Effect of using wire coils and aluminum oxide nanofluid on heat transfer in a double-pipe heat exchanger and predicting data with artificial neural networks
The present study aims to experimentally investigate the Nusselt number and friction factor in a double-pipe heat exchanger equipped with wire coils and aluminum oxide nanofluid, with a particle size of approximately 55 nm, in Reynolds numbers from 4000 to 14000, volume fractions of 0.02, 0.04, and...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-07-01
|
| Series: | Case Studies in Thermal Engineering |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2214157X25004927 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The present study aims to experimentally investigate the Nusselt number and friction factor in a double-pipe heat exchanger equipped with wire coils and aluminum oxide nanofluid, with a particle size of approximately 55 nm, in Reynolds numbers from 4000 to 14000, volume fractions of 0.02, 0.04, and 0.06 %, and pitch ratios of 0, 1, 1.6, and 2.4. Then, a proposed correlation for the Nusselt number is presented, and finally, the experimental data are evaluated using an artificial neural network. The optimum increase of 135.6 % in the Nusselt number with aluminum oxide nanofluid occurs at a volume fraction of 0.06 %, a Reynolds number of 14000, and a pitch ratio of 1. The increase in the friction factor with nanofluid and wire coils, compared to the base fluid (water) without the wire coils, is approximately 7.06 %. The correlation coefficient, mean squared error, root mean squared error, and mean absolute error are calculated for the proposed correlation and artificial neural network. Furthermore, the maximum and minimum deviation margins obtained are +3.4211 and −3.2120, respectively. The results indicated that perceptron neural network of a 3-22-1 topology with Levenberg-Marquardt algorithm has successfully predicted the experimental data. |
|---|---|
| ISSN: | 2214-157X |