Artificial intelligent computing of hybrid nano suspension for enhanced thermal mechanism in Jeffery Hamel configuration

Nanofluids have potential role in the heat transfer enhancement for engineering applications. These fluids together with physical controls change the dynamics of flow inside oblique channel. Thus, this work focuses on the organization of hybrid model using dissipation function and enhanced properties nanoparticles (NiZnFe2O4, and MnZnFe2O4). The similarity transforms used in the governing model and obtain a new model for hybrid nanofluids. The output estimated in the form of graphical results using Levenberg Marquardt Back Propagation scheme under multiple variations. It is noticed that the velocity is maximum towards the middle line of the channel and backflow observed in the surrounding of walls. Higher concentration declines the velocity due to improved density while Reynolds number enhances it. Further, the temperature is also scrutinized optimum about \:η\:=0.0 for Eckert and Reynolds number, which represents the central part of the physical domain. The skin friction improves from 3.87182 to 3.88703, 0.76311 to 0.77658 against \:φ\:₁=1.0% to \:φ\:₁=4.0% . The significant increase (7.63195 to 8.09306, and 4.68020 to 4.96782) in the heat transfer gradient is observed for 1.0% to 4.0% concentration factor. The convergence of the model achieved excellently in the form of error histogram ( \:2.47e⁻⁰⁶,\:4.6e⁻⁰⁷,\:6.01e⁻⁰⁶,\:7.02e⁻⁰⁶,\:4.17e⁻⁰⁶,\:1.06e⁻⁰⁶ ), MSE from \:10⁻¹⁰ to \:10⁻¹² , best validation performance from \:4.0559e⁻¹⁰ to \:8.8097e⁻¹² (velocity case), \:3.684e⁻¹⁰ to \:8.1913e⁻¹³ (temperature case) at epochs 1000. The heat transport through such configuration has potential relevance to the thermal transport in cooling of drilling, mining machinery, and mines equipment. Such problems can also be useful for heat transport systems associated with deep mine energy and waste heat recovery which required effective heat removal.