Predictions of Minimum Fluid Film Thickness of Journal Bearing Using Feed-Forward Neural Network

Abstract

Neural network approach plays a significant role in the performance predictions of mechanical systems due to its fast response and ability to solve complex problems. In present work, feed-forward neural network model is developed for the predictions of minimum fluid film thickness of journal bearing. Literature data is used for the validation of the model. The results obtained by neural network model are highly accurate and precise. The accuracy of the model is achieved through gradient descent algorithm. The decrease in fluid film thickness is observed with increase in external applied radial load. The predictions are made within and out of the prescribed range for radial load as input parameter. This model is best suited for the predictions of performance characteristics of journal bearings.