Scholarlink

<p>Electric vehicles (EVs) use regenerative braking systems (RBS) to increase the driving energy efficiency by generating energy on deceleration. Current control strategies, however, tend to use fixed gear ratios or rule-based controllers, which limit energy recovery in changing driving conditions. To overcome these limitations, this work introduces a Feed Forward Bobcat Neural Controller (FFBNC) for adaptive Continuously Variable Transmission (CVT) gear-ratio control in EVs. The system combines the EV powertrain model, which includes a motor, battery, and CVT, with an intelligent controller that determines the best gear ratio based on vehicle states such as motor torque, speed, Battery State of charge (SOC), and braking distance. The feedforward neural network (FFNN) captures the nonlinear relationship between driving states and optimal gear ratios, while the Bobcat optimization algorithm (BOA) adjusts its weights to attain global optimality. The developed controller ensures the motor runs at its maximum efficiency range during braking. Simulation results verify that the newly proposed FFBNC notably improves regenerative braking effectiveness, reduces reliance on hydraulic braking, and enhances braking smoothness compared with traditional control strategies. This developed controller provides an accessible way to improve the energy use efficiency of EVs.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

An adaptive energy-efficient regenerative braking control system in an electric vehicle

  • 期刊论文

  • 发表日期：

    2026年1月13日

查看全文
  • Priya Gajjal,
  • M. R. Dahake,
  • S. P. Komble,
  • M. H. Attal

摘要

Electric vehicles (EVs) use regenerative braking systems (RBS) to increase the driving energy efficiency by generating energy on deceleration. Current control strategies, however, tend to use fixed gear ratios or rule-based controllers, which limit energy recovery in changing driving conditions. To overcome these limitations, this work introduces a Feed Forward Bobcat Neural Controller (FFBNC) for adaptive Continuously Variable Transmission (CVT) gear-ratio control in EVs. The system combines the EV powertrain model, which includes a motor, battery, and CVT, with an intelligent controller that determines the best gear ratio based on vehicle states such as motor torque, speed, Battery State of charge (SOC), and braking distance. The feedforward neural network (FFNN) captures the nonlinear relationship between driving states and optimal gear ratios, while the Bobcat optimization algorithm (BOA) adjusts its weights to attain global optimality. The developed controller ensures the motor runs at its maximum efficiency range during braking. Simulation results verify that the newly proposed FFBNC notably improves regenerative braking effectiveness, reduces reliance on hydraulic braking, and enhances braking smoothness compared with traditional control strategies. This developed controller provides an accessible way to improve the energy use efficiency of EVs.