Design of neural model for analysing journal bearings considering effects of transverse and longitudinal profile

Abstract

Purpose

The purpose of this paper is to study the effects of shaft surface profiles on the load carriage capacity of journal bearings using an experimental and neural network approach. The paper aims to inspect the performance characteristics of journal bearing systems; the presence of transverse and longitudinal roughness on journal‐shaft surfaces is studied using the proposed neural network.

Design/methodology/approach

The collected experimental data such as pressure variations are employed as training and testing data for an artificial neural network (ANN). Quick propagation algorithm is used to update the weight of the network during the training.

Findings

As a result, a shaft with a transverse profile displays a favorable performance as far as load carriage capacity is concerned. Moreover, the proposed neural network structure outperforms the available experimental model in predicting the pressure as well as the load carriage capacity.

Originality/value

The paper discusses a new modelling scheme known as ANN. A neural network predictor has been employed to analyze the effects of shaft surface profiles in the hydrodynamic lubrication of journal bearings.

Citation

Sinanoğlu, C. (2009), "Design of neural model for analysing journal bearings considering effects of transverse and longitudinal profile", Industrial Lubrication and Tribology, Vol. 61 No. 3, pp. 132-139. https://doi.org/10.1108/00368790910953640