多方向螺旋搜索的混沌海鸥优化算法

Multi-directional Exploring Seagull Optimization Algorithm Based on Chaotic Map

针对海鸥优化算法(SOA)寻优路径单一、寻优精度较低、易陷入局部最优等问题,提出新的多方向螺旋搜索的混沌海鸥优化算法(Multi-directional Exploring Seagull Optimization Algorithm Based On Chaotic Map MESOA).首先,利用混沌序列对海鸥种群进行初始化,令海鸥个体分布更加均匀,能够更加准确地接近目标;其次,让海鸥选择不同方向的螺旋飞行路径,使海鸥飞行路径不再单一,增加算法多样性;最后,根据算法收敛情况进行围绕目标的小范围搜索,避免算法过早收敛,提高算法跳出局部最优的能力.本文选取了8个基准测试函数对算法进行了实验,以不同角度对于算法的性能进行测试,并使用Wilcoxon秩和检验来证明算法的性能,结果表明了MESOA算法改进在寻优能力、稳定性、鲁棒性等方面均有提升.

The seagull optimization algorithm(SOA)has problems of single searching path, low accuracy, easily falling into local optima, multi-directional exploring seagull optimization algorithm based on chaotic map(MESOA)is proposed.First, the original seagull population is created according to the chaotic sequences and seems to be more even-distributed, so that the individuals approach the target more accurately;secondly, the seagulls have different spiral flight trajectories, which enrichs searching diversity of the algorithm;finally, a local searching strategy is adopted according to the the positions of individual seagulls, which helps avoid premature convergence of the algorithm.This paper selects 8 benchmark test functions to conduct experiments on the algorithm, test the performance of the algorithm from different angles, and use the Wilcoxon rank sum test to prove the performance of the algorithm.The results show that the improvement of the MESOA algorithm is in the optimization ability, stability, Robustness and other aspects have been improved.