This content is not included in your SAE MOBILUS subscription, or you are not logged in.

Energy-Efficient Braking Torque Distribution Strategy of Rear-Axle Drive Commercial EV Based on Fuzzy Neural Network

Journal Article
2021-01-0783
ISSN: 2641-9637, e-ISSN: 2641-9645
Published April 06, 2021 by SAE International in United States
Energy-Efficient Braking Torque Distribution Strategy of Rear-Axle Drive Commercial EV Based on Fuzzy Neural Network
Sector:
Citation: Ge, G., Wang, T., Lv, Y., Zou, X. et al., "Energy-Efficient Braking Torque Distribution Strategy of Rear-Axle Drive Commercial EV Based on Fuzzy Neural Network," SAE Int. J. Adv. & Curr. Prac. in Mobility 3(4):2136-2145, 2021, https://doi.org/10.4271/2021-01-0783.
Language: English

Abstract:

Regenerative braking is identified as an essential step toward extending cruising mileage for electric vehicle (EV). Braking energy recovery strategies usually focus on passenger EV and commercial EV is ignored. In this paper, an energy-efficient braking torque distribution strategy is proposed for a rear-axle drive commercial EV to improve braking energy recovery and safety. Firstly, the braking force distribution curve is determined referring to the EU braking law for commercial vehicle and the ideal braking distribution curve. Secondly, a novel braking torque distribution strategy is established adopting fuzzy control algorithm, where the ratio between hydraulic braking torque and regenerative braking torque is updated instantaneously according to vehicle velocity, braking strength and state of charge of battery. Then, the corresponding controller is synthesized on ideal braking condition and several classic cycles. To further enhance the performance of the controller, a neural network based framework is established to optimize the membership function in fuzzy controller. Simulations on ideal braking condition demonstrate the controller can always meet emergency braking needs. For the standard cycles, including NEDC and WLTC, the energy-efficient strategy based on fuzzy control can recover up to 18.88% and 16.56% of energy under NEDC and WLTC cycles, and on this basis, the optimized strategy based on adaptive neuro-fuzzy control can improve energy recovery by 2.84% and 3.6% under these two cycles. The developed braking torque distribution strategy can potentially be embedded in real-time supervisory systems to realize the energy saving and increase the cruising mileage for commercial EV.