Braking control strategy of distributed electric vehicle based on vehicle velocity prediction

To enhance the efficiency of braking energy recovery in distributed electric vehicles, this study investigates the impact of vehicle velocity prediction on braking performance. A regenerative braking control strategy that incorporates vehicle velocity prediction is proposed. Initially, a vehicle velocity prediction model is developed using a backpropagation neural network, which is trained under various operational conditions, and its validity is confirmed through testing. Subsequently, a braking control strategy that integrates vehicle velocity prediction is formulated based on a parallel braking control approach to optimize braking energy recovery. Finally, simulations of the proposed control strategy are conducted and compared with a strategy that does not utilize velocity prediction, under both joint working conditions and the Worldwide Harmonized Light Vehicles Test Cycle (WLTC). The findings indicate that the control strategy utilizing vehicle velocity prediction achieves greater energy recovery in both scenarios, with improvements in braking energy recovery efficiency of 9.2% and 6.13% over the non-predictive strategy, respectively. These results demonstrate the efficacy and rationale of the proposed approach.