Model Predictive Control Considering Engine On/Off and Mode Switch for Dual-mode Hybrid Electric Vehicle

For dual-mode hybrid electric vehicle (HEV), the design of switch strategy and energy management strategy (EMS) are always independent. However, due to the complex and changeable driving condition hybrid electric vehicle encountering, the vehicle performance, including fuel economy and battery lifetime, will be affected by the inconsistency between the two strategies. It's still open to study how to combine switch strategy with EMS. Aiming at this problem, this paper proposes a model predictive control (MPC) considering engine on/off and mode switch for dual-mode hybrid electric vehicle. Firstly, in order to get the precise input sequence, Back Propagation Neural Network is used to forecast future velocity. Secondly, a cost function, considering engine on/off and mode switch, which is solved with Pontryagin's minimum principle (PMP), is put forward. To avoid frequently mode switch and engine on/off during the process of velocity fluctuation, a smooth method is applied in the prediction domain of MPC. Finally, the proposed EMS is verified in simulation. Simulation results show that this strategy can achieve the desired goals.

1. Introduction

Hybrid electric vehicles (HEV) are considered to be an effective way to realize energy saving and emissions reduction[1]. For medium HEV, to meet the power and economic requirement, multiple modes are set to enable the engine to operate in its best efficiency region and provide a better economy[2]. The vehicle mode including engine on/off condition and power train connection is selected according to the current driving condition[3]. Since driving conditions and vehicle movement state are changeable, the switch time must be optimized in realtime to ensure the current mode that best suits immediate needs[4].