An Efficient Hybrid STO-MFDNN Approach for Induction Effect in Nanoparticle with Magneto-Convective Nanofluid Boundary Layer Flow

The accurate synthesis of bespoke design requires complicated chemical reactions and high temperature conditions in modern nanomaterial coating techniques. These flow processes are quite complicated, involving not just viscous behavior but also mass and heat transport. Magnetic nanoparticles are utilized by nanocoatings, which are under manipulation by external magnetic fields. Mathematical models offer a low-cost window into the fundamental properties of these coating dynamics processes. This paper proposes a hybrid approach for induction effect in nanoparticle with magneto-convective nanofluid boundary layer. The hybrid technique that is being proposed involves the simultaneous application of Siberian Tiger Optimization (STO) and Multi-Fidelity Deep Neural Network (MFDNN). Hence, it is named as STO-MFDNN technique. The primary goal of this proposed method is to enhance nanoparticle solid volume fraction and minimize error during validation. The proposed technique — STO approach is utilized to optimize the process parameter and the MFDNN approach is utilized to accurately predict the nanoparticle behavior. By then, the MATLAB platform has the proposed approach implemented, and the present method is used to calculate the execution. The proposed technique displays superior outcomes in all existing systems like Wild Horse Optimizer (WHO), Fertile Field Algorithm (FFA) Seagull Optimization Algorithm (SOA). The existing technique displays the errors of 0.24%, 0.21%, 0.18% and the proposed technique displays an error of 0.15% concluding low error when compared to the existing approach.