Brake blending in electric vehicles is highly critical for enhancing the overall driving experience, energy recovery and safety. In this paper, a control strategy based on fuzzy rules is proposed for brake blending in EVs to integrate regenerative braking with conventional friction braking. Existing braking control approaches, such as machine learning (ML) and model predictive control (MPC), have several limitations compared to fuzzy rule-based control (FRBC). ML-based braking systems require extensive datasets for training and are computationally intensive. Their performance heavily depends on data quality, making real-time adaptation difficult. Therefore, FRBC strategy is proposed.
Simulations and hardware-in-the-loop experiments validate the performance of the proposed fuzzy rule-based system by indicating a higher braking efficiency with better energy recovery and lesser wear on the braking components.
The fuzzy control system shows a steeper activation curve between 1 and 3 s and reaches 85% activation within the third second. This study, therefore, introduces novel directions in flexible and adaptive frameworks to optimize brake blending in EVs and supports the development of sustainable and efficient vehicle operation.
This work is aimed at pushing the state-of-the-art in EV braking systems through rigorous simulations and experimental validation, with a view to improving energy recovery, safety and overall driving experience.
