Adaptive neuro-fuzzy inference system prediction of thermal transport in Casson ternary hybrid nanofluid thin film flow for biomedical applications

Abstract

In this work, the adaptive neuro-fuzzy inference system (ANFIS) and particle swarm optimization (PSO) were used to anticipate the behavior of heat transfer in ternary hybrid nanofluid flow through thin films. For intricate heat transfer processes in nanofluid applications, the combined ANFIS-PSO model improved forecast accuracy. The simulated PDEs were converted into ODEs by varying similarity factors. A ternary hybrid nanofluid across the surface, radiation, heat source/sink, and non-uniform magnetic field were used to theoretically examine the unique properties of unstable thin film flow; nanoparticles, namely MWCNT, SWCNT, and 2TiO, were used in both the Casson and non-Casson scenarios, using the base fluid, blood. The ANFIS-PSO models were trained using the numerical results from MATLAB's built-in BVP4C function to control the complexity and predict the results. Plotting and analysis were done to see how various flow parameters affect temperature, velocity, and heat transfer. A ternary hybrid nanofluid without a Casson scenario had a higher Nusselt number, per the study's conclusions, than a Casson ternary nanofluid with blood as the base fluid with nanoparticles SWCNT +TIO2 +MWCNT. It was discovered that the truth values are accurately predicted by ANFIS-PSO models, and in most of the runs in Table 3, in Case-2, the rate of heat transmission is 1% higher than in Case-1.