Optimal Analysis of MHD Peristaltic flow of Viscoplastic nanofluid with chemical reaction and Electro-Osmotic effect via artificial neural networks and response surface methodology

Abstract In this study the authors aim to investigate sensitivity analysis of pressure rise per wavelength and frictional forces in magnetohydrodynamics peristaltic flow of viscoplastic nanofluid with the effect of electro-osmotic and chemical reaction in tapered channel. To achieve this aim, we have adopted an empirical modeling methodology that establishing pressure rise per wavelength and frictional forces on upper and lower wall relationships as functions that vary according to the governing parameters essential to the problem. To analyze the sensitivity of transport parameters, including pressure rise per wavelengthand frictional forces on upper and lower wall, we initially develop an empirical model for key responses, (pressure rise per wavelengthand frictional forces on upper and lower wall). To my best knowledge, no empirical correlations for pressure rise per wavelength and frictional forces associated with viscoplastic nanofluid peristaltic flow in tapered channels have addressed in the literature. To develop this empirical model, we used the approach of Response Surface Methodology and Artificial Neural Networks. The appropriateness of empirical model is evaluated by determining the coefficient of determination and conducting an analysis of variance to ensure accuracy and reliability. The empirical model display good fit, with the coefficient of determination of 97.65% for all resulting responses, (pressure rise per wavelengthand frictional forces on upper and lower wall). Residuals are plotted to verify accuracy and reliability of empirical correlation. A sensitivity analysis was performed to identify the most influential input parameters. The results, presented graphically, show that the values of resulting responses (pressure rise per wavelengthand frictional forces on upper and lower wall) are highly sensitive to variation in Brownian motion parameter in most situations. This innovative approach contributes valuable insights for optimizing flow performance in biomedical and engineering applications, filling a critical gap in the existing body of knowledge.