Safety-constrained Bayesian-optimized fuzzy regenerative braking control for a front-wheel-drive battery electric vehicle
Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering · 16.11.2026
DOI: 10.1177/09544070261506522 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering, online erschienen am 16.11.2026. Verfasst von Zhiqiang Xi und 4 weiteren Autor:innen. Die Arbeit nennt 29 Literaturangaben.
Abstract (Originalsprache)
Regenerative braking is a key technology for improving the energy efficiency of battery electric vehicles. For front-wheel-drive battery electric vehicles, the drive motor is mounted on the front axle, so regenerative braking must be coordinated with hydraulic braking while satisfying front–rear braking-force distribution, tire-adhesion, motor-torque, and battery charging-power constraints. To reduce the dependence of conventional fuzzy controllers on empirically selected membership-function parameters, this study proposes a safety-constrained Bayesian-optimized fuzzy regenerative braking strategy. A Mamdani fuzzy controller uses braking intensity, battery state of charge (SOC) and vehicle speed as inputs and the front-axle regenerative braking ratio as the output. A dimensionless multi-cycle objective combines normalized SOC depletion, braking-opportunity-normalized recovered energy, demand-normalized force-transition characteristics, and normalized command-to-executable regenerative-force mismatch. The final controller is obtained after 300 Bayesian-optimization evaluations and is assessed on six optimization cycles together with independent low-SOC, high-braking-intensity, unseen-cycle and repeated-braking validations. Across NEDC, WLTC, FTP-72, FTP-75, CLTC-P, and NYCC, the mean of the six cycle-wise recovered-energy improvements is 25.48%, while the mean of the six cycle-wise reductions in specific net battery-energy consumption is 12.13% relative to the original fuzzy controller. Without retuning, recovered energy on the unseen HWFET and US06 cycles increases by 40.03% and 30.11%, respectively, with specific energy-consumption reductions of 8.31% and 9.79%. Low-SOC and high-intensity tests satisfy the prescribed force, motor and battery-power constraints, and dynamic threshold-crossing tests show continuous total front-axle braking demand during regenerative-to-hydraulic redistribution. Repeated-braking simulations further show that the energy-recovery benefit is accompanied by increased battery electrical stress, highlighting an energy-recovery/electrical-stress trade-off.
Zitieren
APA
Xi, Z., Zhang, H., Chen, S., Niu, L., & Gao, T. (2026). Safety-constrained Bayesian-optimized fuzzy regenerative braking control for a front-wheel-drive battery electric vehicle. Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering. https://doi.org/10.1177/09544070261506522
Vancouver
Xi Z, Zhang H, Chen S, Niu L, Gao T. Safety-constrained Bayesian-optimized fuzzy regenerative braking control for a front-wheel-drive battery electric vehicle. Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering. 2026. doi:10.1177/09544070261506522
BibTeX
@article{Xi2026,
author = {Xi, Zhiqiang and Zhang, Haibo and Chen, Shixuan and Niu, Lianbo and Gao, Tianhang},
title = {Safety-constrained Bayesian-optimized fuzzy regenerative braking control for a front-wheel-drive battery electric vehicle},
journal = {Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering},
year = {2026},
doi = {10.1177/09544070261506522}
}
EUVIMED