自适应爆炸半径和差分变异的增强烟花算法

Abstract

增强烟花算法对烟花算法存在的性能缺陷进行改进，并在许多优化问题中显示出不错的效果。然而，增强烟花算法在某些函数上寻优精度低、且容易过早地陷入局部最优解。为了改善这种缺陷，将其最大爆炸半径设置为模拟退火因子来自适应地加速其搜索过程，并借鉴差分变异中的思想以增加种群的多样性，达到跳出局部最优解的目的。改进后的算法在10个Benchmark函数上进行测试，实验结果表明改进后的算法效果明显优于增强烟花算法。 Enhanced fireworks algorithm improves the performance defect of the fireworks algorithm, and shows good effect in many optimization problems. However, it still has low precision and is easy to fall into local optimal solution prematurely in some functions. In order to improve the above men-tioned problems, the maximum explosion amplitude is set as the simulated annealing factor to accelerate its search process, and we learn from the idea of differential variation to increase the diversity of the population, so as to jump out of the local optimal solution. The proposed algorithm is tested on 10 benchmark functions and the experimental results show that the proposed algorithm significantly outperforms the enhanced fireworks algorithm and the fireworks algorithm.