基于粒子群算法的椭偏法测量数据反演

Ellipsometry Data Inverting Based on PSO

为解决椭偏法测量薄膜厚度和折射率实验数据处理较为复杂的问题,采用一种新的基于群体智能的优化算法—粒子群算法处理实验数据.以单层吸收薄膜的测量为例,利用该算法进行数据处理.实验结果表明,三个薄膜参数(折射率n,消光系数k和薄膜厚度d)可以同时获得,而且在未知参数确切范围情况下,大范围内进行搜索仍然能保证快速收敛到最优解.文中算法和遗传模拟退火算法以及利用椭偏仪数据处理软件得出的结果相比较,表明该算法计算精度高,收敛速度快.

Aimed to solve the complexity of data processing of the film＇s parameter by ellipsometry,the paper introduced PSO, a optimization arithmetic based on swarm intellect theory, to process the experimental data. Test data of one layer absorbing film were processed with this arithmetic. Results showed that not only three film parameters（reflective index, extinction coefficient, film thickness） were derived at the same time but also the best solution were convergenced quickly searching in large range. Compared the results of this method with those from ellipsometer data process software, the precision and inversing speed result were better than those of genetic simulated annealing optimization.