Hierarchical Energy Management Strategy for Plug-in HEVs Based on Historical Data and Real-Time Speed Scheduling

The rapid development of the Internet of Things (IoT) and cloud computing has presented unprecedented opportunities and challenges for the automotive industry. As additional traffic data becomes accessible, the automotive powertrain has the capability to employ sophisticated control algorithms in order to enhance fuel efficiency. This study presents a hierarchical control strategy for commuter plug-in hybrid electric vehicles (PHEVs), incorporating historical driving data and real-time traffic conditions. Firstly, in the upper layer, an iterative learning algorithm is introduced to analyze the historical driving patterns, enabling derivation of a battery state-of-charge (SOC) reference on the commuter route through solving a global energy optimization problem, that is subsequently provided to the lower layer controller. In the lower layer, real-time influence from preceding vehicles is considered by designing a short-term speed prediction algorithm based on Gaussian process regression (GPR) model. Then, two MPC controllers are designed: one for speed scheduling and another for power distribution. Finally, the simulation validations have been conducted to demonstrate the effectiveness of the proposed strategy.