摘要
果蔬农药残留严重危害人类健康。实验使用农药为敌敌畏,用自来水分别配制不同浓度的农药溶液。将400个喷施不同浓度农药的脐橙样品风干后采集其荧光光谱,光谱范围是350～1 800 nm,然后用气相色谱法检测脐橙表皮的农药残留量。把采集的光谱数据分成30个光谱区间,通过间隔偏最小二乘法对数据进行分析计算得出5个最佳的特征光谱区间。基于5个特征光谱区间建立农药残留的偏最小二乘法预测模型,其校正组和预测组的相关系数r分别为0.860 7和0.837 5。试验结果表明应用荧光光谱方法检测脐橙表皮敌敌畏农药残留的方法是可行的,而且通过特征光谱区间的选择还能有效地简化模型的复杂程度,为今后农药残留的快速无损检测提供理论基础。
Pesticide residue in fruits and vegetables seriously harms human health.In the experiment,pesticide solutions were of different concentrations prepared with dichlorvos and running water.400 navel oranges were sprayed with different pesticide solutions.After air-dried,fluorescence spectra were collected from the navel oranges,and the spectra range was from 350 to 1 800 nm;then the pesticide residues on navel orange surface were determined by gas chromatography.The collected spectral data were divided into 30 spectral intervals,the data were analyzed by interval partial least squares method and 5 optimal characteristic spectral intervals were obtained.Based on the 5 spectral intervals,partial least squares models of pesticide residue determination were built,and the correlation coefficients of calibration set and prediction set were 0.860 7 and 0.837 5,respectively.The results show that the method of using fluorescence spectrum to determinate dichlorvos residue on navel orange surface is feasible,meantime,the determination models can be simplified by choosing characteristic spectral interval and this provides theoretical basis for fast non-destructive detection of pesticide residue.
出处
《江西农业大学学报》
CAS
CSCD
北大核心
2011年第2期394-398,共5页
Acta Agriculturae Universitatis Jiangxiensis
基金
国家自然科学基金资助项目(30760101)
新世纪优秀人才支持计划资助项目(NCET-09-0168)
江西省科技厅农业科技支撑计划(2009BNB05705)
江西省教育厅科学技术研究(GJJ08513)
关键词
荧光光谱
敌敌畏残留
无损检测
偏最小二乘法
脐橙
fluorescence spectrum
dichlorvos residue
non-destructive detection
partial least squares
navel orange
