Scholarlink

<p>As a multi-energy coordinated mobility platform, hybrid off-road vehicles driven by in-wheel motors (IWMs) operate on complex and variable driving conditions. When the auxiliary range extender (APU) fails to promptly respond to substantial transient power demand fluctuations, vehicle dynamic performance and fuel economy deteriorate, further compromising battery safety and lifespan. To address this issue, this paper proposes a predictive energy management strategy based on road condition identification. A bidirectional long short-term memory (Bi-LSTM) network is constructed as a demand power cluster prediction model, taking velocity/slope characteristics within sliding windows and off-road condition identification results as inputs, while outputting demand power sequences in the further time domain. Integrated with model predictive control (MPC) framework, this strategy establishes adaptive energy flow allocation to ensure optimal energy distribution and achieve enhanced fuel economy and state-of-charge (SOC) stability. Comparative validation with condition-adaptive rule-based strategies demonstrates the proposed method’s effectiveness. Field vehicle tests indicate that compared to the rule-based strategy, the predictive strategy reduces 0–80&#xa0;km/h and 0–50&#xa0;m acceleration times by 15.1 and 4.8% respectively, increases mean climbing speeds by 56.1 (40% slope) and 42.2% (60% slope), while decreasing average traversal times by 14.6 and 20.3% correspondingly.</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

Energy management strategy for hybrid electric off-road vehicles based on model predictive control

  • 期刊论文

  • 发表日期：

    2025年10月15日

查看全文
  • Xiang Fu,
  • Xilong Zhang,
  • Yuhao Tan,
  • Shuiyan Yang,
  • Jiaqi Wan,
  • Yipeng Yin,
  • Qianfeng Ruan,
  • Zitai Xiao,
  • Tianqi Yang

摘要

As a multi-energy coordinated mobility platform, hybrid off-road vehicles driven by in-wheel motors (IWMs) operate on complex and variable driving conditions. When the auxiliary range extender (APU) fails to promptly respond to substantial transient power demand fluctuations, vehicle dynamic performance and fuel economy deteriorate, further compromising battery safety and lifespan. To address this issue, this paper proposes a predictive energy management strategy based on road condition identification. A bidirectional long short-term memory (Bi-LSTM) network is constructed as a demand power cluster prediction model, taking velocity/slope characteristics within sliding windows and off-road condition identification results as inputs, while outputting demand power sequences in the further time domain. Integrated with model predictive control (MPC) framework, this strategy establishes adaptive energy flow allocation to ensure optimal energy distribution and achieve enhanced fuel economy and state-of-charge (SOC) stability. Comparative validation with condition-adaptive rule-based strategies demonstrates the proposed method’s effectiveness. Field vehicle tests indicate that compared to the rule-based strategy, the predictive strategy reduces 0–80 km/h and 0–50 m acceleration times by 15.1 and 4.8% respectively, increases mean climbing speeds by 56.1 (40% slope) and 42.2% (60% slope), while decreasing average traversal times by 14.6 and 20.3% correspondingly.