全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...
环境化学  2015 

取代芳烃的生物降解性与结构相关性研究

Keywords: 生物降解性,线性回归,神经网络,预测

Full-Text   Cite this paper   Add to My Lib

Abstract:

采用量子化学MOPAC-AM1法计算了42种取代芳烃的生成热Hf、分子最高占有轨道能EHOMO、分子量MW、分子总表面积TSA及偶极矩μ.分别采用线性回归分析法和人工神经网络法对所研究化合物的生物降解性参数BOD进行QSBR研究.对训练组而言,线性方法和神经网络法的平均预测误差分别为15.9%和11.4%;而测试组化合物的平均百分误差分别为14.5%和13.0%.无论对于测试组还是训练组,神经网络法的预测都更精确.

References

[1]  Layton AC,Lajoie,Molecular Diagnostics for Polychlorinated Biphenyl Degradation in Contaminated Soils.Ann NY Acad.Sci.,1994,721
[2]  叶常明.多介质环境问题研究.北京:化学工业出版社,1998,100
[3]  Dearden J C,Descriptors and Techniques for Quantitative Structure-Biodegradability Studies.SAR and QSAR in Environmental Research,1996,17-26
[4]  Dearden J C,Nicholson R M,The Prediction of Biodegradability by Use of Structure-Activity Relationships:Correlation of Biological Oxiiygen Demand with Atomic Charge Difference.Pestic.Sci.,1986,17∶305-310
[5]  刘次全,量子生物学及其应用.北京:高等教育出版社,1990,16
[6]  Dearden J C,Cronin M T D,Quantitative Structure Biodegradability Studies:An Investigation of the Miti Aromatic Compound Data-Base.Biodegradation Prediction.Peijnenburg W J G M and Damborsky J(ed.),Academic Press,1996,93-105
[7]  郭明,许禄,酚类化合物QSAR研究.环境科学学报,1998,18(21)∶22-26
[8]  张爱茜,韩朔睽,沈洲等,运用回归分析与人工神经网络预测含硫芳香族化合物好氧生物降解速率常数.环境科学,1998,19(1)∶37-40
[9]  Tabak H H,Rakesh G,Prediction of Biodegradation Kinetics Using a Nonlinear Group Contribution Method.Environmental Toxicology and Chemistry,1993,12∶251-260
[10]  胡守仁,神经网络导论.北京:国防科技大学出版社.1993
[11]  Covind R,Lei L,Development of Structure Biodegradability Relationships(SBRs)for Estimating Half-Lives of Organic Contaminants in Soil Stystems.Peijnenburg W J G M and Damborky J(Ed),Biodegradation Prediction,Academic Press,1996,115-138

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133