基于直觉模糊熵的混合粒子群优化算法

Hybrid Particle Swarm Optimization Algorithm Based on Intuitionistic Fuzzy Entropy

为了提升粒子群算法的全局寻优与局部精细搜索能力并加快收敛速度,提出了基于直觉模糊熵的混合粒子群优化算法.该算法采用粒子的历史最优解信息构造直觉模糊熵的自适应函数,并将熵值作为扰动因子动态调节惯性权重,同时建立自适应全局最优粒子学习策略对扰动后的粒子进行训练,在保持多样性传播的基础上选择学习对象,使粒子探索更多新区域,实现种群间的协作与并行进化.通过仿真实验,将本文算法与两种衍生算法以及其他改进粒子群算法在11个测试函数上进行比较,结果表明,本算法在求解精度、收敛速度和寻优效率上均有更好表现.

In order to improve the global and local fine search capabilities of the particle swarm algorithm and ac⁃celerate the convergence speed,hybrid particle swarm optimization algorithm based on intuitive fuzzy entropy is pro⁃posed.The algorithm constructs an adaptive function of intuitive fuzzy entropy by using the information of the historical optimal solution of particles,and uses the entropy value as a disturbance factor to dynamically adjust the inertia weight.At the same time,it establishes an adaptive global optimal particle learning strategy to train the disturbed particles,chooses learning objects based on maintaining the diversity of propagation,enables the particles to explore more new ar⁃eas,and realizes the cooperation and parallel evolution among populations.Through simulation experiments,the algo⁃rithm is compared with two derivation algorithms and other improved particle swarm optimization algorithms on 11 test functions.The results show that the algorithm performs better in solving accuracy,convergence speed and optimization efficiency.