Key points are not available for this paper at this time.
In this research, we employ artificial neural networks (ANNs) with the Levenberg–Marquardt backpropagation method (ANN-LMBM) to effectively model and solve the nonlinear heat transfer problem of micropolar (MP) fluid through porous media under the influence of thermal radiation. This approach offers a fast, cost-effective alternative to traditional empirical methods, delivering high-precision results without the need for laborious testing (TT). By using a dataset of 1000 points in the range 0, 8, the ANN is trained to analyze the impact of six key physical parameters on the flow dynamics. We examine critical outcomes such as velocity, angular microstructure velocity, temperature distribution, heat flux coefficient, and shearing stress at the plate. Performance is rigorously evaluated through mean square error (MSE), training (TR), validation (VL), TT, and fitting (FT) metrics, all of which affirm the model’s robustness. The results are further validated with error histograms (ES) and regression analysis (RG), showcasing remarkable accuracy with error margins between E-4 and E-8. This study highlights the efficiency and reliability of ANN-LMBM in tackling complex heat transfer problems, providing deep insights into how physical parameters affect fluid behavior. The method’s high stability and precision make it a valuable tool for researchers and engineers, advancing the study of heat transfer in MP fluids.
Mahariq et al. (Wed,) studied this question.