Context and Driver Dependent Hybrid Electrical Vehicle Operation

Abstract:

This paper studies the driver and context changes during the operation of a hybrid electric vehicle (HEV) and their influence on fuel consumption. Firstly, a context estimation model to recognize driving styles is developed based on machine learning techniques, for which a realistic scenario with simulation of urban mobility (SUMO) and car modeling platform (IPG Carmaker) integration is designed. Secondly, a novel context-aware control strategy based on model predictive control with extended prediction self-adaptive control (MPC-EPSAC) strategy is proposed. The control objective is to achieve optimal torque-split distribution, while optimizing fuel consumption in the parallel HEV. The simulation results suggest that an improvement in fuel economy can be achieved when the driving style in the control loop is adequately considered.

Año de publicación:

2020

Keywords:

• Human-in-the-loop control
• Machine learning
• context estimation
• Cyber Physical Systems
• model pbkp_redictive control (MPC)
• Self-optimization
• context aware control

Fuente:

scopus

google

orcid