a reconfigurable pixel antenna with non-uniform size pixel. And this antenna can operate at two frequencies and realize radiation pattern reconfigurability on six tilt directions by optimizing switch states with self-adaptive NSGA-II.