Abstract: This paper proposes an improved quantum delta-potential-well-model particle swarm optimization (QDPSO) algorithm based on simulated annealing (SA) theory. The strong searching ability of SA is employed to overcome the precocious shortcoming of QDPSO, and a performance test according to benchmarks functions is taken to confirm astringency and rapidity of the algorithm compared with SPSO and QDPSO. Simulation results show that the algorithm with better stability and astringency is an excellent global optimization method.

Key words: optimization arithmetic, simulated annealing, particle swarm optimization(PSO)

摘要：

提出了一种基于模拟退火技术的量子空间模型粒子群优化(QDPSO)改进算法，利用模拟退火算法(SA)的搜索能力克服QDPSO算法在寻优过程中早熟的缺点，通过标准测试函数进行性能测试，验证了算法的收敛性和快速性，并和标准PSO及QDPSO进行了比较。仿真结果表明，该算法具有更好的稳定性和收敛性，是一种良好的全局优化方法。

关键词: 优化算法, 模拟退火, 粒子群算法

CLC Number: 

• TP18