Lu Cuiping, Lv Gang, Shi Chaoyi, Qiu Duoyang, Jin Feixiang, Gu Man, Sha Wen
Appl Opt. 2020 Oct 1;59(28):8582-8587. doi: 10.1364/AO.401405.
The quantitative analyses of pH value in soil have been performed using laser-induced breakdown spectroscopy (LIBS) technology. The aim of this work was to obtain a reliable and accurate method for rapid detection of pH value in soil. Seventy-four samples were used as a calibration set, and 24 samples were used as a prediction set. To eliminate the matrix effect, the multivariate models of partial least-squares regression (PLSR) and least-squares support vector regression (LS-SVR) were used to construct the models. The intensities of nine emission lines of C, Ca, Na, O, H, Mg, Al, and Fe elements were used to fit the models. For the PLSR model, the correlation coefficient was 0.897 and 0.906 for the calibration and prediction set, respectively. Furthermore, the analysis accuracy was improved effectively by the LS-SVR method, and the correlation coefficients for calibration and prediction set were improved to 0.991 and 0.987. The prediction mean absolute error was pH 0.1 units, and the root mean square error of the prediction was only 0.079. The results indicated that the LIBS technique coupled with LS-SVR could be a reliable and accurate method for determining pH value in soil.
利用激光诱导击穿光谱（LIBS）技术对土壤中的pH值进行了定量分析。这项工作的目的是获得一种可靠且准确的土壤pH值快速检测方法。74个样本用作校准集，24个样本用作预测集。为消除基体效应，采用偏最小二乘回归（PLSR）和最小二乘支持向量回归（LS-SVR）的多元模型构建模型。使用C、Ca、Na、O、H、Mg、Al和Fe元素的九条发射线强度来拟合模型。对于PLSR模型，校准集和预测集的相关系数分别为0.897和0.906。此外，通过LS-SVR方法有效提高了分析精度，校准集和预测集的相关系数分别提高到0.991和0.987。预测平均绝对误差为pH 0.1个单位，预测均方根误差仅为0.079。结果表明，LIBS技术结合LS-SVR可成为一种可靠且准确的土壤pH值测定方法。
