🤖 AI Summary
Discrete multi-objective topology optimization for antenna inverse design suffers from slow convergence and poor Pareto front quality. Method: This paper proposes a hybrid memetic multi-objective optimization algorithm integrating gradient-based local search with heuristic global optimization. It introduces an adaptive objective weighting mechanism, couples rank-1 gradient approximation for local search with the NSGA-II global framework, and establishes a dual-modality evaluation metric jointly assessing electromagnetic performance and topological structure. Results: Evaluated on three canonical antenna design tasks, the method achieves 10×–100× speedup in optimization convergence while significantly outperforming state-of-the-art approaches in both Pareto front convergence and diversity. Moreover, it delivers high-quality solution sets that facilitate machine learning–driven design data mining and knowledge discovery.
📝 Abstract
This paper describes the modification of a single-objective algorithm into its multi-objective counterpart. The outcome is a considerable increase in speed in the order of tens to hundreds and the resulting Pareto front is of higher quality compared to conventional state-of-the-art automated inverse design setups. This advancement is possible thanks to a memetic algorithm combining a gradient-based search for local minima with heuristic optimization to maintain sufficient diversity. The local algorithm is based on rank-1 perturbations; the global algorithm is NSGA-II. An important advancement is the adaptive weighting of objective functions during optimization. The procedure is tested on three challenging examples dealing with both physical and topological metrics and multi-objective settings. The results are compared with standard techniques, and the superb performance of the proposed technique is reported. The implemented algorithm applies to antenna inverse design problems and is an efficient data miner for machine learning tools.