基于膜系统的粒子群算法
Particle Swarm Optimization Algorithm Based on Membrane System

DOI: 10.12677/CSA.2019.97158, PP. 1406-1415

Keywords: 粒子群算法，膜系统，多目标优化，Pareto前沿，非支配排序
Particle Swarm Optimization, Membrane System, Multi-Objective Optimization, Pareto Frontier, Non-Dominated Sorting

Full-Text   Cite this paper   Add to My Lib

Abstract:

由于粒子群算法具有易于理解实现等特点，受到科学与工程领域的广泛关注，但其搜索策略较为单一，使得算法很难获得Pareto前沿且容易陷入局部最优和无限迭代，在全局搜索和收敛性方面还有一定的不足。受P系统理论启发，本文提出了一种基于膜系统框架的多目标粒子群优化算法用于解决多目标优化问题。根据P系统的层次结构、对象和规则，在基本膜中采用粒子群算法实现并行搜寻策略，在表层膜中采用非支配排序和拥挤距离机制来提高算法收敛速度，并通过进化规则以保持解的多样性。仿真实验采用KUR、ZDT系列和DTLZ系列测试函数对MPSO算法进行测试，并与其他多目标优化算法进行对比，包括MOPSO、dMOPSO、SMPSO、MMOPSO、MOEA/D、SPEA2、PESA2、NSGAII。实验结果表明：新算法能够保证解的多样性，快速收敛并接近于真实的Pareto前沿。
Due to the particle swarm algorithm has the characteristics of easy understanding and implementation, it is widely concerned in the field of science and engineering, but its search strategy is relatively simple, making it difficult for the algorithm to obtain the Pareto frontier and easy to fall into local optimal and infinite iteration. There are still some deficiencies in the aspect of global search and convergence. Inspired by the P system theory, this paper proposes a multi-objective particle swarm optimization algorithm based on the membrane system framework to solve the mul-ti-objective optimization problem. According to the hierarchical structure, objects and rules of P system, the particle swarm algorithm is used to implement the parallel search strategy in the base film. The nondominated sorting and congestion distance mechanism are used in the surface film to improve the convergence speed of the algorithm, and the evolution rule is used to maintain the diversity of solution. The simulation experiment uses the KUR, ZDT series and DTLZ series test functions to test the MPSO algorithm and compare it with other multi-objective optimization algorithms, including MOPSO, dMOPSO, SMPSO, MMOPSO, MOEA/D, SPEA2, PESA2, NSGAII. The experimental results show that the new algorithm can guarantee the diversity of the solution, is fast convergence and close to the real Pareto frontier.

References

[1]	Deb, K. (2001) Multi-Objective Optimization Using Evolutionary Algorithms. John Wiley & Sons, Chichester.
[2]	宁伟康. 进化多目标优化算法研究及其应用[D]: [博士学位论文]. 西安: 西安电子科技大学, 2018.
[3]	Li, X. (2003) A Non-Dominated Sorting Particle Swarm Optimizer for Multi-Objective Optimization. Genetic and Evolutionary Computation, Genetic and Evolutionary Com-putation Conference, Chicago, 12-16 July 2003, 37-48. https://doi.org/10.1007/3-540-45105-6_4
[4]	Coello, C.A.C., Pulido, G.T. and Lechuga, M.S. (2004) Handling Multiple Objec-tives with Particle Swarm Optimization. IEEE Press, Piscataway. https://doi.org/10.1109/TEVC.2004.826067
[5]	Zhang, Q. and Li, R. (2013) Cloud Resource Scheduling Based On Improved Particle Swarm Optimization Algorithm by Membrane Computing. Computer Engineering and Applications, 49, 40-44.
[6]	Liu, C., Han, M. and Wang, X. (2011) A Multi-Objective Evolutionary Algorithm Based on Membrane Systems. The 4th International Workshop on Advanced Computational Intelligence, Wuhan, 103-109. https://doi.org/10.1109/IWACI.2011.6159983
[7]	Parsopoulos, K.E. and Vrahatis, M.N. (2005) Unified Particle Swarm Opti-mization for Tackling Operations Research Problems. Swarm Intelligence Symposium, Pasadena, 8-10 June 2005, 53-59.
[8]	Hu, X. and Eberhart, R.C. (2002) Multi-Objective Optimization Using Dynamic Neighborhood Particle Swarm Optimization. Congress on Evolutionary Computation, Honolulu, 12-17 May 2002, 1677-1681.
[9]	Dou, Z. and Gao, L. (2012) Feature Selection in Conditional Random Fields Using a Membrane Particle Swarm Optimizer. Journal of Xidian University, 39, 107-112.
[10]	Kennedy, J. and Eber-hart, R. (1995) Particle Swarm Optimization. Proceedings of IEEE International Conference on Neural Networks, Perth, 27 Novem-ber-1 December 1995, Vol. 4, 1942-1948. https://doi.org/10.1109/ICNN.1995.488968
[11]	Paun, G., Rozenberg, G. and Salomaa, A. (2009) Handbook of Membrane Computing. Oxford University Press, Oxford. https://doi.org/10.1007/978-3-642-11467-0
[12]	韩丽莎. 类细胞P系统在划分聚类中的研究[D]: [硕士学位论文]. 济南: 山东师范大学, 2015.
[13]	葛平平. 可满足性问题的改进型类组织P系统的求解研究[D]: [硕士学位论文]. 淮南: 安徽理工大学, 2015.
[14]	王康. 细胞型和神经型P系统的应用问题研究[D]: [硕士学位论文]. 成都: 西华大学, 2011.
[15]	张葛祥, 潘林强. 自然计算的新分支——膜计算[J]. 计算机学报, 2010, 33(2): 208-214.
[16]	Pan, L.Q., Wang, J. and Hoogeboom, H.J. (2011) Spiking Neural P Systems with Astrocytes. Neural Computation, 24, 805-825. https://doi.org/10.1162/NECO_a_00238
[17]	Ye, L. and Guo, P. (2016) Design a Membrane System for Solving Linear Equa-tions. Journal of Computational and Theoretical Nanoscience, 13, 1561-1568. https://doi.org/10.1166/jctn.2016.5081
[18]	公茂果, 焦李成, 杨咚咚, 等. 进化多目标优化算法研究[J]. 软件学报, 2009, 20(2): 271-289.

Full-Text