基于粒子群优化算法的散乱数据光顺拟合

Smooth fitting of scattered data based on particle swarm optimization algorithm

在散乱数据光顺拟合问题的计算中,引入了群体智能粒子群优化算法,增加了计算的自动程度以及客观性,优化中提出了优于传统罚函数法的模糊罚函数法,将模糊集合理论和粒子群优化算法有机地结合起来,并通过对节点序列内在关联性的分析,提出了适合邻域搜索类算法实施的邻域结构,以获得目标函数的全局解,最后给出了数值仿真实例。

A particle swarm optimization （PSO） algorithm is introduced to improve the automation and objectivity of the calculation in the fitting and smoothing of scattered data. A fuzzy punish func- tion superior to the traditional one is presented to combine the fuzzy clustering theory and PSO algorithm. By analyzing the intrinsic relationship of the nodal point series, the neighborhood structure fitted into the stochastic algorithm is proposed to obtain the global solutions of the objective function. A numerical simulation is also given.