Scholarlink

Bioethanol is a quick path to lower carbon fuel, yet its high concentration blend specific effects on spark ignition engines have received only limited study. Combined bench tests, machine learning and cycle simulation to map a Nissan HR16DE from pure petrol to pure bioethanol. Pressure, fuel and air flow were measured at 2,000 rpm, 80 Nm, λ = 1.0 and 24°CA BTDC while bioethanol content was used up to 70% in real life. AVL Burn heat release and other metrics trained the artificial neural network, the validated model predicted ignition timing, burn duration and Vibe function shape for mixtures from E70 to E100. These predictions added to AVL BOOST model to calculate brake specific fuel consumption and brake thermal efficiency. Up to E70, ignition timing fell while total burn time stayed roughly constant. Artificial neural network (ANN) estimates a further cut 1.8°CA when using E100. The Vibe shape factor climbed from 1.60 (E0) to 1.98 (E70) then declined to 1.93 (E100), predicting earlier, sharper heat release. BOOST matched measured pressure for E70 and predicted earlier yet cooler combustion for bioethanol richer blends, confirming bioethanol cooling benefit. The integrated bench/ANN/BOOST tests offers a fast, low cost predictions for high bioethanol strategies in today’s engines.

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

Integrated Experimental and Numerical Prediction of Combustion in High-Concentration Bioethanol–Gasoline Blends in SI Engine

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Gabrielius Mejeras,
  • Alfredas Rimkus,
  • Alytis Gruodis,
  • Jonas Matijošius,
  • Martynas Grybauskas

摘要

Bioethanol is a quick path to lower carbon fuel, yet its high concentration blend specific effects on spark ignition engines have received only limited study. Combined bench tests, machine learning and cycle simulation to map a Nissan HR16DE from pure petrol to pure bioethanol. Pressure, fuel and air flow were measured at 2,000 rpm, 80 Nm, λ = 1.0 and 24°CA BTDC while bioethanol content was used up to 70% in real life. AVL Burn heat release and other metrics trained the artificial neural network, the validated model predicted ignition timing, burn duration and Vibe function shape for mixtures from E70 to E100. These predictions added to AVL BOOST model to calculate brake specific fuel consumption and brake thermal efficiency. Up to E70, ignition timing fell while total burn time stayed roughly constant. Artificial neural network (ANN) estimates a further cut 1.8°CA when using E100. The Vibe shape factor climbed from 1.60 (E0) to 1.98 (E70) then declined to 1.93 (E100), predicting earlier, sharper heat release. BOOST matched measured pressure for E70 and predicted earlier yet cooler combustion for bioethanol richer blends, confirming bioethanol cooling benefit. The integrated bench/ANN/BOOST tests offers a fast, low cost predictions for high bioethanol strategies in today’s engines.