Open AccessOptimization of crankshaft main bearing lubrication performance considering bearing profiles
Author(s) -
Qingchuan Du,
Ying Cheng,
Peirong Ren,
Zhongwei Zhang
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1601/6/062051
Subject(s) - lubrication , crankshaft , bearing (navigation) , lift (data mining) , fluid bearing , materials science , work (physics) , genetic algorithm , mechanical engineering , artificial neural network , multi objective optimization , power loss , power (physics) , computer science , control theory (sociology) , mechanics , engineering , physics , artificial intelligence , thermodynamics , control (management) , machine learning , data mining
It is the aim of this work to reduce friction power loss of main bearings by optimization. To this purpose, elastohydrodynamic (EHD) model is used for EHD calculations for different main bearings. BP neural network is implemented to establish the approximation model for bearings. Then, multi-objective optimization of bearings using genetic algorithm is formulated and conducted. It is found that a more compliant bearing profile can provide hydrodynamic lift during film lubrication while bearing profiles have more significant impact on lubrication performance in comparison to other key parameters. The results of the BP network model using the genetic algorithm agree closely with the calculated value based on EHD-MBD model. The presented approach allows reliably to conduct the optimization of bearings. After optimization, the friction power loss is significantly reduced while the minimum oil film thickness increases and the total pressure drops.