Optimal Design of High Speed Angular Contact Ball Bearing Using a Multiobjective Evolution Algorithm

Objectives like long fatigue life and low heat generation need to be achieved collectively in design of high speed angular contact ball bearing. The traditional multiobjective optimum method needs explicit objective functions and weighting factors which are not easy to determine. These disadvantages can be overcome by a multiobjective evolution algorithm. The commonly used NSGA-II(non-dominated sorting genetic algorithm-II)is adopted. The objectives are rating life and spin frictional power loss. The optimized maximum rating live of 7007AC bearing operating at a DN (bore/mm × shaft speed/rpm) of 2.2 million is 7.5 times more than live of current design. When the spin frictional power loss is the same as current design´s, the optimized design´s rating life is 36% more than that of current design.