Supervised learning approaches in neural networks to optimize and understand radiative heat transfer in chemically reactive front and rear stagnation point flow over backward and forward moving surface

Heat and mass transfer analysis perform essential roles in engineering applications such as polymer extrusion, nanofluid cooling, coating processes, and chemical reactors. Using Brownian motion and thermophoresis effects, the present study examines how a magnetic field, thermal radiation, chemical reaction, blowing, and suction affect front and rear stagnation point flow across a moving surface. The Runge-Kutta-Fehlberg fourth and fifth order (RKF-45) technique is used to get numerical results, which are used to assess the model’s performance. To predict velocity, temperature, and concentration profiles, an artificial neural network (ANN) model is used. Furthermore, the heat transfer rate is optimized using the Taguchi technique, and the importance of controlling factors is determined using analysis of variance (ANOVA). The findings indicate that the front and rear stagnation points have the highest heat transfer rates of 1.72498273340198 and 1.61755052910131, respectively. According to ANOVA, the suction/blowing parameter dominates at the rear stagnation point with a contribution of 38.46%, while thermal radiation contributes 30.17% to the variance in heat transfer at the front stagnation point. For the analysis and optimization of complicated heat and mass transfer systems, the suggested ANN-Taguchi framework offers a reliable and efficient strategy.