基于混合优化策略的麻雀搜索算法研究  

Research on Sparrow Search Algorithm Based on Hybrid Optimization Strategy

作　　者：易华辉 王雨璇 黄金香 宋文治 李垒 Yi Huahui;Wang Yuxuan;Huang Jinxiang;Song Wenzhi;Li Lei(College of Ordnance Science and Technology,Xi'an Technological University,Xi'an 710021,China)

机构地区：[1]西安工业大学兵器科学与技术学院,西安710021

出　　处：《机电工程技术》2023年第2期93-97,176,共6页Mechanical & Electrical Engineering Technology

基　　金：西安市智能兵器重点实验室项目(编号:2019220514SYS020CG042)。

摘　　要：针对传统麻雀搜索算法寻优精度低,搜索速度差的缺陷,提出了一种基于混合优化策略的麻雀搜索算法(ISSA)。首先,采用Piecewise混沌映射初始化种群的位置来增加种群的丰富性;其次,引入黄金搜索策略优化生产者的位置公式,提高算法的全局寻优能力;然后,采用自适应的优化步长控制参数,平衡个体间的全局搜索与局部开发的性能;最后,针对原算法陷入局部最优的缺陷,引入模拟退火技术以提高算法的寻优精度,降低遭遇局部最优的可能性。在此基础上,将改进算法与12个典型基准函数进行对比来评价算法的优越性。实验结果表明,改进的ISSA体现出精确性和高效性,效果更优。The traditional sparrow search algorithm has the defects of low optimization accuracy and poor search speed.For this problem,a sparrow search algorithm(ISSA)based on hybrid optimization strategy was presented.First of all,Piecewise chaotic map was used to initialize the position of the population to increase the richness of the population.Next,the golden search strategy was introduced to optimize the location formula of producers to improve the global optimization ability of the algorithm.Then,adaptive optimal step size control parameters were used to balance the performance of global search and local development among individuals.Eventually,aiming at the defect that the original algorithm falls into local optimization,simulated annealing technology was introduced to improve the optimization accuracy of the algorithm and reduce the possibility of encountering local optimization.Basis of this,improved algorithm was compared with 12 typical benchmark functions to evaluate the superiority of the algorithm.The experimental results show that the improved ISSA is accurate and efficient,and the effect is better.

关 键 词：麻雀搜索算法 黄金搜索 自适应惯性权重 模拟退火 局部最优 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]