Real-Time Model Predictive Powertrain Control for a Connected Plug-In Hybrid Electric Vehicle

Abstract

The continued development of connected and automated vehicle technologies presents the opportunity to utilize these technologies for vehicle energy management. Leveraging this connectivity among vehicles and infrastructure allows a powertrain controller to be predictive and forward-looking. This paper presents a real-time predictive powertrain control strategy for a Plug-in Hybrid Electric Vehicle (PHEV) in a connected vehicle environment. This work focuses on the optimal energy management of a multi-mode PHEV based on predicted future velocity, power demand, and road conditions. The powertrain control system in the vehicle utilizes vehicle connectivity to a cloud-based server in order to obtain future driving conditions. For predictive powertrain control, a Nonlinear Model Predictive Controller (NMPC) is developed to make torque-split decisions within each operating mode of the vehicle. The torque-split among two electric machines and one combustion engine is determined such that fuel consumption is minimized while battery SOC and vehicle velocity targets are met. The controller has been extensively tested in simulation across multiple real-world driving cycles where energy savings in the range of 1 to 4% have been demonstrated. The developed controller has also been deployed and tested in real-time on a test vehicle equipped with a rapid prototyping embedded controller. Real-time in-vehicle testing confirmed the energy savings observed in simulation and demonstrated the ability of the developed controller to be effective in a real-time environment.