基于禁忌搜索的自适应人工鱼群优化算法

针对人工鱼群算法在函数优化过程中存在易陷入局部最优、后期收敛速度慢和寻优精度低等问题,提出了一种基于禁忌搜索的自适应人工鱼群优化算法。由于较大的视野范围有利于进行全局探索,较小的视野范围有助于进行局部寻优,该算法引入了分段函数自适应地调整视野,保证了视域在一定范围内随着迭代的进行逐渐减小;利用正态分布函数以及鱼群间距的大小对步长进行了改进,来协调寻优速度与解精度之间的平衡;为了更加贴合生物觅食的本能,在随机行为中加入了具有levy飞行机制的自由游动算子,不仅加强了鱼的全局搜索能力,还降低了随机行为因盲目性而导致解退化的风险;为了改善鱼群因陷入局部极值而出现寻优停滞不前的状况,引入了禁忌搜索思想。实验结果表明,改进后的算法具有明显的寻优优势。

An Adaptive Artificial Fish Swarm Optimization Algorithm Based on Taboo Search

In the process of function optimization,artificial fish swarm algorithm has many problems,such as easy to fall into local optimum,slow convergence speed and low optimization precision,an adaptive artificial fish swarm optimization algorithm based on Taboo search is proposed.Since a larger field of view is conducive to global exploration and a smaller field of view is helpful for local optimization,the algorithm introduces a piecewise function to adaptively adjust the field of view,which ensures that the field of view gradually decreases in a certain range with the progress of iteration.The step size is improved by using the normal distribution function and the distance of the fish to coordinate the balance between optimization speed and solution accuracy.In order to fit the nature of biological foraging,a free swimming operator with a levy flight mechanism is added to the random behavior,which not only strengthens the global search ability but also reduces the risk of random behavior causing solution degradation due.In order to improve the situation that fish get stuck in the local extremum,the idea of Taboo search is introduced.The experiment shows that the improved algorithm has obvious advantages in optimization.