主办：陕西省汽车工程学会
ISSN 1671-7988  CN 61-1394/TH
创刊：1976年
汽车实用技术

汽车实用技术 ›› 2025, Vol. 50 ›› Issue (13): 18-24,35.DOI: 10.16638/j.cnki.1671-7988.2025.013.004

• 新能源汽车 • 上一篇    

基于模糊神经网络的 EV 混合储能能量 管理策略

卢铁军，韦文祥*，胡天祥，赵帮涛   

  1. 湖南科技大学 信息与电气工程学院
  • 发布日期:2025-07-11
  • 通讯作者: 韦文祥
  • 作者简介:卢铁军（2001－），男，硕士，研究方向为新能源汽车混合储能系统 通信作者：韦文祥（1977－），男，博士，研究方向为矿山运输装备电传动控制技术、新能源技术

Energy Management Strategy for EV Hybrid Energy Storage Based on Fuzzy Neural Networks

LU Tiejun, WEI Wenxiang* , HU Tianxiang, ZHAO Bangtao   

  1. School of Information and Electrical Engineering, Hunan University of Science and Technology
  • Published:2025-07-11
  • Contact: WEI Wenxiang

PDF

摘要： 随着电动汽车对高效能量管理需求的日益提升，传统能量管理策略在复杂工况下的自 适应性与精确性短板逐渐凸显。为此，文章提出一种基于模糊神经网络的混合储能能量管理 策略，旨在优化动力电池与超级电容的协同工作机制。通过 Simulink 与 Cruise 联合仿真平台 验证表明，该策略可在高功率需求场景下优先调用超级电容，有效减轻电池负荷并延长其使 用寿命；在制动能量回收阶段，超级电容可快速吸收能量，显著提升能量回收效率。仿真数 据显示，相较于模糊控制和逻辑门限控制策略，该方法使电池荷电状态（SOC）分别提升 0.51% 和 1.96%，不仅延长了车辆续航里程，还能有效抑制电流波动，对电动汽车能量管理具有重 要应用价值。

关键词: 电动汽车；混合储能系统；模糊神经网络；自适应控制

Abstract: With the increasing demand for efficient energy management in electric vehicles, the shortcomings of traditional energy management strategies in adaptability and accuracy under complex working conditions have gradually become prominent. Therefore, this paper proposes a hybrid energy storage management strategy based on fuzzy neural network, aiming to optimize the collaborative working mechanism of power battery and supercapacitor. Verification through the joint simulation platform of Simulink and Cruise shows that the strategy can preferentially call supercapacitors in high-power demand scenarios, effectively reducing battery load and extending its service life; during braking energy recovery, supercapacitors can quickly absorb energy, significantly improving energy recovery efficiency. Simulation data shows that compared with the fuzzy control and logic threshold control strategies, this method increases the battery state of charge (SOC) by 0.51% and 1.96% respectively, not only extending the vehicle's driving range, but also effectively suppressing current fluctuations. The above achievements have important application value for energy management of electric vehicles.

Key words: electric vehicle; hybrid energy storage system; fuzzy neural network; adaptive control