Scholarlink

<p>Fast and accurate flow field prediction is crucial for evaluating the hydrodynamic performance of autonomous underwater vehicles (AUVs). Establishing a reliable database and determining which computational fluid dynamics (CFD) model is most suitable requires in-depth research. Therefore, one of the purposes of this paper is to verify the accuracy of the commonly used Reynolds-averaged Navier–Stokes (RANS) turbulence model in calculating the flow around AUVs with SUBOFF bare hull as a typical example. The established database was decomposed using the proper orthogonal decomposition (POD) method, and the cubic spline interpolation method was used to predict the basis coefficients of unknown operating conditions. Thereafter, flow field of the unknown cases was predicted. The results show that the standard <i>k–ε</i> model performs well in both the straight-line and the oblique motions. When taking the first three POD bases and modes, the maximum errors between reconstructed velocity (two components) and pressure of the representative two-dimensional horizontal flow field and the CFD results are 2.23%, 1.02%, and 0.45%, respectively. The maximum error of using the first three orders to predict the two velocities and pressures of the plane flow field under unknown operating conditions is less than 1.5%. In addition, using the POD method to decompose and predict the two-dimensional flow field of AUVs only takes 0.83 and 0.54&#xa0;s, while using the CFD method takes over 1200&#xa0;s.</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

Rapid prediction of flow field in bare hull SUBOFF submarine with drift angles based on POD method

  • 期刊论文

  • 发表日期：

    2025年6月16日

查看全文
  • Chao Yuan,
  • Lixia Chen,
  • Yong Lin,
  • Dibo Dong,
  • Xiujing Gao,
  • Junlong Zhao

摘要

Fast and accurate flow field prediction is crucial for evaluating the hydrodynamic performance of autonomous underwater vehicles (AUVs). Establishing a reliable database and determining which computational fluid dynamics (CFD) model is most suitable requires in-depth research. Therefore, one of the purposes of this paper is to verify the accuracy of the commonly used Reynolds-averaged Navier–Stokes (RANS) turbulence model in calculating the flow around AUVs with SUBOFF bare hull as a typical example. The established database was decomposed using the proper orthogonal decomposition (POD) method, and the cubic spline interpolation method was used to predict the basis coefficients of unknown operating conditions. Thereafter, flow field of the unknown cases was predicted. The results show that the standard k–ε model performs well in both the straight-line and the oblique motions. When taking the first three POD bases and modes, the maximum errors between reconstructed velocity (two components) and pressure of the representative two-dimensional horizontal flow field and the CFD results are 2.23%, 1.02%, and 0.45%, respectively. The maximum error of using the first three orders to predict the two velocities and pressures of the plane flow field under unknown operating conditions is less than 1.5%. In addition, using the POD method to decompose and predict the two-dimensional flow field of AUVs only takes 0.83 and 0.54 s, while using the CFD method takes over 1200 s.