Theoretical study of effect of the geometrical parameters on the dynamic properties of the elastic rings of an air journal bearing.

The paper investigates theoretically the effect of the geometry of the elastic rings of an air journal bearing on the elastic rings dynamic coefficients. The physical finite element method (FEM) model used to obtain the dynamic coefficients of the rings is discussed. A theoretical model is implemented to predict the effect of the geometrical parameters on the dynamic coefficients of the elastic rings. The effect of the geometrical parameters on the dynamic coefficients at different frequencies is studied using FEM. The elastic geometry that result in desired dynamic coefficients is demonstrated. Since predicting the dynamic coefficients for all possible ring geometries using FEM would be computationally expensive. A neural network (NN) is trained to predict the dynamic coefficients for all possible ring geometries generated by the different ring geometrical parameters within a given input domain. The NN results are compared to the experimentally verified FEM results and the results are in good agreement.

Using neural networks to predict the effect of the preload location on the natural frequencies of a cantilever beam.

With the evolution of computational power of computers in 20th century, neural networks (NNs) are becoming more popular in different engineering applications because of its ability to approximate static and dynamic, linear and non-linear, multi-dimensional systems. For example, they are used in industrial processes to automate assembly lines which increases its productivity and in automotive to reduce gas consumption of an engine. In this paper, NNs are utilized to reduce the computational power needed for finite element methods (FEM) simulations. A case study is taken for which NNs are used to predict the effect of the preload position and magnitude on the natural frequencies of the prestressed cantilever beam. A simple FEM model is implemented to generate the data set required to train the NN. The steps done to construct the FEM are discussed and the FEM model results are verified. The effect of the preload position on the natural frequencies of the beam is studied. A NN is then implemented to predict the natural frequencies of the beam for different beam cross-section geometries and different preload magnitudes and positions. The NN architecture, data processing and training methodology are explained. The NN and FEM results are compared to show the accuracy of the NN predictions. The results are shown to be in good agreement.