Abstract
<div class="section abstract"><div class="htmlview paragraph">This paper introduces a novel approach to modeling Torque Converter (TC) in conventional and hybrid vehicles, aiming to enhance torque delivery accuracy and efficiency. Traditionally, the TC is modelled by estimating impeller and turbine torque using the classical Kotwicki’s set of equations for torque multiplication and coupling regions or a generic lookup table based on dynamometer (dyno) data in an electronic control unit (ECU) which can be calibration intensive, and it is susceptible to inaccurate estimations of impeller and turbine torque due to engine torque accuracy, transmission oil temperature, hardware variation, etc. In our proposed method, we leverage an understanding of the TC inertia – torque dynamics and the knowledge of the polynomial relationship between slip speed and fluid path torque. We establish a mathematical model to represent the polynomial relationship between turbine torque and slip speed. The mathematical model is used in the forward torque converter model to calculate current impeller torque based on known input speed and turbine speed and reverse torque converter model to calculate target input speed to deliver driver torque request. The parameters of the polynomial torque converter model are online identified with a Kalman Filter to adapt the model to the varying transient operating conditions of the powertrain. The effectiveness of this approach is demonstrated through vehicle results, showcasing improved performance under changing powertrain conditions.</div></div>