摘要
针对人工鱼群算法在寻优过程中接近最优点时收敛速度下降而难以得到精确解,优化复杂问题时易陷入局部极值的缺点,提出了一种复合混沌搜索技术与改进人工鱼群算法相结合的混合算法。该算法采用更具遍历性的组合映射产生复合混沌局部搜索方法,来避免人工鱼长时间陷入局部极值区域,从而更加精确地达到全局最优点;同时,对人工鱼引入反馈-吞食行为进行改进,改进的人工鱼群算法降低了优化后期的复杂度,并提高了优化精度,保证了收敛效率。实验结果表明,在相同参数条件下,该混合算法的收敛速度、优化精度和全局寻优能力均优于基本人工鱼群算法,实例验证了算法的有效性。
When artificial fish swarm algorithm is close to the optimal point during its optimization process, the convergence rate declines so that it is difficult to get exact solutions. Besides, the algorithm is easy to fall into local minima in complex issues. Aiming at the aforementioned disadvantages, a hybrid algorithm is proposed, which combines the compound chaotic search technology and the improved artifi- cial fish swarm algorithm. It adopts the mapping combination with more ergodicity to generate the local search method. The method can avoid that artificial fish are into local extremum area for a long time, so that it reaches the global extreme points more precisely. Meanwhile, the artificial fish swarm algorithm is improved by introducing feedback-swallowed behavior of artificial fish. The improved algorithm re- duces optimization complexity at late stage, improves accuracy and guarantees convergence efficiency. Experimental results show that, under the same parameter conditions, the proposed hybrid algorithm outperforms the basic artificial fish swarm algorithm in convergence rate, optimization accuracy and global optimization ability. Experiments demonstrate the efficiency of the proposed method.
出处
《计算机工程与科学》
CSCD
北大核心
2013年第8期89-95,共7页
Computer Engineering & Science
关键词
人工鱼群算法
复合映射
混沌搜索
全局寻优
反馈策略
吞食行为
artificial fish swarm algorithm
compound map
chaotic search
global optimization
feedback strategy
swallowed behavior
