.png){kind=icon}
.png){kind=icon}
Return
Assessment of chemically reactive flow of hybrid nanofluid with variable properties using machine learning approach
.png){kind=icon}
OA
AI
S
Z
A
A
DOI：10.1080/02286203.2026.2683652.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The ability of artificial neural networks (ANNs) to accurately simulate nonlinear functions has made them incredibly helpful for modeling complex phenomena. Their flexibility and adaptability are ideal for addressing complex issues in various domains, including epidemiology, engineering, and the applied sciences. To model the behavior of hybrid nanofluids, this work presents a novel computational framework incorporating the Morlet Wavelet Neural Network (MWNN) with the Hybrid Cuckoo Search Algorithm (HCSA). The heat transportation mechanism is explored utilizing the Cattaneo–Christov heat flux model, which incorporates thermal stratification, energy sources, and Thomson−Troian boundary conditions. Using sophisticated wavelet theory and stochastic optimization approaches, the MWNN-HCSA solver is made to capture the intricate nonlinear dynamics present in hybrid nanofluid systems. A comprehensive performance evaluation is ensured by carefully assessing the MWNN-HCSA framework’s efficacy using a variety of error metrics, such as mean absolute error (10°-10−07), fitness curves, root mean square error (10−1−10−05), and the ENSE measure. The MWNN-HCSA model’s effectiveness in handling challenging modeling issues is confirmed by extensive validation against reference solutions, which shows its precision, convergence, and reliability. Moreover, statistical analyses demonstrate that the solver can accurately utilize an exponential incidence function to represent the nonlinear evolution of hybrid nanofluids.
Keywords:
Hybrid nanofluid
neural network
Thomson and Troian conditions
thermal stratification
energy source
Journal
I
IF:
3.9
Papers:
596
Citations:
1.5K
Organization
Cited Papers