摘要
为克服基本人工蜂群算法在求解复杂函数优化问题时,存在求解精度低且搜索盲目性大的缺点,提出一种改进的人工蜂群算法。上述算法在基本人工蜂群算法的跟随蜂阶段,引入局部搜索性能较强的共轭梯度法改变搜索策略,用确定性搜索代替盲目性搜索,减少随机性,增强跟随蜂的局部寻优能力,确保食物源的每次更新都会得到改善。将改进后的算法用于传统登革病毒传播模型的参数优化。经过标准测试函数问题的仿真表明,所得改进的人工蜂群算法较基本人工蜂群算法具有更高的求解精度,所得的参数对应的模型输出与实际数据拟合情况较好。
In order to overcome the shortcomings of the basic artificial bee colony algorithm(ABC)in solving the complex function optimization problem with low accuracy and large search blindness,an improved artificial bee colony algorithm(FABC)is proposed.The FABC algorithm uses the conjugate gradient method with strong local search performance to replace the search strategy in the following bee stage.The FABC algorithm reduces the blind search with deterministic search,enhances the local optimization ability of the following bee by reducing the randomness,which ensures each update of the food source will be improved.The improved algorithm is used to optimize the parameters of the traditional Dengue virus propagation model.The simulation results on 5 standand test problems shows that FABC algorithm can obtain more accuracy solution than ABC algorithm,and the model output corresponding to the parameters obtained by the improved algorithm is better fit with the actual data.
作者
赵旭芳
梁昔明
ZHAO Xu-fang;LIANG Xi-ming(School of Science,Beijing University of Civil Engineering&Architecture,Beijing 100044,China)
出处
《计算机仿真》
北大核心
2019年第9期320-325,共6页
Computer Simulation
基金
国家自然科学基金(61463009)
北京自然科学基金项目(4122022)
中央支持地方科研创新团队项目(PXM2013-014210-000173)
贵州省科学技术基金(黔科合基础[2016]1022)
关键词
人工蜂群算法
共轭梯度法
参数优化
登革病毒传播模型
数值试验
Artificial bee colony algorithm
Conjugate gradient method
Parameters optimization
Dengue virus propagation model
Numerical experiments
