摘要
基于航空发动机轴承腔润滑中所存在的气液两相流问题, 采用基于神经网络的理论方法建立预测模型, 以便进行轴承腔内气液两相流流型的识别。研究以管道气液两相流为原型, 采用 3种典型的神经网络对流型进行模式识别,通过考察 3种网络的辨识率, 发现BP网络的识别方法具有较高的准确性。
Based on the problem of the gas-liquid two-phase flow in aero-engine bearing chamber lubrication, the model was bulit on the basis of theoretical method of using neural network, from which the gas-liquid two-phase flow regime can be identified in bearing chamber. The gas-liquid two-phase flow in pipes was adopted as the prototype of study, three types of neural network were used to identify the flow regime, and the recognition possibility of three types of network was compared. The results show that BP network is more precisely to identify the flow regime.
出处
《润滑与密封》
EI
CAS
CSCD
北大核心
2005年第2期44-46,共3页
Lubrication Engineering
