Scholarlink

<p>In this study, the thermal–hydraulic performance of twisted oval double-pipe heat exchangers was numerically investigated for various geometries and flow conditions. Parameters included the aspect ratio (AR), pitch ratio (S), Reynolds number (Re), and flow direction (parallel or counterflow). Results showed that lower AR and S intensified flow disturbance due to geometric effects, promoting turbulence and secondary flow, which enhanced heat transfer but increased pressure drop. Under comparable pressure-drop conditions for parallel and counter-flow arrangements, the counterflow consistently exhibited higher Nusselt numbers and performance evaluation criteria (PEC). An artificial neural network (ANN) model was developed using simulation data, accurately predicting the friction factor, Nusselt number, and PEC with R<sup>2</sup> above 0.999 and a maximum error of 1.6 %. The optimal configuration, Re = 17500, AR = 0.2, and S = 2.5, achieved a PEC of 1.9577. These results provide useful guidelines for compact, high-efficiency heat exchanger design.</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

CFD simulation for thermal performance analysis of twisted oval double-pipe heat exchangers with ANN prediction

  • 期刊论文

  • 发表日期：

    2026年5月12日

查看全文
  • Yoo Hoon Shin,
  • Jeong Geun Gwon,
  • Hoon Ki Choi,
  • Yong Gap Park

摘要

In this study, the thermal–hydraulic performance of twisted oval double-pipe heat exchangers was numerically investigated for various geometries and flow conditions. Parameters included the aspect ratio (AR), pitch ratio (S), Reynolds number (Re), and flow direction (parallel or counterflow). Results showed that lower AR and S intensified flow disturbance due to geometric effects, promoting turbulence and secondary flow, which enhanced heat transfer but increased pressure drop. Under comparable pressure-drop conditions for parallel and counter-flow arrangements, the counterflow consistently exhibited higher Nusselt numbers and performance evaluation criteria (PEC). An artificial neural network (ANN) model was developed using simulation data, accurately predicting the friction factor, Nusselt number, and PEC with R2 above 0.999 and a maximum error of 1.6 %. The optimal configuration, Re = 17500, AR = 0.2, and S = 2.5, achieved a PEC of 1.9577. These results provide useful guidelines for compact, high-efficiency heat exchanger design.