Scholarlink

<p>To overcome the limitations of conventional energy management strategies (EMS) in P1 + P2 plug-in hybrid electric light commercial vehicles (PHEV-LCVs), this paper proposes an adaptive equivalent consumption minimization strategy (ECMS) incorporating vehicle speed prediction. In this strategy, the equivalent factor (EF) is dynamically adjusted based on real-time vehicle speed patterns and battery state of charge (SOC) to improve fuel economy and energy efficiency. A radial basis function (RBF) neural network predicts short-term vehicle speed using historical data, while principal component analysis (PCA) extracts key features for input into a self-organizing map (SOM) neural network for speed pattern classification. Based on the identified patterns and a reference SOC trajectory designed to maintain SOC within allowable limits, a PI-fuzzy controller adaptively updates the EF in real time. This coordinated approach optimizes torque distribution and enhances energy management performance under varying driving conditions. Joint simulations under the New European Driving Cycle (NEDC) and Worldwide Harmonized Light Vehicle Test Cycle (WLTC) demonstrate that the proposed method reduces fuel consumption by 12.23% and 6.64% compared to conventional ECMS and SOC-based adaptive ECMS under NEDC and by 26.15% and 11.94% under WLTC, respectively. These findings confirm that the proposed strategy effectively improves fuel economy across different driving scenarios.</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 Optimization for P1 + P2 Plug-in Hybrid Light Commercial Vehicles Based on Speed Prediction

  • 期刊论文

  • 发表日期：

    2025年9月13日

查看全文
  • Xuanyao Wang,
  • Qing Wang

摘要

To overcome the limitations of conventional energy management strategies (EMS) in P1 + P2 plug-in hybrid electric light commercial vehicles (PHEV-LCVs), this paper proposes an adaptive equivalent consumption minimization strategy (ECMS) incorporating vehicle speed prediction. In this strategy, the equivalent factor (EF) is dynamically adjusted based on real-time vehicle speed patterns and battery state of charge (SOC) to improve fuel economy and energy efficiency. A radial basis function (RBF) neural network predicts short-term vehicle speed using historical data, while principal component analysis (PCA) extracts key features for input into a self-organizing map (SOM) neural network for speed pattern classification. Based on the identified patterns and a reference SOC trajectory designed to maintain SOC within allowable limits, a PI-fuzzy controller adaptively updates the EF in real time. This coordinated approach optimizes torque distribution and enhances energy management performance under varying driving conditions. Joint simulations under the New European Driving Cycle (NEDC) and Worldwide Harmonized Light Vehicle Test Cycle (WLTC) demonstrate that the proposed method reduces fuel consumption by 12.23% and 6.64% compared to conventional ECMS and SOC-based adaptive ECMS under NEDC and by 26.15% and 11.94% under WLTC, respectively. These findings confirm that the proposed strategy effectively improves fuel economy across different driving scenarios.