An improved particle swarm optimization algorithm for adaptive inertial weights
-
摘要:
针对传统粒子群算法容易陷入局部最优、收敛速度快, 导致收敛精度低等弊端, 一种改良的自适应惯性权重的粒子群算法在本文中被提出.通过对粒子飞行速度和位置变化的分析, 并结合粒子的自适应值动态调整惯权重, 使得算法能够在全局空间和局部空间搜索之内达到良好的均衡.选择典型的测试函数, 将改进后的粒子群算法(PSO-A)、带收缩因子的粒子群算法(PSO-X)和惯性权重线性递减粒子群算法(PSO-W)的性能进行了对比分析.最后采用MATLAB软件进行算法仿真, 从结果得出, 本文所提出的自适应改变惯性权重的粒子群算法在收敛精度、收敛速度上都取得了明显的改善.
Abstract:In view of the traditional particle swarm optimization (pso) algorithm convergence speed, it's easy to fall into local optimum and cause disadvantages such as low convergence accuracy and it is not easy to converge, an improved particle swarm algorithm of adaptive inertia weight is proposed. Through the analysis of particle flying speed and position change, combined with the adaptive value of particles used to dynamically adjust the weight, enables the algorithm to the to achieve a good balance between global search and local search. Choosing typical test functions, the improved particle swarm optimization (PSO-A), with compression factor of particle swarm optimization (PSO-X)and inertia weight linear decreasing of the particle swarm optimization (PSO-W) performance is analyzed. Finally, MATLAB software is used for simulation. The results show that the improved particle swarm optimization algorithm has apparent improved its convergence speed and accuracy.
-
Key words:
- particle swarm optimization /
- inertia weight /
- adaptive /
- convergence accuracy
-
表 1 测试结果
PSO-A PSO-X PSO-W F1 a 6.450 3e-025 5.274 5e-038 7.420 5e-006 b 2.340 3e-035 2.782 4-002 5.345 0e-033 F2 a 34.653 2 28.323 0 32.765 3 b 24.003 2 12.563 5 30.056 0 F3 a 8.340 5e-012 5.250 3e-023 0.003 4 b 0.002 3 0.081 2 0.095 7 F4 a 6.723 4e-023 0.135 4 0.150 2 b 0.043 4 0.145 3 5.324 0e-034 -
[1] KENNEDY J, EBERHART R C. Particle swarm optimization[C]//Proceedings of the IEEE International Conference on Neural Networks. Perth, 1995: 1942-1948. [2] SHI Y, EBERHART R C. A modified particle swarm optimizer[C]//Proceedings of the IEEE Congress on Evolutionary Computation. Anchorage, 1998: 69-73. [3] CLERC M, KENNEDY J. The particle swarm:explosion stability and convergence in a multi-dimensional complex space[J]. IEEE Transactions on Evolutionary Computation, 2002, 6(1):58-73. doi: 10.1109/4235.985692 [4] 黄廷林, 戴雪峰, 张卉, 等.改进PSO算法在多水源供水系统优化调度中的应用[J].中国给水排水, 2013, 28(23): 64- 68. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zgjsps201323015 [5] 范成礼, 邢清华, 范海雄, 等.带审敛因子的变邻域粒子群算法[J].控制与决策, 2014, 28(4):696-700. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=kzyjc201404020 [6] 余胜威.MATLAB优化算法案例分析与应用[M].北京:清华大学出版社, 2014:201- 202. [7] 朱正伟, 郭晓, 刁小敏.基于混合免疫粒子群算法的WSN移动sink路径研究[J].微电子学与计算机, 2018, 35(5):89-94. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=wdzxyjsj201805017 [8] 崔丽珍, 李晓宇, 胡海东, 等.基于自然选择粒子群算法的WSN覆盖优化策略[J].微电子学与计算机, 2018, 35(4):73-78. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=wdzxyjsj201804015 [9] 李军民, 王洪涛.一种改进惯性权重策略的粒子群优化算法[J].西安科技大学学报, 2010, 30(5):604-608. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=xakyxyxb201005020 [10] 邵鹏, 吴志健.基于改进PSO算法的Rosenbrock函数优化问题的研究[J].计算机科学, 2013, 40(9):194-197. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjkx201309041 [11] 董平平, 高东慧, 田雨波, 等.一种改进的自适应惯性权重粒子群优化算法[J].计算机仿真, 2012, 29(12):283-286. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjfz201212067 [12] 罗华.一种惯性权重自适应的粒子群优化算法[J].电子科技, 2017, 30(3):30-32. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dzkj201703009 -