<p>The present study aims to investigate the effect of triple-channel twisted tape with separating blades with an angle of 80 degrees, in two states with pitches of 0.1 and 0.05 m, in the range of Reynolds numbers from 4000 to 12,000 in a double-tube heat exchanger with <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\({\text{ZrO}}_{{2}} {-}{\text{Cu}}\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msub> <mtext>ZrO</mtext> <mn>2</mn> </msub> <mo>-</mo> <mtext>Cu</mtext> </mrow> </math></EquationSource> </InlineEquation> /water nanohybrid in volume fractions of 0.2 and 0.4% v/v on the convective heat transfer coefficient and Nusselt number. Based on the results obtained from the heat transfer tests, compared to the water base fluid, the use of <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\({\text{ZrO}}_{{2}} {-}{\text{Cu}}\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msub> <mtext>ZrO</mtext> <mn>2</mn> </msub> <mo>-</mo> <mtext>Cu</mtext> </mrow> </math></EquationSource> </InlineEquation> /water hybrid nanofluid improves the thermal performance. The experiments were designed using the full factorial design method. By using a triple-channel twisted tape with a pitch of 0.05 m and nanofluid with a concentration of 0.4%, the most optimal value obtained for the convective heat transfer coefficient and Nusselt number is 4684 (W/m<sup>2</sup>.K) and 153, respectively. A quadratic equation with single, double, and interaction parameters was presented using the two factorial method, which can be used for other ranges for the heat transfer coefficient with correlation coefficient and <i>p</i>-value equal to 0.92 and zero, respectively. By using Levenberg–Marquardt algorithm, the most optimal topology was obtained for the perceptron neural network with the mean squared error, root mean error, and correlation coefficient values of 1.5216, 1.2335, and 0.92, respectively.</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

Investigation of the Effect of Adding Hybrid Cu-ZrO2 Nanoparticles on the Convection Coefficient in a Heat Exchanger with Disturber and Proposing a Model for Predicting it Using Artificial Neural Network

  • Narges Ahmadvand,
  • Samira Asleshirin,
  • Hossein Mazaheri,
  • Ali Hassani Joshaghani,
  • Parvin Shahdousti

摘要

The present study aims to investigate the effect of triple-channel twisted tape with separating blades with an angle of 80 degrees, in two states with pitches of 0.1 and 0.05 m, in the range of Reynolds numbers from 4000 to 12,000 in a double-tube heat exchanger with \({\text{ZrO}}_{{2}} {-}{\text{Cu}}\) ZrO 2 - Cu /water nanohybrid in volume fractions of 0.2 and 0.4% v/v on the convective heat transfer coefficient and Nusselt number. Based on the results obtained from the heat transfer tests, compared to the water base fluid, the use of \({\text{ZrO}}_{{2}} {-}{\text{Cu}}\) ZrO 2 - Cu /water hybrid nanofluid improves the thermal performance. The experiments were designed using the full factorial design method. By using a triple-channel twisted tape with a pitch of 0.05 m and nanofluid with a concentration of 0.4%, the most optimal value obtained for the convective heat transfer coefficient and Nusselt number is 4684 (W/m2.K) and 153, respectively. A quadratic equation with single, double, and interaction parameters was presented using the two factorial method, which can be used for other ranges for the heat transfer coefficient with correlation coefficient and p-value equal to 0.92 and zero, respectively. By using Levenberg–Marquardt algorithm, the most optimal topology was obtained for the perceptron neural network with the mean squared error, root mean error, and correlation coefficient values of 1.5216, 1.2335, and 0.92, respectively.