摘要
为了更有效地对润滑油中的磨损磨粒进行识别,探讨了基于粗糙集和神经网络的磨粒识别。它首先利用粗糙集理论对磨粒特征参数进行约简,这样能够大大减少了神经网络的输入维数。然后介绍了一种径向基神经网络,并利用它对磨粒进行分类。对20个磨粒进行识别,磨粒分类分对14个,分错6个,识别率达到70.0%。
In order to classify wear debris in lubricating oil more effectively, a wear debris recognition method based on rough set and neural network was put forward. At first, debris feature parameters are simplified based on rough sets theory, and the input information-dimensions is reduced. Then a radius basis function neural network was introduced, and it is used to classify wear debris. 20 wear debris was classified by the method,and result shows that 14 is right,6 is fault, the ratio of recognition reaches 70.0%.
出处
《润滑与密封》
CAS
CSCD
北大核心
2007年第1期162-164,共3页
Lubrication Engineering
