Neural Networks Based Study of Friction Coefficient Variation in Sliding Bearings

Abstract

The paper presents the possibility of using the neural networks approach for the analysis of friction coefficient evolution in case of sliding bearings. Several non-metallic bearing materials are investigated, both with water and emulsion as lubricant. The results show that the neural networks can be successfully used for prediction of friction coefficient evolution during bearing service. This way the working life of the bearing can be predicted with higher accuracy, leading to preventing the mechanical systems failure.