Predictions of Friction Coefficient in Hydrodynamic Journal Bearing Using Artificial Neural Networks

This paper explores the influence of frequency shaft sleeve rotation and radial load on a journal bearing made tin-babbitt alloy (Tegotenax V840) under hydrodynamic lubrication conditions. An experimental test frictional behaviour plain was performed an originally developed device for testing rotating elements: bearings. Using back-propagation neural network, based data, artificial network models were to predict dependence friction coefficient temperature in relation speed. data measured with which trained, well-trained networks mean absolute percentage error training 0.0054 % 0.0085 %, respectively, obtained. Thus, model can depending