Multiobjective optimization of journal bearing using mass conserving and genetic algorithms

Abstract

An optimal design of hydrodynamic journal bearing using mass conserving thermal analysis and genetic algorithms is presented. Simultaneous minimization of power loss and oil flow, subjected to constraints on film thickness, film pressure, and temperature rise between the bearing surfaces, is the objective of this study. The radial clearance, L/D ratio, oil groove location, feed pressure, and the oil viscosity are the design variables. The rank-based genetic algorithm is used to deal with the discrete variables and multimodal objective functions and to capture Pareto optimal points. In view of computation economics and robustness, initial guesses of oil film pressure distribution, eccentricity ratio, and attitude angle obtained by two-dimensional analytical approach are provided for mass conserving thermal analysis. The complete optimization strategy is illustrated by a step-by-step (in four steps) approach. A comparative study of thermal and isothermal analyses is illustrated. Effects of constraints on temperature, pressure, and film thickness on the design vector are enlightened. The mass conserving thermal analysis is validated against experimental results. Pareto optimal fronts for various operating conditions are presented.