Predictive Modelling and Optimization of Entropy Generation in Power-Law Hybrid Nanofluid Flow Over Rotating Disk Using Adaptive Neuro-Fuzzy Inference System-Particle Swarm Optimization

Modelling hypothetical phenomena with a power-law hybrid fluid over a spinning disk is the main objective of the present effort. In this phase of progress, optimization of a numerical approach that integrates the Adaptive Neuro-Fuzzy Inference System (ANFIS) and Particle Swarm Optimization (PSO) to fabricate a heat source/sink, non-linear thermal radiation and entropy generation has been facilitated. To make sure that the appropriate self-similarity variables have been used to convert a non-linear PDE set of s into an ODE. The model’s study findings, with a few notable exceptions, generally align with those from earlier studies included in the dataset that trained the ANFIS-PSO model. The visually appealing outcomes for various profiles reflect the effects of the active elements. This illustrates how the velocity and temperature profiles grow abruptly with increasing magnetic field and radiation, creating a paradox with decreasing electric field inputs. Also, the temperature profile’s tilting of the heat source or sink indicates inclination. Additionally, ANFIS-PSO training was used to examine the approximations for certain circumstances, and the effectiveness of the created ANFIS-PSO was assessed by comparing it to a testing dataset. The nanoparticles used here, with their longer render, might be suitable for use in biological and industrial applications, including heat exchanger simulation, the glass and polymer industries, and metal plate cooling.