Analysis of Oil Film Thickness in Hydrodynamic Journal Bearing Using Artificial Neural Networks

Abstract

Journal bearings are allowed for transmission of large loads at mean speed of rotation. These bearings are susceptible to large amplitude lateral vibration due to self exited instability which is known as oil whirl or Synchronous whirl. This oil whirl depends on many parameters such as oil film thickness, viscosity of lubricant; load on bearing, Inertia of fluid etc. out of which oil film thickness plays an important role in operation of journal bearings. As oil film thickness decreases metal to meal contact occurs, this further can damage the journal bearing. So during the operation minimum oil film thickness should be maintained which can avoid the metal to metal contact and further increases the life of bearing. This paper presents a theoretical calculation of oil film thickness and experimental verification of same on journal bearing test rig, different journal speeds and loads are considered for the analysis. The collected experimental data of oil film thickness is used for training and testing an artificial neural network. The neural network is a feed forward network. Back propagation algorithm is used to update the weight of the network during the training. Finally, neural network predictor has predicted oil film thickness which is in close agreement with experimental oil film thickness by test rig.