摘要
用反向传播神经网络方法研究了润滑油旋转氧弹时间与基础油化学组成之间的关系。研究表明，网络方法比常用的线性或指数关联法更能准确地关联和预报润滑油基础油的抗氧性。
A back propagation neural network model was used to study the relationship between the time of rotory bomb oxidation of lube base oil and its chemical composition.It is shown that ANN method can correlate and predict oxidation stability with quite better accuracy than linear or nonlinear regression methods.The methods for selecting input parameters and avoiding overfitting phenomena were also studied.
出处
《润滑油》
CAS
1996年第6期41-43,共3页
Lubricating Oil
关键词
预测
润滑油
基础油
抗氧性
人工神经网络
network analysis,prediction,lubricating oil,base oil,oxidation stability
