基于LIBS技术的脐橙中铜含量联合区间偏最小二乘法定量分析  

Quantitative analysis of copper content in navel orange with synergy interval partial least squares by LIBS

作　　者：何秀文[1] 陈添兵[1] 黄林[1] 周华茂[1] 姚明印[1] 黎文兵[1] 刘木华[1] 

机构地区：[1]江西农业大学工学院生物光电及应用重点实验室,江西南昌330045

出　　处：《南昌大学学报（理科版）》2014年第4期377-380,共4页Journal of Nanchang University(Natural Science)

基　　金：国家自然科学基金资助项目(31271612);江西省教育厅科技计划资助项目(GJJ10681);江西省学术带头人计划资助项目(09004004)

摘　　要：选用60个果形相当的脐橙样品进行激光诱导击穿光谱（LIBS）试验，引入联合区间偏最小二乘法（siPLS）的分析方法，实现水果脐橙中重金属铜元素的 LIBS定量分析。先对脐橙在320～340 nm波段内激光诱导击穿光谱进行九点平滑处理，并提取特征光谱CuI324．75 nm为分析线。采用siPLS方法将光谱划分为25个子区间，当选择其中5、14、16、224个子区间建立的模型时效果最佳，其校正集相关系数和均方根误差分别为0．9883和5．61，预测集相关系数和均方根误差分别为0．9792和8．62。研究表明，与偏最小二乘法（PLS）相比，改进的 siPLS模型更能有效利用光谱中的信息，减少模型计算量，提高预测能力。为进一步实现水果中痕量重金属元素的快速定量分析提供了方法。A set of 60 navel orange samples with sizable shape were used for laser induced breakdown spec-troscopy （LIBS）experiment.The synergy interval partial least square （siPLS）methods were employed for a quantitative analysis of heavy metal element Cu in navel orange,The LIBS spectrum （320-340nm ）of na-vel orange were preprocessed by combining the nine smoothing method,and the characteristic spectrum CuI：324.75nm line was verified.The optimal siPLS model was obtained with 25 intervals and the optimal combinations of intervals were number 5,14,16 and 22 respectively.The model was achieved with correla-tion coefficient 0.9883 and root mean square error of cross validation （RMSECV）5.61 in calibration set and the correlation coefficient 0.9792 with a root mean squared error of prediction （RMSEP）8.62 in pre-diction set.The result showed that：siPLS mode could utilize the information more about the spectra,re-duce the computation involved in,and also improve the forecast ability compared with PLS mode.There-fore,the siPLS model was proved to be feasible in improving the prediction accuracy of LIBS.

关 键 词：激光诱导击穿光谱 联合区间偏最小二乘法 脐橙 铜元素 

分 类 号：TN247[电子电信—物理电子学]