Investigation of The Effects of Design Parameters on The Performance of Gas Lubricated Two Lobe Journal Bearings Using Artificial Neural Networks

Artificial Neural Networks (ANNs) are information processing systems which in recent years have attracted the attention of many researchers for modeling of the systems and also finding solution for linear and nonlinear problems in various fields. In the present work, using ANNs method, the performance of two lobe gas lubricated noncircular journal bearing are investigated. For this purpose, the effect of compressibility number (bearing number), mount and tilt angles on the static and dynamic performance characteristics of these bearings, such as attitude angle, load carrying capacity, power loss and stability margines, are studied. The present results of ANNs method compared with earlier reported results obtained by finite element method (FEM), are in good match. It is while the ANNs method is simple and consumes not much computer time. The results also show that tilling of the two lobe bearings can improve the stability margines and this effect is more at higher compressibility numbers.