A Multi-Strategy Sparrow Search Algorithm with Selective Ensemble

Aiming at the deficiencies of sparrow search algorithm (SSA), such as being easily disturbed by local optimal and deficient optimization accuracy, a multi-strategy with selective ensemble (MSESSA) is proposed. Firstly, three novel strategies in strategy pool are proposed: variable logarithmic spiral saltation learning enhances global capability, neighborhood-guided accelerates convergence, adaptive Gaussian random walk coordinates exploration exploitation. Secondly, idea adopted to select an appropriate current stage aid priority roulette selection method. In addition, modified boundary processing mechanism adjusts transgressive sparrows’ locations. The relocation method for discoverers alerters conduct large range, based on suboptimal population scroungers better search. Finally, MSESSA tested CEC 2017 suites. function test, Wilcoxon ablation experiment results show that achieves comprehensive performance than 13 other advanced algorithms. four engineering problems, stability, effectiveness, superiority systematically verified, which has significant advantages can reduce design cost.