Scholarlink

<p>In this work, Direct Numerical Simulations (DNS) of a NACA65 midspan compressor configuration are used as a reference case to assess the reliability of the URANS simulation by evaluating the mechanisms that govern the unsteadiness via the interaction of deterministic and stochastic scales. To this end, in the present work, both RANS and URANS simulations have been performed to assess how lower order model with respect to DNS are able to predict flows with high level of unsteadiness. Care has been posed on boundary condition imposition for URANS calculation. In addition, a series of URANS simulations have been performed for a similar configuration by finding the most proper inlet condition that best matches the high-fidelity simulation results. A grid dependency analysis was conducted to identify the appropriate mesh density to ensure good calculation reliability and at the same time save time during computations. Several turbulence models have been tested in RANS calculations to identify the most appropriate to be adopted in the following calculations. The results obtained from the RANS and URANS calculations were compared with those obtained previously from the DNS analysis. A two-scale turbulent kinetic energy production (<i>P</i><sub>TKE</sub>) terms decomposition, based on wake-related and the stochastic velocity fluctuations evaluation, was employed both for URANS and DNS to identify similarities and differences between the two methods.</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

Quantification of Turbulent Kinetic Energy Production in Multi-Stage Compressors

  • 期刊论文

  • 发表日期：

    2026年4月10日

查看全文
  • Dario Barsi,
  • Matteo Dellacasagrande,
  • Davide Lengani,
  • Daniele Simoni,
  • Pawel Jan Przytarski,
  • Yueliang Zhang

摘要

In this work, Direct Numerical Simulations (DNS) of a NACA65 midspan compressor configuration are used as a reference case to assess the reliability of the URANS simulation by evaluating the mechanisms that govern the unsteadiness via the interaction of deterministic and stochastic scales. To this end, in the present work, both RANS and URANS simulations have been performed to assess how lower order model with respect to DNS are able to predict flows with high level of unsteadiness. Care has been posed on boundary condition imposition for URANS calculation. In addition, a series of URANS simulations have been performed for a similar configuration by finding the most proper inlet condition that best matches the high-fidelity simulation results. A grid dependency analysis was conducted to identify the appropriate mesh density to ensure good calculation reliability and at the same time save time during computations. Several turbulence models have been tested in RANS calculations to identify the most appropriate to be adopted in the following calculations. The results obtained from the RANS and URANS calculations were compared with those obtained previously from the DNS analysis. A two-scale turbulent kinetic energy production (PTKE) terms decomposition, based on wake-related and the stochastic velocity fluctuations evaluation, was employed both for URANS and DNS to identify similarities and differences between the two methods.