论文部分内容阅读
针对粒子群算法由于算法可调参数较少,在求解多维函数时极易陷入早熟收敛的问题,提出一种改进的变参数粒子群算法。根据粒子运动特性,对粒子速度更新公式进行改进,使各项都融入相应的权重因子,通过权重因子调整粒子寻优性能。通过3个标准测试函数进行验证,并与其他算法进行比较。仿真结果表明,通过设置不同的权重因子,所提算法具有更好的寻优精度和执行能力,在求解多维函数时亦能取得较好的效果。
Because the algorithm of particle swarm optimization has fewer adjustable parameters, it is easy to fall into premature convergence when solving multidimensional functions. An improved variable parameter particle swarm optimization algorithm is proposed. According to the movement characteristics of particles, the updating formula of particle velocity is improved, all the factors are integrated into the corresponding weight factors, and the particle optimization performance is adjusted by the weighting factors. Validated by 3 standard test functions and compared with other algorithms. The simulation results show that by setting different weight factors, the proposed algorithm has better precision and execution ability, and can achieve better results when solving multidimensional functions.