引导小生境回溯优化算法

发布时间：2018-05-14 07:48

本文选题：回溯搜索优化算法 + 引导机制　；参考：《计算机工程与应用》2017年21期

【摘要】：回溯搜索优化算法(BSA)是近年提出的一种新型优化算法,针对其收敛速度较慢、易陷于局部最优的缺点,提出了一种基于最优个体引导和小生境技术相结合的改进BSA算法。本方法首先在BSA的变异操作中引入向最优个体学习的策略,以提高算法的收敛速度;其次,设计一种新的小生境排挤技术,根据每个个体到其他个体距离的平均最小值确定小生境半径,排除部分相似性较高的个体;结合群体当前的最差信息,设计一种新的变异方法产生一定数量的新个体补充到新的种群中,维持群体数量的恒定并增强群体多样性。改进的BSA算法充分考虑了算法的收敛速度和群体的多样性,较大地提高了传统BSA算法的性能。对10个典型函数进行仿真测试,并与其他算法结果进行对比,实验结果表明,改进算法在收敛速度与精度方面具有较好的效果。
[Abstract]:The backtracking search optimization algorithm (BSA) is a new optimization algorithm proposed in recent years. In view of its slow convergence and local optimum, an improved BSA algorithm based on the combination of optimal individual guidance and niche technology is proposed. This method first introduces the strategy of learning to the optimal individual in the variant operation of BSA. The convergence speed of the high algorithm is obtained. Secondly, a new niche extrusion technique is designed to determine the niche radius according to the average minimum distance of each individual to other individual distances and eliminate some individuals with higher similarity. In the population, the number of groups is maintained and the population diversity is enhanced. The improved BSA algorithm fully considers the convergence speed of the algorithm and the diversity of the group, and greatly improves the performance of the traditional BSA algorithm. 10 typical functions are simulated and compared with other algorithms. The experimental results show that the improved algorithm is convergent. The speed and precision have good effect.

【作者单位】：淮北师范大学物理与电子信息学院;
【基金】：国家自然科学基金(No.61572224,No.61304082) 安徽省高校自然科学研究重大项目(No.KJ2015ZD36);安徽省高校自然科学研究重点项目(No.KJ2016A639) 安徽省国际科技合作计划项目(No.10080703003) 安徽省第七批”115”产业创新团队皖人才[2014]4号
【分类号】：TP18

【相似文献】