A multi-objective optimal torque distribution strategy for four in-wheel-motor drive electric vehicles

Cheng Lin, Sheng Liang, Jian Chen*, Xiang Gao

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

37 Citations (Scopus)

Abstract

Since four in-wheel-motor drive electric vehicles (4IDEVs) are overactuated systems, the torque distribution strategy is crucial for improving the system efficiency, lateral stability, and safety. Hence, this paper proposes a multiobjective optimal torque distribution strategy for 4IDEVs to improve the vehicle yaw stability performance and energy efficiency. First, a motor energy loss model is built to describe the motor power loss characteristics, and an energy efficiency control allocation (EECA) method over the NEDC is proposed to analyze the model accuracy. Then, a hybrid model predictive control (hMPC)-based nonlinear yaw stability controller is employed to calculate the reference yaw moment and the active steering angle. Finally, a multiobjective controller is designed to minimize the drivetrain power loss while ensuring the vehicle stability, in which the four wheels torques are allocated to track the reference yaw moment. The proposed strategy is evaluated on the dSPACE-based platform over the single lane change test and fishhook steering test. The results indicate that the suggested torque distribution strategy can improve the vehicle stability on different conditions and the energy consumption is significantly reduced compared to an electric stability control (ESC) method.

Original languageEnglish
Article number8716662
Pages (from-to)64627-64640
Number of pages14
JournalIEEE Access
Volume7
DOIs
Publication statusPublished - 2019

Keywords

  • Electric vehicles
  • hybrid model predictive control (hMPC)
  • multi-objective control
  • torque distribution strategy

Fingerprint

Dive into the research topics of 'A multi-objective optimal torque distribution strategy for four in-wheel-motor drive electric vehicles'. Together they form a unique fingerprint.

Cite this