The optimal design model of double-row angular-contact ball bearing was established. Dynamic load rating

static load rating and the weighted sum of these two load capacities were set as objective functions respectively

the bearings were optimally designed with immune algorithm. The design parameters included number of rolling elements in one row

the rolling element diameter

inner race groove curvature radii

outer race groove curvature radii and the pitch diameter. The optimal results of immune method were compared with those of genetic algorithm

and the immune algorithm achieved more efficient results. Furthermore

compared with the designs in the handbook of rolling bearings

dynamic load rating and static load rating of the optimized bearings(3210 and 3218) had an increase of more than 60%.