Bending fatigue is an essential parameter that needs to be considered in the improvement process of the power density and reliability of gear drives. Quantitative relations among the manufacturing parameters, surface ...Bending fatigue is an essential parameter that needs to be considered in the improvement process of the power density and reliability of gear drives. Quantitative relations among the manufacturing parameters, surface integrities, and fatigue performance are not clear, which seriously limits the effectiveness of an anti-fatigue design. For this work, tooth-bending fatigue tests of carburized gears with different surface integrities were performed using a pulsator. The effects of the manufacturing parameters and surface integrities on the gear fatigue, such as surface hardness and residual stress, were investigated. The experimental results revealed that due to the improvement of surface integrities after shot peening, the nominal bending stress number(fatigue limit) increased by 6.3%–31.1%, with an amplitude range of 39–143 MPa. A supervised learning algorithm of a random forest was implemented to determine the contribution of the surface hardness and surface residual stress to the nominal stress number. An empirical formula was proposed to predict the nominal stress number considering the surface integrities. The prediction error was less than 7.53%, as verified by several gear-bending fatigue tests. This provided theoretical support for the modern, anti-fatigue design of the gears.展开更多
基金supported by the National Key R&D Program (Grant No.2020YFB2008200)the National Science and Technology Major Project(Grant No. 2019-VII-0017-0158)。
文摘Bending fatigue is an essential parameter that needs to be considered in the improvement process of the power density and reliability of gear drives. Quantitative relations among the manufacturing parameters, surface integrities, and fatigue performance are not clear, which seriously limits the effectiveness of an anti-fatigue design. For this work, tooth-bending fatigue tests of carburized gears with different surface integrities were performed using a pulsator. The effects of the manufacturing parameters and surface integrities on the gear fatigue, such as surface hardness and residual stress, were investigated. The experimental results revealed that due to the improvement of surface integrities after shot peening, the nominal bending stress number(fatigue limit) increased by 6.3%–31.1%, with an amplitude range of 39–143 MPa. A supervised learning algorithm of a random forest was implemented to determine the contribution of the surface hardness and surface residual stress to the nominal stress number. An empirical formula was proposed to predict the nominal stress number considering the surface integrities. The prediction error was less than 7.53%, as verified by several gear-bending fatigue tests. This provided theoretical support for the modern, anti-fatigue design of the gears.
