基于混合策略机制的人工鱼群算法

Artificial fish-school algorithm based on hybrid strategy mechanism

针对人工鱼群算法在处理多峰函数问题时存在一部分人工鱼处于漫无目的的随机移动、易陷入"早熟收敛"情况造成的收敛速度减慢的缺点,提出了一种基于混合策略机制的人工鱼群算法。它借鉴群体位置方差的早熟判断机制,把云发生器产生的杂交和变异算子引入到该算法中,为减少算法计算量,而采用耗散的人工鱼群算法结构。实验表明:该算法比只有一个适应值的人工鱼群算法具有更快的收敛速度。且具有很强的避免局部极小能力,其性能远远优于单一优化方法。

Artificial fish swarm algorithm for multi-modal function in dealing with the issue as part of the artificial fish in the presence of random aimlessly move easily into a ＂premature convergence＂ situation due to the shortcomings of slow convergence rate,a mechanism based on the artificial hybrid strategy fish algorithm.It draws on the variance of the early groups to determine the location mechanisms to cloud conversion and mutation operators are introduced into the algorithm,Experiments show that the algorithm is only one fitness value than the artificial fish school algorithm has faster convergence speed.And has a strong ability to avoid local minimum,its performance is far superior to the single optimization method.