A novel downshifting strategy based on medium-time-distance information for hybrid electric bus

Vehicle downshifting during braking for the hybrid electric vehicle (HEV) equipped with the automatic mechanical transmission (AMT) could adjust work points of the motor. Thus, downshifting has great potential to effectively improve the efficiency of braking energy recovery. However, the power interruption during shifting could cause some loss of regenerative energy meanwhile. Hence, the choice of the downshifting point during vehicle braking which has crucial effect on energy recovery efficiency needs to be intensively studied. Moreover, the real-time application of the high-efficiency braking energy recovery strategy is a challenging problem to be tackled. Therefore, this paper proposes a dynamic-programming-rule-based (DPRB) downshifting strategy for a specific hybrid electric bus (HEB) driving condition. Firstly, the braking characteristic of the HEB during the process of pulling in is analyzed. Secondly, the medium-time-distance (MTD) demonstrating the dimension of time and space is proposed to define the boundary condition of the running bus. Then, look-up tables are established based on a dynamic programming algorithm offline using multiple sets of historical data. Thus, Based on the real-time driving data, whether to enter the optimal gear selection process can be decided online. Finally, simulations and hardware-in-the-loop (HIL) tests are carried out, and the results show that the proposed method can be indeed effective for braking energy recovery.

Automated summary

The study proposes a dynamic-programming-rule-based (DPRB) downshifting strategy for a specific hybrid electric bus (HEB) driving condition. By analyzing the braking characteristics of the HEB during the process of pulling in, a medium-time-distance (MTD) representing the dimension of time and space is proposed to define the boundary condition of the running bus. Look-up tables are established based on a dynamic programming algorithm offline using multiple sets of historical data, allowing real-time decision-making on whether to enter the optimal gear selection process based on driving data.