Multivariable model predictive control of a compound power-split hybrid electric vehicle during the tip-in/out process

Abstract

For a compound power split hybrid system, such as the Corun hybrid system (CHS), a frequently encountered problem is determining how to compensate for the inertia torque of the transmission via the torque distribution of the two motors. The equivalent inertia of the transmission is variable and depends on the angular acceleration. According to the traditional control scheme, the target torque of the motors is distributed by simplifying the output angular acceleration to 0. This simplification introduces deviations in the torque distribution, and it is difficult to guarantee the quality of the two control quantities simultaneously. Recent research has also focused only on the control quality of one of the outputs. To solve this problem, a direct MIMO control scheme is proposed in this paper that directly calculates the target torques of the two motors based on control deviations. A multivariable model predictive controller (MPC) was designed for the control of transient processes. The simulation results showed that the proposed control method can improve the control accuracy of two objectives significantly, and it has an excellent ability to reject torque disturbances.