近红外光谱技术结合支持向量回归法测定非法对硫磷的含量  

Determination of the Content of Illegal Parathion by Near-infrared Spectroscopy combined with Support Vector Regression

作　　者：吴厚斌[1,2] 于荣[2] 刘苹苹[2] 穆兰[2] 刘丰茂[1] 张延秋[2] 江树人[1] 闵顺耕[1] 

机构地区：[1]中国农业大学理学院,北京100193 [2]农业部农药检定所,北京100125

出　　处：《农药》2012年第4期270-272,共3页Agrochemicals

基　　金：国家自然科学基金资助项目(20575076)

摘　　要：[目的]采用近红外光谱(N I R)技术结合支持向量回归法(S V R)建立乳油中非法对硫磷的定量分析方法。[方法]通过向1.8%阿维菌素乳油中加入对硫磷原药和二甲苯溶剂来配制不同质量分数的校正集,采用SVR法建立非法对硫磷的定量分析模型。[结果]模型的决定系数(R2)、校正集均方根误差(RMSEC)、检验集均方根误差(RMSEV)、预测集均方根误差(RMSEP)分别为0.9994、0.4377、0.5065、0.7482。[结论]该法可以定量测定乳油中非法对硫磷的含量。[Aims] The method of determination of the content of illegal parathion in EC was established by near-infrared spectroscopy（NIR）, combined with support vector regression（SVR）. [Methods] The calibration set was composed of samples with different content of parathion by adding parathion TC and xylene to abamectin 1.8% EC. The model for the content of the illegal parathion was established by SVR. [Results] The determination coefficient（R2）, root mean squared error of calibration（RMSEC）, root mean squared error of validation（RMSEV）, and root mean squared error of prediction（RMSEP） for the model were 0.9994, 0.4377, 0.5065 and 0.7482 respectively. [Conclusions] The results showed the method could be used to analyze the content of illegal parathion in EC quantitatively.

关 键 词：近红外光谱 支持向量回归 对硫磷 定量 分析 

分 类 号：TQ450.7[化学工程—农药化工]