CAJ | 학술논문

由于标准粒子群算法易于陷入局部最优和收敛速度慢等问题，提出了一种引入人工蜂群搜索策略和混合蛙跳搜索策略的粒子群算法（ABCSFL-PSO）。使用人工蜂群的搜索策略提高算法的探索能力，避免算法陷入局部最优；使用蛙跳算法中更新最差粒子的策略，来加快算法收敛速度，并进一步提高求解精度。在12个标准测试函数上的仿真实验结果表明，算法性能优良，不仅能够避免陷入局部最优，而且显著提升了收敛速度。
유우표준입자군산법역우함입국부최우화수렴속도만등문제，제출료일충인입인공봉군수색책략화혼합와도수색책략적입자군산법（ABCSFL-PSO）。사용인공봉군적수색책략제고산법적탐색능력，피면산법함입국부최우；사용와도산법중경신최차입자적책략，래가쾌산법수렴속도，병진일보제고구해정도。재12개표준측시함수상적방진실험결과표명，산법성능우량，불부능구피면함입국부최우，이차현저제승료수렴속도。
Since the standard Particle Swarm Optimization(PSO)algorithm is easy to fall into local optimum and converges slowly, a novel particle swarm optimization algorithm combined with the search operator of artificial bee colony algorithm and that of shuffled frog leaping algorithm is proposed. Firstly, use the search operator of artificial bee colony algorithm to improve the ability of explore to avoid falling into local optimum;Secondly, introduce the update operator of shuffled frog leaping algorithm into PSO to enhance the accuracy of convergence. Through the twelve standard test functions simu-lation experiments and compared with other algorithms, experiment results show that the proposed algorithm can avoid falling into local optimization and significantly improve convergence speed.