Scholarlink

<p>Hydrogen vehicles are facing significant challenges, including safety risks in the event of leakage, high costs associated with physical testing, and fuel cell inefficiency under dynamic conditions; to address these concerns, we propose VR-H2Safe, a cyber-physical platform that combines high-fidelity simulation, physics-informed reinforcement learning (RL), and secure vehicle-to-everything (V2X) communication. Using ANSYS Fluent Large Eddy Simulation (LES) to train real-time predictive models, the platform’s digital twin simulates hydrogen leakage with 92% spatio-temporal correlation compared to high-fidelity LES-CFD benchmarks at key monitoring points. Furthermore, we introduce the H2-LEAK protocol, which extends the SAE J2735 standard with a 128-bit risk vector and blockchain authentication, which provides hazard warnings in less than 2&#xa0;s with 95% spoofing resistance. In addition, our physics-informed AI system optimizes the fuel cell performance, achieving a 10% increase in energy efficiency and a 58% reduction in thermal violations during hardware verification. The experimental findings of urban, tunnel, and road conditions prove that VR-H2Safe accelerates the process of safety verification and reduces the engineering validation phase from 14 to 10 weeks.</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

VR-H2Safe: a VR-based hydrogen vehicle safety & efficiency optimization platform with V2X integration

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月5日

查看全文
  • Mohammad BaniSalman,
  • Mohammad Aljaidi,
  • Ahmad Abdullah Alshammari,
  • Ali Fayez Alkoradees,
  • Samy Abdelwahab Safaan,
  • Mohammad Khishe,
  • Elham Askari

摘要

Hydrogen vehicles are facing significant challenges, including safety risks in the event of leakage, high costs associated with physical testing, and fuel cell inefficiency under dynamic conditions; to address these concerns, we propose VR-H2Safe, a cyber-physical platform that combines high-fidelity simulation, physics-informed reinforcement learning (RL), and secure vehicle-to-everything (V2X) communication. Using ANSYS Fluent Large Eddy Simulation (LES) to train real-time predictive models, the platform’s digital twin simulates hydrogen leakage with 92% spatio-temporal correlation compared to high-fidelity LES-CFD benchmarks at key monitoring points. Furthermore, we introduce the H2-LEAK protocol, which extends the SAE J2735 standard with a 128-bit risk vector and blockchain authentication, which provides hazard warnings in less than 2 s with 95% spoofing resistance. In addition, our physics-informed AI system optimizes the fuel cell performance, achieving a 10% increase in energy efficiency and a 58% reduction in thermal violations during hardware verification. The experimental findings of urban, tunnel, and road conditions prove that VR-H2Safe accelerates the process of safety verification and reduces the engineering validation phase from 14 to 10 weeks.