摘要
虽然相比较粒子群算法而言,量子粒子群优化算法有着更好的性能,但其仍然面临因过早收敛而陷入局部最优的问题。因此尝试将量子粒子群算法与蝙蝠算法相混合,一方面利用蝙蝠算法中的随机游走策略来避免过早地陷入局部最优,另一方面学习蝙蝠算法中发声速度的变化方式来改变量子粒子群算法中的因子。将所提算法与粒子群优化算法和量子粒子群优化算法经过5个标准测试函数和一个实际模具车间的调度模型的仿真验证,并与粒子群算法和量子粒子群算法进行对比,仿真结果表明了该算法在求解连续型问题和离散型问题的有效性和优越性。
Quantum-behaved particle swarm optimization has better performance compared with particle swarm optimization, but it still has the problem of getting trapped into local optimum with premature convergence. According to the above problem, a hybrid algorithm included quantum-behaved particle swarm optimization and bat algorithm was proposed. On the one hand, the random walk strategy of bat algorithm was used to avoid getting into local optimum, on the other hand, the speed changing of bats＇ sound was learned to transform the factor of quantum-behaved particle swarm optimization. The proposed algorithm was tested on five benchmark functions and a mould job shop scheduling example,compared with PSO（Particle Swarm Optimization） and QPSO（Quantum-Behaved Particle Swarm Optimization）. The simulated experimental results indicate the validity and superiority of the hybrid algorithm.
出处
《系统仿真学报》
CAS
CSCD
北大核心
2016年第6期1247-1254,1280,共9页
Journal of System Simulation
基金
国家自然科学基金(61572238)
国家863计划(2014AA041505)
关键词
蝙蝠算法
量子粒子群算法
调度优化
模具车间
bat algorithm
quantum-behaved particle swarm optimization
scheduling optimization
mould job shop
