Based on the statics and quasi-statics analysis methods, the thermal deformation calculation model of a deep-groove ball bearing was constructed for the helical gear transmission system of a high speed electric drive, and the radial and axial bearing stiffness values of the bearing were calculated under the thermal deformation in this study. The obtained radial and axial stiffness values were introduced into the established dynamics model of helical gear system, and the influence of changed bearing stiffness, resulting from the thermal deformation, on the nonlinear dynamic characteristics of gear pair was analyzed using the Runge–Kutta method. The results show that the axial and radial deformations of bearing occur due to the increase of working speed and temperature, in which the axial stiffness of bearing is improved but the radial stiffness is reduced. The decreasing degree of axial stiffness and the increasing degree of radial stiffness decrease with the gradually increasing working rotational speed. When considering the influence of thermal deformation on the bearing stiffness, the helical gear system will have nonlinear behaviors, such as single periodic, double periodic, and chaotic motion with the change of working speed. Therefore, in order to improve the nonlinear dynamic characteristics of high speed electric drive gear systems, the influence of bearing stiffness change on the dynamic performance of a gear system should be considered in the industrial applications.

Purpose: The lubrication of the high-speed reducer of an electric vehicle is investigated. The specific contents include visualization of the flow field inside reducer, lubrication evaluation of bearings and efficiency experiment. Design/methodology/approach: The flow field inside reducer at five working conditions: straight, uphill, downhill, left lean and right lean is simulated by smoothed particle hydrodynamics (SPH). According to the instantaneous number of particles through bearings, the lubrication states of bearings are evaluated. The test platform is set up to measure the efficiency of the reducer. Findings: The flow field inside the reducer is obtained, the lubrication of bearings needs to be improved, the efficiency of the electric vehicle reducer meets the requirement. Originality/value: The SPH method is used to simulate lubrication instead of using the traditional grid-based finite volume method. A novel method to evaluate the lubrication of bearings is proposed. The method and conclusions can guide electric vehicle reducer design.