Polynomial neural network based probabilistic hydrodynamic analysis of two-lobe bearings with stochasticity in surface roughness. (October 2022)

• Record Type: Journal Article
• Authors: Roy, B.; Mukhopadhyay, T.; Dey, S.
• Is Part Of: Tribology international. Volume 174(2022)
• Publication Date: 2022-10
• Subjects: Two-lobe journal bearing -- Machine learning based analysis of bearings -- Stochasticity in surface roughness, Polynomial neural network; Tribology -- Periodicals; 621.89
• DOI: 10.1016/j.triboint.2022.107733
• Languages: English
• ISSNs: 0301-679X

Abstract

Abstract: This paper investigates the probabilistic response of two-lobe bearings considering the uncertainty in eccentricity ratio, preload value, bearing clearance, supply pressure, oil viscosity and surface roughness. To simulate stochasticity in input variables, Monte Carlo simulation (MCS) is carried out in conjunction with the Reynolds equation using finite difference method. Polynomial neural network based machine learning model is used as a surrogate model to increase the efficiency of MCS. To assess the relative importance of the stochastic input parameters, a sensitivity analysis is carried out. The physically insightful new probabilistic results, presented here covering a wide spectrum of uncertainty sources including surface roughness, make it evident that different forms of source-uncertainties have a significant effect on the critical performance of bearings. Highlights: The compound effect of uncertainty in eccentricity ratio, preload value, bearing clearance, supply pressure, oil viscosity and surface roughness is investigated on the critical system performance of two-lobe bearings. The results reveal that a compound effect could trigger a pronounced random variation of the performance parameters due to their stochastic interaction. For the Monte Carlo simulation assisted stochastic analysis, a PNN based machine learning approach is coupled with the finite difference method in conjunction with Reynolds equation. Sensitivity analysis is carried out to assess the relative importance of the stochastic input parameters. It is concluded that dynamic response of two lobe bearings is significantly influenced by source-uncertainties, leading to the realization that such inevitable effects must be included in the design procedure for safe and reliable system performance.