改进的萤火虫算法及其在函数优化中的应用

Improved Firefly Algorithm and Its Application in Function Optimization

萤火虫算法是一种简单、高效的启发式搜索方法,能够广泛应用到各类优化问题中,针对传统的萤火虫算法容易陷入局部最优,演化后期收敛速度偏慢等问题,提出了改进的Logistic映射策略和定时逆向学习算子相结合的初始化种群方法,其目的是改进种群的多样性、加快算法的收敛速度、避免算法过早的陷于局部最优.通过对6个标准测试函数进行测试验证,改进的算法在最优解的质量与稳定性优于其他被比较算法.

Firefly algorithm is a simple and efficient heuristic search method,which can be widely applied to various optimization problems.In view of the problems that the traditional firefly algorithm is easy to fall into local optimum and convergence rate is slow in the late evolution period,an initialization population method based on improved Logistic mapping strategy and timed reverse learning operator is proposed,with an aim to improve the diversity of the population,speed up the convergence of the algorithm,and avoid the prematurely falling into local optimum.By testing and verifying 6 standard test functions,it is found that the quality and stability of the improved algorithm are better than those of other compared algorithms in optimal solution.