Machine learning-based performance analysis of two-axial-groove hydrodynamic journal bearings
The precise prediction of a rotor against instability is needed for avoiding the degradation or failure of the system’s performance due to the parametric variabilities of a bearing system. In general, the design of the journal bearing is framed based on the deterministic theoretical analysis. To map the precise prediction of hydrodynamic performance, it is needed to include the uncertain effect of input parameters on the output behavior of the journal bearing. This paper presents the uncertain hydrodynamic analysis of a two-axial-groove journal bearing including randomness in bearing oil viscosity and supply pressure. To simulate the uncertainty in the input parameters, the Monte Carlo simulation is carried out. A support vector machine is employed as a metamodel to increase the computational efficiency. Both individual and compound effects of uncertainties in the input parameters are studied to quantify their effect on the steady-state and dynamic characteristics of the bearing.