Improved self-adaptive genetic algorithm with quantum scheme for electromagnetic optimisation

In this study, an accurate and efficient quantum genetic algorithm (QGA) combined with an improved self-adaptive (SA) scheme is proposed to solve electromagnetic optimisation problems. QGA is employed as the main optimisation frame because of its wider search range and higher efficiency than the conventional genetic algorithm. By introducing an improved SA scheme, the population at each generation is divided into two groups for crossover operation according to the magnitudes of individual fitness values. The crossover probability and mutation rate remain unchanged at the early stage of iterative process while the SA scheme will be carried out for the rest of the iterative process. Moreover, the elitist model is introduced to save the optimal father-individuals and abandon the worst ones. All these strategies make the whole population nearly converge to the optimal solution very fast. In two numerical examples of filter design and linear array synthesis, the effectiveness of the author's proposed optimisation algorithm, combined with the finite-difference time-domain method and finite-element method in HFSS, respectively, is verified.