摘要
人工蜂群算法作为一种新生代的优化算法,近年来在众多科学领域中表现出一定的优势,但是其收敛速度并不高效,并且容易过早地陷入局部最优。首先通过对适应度选择进行改进,提高算法的收敛速度,同时结合模拟退火算法,一定程度上避免过早陷入局部最优。最后用一组基准函数进行实验,证明改进后的人工蜂群算法有更好的优化性能。
As a new optimization algorithm, artificial bee colony algorithm shows a certain advantages in many scientific fields over the years. But its convergence speed is not efficient, and it can get into local optimization easily. Improves the convergence speed through the improve- ment of fitness selection. The algorithm is combined with simulated annealing algorithm at the same time, which avoids falling into local optimum easily in some extent. The experiments which use a set of benchmark functions show that the improved artificial bee colony algorithm can get better optimization performance.
关键词
人工蜂群算法
适应度
模拟退火算法
函数优化
Artificial Bee Colony Algorithm
Fitness
Simulated Annealing Algorithm Simulated Annealing Algorithm
Numerical Optimization
