论文部分内容阅读
针对连续域函数优化问题,提出了一种新的全局极大值搜索方法———多感官群集智能算法(multi-sense swarmintelli-gence algorithm,MSA).受鱼群算法(artificial fish-swarmalgorithm,AFA)和FS算法(free search algorithm,FSA)的启发,MSA的搜索机制将大范围勘察和小范围精确搜索相结合,个体在使用视觉信息快速逼近局部较优解的同时,利用嗅觉信息避免群体过于集中并引导个体向全局较优解方向移动.仿真结果证明:MSA鲁棒性较强,全局收敛性好,收敛速度较快,收敛精度较高.最后,将该方法应用于前向神经网络训练,结果表明满足应用要求.
In order to solve the continuous domain function optimization problem, a new global maximum search method (MSA) is proposed, which is composed of fish-swarm algorithm (AFA ) And FS algorithm (FSA), the MSA search mechanism combines large-scale survey with small-scale accurate search. While using visual information to quickly approximate the local optimal solution, the individual uses the olfactory information to avoid the over-population The simulation results show that the MSA has strong robustness, good global convergence, fast convergence rate and high convergence precision.Finally, the method is applied to the training of forward neural networks , The results show that meet the application requirements.