Robust Feedforward and Feedback Control Integration Using H-Infinity Loop Shaping for Electrified Vehicle Powertrains

Abstract

This paper introduces a novel robust control strategy that combines model inversion-based feedforward control with H-infinity loop shaping feedback control. Feedforward control is widely utilized across various applications and aims to enhance tracking accuracy in control systems. However, its efficiency and impact on system stability are significantly susceptible to model inaccuracies. Conventionally, feedforward control inputs are appended to pre-existing feedback controls without proper evaluation of their effects on the overall system stability. In this study, we reformulate the inversion-based feedforward control law as an equivalent pre-compensator, seamlessly integrating it into the feedback control framework to ensure system stability. We employ H-infinity loop shaping to robustly stabilize the complete control architecture. We demonstrate the effectiveness of this strategy through its application to gear shift management in an electrified vehicle with a dual-clutch transmission (DCT). Simulations performed on a rigorously validated MATLAB/SimDriveline model of an electrified powertrain verify that our approach significantly improves tracking performance, particularly during transient states, while maintaining robust stability.