Scholarlink

With the development of sustainable agriculture, traditional diesel tractors are increasingly phased out due to environmental pollution, while pure electric tractors are limited by battery endurance. This paper proposes an adaptive physics-informed neural network-based boundary value problem solver (APINN-BVP) to optimize the energy management strategy (EMS) of a series hybrid electric tractor (HET). By reformulating the energy management problem as a boundary value problem and solving it using APINN-BVP, an optimal solution can be effectively obtained under limited dataset conditions. The integration of physical laws into the neural network architecture ensures compliance with system dynamics while enhancing computational efficiency. Simulations validate the effectiveness and reliability of the proposed method, demonstrating its potential for application in energy management systems within agricultural machinery and beyond.

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

Utilizing Adaptive Physics-Informed Neural Networks for Energy Management in Hybrid Electric Tractor

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Chenhui Li,
  • Fangyuan Li,
  • Lefei Gao,
  • Yanhong Liu

摘要

With the development of sustainable agriculture, traditional diesel tractors are increasingly phased out due to environmental pollution, while pure electric tractors are limited by battery endurance. This paper proposes an adaptive physics-informed neural network-based boundary value problem solver (APINN-BVP) to optimize the energy management strategy (EMS) of a series hybrid electric tractor (HET). By reformulating the energy management problem as a boundary value problem and solving it using APINN-BVP, an optimal solution can be effectively obtained under limited dataset conditions. The integration of physical laws into the neural network architecture ensures compliance with system dynamics while enhancing computational efficiency. Simulations validate the effectiveness and reliability of the proposed method, demonstrating its potential for application in energy management systems within agricultural machinery and beyond.