Hybrid Evolutionary Compact Optimization with Archive and Dual Population

This paper presents an approach to enhancing traditional evolutionary compact algorithms using archives and dual populations to avoid premature convergence by balancing exploitation and exploration. An archive is adopted to keep a set of promising solutions. The hybrid approach combines compact particle swarm optimization and differential evolution to update the dual populations and maintain the archive. We compare the proposed algorithm with standard compact algorithms using CEC2021 benchmarks to evaluate the performance. The experimental results confirm the effectiveness of the proposed approach.