基于微粒群算法的l_p数据拟合及其应用

Particle Swarm Optimization Based on l_p Data Fitting and Its Applications

在动态测试数据处理中,常常要进行稳健回归分析和最小最大值回归分析,讨论了微粒群算法及其在lp数据拟合中的应用.微粒群算法通过多个粒子在解空间中根据自身的信息和群体的信息不断调整自己的位置进行寻优,在寻优过程中粒子间不断地进行信息交流,使得算法收敛速度很快,特别适合用于函数优化,从而能够在lp数据拟合中得到很好的应用.实例计算结果表明,该方法能够更准确地进行lp数据拟合,理论上可以以任意逼近真实值,从而减小了计算误差,并且有更快的收敛速度,可以快速收敛到全局最优解,因而具有一定的理论意义和现实意义.

Robust regression analysis and minimal residual error analysis are two aspects of data processing of dynamic measurement. The Particle Swarm Optimization （PSO） algorithm and its application on lp data fitting are described, In PSO algorithm, every particle adjusts its position to find good results through its own information and particle swam. Every particle communicates with the others in every iteration, and PSO algorithm converges quickly, PSO algorithm has some advantages in function optimization and can be applied to lp Data Fitting. At last, examples and related results prove its validity. This method can make lp Data Fitting very precise, and decrease the calculation error, meanwhile global optimum solution can be obtained more rapidly than genetic algorithm, This method has the theoretical and practical signifieanees.