基于扰动的自适应粒子群优化算法

An Adaptive Particle Swarm Optimization Algorithm based on Disturbances

【目的】针对标准粒子群优化算法在应用中暴露出的缺点,如在迭代后期收敛速度慢、搜索精度不高、容易陷入局部最优等,提出一种基于扰动的自适应粒子群优化算法。【方法】该算法将扰动因子加入速度更新公式中,使种群搜索范围扩大;采用自适应的惯性权重,以起到平衡全局和局部寻优能力的作用;对最优粒子进行自适应的柯西变异,拓展最优粒子的搜索空间,降低粒子陷入局部最优的可能性;最后对算法进行仿真实验。【结果】新算法能够增强全局搜索能力,有效避免局部最优,具有更快的收敛速度。【结论】新算法克服了标准粒子群优化算法的缺点,为进一步研究粒子群优化算法的改进和应用提供科学依据。

【Objective】Aiming at the shortcomings of the standard particle swarm algorithm in application?such as slow convergence speed?low precision and easy to fall into local optimum at the end of iteration,an adaptive particle swarm optimization algorithm based on disturbances is proposed.【MethodslThe main improvement strategies were as follows:1)The disturbance fac-tor was added to the velocity updating formula?so that the population search range was expand-ed;2)The adaptive inertia weight was exponentially decreased in order to balance the global and local optimization;3)An adaptive Cauchy mutation on the best particle was added to expand the search space and reduce the possibility of local optimum.【Results】Through the experimen-tal simulation and comparison,the proposed algorithm could enhance global search capability?and had a higher optimization performance?so the convergence precision and convergence speed of the PSO were improved obviously.【Conclusion】The proposed algorithm could overcome the shortcomings of the standard particle swarm algorithm and provided a scientific basis for fur-ther research on the improvement and application of particle swarm algorithm.