Scholarlink

Range-extended electric vehicles (REEVs), which combine the advantages of both electric and fuel-powered systems, have emerged as a key technology for sustainable transportation. However, efficient energy management (EM) under complex operating conditions remains a major challenge. To address this problem, this paper proposes an optimization approach based on physics-informed machine learning (PIML), which reformulates the EM problem as a Hamilton–Jacobi–Bellman (HJB) equation. A PIML framework is designed to embed physical system constraints and solve for the optimal control strategy. Simulations are conducted to verify the efficiency of the proposed method on reducing energy consumption and optimizing power allocation. Errors are further analyzed to show the accuracy of the PIML approach.

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

Physics-Informed Machine Learning for Energy Management in Electric Drive System of Range-Extended Electric Vehicles

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Fengxin Zheng,
  • Fangyuan Li,
  • Yanni Wan,
  • Yanhong Liu

摘要

Range-extended electric vehicles (REEVs), which combine the advantages of both electric and fuel-powered systems, have emerged as a key technology for sustainable transportation. However, efficient energy management (EM) under complex operating conditions remains a major challenge. To address this problem, this paper proposes an optimization approach based on physics-informed machine learning (PIML), which reformulates the EM problem as a Hamilton–Jacobi–Bellman (HJB) equation. A PIML framework is designed to embed physical system constraints and solve for the optimal control strategy. Simulations are conducted to verify the efficiency of the proposed method on reducing energy consumption and optimizing power allocation. Errors are further analyzed to show the accuracy of the PIML approach.