基于粒子群算法的分段波长选择方法

Piecewise Wavelength Selection Based on Particle Swarm Optimization Algorithm

选择合适的波长数据点建立分析校正模型是克服仪器、人员和溶液杂质等因素的影响,提高光谱分析精度的一种有效方法。特别是在多组分的情况下,组分间的交互影响使得对参与建模的波长数据点进行筛选显得尤为必要。文章在概述粒子群算法基本原理的基础上,探讨了粒子群算法在光谱波长筛选中的应用,提出了基于粒子群算法的分段波长选择方法,最后采用计算机模拟的方式对该方法进行了举例说明。

The method of selecting good wavelength data to build an analysis model is an effective method for eliminating the influences of instrument, personnel, impurity of solution, and other factors, and for improving spectral analysis precision. Especially in multicomponent condition, it is quite necessary of selecting good wavelength data for the influence among many components. A basic method of particle swarm optimization algorithm is introduced briefly, then discussed is its applications in wavelength selection. A piecewise wavelength selection method based on particle swarm optimization algorithm was proposed, an example is given for explaination by making use of computer simulation.