Adaptive Multi-Objective Optimum Design of Journal Bearing Based on a Back-Stepping Approach

Abstract

This article describes an adaptive multi-objective optimization journal bearing design method based on a back-stepping approach. Thermohydrodynamic lubrication analysis was used for accurate prediction of lubricant flow and power loss. The multi-objective optimization design method based on a back-stepping method was used to search for the optimum journal bearing design that can adaptively identify the equilibrium position of journal bearings and avoid iteration for the calculation of the equilibrium position of journal bearings. In the optimum journal bearing design, design variables such as radius, bearing length, radial clearance, journal eccentricity ratio, and attitude angle are used to simultaneously minimize oil flow and power loss. Comparisons between the adaptive optimization method and the conventional iteration method show that the proposed method can reduce the computational time of the optimization procedure, and the ability to find a diverse set of solutions was improved with the adaptive optimization methodology.

KEY WORDS:

• Multi-Objective Optimum Design
• Back-Stepping Method
• Genetic Algorithm
• Thermohydrodynamic Lubrication Model
• Journal Bearing

ACKNOWLEDGEMENT

This work was supported by the National Nature Science Foundation of China (Nos. 51175424 and 50905143) and China Postdoctoral Science Foundation (No. 20100471634). This support is gratefully acknowledged.

Review led by David Burris