Scholarlink

<p>Conventional airfoils suffer from aerodynamic limitations, including early stall onset and increased drag at higher angles of attack, which restrict their operational efficiency and versatility. Active flow control methods, such as the Co Flow Jet (CFJ), have shown promise in overcoming these challenges. This study investigated whether a Dual Co Flow Jet (DCFJ) system with independently controlled suction and blowing slots may extend the stable working range of the airfoil. Building on these foundations, airfoils were tested at various free stream velocities and angles of attack, ranging from 0° to 24°, each corresponding to multiple coefficients of jet momentum. Compared to the baseline airfoil, the DCFJ configuration not only greatly improves lift-to-drag ratios but also raises maximum lift coefficients and delays stall. To further reduce the cost and complexity of full-scale testing, numerous machine learning (ML) models, including multi-layer artificial neural networks, were trained on experimental datasets. These models achieved minimal prediction errors for lift and drag coefficients, providing a reliable and computationally efficient alternative to conventional experimental workflows.</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

Experimental Insights and Machine Learning-Based Predictions of Momentum-Controlled Dual Co-Flow Jet Airfoil

  • 期刊论文

  • 发表日期：

    2026年3月27日

查看全文
  • Anees Waqar,
  • Muhammad Hammad Ajmal,
  • Muhammad Umer Sohail

摘要

Conventional airfoils suffer from aerodynamic limitations, including early stall onset and increased drag at higher angles of attack, which restrict their operational efficiency and versatility. Active flow control methods, such as the Co Flow Jet (CFJ), have shown promise in overcoming these challenges. This study investigated whether a Dual Co Flow Jet (DCFJ) system with independently controlled suction and blowing slots may extend the stable working range of the airfoil. Building on these foundations, airfoils were tested at various free stream velocities and angles of attack, ranging from 0° to 24°, each corresponding to multiple coefficients of jet momentum. Compared to the baseline airfoil, the DCFJ configuration not only greatly improves lift-to-drag ratios but also raises maximum lift coefficients and delays stall. To further reduce the cost and complexity of full-scale testing, numerous machine learning (ML) models, including multi-layer artificial neural networks, were trained on experimental datasets. These models achieved minimal prediction errors for lift and drag coefficients, providing a reliable and computationally efficient alternative to conventional experimental workflows.