Flow visualization and analysis of thermal distribution for the nanofluid by the integration of fuzzy c-means clustering ANFIS structure and CFD methods
Abstract: A nanofluid containing copper (Cu) nanoparticles was simulated in a rectangular cavity using computational fluid dynamic (CFD). The upper and lower walls of the cavity were adiabatic, while the right and left walls had warm and cold temperatures, respectively. This temperature difference c...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Published: |
Springer Verlag
2020
|
| Subjects: | |
| Online Access: | http://eprints.um.edu.my/24693/ https://doi.org/10.1007/s12650-019-00614-0 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract: A nanofluid containing copper (Cu) nanoparticles was simulated in a rectangular cavity using computational fluid dynamic (CFD). The upper and lower walls of the cavity were adiabatic, while the right and left walls had warm and cold temperatures, respectively. This temperature difference causes a thermal flow from the right wall to the left wall. The elements of the coordination system in different directions, including velocity in the Y direction (V) and fluid temperature, were obtained using CFD. Adaptive network-based fuzzy inference system (ANFIS) was used to train the CFD outputs and provided artificial flow field and temperature distribution along the cavity domain. The CFD outputs were used as input and output data for the ANFIS method. The position of the fluid layer in X and Y computing directions and fluid velocity (Y axis) were used as three inputs, and the fluid temperature was taken as the output in the ANFIS method training process. The data were categorized using fuzzy c-means clustering, and different numbers of clusters were taken as a key parameter in this method. Using the fuzzy inference system, it is possible to predict the nodes in the cavity not generated through CFD simulation so that different coordination of the fluid at these points can be computed. Using ANFIS method, it is possible to reduce the computation time of CFD method so that more nodes are predicted in a shorter period of time, while clustering method can enhance the computing time for each neural cell. The ANFIS method can also visualize the flow in the cavity and display the thermal distribution along with the heat source. Graphic abstract: [Figure not available: see fulltext.]. © 2019, The Visualization Society of Japan. |
|---|