| 345 | 23 | 337 |
| 下载次数 | 被引频次 | 阅读次数 |
针对粒子群优化算法在求解高维问题时易出现的早熟收敛、停滞现象,提出一种拟随机初始化模拟退火粒子群算法.采用Hammersley方法对算法进行初始化,可以提高算法在高维搜索空间的搜索能力,进一步将模拟退火思想引入到粒子群优化算法中,结合粒子群优化算法的快速寻优能力和模拟退火算法的概率突跳特性,使算法具有跳出局部最优从而实现全局最优的能力.分别在5个经典测试函数上测试算法的性能,仿真实验结果表明,提出的算法有效克服了传统粒子群优化算法在求解高维空间优化问题时易出现的停滞现象,在进化后期仍保持较强的搜索能力,提高了传统粒子群优化算法在高维空间的全局寻优能力.
Abstract：To overcome the shortcomings of particle swarm optimization( PSO) algorithm such as premature convergence and stagnation when solving the high-dimensional problems,a quasi-randomized simulated annealing( SA)-PSO algorithm was proposed. The performance of algorithm in high-dimensional optimization space could be improved by using the Hammersley initialization. And the idea of SA algorithm was introduced into the PSO algorithm,combining with the fast searching ability of PSO and the probabilistic jumping property of SA,to jump out of local optimal algorithm to achieve the global optimum. The proposed algorithm could effectively overcome the stagnation phenomenon,enhance the global search ability in high-dimensional space. The proposed algorithm was then tested on 5 different functions,and the results demonstrated better optimization ability over the traditional PSO algorithm.
[1]GARNIER S,GAUTRAIS J,THERAULAZ G.The biological principles of swarm intelligence[J].Swarm intelligence,2007,1(1):3-31.
[2]刘逸.粒子群优化算法的改进及应用研究[D].西安:西安电子科技大学,2013.
[3]FENG M Y,PAN H.A modified PSO algorithm based on cache replacement algorithm[C]//Tenth International Conference on Computational Intelligence and Security.Kunming,2014:558-562.
[4]LANGDON W B,POLI R.Evolving problems to learn about particle swarm and other optimisers[C]//IEEE Congress on Evolutionary Computation.Edinburgh,2005:561-578.
[5]HIGASHI N,IBA H.Particle swarm optimization with Gaussian mutation[C]//Proceedings of the IEEE Swarm Intelligence Symposium.Indianapolis,2003:72-79.
[6]MIRANDA V,FONSECA N.EPSO:evolutionary particle swarm optimization,a new algorithm with applications in power systems[C]//Transmission and Distribution Conference and Exhibition.Yokohama,2002:2491-2505.
[7]刘爱军,杨育,李斐,等.混沌模拟退火粒子群优化算法研究及应用[J].浙江大学学报(工学版),2013,47(10):1722-1730.
[8]YANG H F,YANG Z Y,YANG Y,et al.An improved particle swarm optimization algorithm based on simulated annealing[C]//10th International Conference on Natural Computation.Xiamen,2014:529-533.
[9]吕丹,童创明,钟卫军.基于粒子群和模拟退火算法的混合算法研究[J].计算机工程与设计,2011,32(2):663-666.
[10]BERGH F,ENGELBRECHT A P.A cooperative approach to particle swarm optimization[J].IEEE transaction on evolutionary computation,2004,8(3):225-239.
[11]LIANG J J,QIN A K,SUGANTHAN P N,et al.Comprehensive learning particle swarm optimizer for global optimization of multimodal functions[J].IEEE transaction on evolutionary computation,2006,10(3):281-295.
[12]HARALD N E,SIANGJING Y A.Halton-type sequences from global function fields[J].Science China mathematics,2013,56(7):1467-1476.
基本信息:
DOI：10.13705/j.issn.1671-6841.2016022
中图分类号:TP18
引用信息:
[1]王杰,李慧慧,彭金柱.一种拟随机初始化模拟退火粒子群算法[J],2016,48(03):75-81.DOI:10.13705/j.issn.1671-6841.2016022.
基金信息:
教育部高等学校博士学科点专项科研基金资助项目(20124101120001);; 河南省教育厅科学技术研究重点项目(14A413009);; 中国博士后科学基金资助项目(2014T70685)