Experimental study for thermal conductivity of water‐based zirconium oxide nanofluid: Developing optimal artificial neural network and proposing new correlation

2020 ◽  
Author(s):  
Andaç Batur Çolak
Keyword(s):  
Neural Network ◽  
Thermal Conductivity ◽  
Experimental Study ◽  
Artificial Neural Network ◽  
Zirconium Oxide ◽  
New Correlation ◽  
Water Based ◽  
Artificial Neural

Development of a New Correlation and Post Processing of Heat Transfer Coefficient and Pressure Drop of Functionalized COOH MWCNT Nanofluid by Artificial Neural Network

2018 ◽  
Vol 14 (2) ◽  
pp. 104-112 ◽  
Author(s):  
Mohammad Hemmat Esfe ◽  
Somchai Wongwises ◽  
Saeed Esfandeh ◽  
Ali Alirezaie
Keyword(s):  
Neural Network ◽  
Heat Transfer ◽  
Thermal Conductivity ◽  
Artificial Neural Network ◽  
Heat Transfer Coefficient ◽  
Pressure Drop ◽  
Transfer Coefficient ◽  
Regression Coefficient ◽  
New Correlation ◽  
The Neural Network

Background: Because of nanofluids applications in improvement of heat transfer rate in heating and cooling systems, many researchers have conducted various experiments to investigate nanofluid's characteristics more accurate. Thermal conductivity, electrical conductivity, and heat transfer are examples of these characteristics. Method: This paper presents a modeling and validation method of heat transfer coefficient and pressure drop of functionalized aqueous COOH MWCNT nanofluids by artificial neural network and proposing a new correlation. In the current experiment, the ANN input data has included the volume fraction and the Reynolds number and heat transfer coefficient and pressure drop considered as ANN outputs. Results: Comparing modeling results with proposed correlation proves that the empirical correlation is not able to accurately predict the experimental output results, and this is performed with a lot more accuracy by the neural network. The regression coefficient of neural network outputs was equal to 99.94% and 99.84%, respectively, for the data of relative heat transfer coefficient and relative pressure drop. The regression coefficient for the provided equation was also equal to 97.02% and 77.90%, respectively, for these two parameters, which indicates this equation operates much less precisely than the neural network. Conclusion: So, relative heat transfer coefficient and pressure drop of nanofluids can also be modeled and estimated by the neural network, in addition to the modeling of nanofluid’s thermal conductivity and viscosity executed by different scholars via neural networks.

Sign in / Sign up

Export Citation Format

Share Document