A Hybrid Algorithm Based on Improved Krill Herd and Particle Swarm Optimization

Number of views： 13

分享到:

摘要:

针对基本磷虾群(KH)算法在求解高位复杂优化问题时容易陷入局部最优、求解精度低等缺点,提出了一种基于改进的磷虾群和粒子群的混合算法(AIPSOKH).该算法首先对KH算法中的觅食权重和诱导权重采用非线性递减策略,然后将其与惯性权重线性递减的粒子群算法(LDWPSO)混合,采用双子种群同时计算的并行策略进行迭代计算,借鉴自然选择中适者生存的进化机制提升母种群中个体的质量,以此来避免算法陷入局部最优,并提升其求解精度.最后通过8个标准测试函数的对比实验表明,在全局搜索能力和求解精度上与提到的2种算法相比都有着显著优势.

Based on improved krill herd and swarm optimization（AIPSOKH） this paper puts forward a hybrid algorithm to solve problems of low convergence and easily falling into local optimization. This algorithm adds nonlinear decreasing strate-gy of foraging weight and induced weight into the basic krill herd algorithm, and then involves the particle swarm optimization with linear decreasing inertia weight （LDWPSO） into it. The introduced AIPSOKH uses the strategy of parallel computing with double sub-populations. The evolution of the survival of the fittest in natural selection mechanism enhances the quality of the individuals in the population. Finally through the experiment of 8 Benchmark standard test functions, the performance of algorithm AIPSOKH is compared with the performances of the other two algorithms mentioned. Experiments demonstrate that the algorithm has a significant advantage in the global search ability and the convergence of the results.

作者:

刘沛高岳林郭伟

Liu Pei Gao Yuelin Guo Wei(Research Institute of Information and System Science, Beifang University of Nationalities, Yinchuan 750021, China)

机构地区:

北方民族大学信息与系统科学研究所

出处:

《betway官方app 学报：自然科学版》 CAS 北大核心 2017年第2期119-124,共6页

基金:

国家自然科学基金(61561001) 北方民族大学重点科研项目资助(2015KJ10)

关键词:

磷虾群算法非线性递减粒子群算法双子种群并行策略自然选择

krill herd algorithm nonlinear decreasing particle swarm optimization strategy of double sub-populations parallel computing natural selection