基于随机过程的粒子群改进算法

Improved Particle Swarm Optimizat iioonn Algorithm Based on Stochastic Process

针对PSO算法在寻优过程中,容易陷入局部最优解和早熟的问题,提出基于随机过程的粒子群改进(SPPSO)算法.该算法从随机变量的角度以及布朗运动对某点值的吸收与反射思想,提出三个方面的改进,从而提高算法的全局搜索能力.最后对6个测试函数进行仿真实验,结果表明SPPSO算法在寻优精度、收敛速度以及寻优正确率等方面的性能优于PSO算法和权重线性递减的PSO算法.

Aiming at the problem that the PSO algorithm is easy to fall into the local optimal solution and premature in the process of optimization,a particle swarm improvement(SPPSO)algorithm based on random process is proposed. This algorithm proposes three improvements to improve the global search ability of the algorithm from the perspective of random variables and the idea that Brownian motion absorbing and reflecting a certain point value. Finally,a simulation experiment is carried out on the six test functions. The results show that the SPPSO algorithm is superior to PSO algorithm and PSO algorithm with linearly decreasing weights in terms of the optimization accuracy,convergence speed and optimization accuracy rate.