温度对复配乳油的近红外光谱定量分析模型的影响  

Influence of Temperature on Near-Infrared Spectroscopic Quantitative Analysis Models of Compound Emulsifiable Concentrate

作　　者：王冬[1] 熊艳梅[1] 黄蓉[1] 吴厚斌[1] 闵顺耕[1] 

机构地区：[1]中国农业大学理学院应用化学系,北京100193

出　　处：《分析化学》2010年第9期1311-1315,共5页Chinese Journal of Analytical Chemistry

基　　金：国家自然科学基金(No20575076)资助项目

摘　　要：研究了样品温度对克螨特-高效氯氰菊酯乳油制剂的近红外光谱定量分析模型预测能力的影响。在20,25,30和35℃温度下分别采集了农药样品的近红外光谱,并采用偏最小二乘算法结合全交互验证的模型验证方法对两种有效成分分别建立了各温度的定量校正模型以及混合温度定量校正模型,以外部检验集的RMSEP值作为模型预测能力的评价指标。结果表明,温度对克满特、高效氯氰菊酯成分的预测结果有一定的影响,混合温度模型对不同温度样品的预测结果表现出了较强的适应性。因此,对于克螨特-高效氯氰菊酯复配乳油制剂,建立混合温度校正模型,使模型具有良好的温度适应性,可以最大限度地降低预测误差,以适应不同温度样品的分析需要。Influence of temperature on near-infrared spectroscopy quantitative analysis model of the compound emulsifiable concentrate of propargite and beta cypermethrin was studied.The near-infrared spectra of the compound emulsifiable concentrate were collected by FT-NIR spectrometer at temperatures of 20,25,30 and 35℃,respectively.The algorithm of partial least square regression was employed to develop the calibration models of the two active ingredients under the different temperature and the all-temperature respectively,while the validation method was full cross validation.The root mean square error of prediction（RMSEP） of the external validation sets were regarded as the criterion of prediction ability of the calibration models.The result indicated that the variation of temperatures will influence the prediction result of propargite and beta cypermethrin partly.The all-temperature calibration models of the two ingredients show the better suitability respectively.Therefore,it will meet the needs of the analysis demand of the samples under different temperatures for the emulsifiable concentrate of propargite and beta cypermethrin to develop the all-temperature calibration model which has the better suitability for temperature,which will decrease the prediction error furthest.

关 键 词：温度 近红外光谱 偏最小二乘法 克螨特 高效氯氰菊酯 

分 类 号：TQ450.7[化学工程—农药化工]