Computational design and optimization of rough elliptic bore journal bearing

Abstract

Journal bearing is a hydrodynamic bearing most widely used in rotating types of machinery, especially deployed in engines and power plants. Such a bearing consists of a rotating journal supported in a stationary bore of a sleeve. Due to the converging-diverging film profile and incompressible damping offered by flowing lubricant, it develops a reactive force after the shaft achieved the desired speed. In this research, the realistic assumption of journal bearing considering bore ellipticity, shaft misalignment effect, and surface roughness, adds much geometric as well as operational information. Such information in terms of coefficients largely affects the tribological performance output. The objective is to understand the contributions of such irregularities, as ignoring such effects may lead to improper design and manufacturing practices. Regulating such irregularities in a particular range may help reduce friction, and enhance operational life through smooth service. Hence it is essential to evaluate the operational efficiency of journal bearing through computational design and optimization considering all these irregularities in the variable range. Such that the knowledge of bearing critical parameters adjudged through simulation can be considered during the manufacturing process. The decision-making in journal bearing design in this research is achieved through various optimization techniques such as Taguchi-GRA, Fuzzy-GRA, Artificial Neural Network, and Response Surface Methodology, and the accuracy of such decision is verified through proper validation. newline