水体重金属激光诱导击穿光谱定量分析方法对比研究

建立了适用于激光诱导击穿光谱探测的多元线性回归、神经网络回归和支持向量机回归三种定量反演算法模型, 以水体重金属Ni为例进行了回归实验测试和对比分析. 多元线性回归、神经网络回归和支持向量机回归的平均相对标准偏差分别为7.60%, 4.86%, 2.35%; 最大相对标准偏差分别为23.35%, 15.20%, 8.29%; 平均相对误差分别为25.98%, 10.58%, 2.72%, 最大相对误差分别为116.47%, 47.38%, 9.89%. 研究为进一步实现水中痕量金属元素的快速定量分析提供了方法和数据参考. 关键词： 光谱学 激光诱导击穿光谱 支持向量机回归 重金属

Comparative analysis of quantitative method on heavy metal detection in water with laser-induced breakdown spectroscopy

The quantitative analysis models of multiple linear regression, neural network regression and support vector machine regression with laser-induced breakdown spectroscopy are established in this paper. Heavy metal Ni in water selected as research object is tested and comparativly analyzed. The average relative standard deviations of multiple linear regression, neural network regression and support vector machine regression are 7.60%, 4.86% and 2.35%, and the maximum standard deviations are 23.35%, 15.20% and 8.29% respectively, the average relative errors are 25.98%, 10.58% and 2.72%, and the maximum relative errors are 116.47%, 47.38% and 9.89% respectively. Methods and reference data are provided for the further study of fast measurement of tracing heavy metals in water by laser induced breakdown spectroscopy technique.
Keywords： spectroscopy laser-induced breakdown spectroscopy support vector machine regression heavy metals