烃类物质自燃点的QSPR预测研究

Study on QSPR prediction of auto-ignition temperature of hydrocarbons

应用定量构效关系(QSPR)方法对烃类物质的自燃点开展了预测研究。选取国际电工委员会数据库中的39种烃类物质作为样本集,随机选择34种作为训练集,5种作为测试集。采用遗传算法(GA)对变量进行筛选,结合线性和非线性方法分别建立多元线性回归(MLR)模型和支持向量机(SVM)模型,理论预测得到了5种烃类物质的自燃点。结果表明,2个预测模型均比较稳定,理论预测值与实验值均较为相符,GA-SVM模型预测结果较GA-MLR模型更接近于实验值,这表明自燃点与其分子结构间具有更强的非线性关系。

The auto-ignition temperature(AIT)were predicted by Quantitative Structure-Pharmacokinetics Relationship(QSPR).Thirty-nine kinds of hydrocarbons in the International Electrotechnical Commission(IEC)database were selected as sample sets,34 kinds were randomly selected as training sets and 5 kinds as test sets.Genetic algorithm(GA)was used to screen variables,multiple linear regression(MLR)model and support vector machine(SVM)model were established by combining linear and nonlinear methods respectively,and the spontaneous ignition points of 5 hydrocarbon substances were predicted theoretically.Finally,the performance and application fields of the model were evaluated.The results showed that the two prediction models are stable and have strong prediction ability and generalization performance.The theoretical predicted values are consistent with the experimental values,and the predicted results of GA-SVM model are closer to the experimental values than GA-MLR model,which indicates that the relationship between auto-ignition temperature and its molecular structure is more nonlinear.