LIBS与变量选择PLS结合的含油土壤中Cu,Ni定量分析

Quantitative Analysis of Cu and Ni in Oil-Contaminated Soil by LIBS Combined With Variable Selection Method and PLS

土壤中重金属元素检测是环境保护事业的重点之一,因此亟需一种能够快速检测土壤重金属浓度的定量分析手段。该研究旨在建立一种基于激光诱导击穿光谱结合偏最小二乘法的含油土壤中重金属元素快速定量检测方法。通过激光诱导击穿光谱(LIBS)获取含油土壤光谱数据,采用偏最小二乘法(PLS)对样品中铜、镍元素进行定量分析预测,并在此基础上,结合区间以及后向区间法对全谱进行变量筛选,构建形成区间偏最小二乘法(iPLS)和后向区间偏最小二乘法(BiPLS)定量分析铜、镍元素含量的模型。结果表明:后向区间偏最小二乘法(BiPLS)在剔除了干扰信息的基础上,保留了更多的有效光谱信息,获得了比PLS和iPLS更好的预测结果:铜元素的测试集预测结果的决定系数(RP^2)和均方根误差(RMSEP)分别为0.944 9和0.036 3,相对分析误差(RPD)为3.0;镍元素的测试集预测结果的RP^2和RMSEP分别为0.933 7和0.041 4, RPD为2.6,两元素的BiPLS预测结果相较于PLS和iPLS方法均有所提升。因此,针对含油土壤重金属元素光谱信息, BiPLS算法相较于iPLS和PLS算法更适合与LIBS光谱相结合,筛选对Cu和Ni两种重金属元素定量分析贡献度较大的特征变量,进而提升模型的预测效果。该方法将促进LIBS技术应用于土壤品质在线评价。

The detection of heavy metals in soil is one of the emphases of environmental protection.This paper aims to establish a fast and quantitative method for the determination of heavy metal elements in soil,based on LIBS and combined with the PLS method.We used PLS model to quantitatively analyze and predict the contents of Cu and Ni elements in oil-contaminated samples.On this basis,the variables of the full spectrum were screened by combining the Interval method and the Backward Interval method,which formed the Interval Partial Least Square(iPLS)and the Backward Interval Partial Least Square(BiPLS).The experimental results showed that the BiPLS method retained more spectral information after removing the interference information,and obtained better-predicted results than PLS and IPLS.The R2P and RMSEP of the predicted results of the test set for the copper element are 0.9449 and 0.0363,respectively,and the RPD reached 3.0.Those of the predicted results of the test set for nickel element are 0.9337 and 0.0414,respectively,and the RPD reached 2.6.Compared with the PLS and iPLS methods,the prediction results of the BiPLS method of the two elements were significantly optimized,the predictive ability was significantly improved,and the accuracy was much better.Therefore,In the analysis of heavy metal elements in oil-contaminated soil by LIBS technique,BiPLS is more suitable than iPLS and PLS for screening the feature variables that contribute greatly to the quantitative analysis of Cu and Ni elements,so as to improve the prediction effect.This method will promote the application of LIBS technology to the online evaluation of soil quality.