Multiobjective Electromagnetic Optimization Based on a Nondominated Sorting Genetic Approach with a Chaotic Crossover Operator


Real-world engineering optimization problems involve multiple design factors and constraints and consist in minimizing multiple noncommensurable and often competing objectives. In recent years, many evolutionary techniques for multiobjective optimization have been proposed. In this context, the Non-dominated Sorting Genetic Algorithm II (NSGA-II) algorithm is an effective methodology to solve multiobjective optimization problems. A modified NSGA-II to seek the Pareto front of electromagnetic multiobjective design problems is proposed in this paper. We propose the use of chaotic sequences based on Zaslavskii map in the NSGA-II crossover operator. The proposed method is tested on TEAM 22 benchmark optimization problem with promising results.