| 许均华
张红莉
贺伟
邓湘波
李勇
刘冠琼
徐文泱.基于RSM和GA-ANN优化超声-顶空气相色谱法测定植物油中六号溶剂残留量前处理条件[J].中国油脂,2020,45(2):132~137.[XU Junhua
ZHANG Hongli
HE Wei
DENG Xiangbo
LI Yong
LIU Guanqiong
XU Wenyang.Optimization of pretreatment conditions for determination of No. 6 solvent residual in vegetable oil by ultrasound-headspace gas chromatography based on response surface methodology and genetic algorithm-artificial neural network[J].China Oils and Fats,2020,45(2):132~137.] |
| 基于RSM和GA-ANN优化超声-顶空气相色谱法测定植物油中六号溶剂残留量前处理条件 |
| Optimization of pretreatment conditions for determination of No. 6 solvent residual in vegetable oil by ultrasound-headspace gas chromatography based on response surface methodology and genetic algorithm-artificial neural network |
| |
| DOI：10.12166/j.zgyz.1003-7969/2020.02.026 |
| 中文关键词: 超声 六号溶剂 响应面法 神经网络 植物油 溶剂残留 |
| 英文关键词:ultrasound No.6 solvent response surface methodology neural network vegetable oil solvent residual |
| 基金项目:湖南省食品药品监督管理局食品药品安全科技项目（湘食药科 R201701） |
|
| 摘要点击次数: 1554 |
| 全文下载次数: 530 |
| 中文摘要: |
| 建立植物油中六号溶剂残留量的超声-顶空气相色谱(UV-HGC)分析方法。采用单因素实验考察超声功率、超声时间、超声温度对六号溶剂残留量检测的影响，并用响应面法(RSM)和遗传算法-人工神经网络(GA-ANN)对UV-HGC条件进行优化。结果表明：RSM和GA-ANN相对误差、决定系数（R2）分别为9.15%、0.891 4和5.95%、0.949 5，说明GA-ANN比RSM具有更强的预测和优化能力。采用GA-ANN优化获得UV-HGC最佳前处理条件为超声功率500 W、超声时间7 min、超声温度50 ℃。在最佳前处理条件下，植物油中的六号溶剂在9.53~200 mg/kg质量浓度范围内线性关系良好，相关系数（r）为0.999 4，在低、中、高3个加标水平下，平均回收率分别为95.5%、98.6%、100.4%，相对标准偏差(RSD)均小于3.5%。方法的检出限(LOD)为1.85 mg/kg，定量限(LOQ)为9.53 mg/kg。该方法操作简单、快速、灵敏、准确，适用于植物油中六号溶剂残留量的检测。 |
| 英文摘要: |
| A new method for the determination of No.6 solvent residual in vegetable oil by ultrasound-headspace gas chromatography (UV-HGC) was established.The effects of ultrasonic power, ultrasonic time and ultrasonic temperature on the determination of No.6 solvent residual in vegetable oil were investigated by single factor experiment. Subsequently, the conditions of UV-HGC were optimized using response surface methodology (RSM) and genetic algorithm-artificial neural network (GA-ANN). The results indicated that the GA-ANN showed better prediction and optimization abilities than RSM with lower relative error value (5.95% versus 915%) and higher determination coefficient (R2) (0.949 5 versus 0.891 4).The optimal pretreatment conditions of UV-HGC were obtained by GA-ANN as follows:ultrasonic power 500 W, ultrasonic time 7 min and ultrasonic temperature 50 ℃.Under the optimal conditions, the No.6 solvent in vegetable oils showed good linear relationships in the range of 9.53-200 mg/kg mass concentration with the correlation coefficients 0.999 4. The average recoveries were 95.5%, 98.6% and 100.4% at low, medium and high standard addition levels respectively, and the relative standard deviation (RSD) was less than 3.5%.The limits of detection (LOD) and quantification (LOQ) of the method were 1.85 mg/kg and 9.53 mg/kg.The method was simple, rapid, sensitive and accurate, and it was suitable for the determination of No.6 solvents residual in vegetable oil. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
