摘要
为解决差分进化算法后期收敛易陷入局部最优和早熟收敛的问题,提出一种群体智能优化算法,即协同智能的蝙蝠差分混合算法。利用蝙蝠个体脉冲回声定位的特点,与差分种群相互协作,在当前最优解gbest附近进行一次详细搜索,有效增加种群的多样性,跳出局部最优。通过蝙蝠种群和差分种群两个种群的相互协作,较好平衡全局搜索和局部开发之间的能力。为验证算法有效性,选用9个常用的基准测试函数和5个0-1背包问题,与标准粒子群算法、带高斯扰动的粒子群算法、蝙蝠算法、差分算法、烟花算法相对比,仿真实验表明,所提算法总体性能优于其它5种算法。
To solve the problem that the late convergence of differential evolution algorithm is easy to fall into local optimum and premature convergence,a swarm intelligence optimization algorithm,namely hybrid bat and differential evolution algorithm was proposed.Taking the advantage of the characteristics of individual pulse echolocation,bat population and differential population cooperated to conduct a detailed search near the current optimal solution gbest.Using the algorithm increases the diversity of the population and prevents the differential population from falling too fast into local optimum.Through the information interaction and of bat population and differential population,the improved algorithm can balance the ability between global search and local development.To verify the effectiveness of the improved algorithm,nine commonly used benchmark functions and five 0-1 knapknack problems were selected for comparing the proposed method with particle swarm optimization,particle swarm optimization with Gaussian mutation,bat algorithm,differential evolution algorithm,fireworks algorithm.The simulation results show that the overall performance of the algorithm proposed is better than the other five algorithms.
作者
赵志刚
曾敏
莫海淼
李智梅
温泰
ZHAO Zhi-gang;ZENG Min;MO Hai-miao;LI Zhi-mei;WEN Tai(School of Computer,Electronics and Information,Guangxi University,Nanning 530004,China)
出处
《计算机工程与设计》
北大核心
2020年第2期402-410,共9页
Computer Engineering and Design
基金
广西自然科学基金项目(2015GXNSFAA139296)
关键词
差分算法
蝙蝠算法
蝙蝠差分混合算法
协同智能
函数优化
0-1背包问题
differential evolution algorithm
bat algorithm
hybrid bat and differential evolution algorithm(BADE)
cooperative intelligence
function optimization
0-1 knapsack problem