人工神经网络用于有机磷酸酯类化合物的定量结构色谱保留相关研究  

QSRR Study of Organophosphate Compounds Using Artificial Neural Network

作　　者：何琴[1] 

机构地区：[1]许昌学院化学化工学院,河南许昌461000

出　　处：《分析科学学报》2013年第4期483-487,共5页Journal of Analytical Science

基　　金：河南省科技厅成果鉴定(豫科鉴委字[2012]第1078号);河南省教育厅自然科学研究计划(No.2009B150023);许昌市科技计划(No.5007);许昌学院校内基金(2013067)

摘　　要：采用误差反传前向人工神经网络(ANN),研究了35种有机磷酸酯类化合物在3种不同极性固定相上的结构与其色谱保留(QSRR)之间的定量关系。以其分子电性距离矢量(或分子拓扑指数)作为输入、色谱保留值作为输出,采用内外双重验证的办法分析和检验所得模型的稳定性和外推能力。结果表明,ANN模型获得了比多元线性回归(MLR)模型更好的拟合效果。使用MLR模型时QSRR模型相关性受色谱固定相极性的影响,而采用ANN模型无此现象。同时,ANN模型解决了QSRR中预测维数为1时耗时较长的问题。通过ANN建模可以同时预测3种不同极性固定相上的色谱保留值,可大大缩短建模和预测所需的时间。The systematic study of the quantitative structure-retention relationship （QSRR） of 35 organophosphate compounds on three different polar stationeries was performed by the artificial neural network（ANN） based on the back propagation algorithm. The stabilization and generalization ability of the model constructed by ANN were verified by the inner-external test when the molecular electronegativity-distance vector（or the topolohical indices） and the chromatographic retention values on three different polar stationeries were used as the inputs and outputs of the neural network, respectively. The results showed that the performance of ANN method was better than that of MLR method. The correlation of QSRR model was influenced by the polarity of the stationary phase using the MLR method. However, there was no such problem using the ANN method. Meanwhile, the ANN model could also overcome the shortcoming that long time was generally needed＇in one-dimensional prediction. The chromatographic retentions on the three different stationary phases were predicted simultaneously, which drastically shortened the time generally needed in modeling and prediction.

关 键 词：有机磷酸酯类化合物 定量结构色谱保留相关 人工神经网络 

分 类 号：O622[理学—有机化学]