New and energy efficient cylindrical roller bearing design using contact mechanics and optimization methods

In this thesis, the roller crown profile and ring raceway of an NJ type cylindrical roller bearing are optimized. It is intended to minimize the friction moment without sacrificing the fatigue life of the existing cylindrical roller bearing. The crown heights in the roller profile and the ring raceway were selected as design variables, and finite element analyzes were performed for various combinations of these variables. Separately developed finite element models for frictional moment and fatigue life estimation were developed to reflect the operating and loading conditions of the bearing in the test mechanism. By means of the finite element method, frictional moment and fatigue life values were obtained for different values of the design variables, and the optimization process was started in line with these data. Since repetitive simulations are required for the optimization work, a surrogate model has been used, thus the computational cost was reduced. Kriging method was used to create a surrogate model, and a genetic algorithm was used to solve the optimization problem. At the end of this process, the optimum design determined was produced; friction moment and fatigue life tests were carried out. When the experimental outputs related to the current design and optimum design are compared; It was found that the frictional moment of the optimum design was 29% lower than that of the existing bearing, while the L10 fatigue life was only 3% less. When the test results were evaluated statistically, it was evaluated that the difference between the frictional test results of the current design and the optimum design was statistically significant, but there was no statistically significant difference between the fatigue test results.