Optimization of Frequency Reconfigurable Antenna Parameters Design Using Genetic and PSO Algorithms Based on Neural Networks
In this paper, we propose a novel mixed-integer optimization formulation for the optimal design of a reconfigurable antenna inspired by methodology to design frequency reconfigurable patch antennas using multi-objective genetic algorithms (MOGA) genetic algorithm trained by recurrent neural networks and nondominated sorting genetic algorithm II (NGSA-II) improve global optimization capability by diversity detection operation to surrogate a model optimized. Experimental validation of Pareto-optimal set miniaturized multiband antenna designs is also provided, demonstrating a new optimization technique. The oriented design here is practiced for improving reflection coefficient S11, and gain specifications at the frequency band that is achieved by sizing the design parameters using our proposed method in the author’s way the performance parameters were predicted by an iterative process of particle swarm optimization based on feed-forward neural networks (FFNN). The proposed optimization technique is successfully attracted as a problem solver for designers to tackle the subject of antenna design. which works in the frequency range from 200 MHz to 224.25MHz (50% impedance bandwidth at operated frequency 200 MHz) sequentially is obtained.
