Synchronized Gear Shift Control Using RLS Estimator for Two Speed Gear System

Abstract

Fixed gears are widely used in electric vehicle (EV) powertrain. Nowadays, multi-speed gear is applied in some special vehicles to achieve a wide speed range or an extremely high torque. It also increases overall efficiency up to 5%. Normally, synchronizing devices are adopted to avoid gear shifting shock. However, the synchronizer adds more cost and space, thereby it is not used in EV powertrain. Instead, a simple sleeve mechanism is used to engage and disengage the gears. In EV, the speed synchronization is performed by the propulsion motor itself. However, the vehicle speed changes during the period of neutral (disengaged) state. Furthermore, the amount of speed variation depends on the load and road conditions. For a better synchronization, the vehicle speed change must be estimated. In this work, the recursive least squares (RLS) algorithm is employed to estimate the vehicle parameters, as well as the vehicle speed. A motor control method is developed that synchronizes the motor speed with that of vehicle side gear. Superiority over the general method is demonstrated by simulation studies. Experimental setup is constructed and some experimental data are shown.