Optimal shape design of steadily loaded journal bearings using genetic algorithms

This paper describes an implementation of a genetic algorithm scheme suitable to the optimal shape design of finite-width, iso-viscous, fluid film journal bearings under steady load and steady journal rotation. Assuming perfect journal cylindricity, the optimization objective is to determine the circumferential variation of sleeve geometry that maximizes bearing load capacity, subject to specified minimum film thickness. Particular emphasis is placed on the selection of the number of design variables, chromosomes per generation, crossover and mutation probabilities, and the number of generations that promote convergence to one or more optimal designs. Optimal film thickness specifications obtained using this genetic algorithm are found to be generally superior to random search methods and comparable with previously published results based on traditional optimization strategies. For finite-width bearings, optimal bearing geometry is found to offer a small improvement in load capacity and a substantial improvement in oil flow over purely cylindrical designs.