The design of intelligent networks for entropy generation in Ree-Eyring dissipative fluid flow system along quartic autocatalysis chemical reactions

In this framework entropy generation optimization in dissipative flow of Ree-Eyring fluid model (EGODF-RFM) with chemical reaction of quartic autocatalysis between two rotating disks is observed by operating the artificial neural networks model backpropagated with Bayesian Regularization technique (ANNM-BBRT). The effects of viscous dissipation and nonlinear thermal radiation on heat transmission are examined. The effects of thermophoresis and Brownian motion are discussed in detail. Entropy generation optimization in nanofluids with homogeneous and heterogeneous chemical reactions is explored in terms of physical features. Partial differential equations of (EGODF-RFM) were transformed to ordinary differential equations using appropriate techniques of similarity conversions. To solve the ODEs and evaluate the reference dataset for ANNM-BBRT, the Homotopy analysis method (HAM) is used. In a MATLAB software this dataset assists to compute the estimated solution of EGODF-RFM. Performance of designed intelligent based computing solver is validated by error histogram, regression analysis, and MSE outcomes. For various parameters, the impacts of flow on the velocity profile, temperature distribution, and concentration profile are investigated. The axial component of velocity is declined with the increase in Weissenberg number but opposite trend is observed in the case of radial and tangential components of velocity. The temperature profile enhances due to higher values of Weissenberg number, thermophoresis parameter, Brownian motion parameter and Eckert number, while the concentration profile associated with the presented fluid flow system is accelerated due to rise of Brownian motion parameter, homogeneous and heterogeneous reaction parameters, respectively. Due to Eckert number, a dual effect is observed for entropy generation and Bejan number, and in the region near the bottom disc, both entropy generation and Bejan numbers are reduced, but their values are gradually improved in the region closer to above disc. Impact of We and Re on the skin friction coefficient is to be observed rising and rate of heat transmission increases as Rd increases, whereas We has the reverse effect.

Automated summary

This study optimized the entropy generation in dissipative flow using artificial neural networks and investigated the effects of various parameters on the flow characteristics.