摘要
人工鱼群算法是一种收敛速度快、全局优化能力强的新型群智能算法。然而,在基本鱼群算法的应用中发现:在迭代前期,算法具有较强的搜索能力;但在运行后期,其搜索能力减弱,易陷入局部极值,且搜索到的最优解精度不高。针对上述弱点,提出对可视域和步长采用自适应变化策略,引入变异算子策略,通过消亡操作对部分个体进行重新初始化或变异,对基本鱼群算法进行改进,并以函数优化和多维变量的非线性优化问题为例进行了实验研究。结果表明:改进后的人工鱼群算法具有较好的优化效果。
Artificial fish swarm algorithm is a new swarm intelligence algorithm with fast convergence speed and good global optimization ability. However, in practical applications, it is found that in the later period of arithmetic operating, the ability of breaking through local points becomes weak and it easily falls into local points. In addition, the solution has low precision. In order to overcome these faults, self-adaptive strategy for the visual field and step, discarding operation and re-initialization were synthetically applied to improve it. As a case, the improved algorithm is used for function optimization and high-dimension nonlinear function optimization. The simulation results show that it has good optimization effects.
出处
《系统工程》
CSCD
北大核心
2009年第12期105-110,共6页
Systems Engineering
基金
国家自然科学基金资助项目(50379003)
安徽省自然科学基金资助项目(070416243)
关键词
改进人工鱼群算法
函数优化
自适应策略
投影寻踪模型
Improved Artificial Fish Swarm Algorithm
Function Optimization
Adaptive Strategy
Projection Pursuit Model