Drivability Optimization by Reducing Oscillation of Electric Vehicle Drivetrains

Abstract

The drivetrain of electric vehicles differs significantly from vehicles with combustion engines. Current concepts of electric vehicle drivetrains usually have a low damping. Typically, there is no clutch to separate the inertial mass of the electric drive machine from the rest of the vehicle drivetrain. External (road unevenness, potholes, etc.) and internal excitation (torque changes of the electric machine, brake interferences, etc.) cause jerk oscillation and sometimes high component stress. These excitations can be reduced by suitable drivability functions, to which a reference filter can also be assigned. A common approach known from conventional drivetrains is to limit the gradient of the demand torque of the drive machine or the driver′s desired torque in order to influence the torque build-up of the drive machine and to reduce the excitation of jerk oscillations. A second approach is the use of a prefilter. The prefilter uses the inverse dynamics of the drivetrain to influence the demand torque of the drive machine. In this paper, the influence of a prefilter based on the inverse dynamics of electric vehicle drivetrains to reduce oscillations is investigated. In addition, an anti-jerk control enhances the drivability function afterwards. All investigations are made on a hardware-in-the-loop test bench to create reproducible results.