摘要
基于DRAGON描述符,采用多元线性回归(MLR)、支持向量机(SVM)和投影寻踪回归(PPR)方法,建立了3种预测模型,预测了270个挥发性有机化合物(VOCs)在大气对流层中与硝基自由基(NO_(3)·)反应的动力学常数(以-log k_(NO_(3))表示),并探讨了对反应降解机理有影响的分子结构因素.与MLR模型、SVM模型相比较,PPR模型稳健可靠且泛化能力更好,说明非线性模型更适用于VOCs与NO_(3)·5反应速率常数的拟合.对于测试集,PPR模型给出R 2=0.956,均方根误差RMSE=0.434,平均绝对相对偏差AARD=2.563%.
Based on DRAGON and CODESSA theoretical molecular descriptors of volatile organic compounds(VOCs),quantitative structure-property relationship(QSPR)models were developed by using multi-linear regression model(MLR),Support Vector Machines(SVM)and Project Pursuit Regression(PPR)to study their reactivity with nitrate radicals in the troposphere.Structural features affecting the reaction between the VOCs and nitrate radicals were also investigated.Rate constant was transformed into negative logarithmic unit,-log k_(NO_(3)).PPR model performed better than the MLR model and the SVM model both in the fitness and in the prediction capacity,indicating its good generalization capability.For the test set,PPR gave a predictive squared correlation coefficient R 2 of 0.956,root mean square error(RMSE)of 0.434 and absolute average relative deviation(AARD)of 2.563%,respectively.The prediction results were in good agreement with the experimental values.
作者
丛建业
楚留意
任月英
CONG Jian-ye;CHU Liu-yi;REN Yue-ying(School of Environmental and Municipal Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)
出处
《兰州文理学院学报（自然科学版）》
2022年第6期84-88,共5页
Journal of Lanzhou University of Arts and Science(Natural Sciences)
