Regenerative braking control strategy for pure electric vehicles based on fuzzy neural network

Abstract

This study investigates the efficiency and safety of regenerative brake energy recuperation systems for electric vehicles. A three-input single-output fuzzy controller is developed to allocate hydraulic and electric braking forces, considering brake intensity, vehicle speed, and battery SOC's impact on regenerative braking performance. Fuzzy neural networks are utilized due to their effectiveness in solving complex, nonlinear, and fuzzy systems, along with their robustness to parameter changes and external disturbances. The fuzzy process of the controller is optimized using a self-tuning algorithm for designing membership function parameters, resulting in a fuzzy neural network controller. Moreover, electric and hydraulic braking forces are redistributed. Simulation using AVL Cruise software is conducted under NEDC and FTP-75 working conditions. The suggested brake energy recovery control approach using fuzzy neural networks successfully recovers braking energy, achieving energy recovery efficiencies of 14.52% and 39.61% under NEDC and FTP-75 conditions, respectively.