Novel computational analysis of entropy generation in chemically reactive non-newtonian fluid over a stretching surface with ohmic and viscous dissipation effect: non-similar solution
摘要
Fluid mechanics theorists and scientists are deeply engaged in efforts to enhance the efficiency of thermal engineering systems. A notable achievement in this field is the use of simulation methods to reduce energy dissipation. Such losses can impair thermal and mechanical functions, stemming from entropy generation and related to irreversible thermodynamic processes. The author concentrates on analysing entropy generation within mechanics problems involving non-Newtonian fluids. Applications span heat exchangers, turbo-machines, combustion systems, nuclear reactor cooling, and many more. The considered problem involves non-similar partial differential equations with various pertained parameters. An approximate solution is obtained using an advanced machine learning approach known as the Multilayer Perceptron Artificial Neural Network (MLP-ANN). The predicted results are compared with those from a numerical solution computed via the implicit finite difference method. The validity and reliability of the proposed MLP-ANN scheme are found to be efficient and accurate. The proposed ANN performance and efficiency are attained at 1.59