Advanced computational modeling of Darcy-Forchheimer effects and nanoparticle-enhanced blood flow in stenosed arteries

Enhancing blood flow in stenosed arteries through the optimization of nanoparticle-based solutions is one potential application of this research to improve medication delivery systems and therapeutic therapies for cardiovascular disorders. This research employed artificial neural networks (ANNs) to...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Engineering applications of artificial intelligence Jg. 152; S. 110737
Hauptverfasser: Hira, Fatima Shafiq, Rubbab, Qammar, Ahmad, Irshad, Majeed, Afraz Hussain
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Ltd 15.07.2025
Schlagworte:
Activation energy
Darcy-Forchheimer flow
Nanoparticles
Soft computing
Stenosed arteries
Darcy-Forchheimer flow
Nanoparticles
Soft computing
Activation energy
Stenosed arteries
ISSN:0952-1976
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Enhancing blood flow in stenosed arteries through the optimization of nanoparticle-based solutions is one potential application of this research to improve medication delivery systems and therapeutic therapies for cardiovascular disorders. This research employed artificial neural networks (ANNs) to analyze the Darcy-Forchheimer flow of hybrid nanoparticles embedded in magnetized blood flow across the stenosed arteries. The ANNs were specifically used to consider the influence of heat generation and activation energy on this complex flow scenario. Heat transfer analysis accounts for multiple factors, such as thermal radiation, viscous dissipation, heat sources, and Joule heating. These elements collectively exert a substantial influence on the overall heat transfer process. Similarity transforms are applied to convert the original Partial Differential Equation (PDE) into a more manageable Ordinary Differential Equation (ODE). This ODE is solved numerically with MATLAB's built-in bvp4c scheme. The Levenberg–Marquardt Algorithm (LMA) is considered in multi-layer perceptron models with 10 neurons in the hidden layers. The ANN model is structured with a configuration that includes 9 input layers for data entry, 3 output layers for results, and 10 hidden layers that process information between the input and output stages. Increased porous parameter values indicate thermal energy is held longer within the flow. Increasing the heat source and thermal radiation leads to a rise in the Nusselt number while increasing the flow parameter causes it to drop.
ISSN:0952-1976
DOI:10.1016/j.engappai.2025.110737