The purpose of this study is to develop an intelligent artificial neural network-based framework for predicting the flow, heat, mass and bioconvection characteristics of Johnson–Segalman nanoliquid transport in a symmetric peristaltic channel. The model incorporates viscous dissipation, wavy boundaries, magnetic field effects, thermal radiation, thermophoresis, Brownian motion and chemical reaction. Numerical solutions obtained via lubrication theory are used to train the neural network, enabling accurate prediction of velocity, temperature, concentration and bioconvection profiles under varying physical parameters.
The governing partial differential equations describing Johnson–Segalman nanoliquid flow in a symmetric peristaltic channel are reduced to a system of dimensionless ordinary differential equations using the lubrication approximation. These equations, incorporating viscous dissipation, magnetic field, thermal radiation, thermophoresis, Brownian motion, chemical reaction and bioconvection effects, are solved numerically using NDSolve in Mathematica. The resulting data sets are used to train an artificial neural network implemented in Python using TensorFlow, with optimized architecture and performance assessed through multiple statistical error metrics.
The results demonstrate that the artificial neural network accurately predicts the velocity, temperature, concentration and bioconvection characteristics of Johnson–Segalman nanoliquid flow in a peristaltic channel. Excellent agreement is observed between numerical and predicted results, confirmed by low error metrics. The magnetic field strength and thermal radiation significantly enhance the thermal distribution, while the Weissenberg number and velocity slip parameter strongly influence the velocity field. The trained model effectively captures the complex nonlinear interactions among physical parameters, confirming the reliability of the proposed intelligent framework for analyzing peristaltic nanofluid transport phenomena.
This study presents a novel integration of peristaltic Johnson–Segalman nanoliquid modeling with an intelligent artificial neural network framework. Unlike conventional numerical or semi-analytical approaches, the proposed methodology combines lubrication theory-based solutions with data-driven learning to efficiently predict flow, thermal, mass and bioconvection behaviors under complex physical effects. The incorporation of viscous dissipation, slip conditions, magnetic field, thermal radiation, thermophoresis, chemical reaction and microbial dynamics within a unified intelligent framework offers new insights and provides a computationally efficient tool for analyzing complex microfluidic transport systems.
