基于拟反向学习的自适应 QPSO 算法及其在工程中的应用
- 分享到:
摘要：
为改善量子粒子群优化(quantum-behaved particle swarm optimization algorithm,QPSO)算法在求解复杂的多模问题时表现出的收敛精度差和易于陷人局部最优的问题,提出了一种基于拟反向学习的自适应 QPSO算法.首先,借鉴拟反向学习的思路,对粒子初始位置进行优化调整,增加算法搜索效率,加快收敛速度;其次,在粒子运动幅度的设置中考虑了种群进化程度和粒子聚集程度,构造了具有自适应特点的收缩-扩张因子,用于增强算法的局部控掘和全局搜索能力:然后,将混沌映射的方法引人到越界粒子的处理上,有助于算法逃离局部最优,接着,基于14个测试函数将改进算法与8种智能优化算法进行对比分析,最后借助2个具体的工程设计问题进一步检验改进算法在实际应用中的效果,实验结果表明改进算法无论在基准测试中还是在工程应用上,其搜索能力更强,整体性能表现更为均衡.
To address the issues of poor convergence accuracy and susceptibility to local optima exhibited by the quantum-behaved particle swarm optimization(QPSO) algorithm in solving complex multi-modal problems, an adaptive QPSO algorithm based on quasi-ropposite learning is proposed. Firstly, drawing on the idea of quasi-opposite learning, the initial positions of particles are optimized and adjusted to increase algorithm search efficiency and accelerate convergence speed. Secondly, in the setting of particle movement range, population evolution degree and particle aggregation degree are taken into account, and a contraction-expansion factor with adaptive characteristics is constructed to enhance the local exploitation and global exploration capabilities of the algorithm. And the method of chaotic mapping is introduced to handle out-of-range particles, which helps the algorithm escape local optima. Then the improved algorithm is compared with eight existing intelligent optimization algorithms on 14 benchmark test funetions. Additionally, the application efectiveness of the improved algorithm is examined through two real- world engineering design problems. The experimental results demonstrate that the improved algorithm exhibits stronger search capability and more balanced overall performance.
作者:
何光
He Guang
机构地区：
重庆工商大学数学与统计学院；统计智能计算与监测重庆市重点实验室
引用本文：
何光。基于拟反向学习的自适应 QPSO算法及其在工程中的应用[J].河南师范大学学报(自然科学版),2025.53 (5) :81-89.( He Guang.Adaptive QPS0 algorithm based on quasi-opposite learning and its applications in engineering[J].Journal of Henan Normal University(Natural Science Edition),2025,53(5):81-89.DO1:10.16366/j.cnki.1000-2367.2024.04.23.0004.)
基金：
重庆市科委项目；重庆市教委项目
关键词：
量子粒子群优化算法；拟反向学习；收缩-扩张因子；混沌映射；工程应用
Quantum-behaved particle swarm optimization algorithm; quasiopposite learning; contraction-expansion factor; chaotic map; engineering application
分类号：
TP301.6
