摘要
采用人工神经网络(ANN)方法建立以磨损体积量度来表征抗磨性能的定量构效关系(QSPR)模型。以载荷因子变量及3个分子描述符(包括润滑油添加剂分子中P原子个数,描述分子的大小及对称的2D自相关描述符,以及描述分子极性与反应活性的3D描述符)作为模型的输入变量,采用误差反向传播的ANN拟合磨损体积量度与4个描述符之间的关系。结果表明:建立的模型具有统计意义,且与文献报道模型相比,模型更为精确;磨损体积量度与3个分子描述符之间存在非线性关系,表明3个分子描述符能反映影响润滑油添加剂抗磨性能的重要结构因素。
By means of artificial neural network(ANN),a quantitative structure-property relationship (QSPR) model was constructed to predict the antiwear properties measured with wear volumes of 36 lubricating additives under three loads,196,294 and 392 N.The load factor of measurement and three molecular descriptors (including the number of phosphorous atoms in lubricating additives,a two-dimension autocorrelation descriptor reflecting molecular size and symmetry,and a three-dimension descriptor denoting molecular polarity and reactivity) were used as input vectors to develop the model.The typical back-propagation (BP) neural network was employed to fit the relationships between the four descriptors and wear volumes.The results show that the model in this paper possesses statistical significance and is more accurate compared with the models reported in the literature.There are non-linear relationships between the wear volumes and the three molecular descriptors,which suggests the molecular descriptors can reflect the important structure factors affecting antiwear properties of lubricant additives.
作者
黄磊
禹新良
刘拥君
HUANG Lei;YU Xinliang;LIU Yongjun(College of Chemistry and Chemical Engineering,Hunan Institute of Engineering,Xiangtan Hunan 411104,China)
出处
《润滑与密封》
CAS
CSCD
北大核心
2019年第1期103-108,共6页
Lubrication Engineering
基金
湖南教育厅重点项目(16A047)
湖南工程学院研究生科技创新项目(2017)
关键词
抗磨性能
人工神经网络
润滑油添加剂
构效关系
antiwear properties
artificial neural network
lubricant additive
structure-property relationship
