A Multi‐Objective Genetic Algorithm for Performance Optimization of a Hybrid Thrust Bearing

ABSTRACT This paper deals with the numerical simulation of texture surface thrust bearing utilizing a Newtonian lubricant. Reynolds' equation is applied to model the lubricant flow within these bearings, which feature square and rectangular micro‐depressions oriented in both circumferential and radial directions. The finite element method is employed to solve Reynolds' equation, enabling the calculation of film pressure, load‐supporting capacity, and frictional power loss. To optimum bearing performance the texture shapes are optimized based on dimple size, depth, and number of dimples. It was found that square dimples outperform rectangular dimples in terms of both load‐carrying capacity and frictional power loss. The performance enhancement of the thrust bearing was achieved by identifying optimal texturing parameters through a fuzzy logic‐based optimization framework. A multi‐objective genetic algorithm (MOGA) is utilized to identify the optimal design parameters of the bearing system, with the objectives of maximizing load‐carrying capacity and minimizing frictional power loss concurrently.