Thermal and mass exchange in a multiphase peristaltic flow with electric-debye-layer effects and chemical reactions using machine learning

Abstract

The present investigation delves into the analysis of mass and thermal transmission during the peristalsis of a two phase flow of a Prandtl fluid, within a symmetric channel by considering entropy generation and minimization. The considered flow is examined by taking into account the impacts of electroosmotic effects through Electric Debye Layer (EDL) and chemical reactions followed by the lubrication theory approximation. Numerical solutions are acquired for the consequential differential equations through artificial neural networks (ANN) based fitness functions. The artificial neural networks (ANN) based fitness function optimized through hybrid evolutionary algorithms especially ABC (artificial bee colony) and NNA (neural network algorithm) like ANN-ABC-NNA. For the efficiency and validation of the algorithm we conduct the one hundred (100) independent runs. The obtained results through ANN-ABC-NNA are compared with the exact solutions for validation of results. In the present examination, the velocity profile is noticed to increase by the rise in the electroosmotic factor and solid particles concentration, while it decreases for the Prandtl fluid parameter. This research holds practical significance in diverse applications, particularly in the design and optimization of microfluidic devices.

Keywords

• Multiphase flow
• Entropy generation
• Chemical reaction
• EDL
• Heat and mass transfer
• Machine learning solutions