欢迎访问《汽车安全与节能学报》，
汽车安全与节能学报

汽车安全与节能学报 ›› 2025, Vol. 16 ›› Issue (5): 736-746.DOI: 10.3969/j.issn.1674-8484.2025.05.008

• 汽车节能与环保 • 上一篇    下一篇

基于分段SOC轨迹预测的PHEV分层能量管理策略

代立宏1(), 金妮妮1(), 莫宗华2, 胡鹏3, 万文俊1, 刘浩业1,*(), 王天友1   

  1. 1.天津大学，先进内燃动力全国重点实验室，天津 300354，中国
    2.广西玉柴机器股份有限公司，玉林 537006，中国
    3.奇瑞捷途事业部，芜湖 300399，中国
  • 收稿日期:2025-01-08 修回日期:2025-09-29 出版日期:2025-10-31 发布日期:2025-11-10
  • 通讯作者: *刘浩业，教授。E-mail：liuhaoye@tju.edu.cn。
  • 作者简介:代立宏（1973—），男（汉），安徽，博士研究生。E-mail：dailihong@mychery.com。
    金妮妮（1998—），女（汉），天津，硕士研究生。E-mail：jinnini@tju.edu.cn。
  • 基金资助:
    国家重点研发计划资助(2024YFB2505303)

Hierarchical energy management strategy for PHEVs based on segmented SOC trajectory prediction

DAI Lihong1(), JIN Nini1(), MO Zonghua2, HU Peng3, WAN Wenjun1, LIU Haoye1,*(), WANG Tianyou1   

  1. 1. State Key Lab of Engine Tianjin University, Tianjin University, Tianjin 300354, China
    2. Guangxi Yuchai Machinery Co., Ltd., Yulin 537006, China
    3. Chery Jetour Automobile Co., Ltd., Wuhu 241100, China
  • Received:2025-01-08 Revised:2025-09-29 Online:2025-10-31 Published:2025-11-10

RichHTML

7

PDF

56

摘要： 为了在实际驾驶条件下实现接近全局的最佳能量分配，该文提出了一种分层能量管理-自适应初始等效因子策略（HEMS-AIEFS）。HEMS-AIEFS采用了 2 层结构：上层实施分段的电池荷电状态（SOC）规划方法，利用动态规划（DP）算法生成的数据训练神经网络模型，再利用该模型在线规划SOC 节点轨迹；下层采用预测等效消耗最小化策略（P-ECMS）跟踪上层规划的 SOC 轨迹，并加入自适应初始等效因子策略（AIEFS）改善跟踪效果。结果表明：相比于传统确定初始等效因子的方法，所提 AIEFS 降低 2.36%~7.69% 油耗；与电量消耗-电量维持策略（CD-CS）相比，HEMS-AIEFS 在不同工况下可节省 1.56%~9.13% 的燃油消耗，所需计算时间是 DP 算法的 4.9%~5.6%。该研究展示了基于速度信息分段 SOC 规划的HEMS-AIEFS在插电式混合动力车（PHEV）能量管理优化方面的巨大潜力。

关键词: 插电式混合动力电动车（PHEV）, 分层能量管理-自适应初始等效因子策略（HEMS-AIEFS）, 电池荷电状态（SOC）轨迹, 等效因子（EF）

Abstract:

A hierarchical energy management strategy-adaptive initial equivalent factor strategy (HEMS-AIEFS) was proposed to achieve near-global optimal energy allocation under real driving conditions. HEMS-AIEFS adopted a two-layer structure: The upper layer implemented a node-split state-of-charge (SOC) planning method for batteries, which used a dynamic programming (DP) algorithm to generate the relevant data for training neural network models. These models can predict the SOC node trajectories of different road sections in real time; In the lower layer, the predicted equivalent consumption minimization strategy (P-ECMS) was used to track the predicted SOC trajectories, in which the adaptive initial equivalent factor strategy (AIEFS) was added to set the initial equivalent factor (EF0). The results show that the proposed AIEFS reduces fuel consumption by 2.36% to 7.69% compared to the conventional method of determining the initial equivalence factor, and that HEMS-AIEFS saves 1.56% to 9.13% of fuel consumption under different operating conditions comparing to the CD-CS strategy and requires 4.9% to 5.6% of the computation time of the DP algorithm. This study provides an effective optimization method for plug-in hybrid elective vehicle (PHEV) energy management optimization and demonstrates the potential application of navigation information in PHEV energy management optimization.

Key words: plug-in hybrid electric vehicle (PHEV), hierarchical energy management strategy-adaptive initial equivalent factor strategy (HEMS-AIEFS), state-of-charge (SOC) trajectory, equivalent factor(EF)

中图分类号: