A Combined Machine Learning and Computational Methodology for Optimum Thrust Bearings’ Behavior in Mixed Lubrication Regime

A hydrodynamic thrust bearing could be forced to operate within the mixed lubrication regime under various circumstances. At this state, the tribological characteristics of the bearing could be affected significantly and the developed phenomena would have a severe impact on the performance of the mechanism. Recent technological advances, especially on the field of computer science, have provided tools that enhance and accelerate the modeling of thrust bearings’ operation. The aim of this study is to combine numerical analysis and machine learning techniques in order to solve the mixed lubrication problem of a tilting pad thrust bearing providing the optimum selection of lubricant and coating for the given operating conditions. For the hydrodynamic analysis of the bearing the 2-D Reynolds equation is solved numerically with the finite difference (central differences) approach. In order to describe the roughness of the profiles, both the flow factors suggested by N. Patir and H. S. Cheng (1978) and the model of J. A. Greenwood and J. H. Tripp (1970) are taken into consideration. Two lubricants, the SAE 10W40 and the SAE 10W60, are tested and compared for a variety of operating velocities and applied coatings. All the numerical analysis data are gathered and used in order to train machine learning algorithms. Three different ML methods are applied in this investigation: Multi-Variable Quadratic Polynomial Regression, Quadratic SVM and Regression Trees. The R2, is calculated and used in order to determine the best fit ML method for the current study.