一种动态调整惯性权重的自适应蝙蝠算法
发布时间：2018-12-11 00:40
【摘要】：为了加快蝙蝠算法的收敛速度并提高寻优精度,提出一种动态调整惯性权重的自适应蝙蝠算法。该算法在速度公式中加入惯性权重,并采用一种服从均匀分布和贝塔分布的随机调整策略,动态地调整惯性权重的大小,以加快算法的收敛速度。另外,引入了速度纠正因子,在每次迭代时,算法可根据当前种群的迭代次数动态地约束每一代蝙蝠的移动步长,从而使算法具有一定的自适应性。仿真实验结果表明,改进后的算法的寻优性能显著提高,具有较快的收敛速度和较高的寻优精度。
[Abstract]:In order to speed up the convergence speed of bat algorithm and improve the accuracy of optimization, an adaptive bat algorithm with dynamically adjusting inertia weight is proposed. In this algorithm, inertia weight is added into the velocity formula, and a random adjustment strategy of uniform distribution and beta distribution is adopted to dynamically adjust the size of inertia weight to accelerate the convergence speed of the algorithm. In addition, the speed correction factor is introduced. In each iteration, the algorithm can dynamically constrain the moving step size of each generation bat according to the number of iterations of the current population, thus making the algorithm self-adaptive. The simulation results show that the improved algorithm has better performance, faster convergence speed and higher optimization accuracy.
【作者单位】： 河南大学计算机与信息工程学院;河南大学复杂智能网络系统研究所;河南大学软件学院;河南大学管理科学与工程研究所;
【基金】：河南省科技厅科技攻关项目(162102110109) 河南省科技攻关重点项目(142102210036)资助
【分类号】：TP18
[Abstract]:In order to speed up the convergence speed of bat algorithm and improve the accuracy of optimization, an adaptive bat algorithm with dynamically adjusting inertia weight is proposed. In this algorithm, inertia weight is added into the velocity formula, and a random adjustment strategy of uniform distribution and beta distribution is adopted to dynamically adjust the size of inertia weight to accelerate the convergence speed of the algorithm. In addition, the speed correction factor is introduced. In each iteration, the algorithm can dynamically constrain the moving step size of each generation bat according to the number of iterations of the current population, thus making the algorithm self-adaptive. The simulation results show that the improved algorithm has better performance, faster convergence speed and higher optimization accuracy.
【作者单位】： 河南大学计算机与信息工程学院;河南大学复杂智能网络系统研究所;河南大学软件学院;河南大学管理科学与工程研究所;
【基金】：河南省科技厅科技攻关项目(162102110109) 河南省科技攻关重点项目(142102210036)资助
【分类号】：TP18
【相似文献】
相关期刊论文 前10条
1 周俊;陈t熁，
本文编号：2371527
本文链接：https://www.wllwen.com/kejilunwen/zidonghuakongzhilunwen/2371527.html
