LI Junjie,ZHOU Yan,HUANG Yan,et al.Research on Particle Number Change based on MQHOA Optimization Algorithm[J].Journal of Chengdu University of Information Technology,2019,(04):352-357.[doi:10.16836/j.cnki.jcuit.2019.04.005]
基于MQHOA优化算法的采样粒子数量变化研究
- Title:
- Research on Particle Number Change based on MQHOA Optimization Algorithm
- 文章编号:
- 2096-1618(2019)04-0352-06
- 关键词:
- 多尺度量子谐振子算法; 函数优化; 采样; 维度; 粒子数
- Keywords:
- multi-scale quantum harmonic oscillator algorithm; function optimization; sampled; dimension; particle number
- 分类号:
- TP301.6
- 文献标志码:
- A
- 摘要:
- 在多尺度量子谐振子算法(MQHOA)的优化迭代过程中,采样粒子数量的多少对算法求解 成功率和计算效能有重要影响。以基准测试函数在2维和多维状态下分别进行研究,找寻不同 函数对应的最佳粒子数。研究发现,结构复杂度较高的目标函数需要较大的采样粒子数进行 求解,而相对简单的单峰凸函数所需采样粒子数较小。最佳粒子数可以作为算法衡量目标函 数结构复杂度的重要参考依据,针对不同的目标函数,采用相对应的最佳粒子数进行求解,能 够以最小的计算代价获取最佳的求解效果。
- Abstract:
- In the optimized iterative process of the multi-scale quantum harmonic oscillator algorithm(MQHOA), the number of sampled particles has an important influence on the algorithm’s success rate and computational efficiency. The benchmark function is studied in the 2-dimensional and multidimensional states respectively to find the optimal particle number corresponding to different functions. It is found that the objective function with higher structural complexity requires a larger number of sampled particles to solve, while the relatively simple one-peak convex function requires a smaller number of sampled particles. The optimal particle number can be used as an important reference for the algorithm to measure the complexity of the objective function structure. For different objective functions, the corresponding optimal particles number can be used to solve the problem, and the best solution can be obtained with the minimum computational cost.
参考文献/References:
[1] Kirkpatrick,S.Optimization by simulated annealing:quantitative studies
[J].Stat.Phys,1984,34(5/6):975-986.
[2] Brooke J,Bitko D,Rosenbaum T F,et al.Quantum Annealing of a Disordered
Spin System[J].Science, 2001,284(5415):779-781.
[3] Kennedy,J,Eberhart, R.Particle swarm optimization[J].In:IEEE ICNS
1995,2002,4:1942-1948.
[4] Tian N. Quantum-Behaved Particle Swarm Optimization with Cooperative
Coevolution for Large Scale Optimization[C]. International Symposium on
Distributed Computing and Applications for Business Engineering and Science.
IEEE, 2016:82-85.
[5] Dorigo M,Maniezzo V, Colorni A.Ant system: optimization by a colony of
cooperating agents[J]. IEEE Trans.Syst.Man.Cybern,1995,26(1):29-41.
[6] R C Eberhart,J Kennedy.A new optimizer using particle swarm theory
[J].Proceedings of the Sixth International Symposium on Micro Machine and Human
Science,1995,1:39-43.
[7] Sun J, Xu W, Feng B.A global search strategy of quantum-behaved
particle swarm optimization[J].IEEE CIS,2004,1:111-116.
[8] Sun J,Fang W,Wu X,et al.Quantum behaved particle swarm
optimization:analysis of individual particle behavior and parameter selection
[J].Evolutionary Computation,2012,20(3):349-393.
[9] Wang P,Huang Y,Ren C,et al.Multi-scale quantum harmonic oscillator for
high-dimensional function global optimization algorithm.Chin
[J].Electron,2013,41(12):2468-2473.
[10] Wang P,Ye XG,Li B,et al.Multi-scale quantum harmonic oscillator
algorithm for global numerical optimization[J].Applied Soft
Computing,2018,69:655-670.
[11] 王鹏,黄焱,袁亚男,等.多尺度量子谐振子算法的收敛特性[J].电子学
报,2016,44(8):1988-1993.
[12] 黄焱,王鹏,程琨,等.多尺度量子谐振子优化算法的并行性研究[J].通信学
报,2016,37(9):68-74.
[13] 王鹏,黄焱.具有能级稳定过程的MQHOA优化算法[J].通信学报,2016,37(7):79-
86.
[14] 王鹏,黄焱.多尺度量子谐振子优化算法物理模型[J].计算机科学与探索,2015,9
(10):1271-1280.
[15] 王鹏,黄焱,任超,等.多尺度量子谐振子高维函数全局优化算法[J].电子学
报,2013,41(12):2468-2473.
备注/Memo
收稿日期:2019-02-25 基金项目:国家自然科学基金资助项目(60702075)