ABSTRACT This study is an analysis of irreversibility using an artificial neural network (ANN) for irreversible irreversibility in Sisko nanofluid flow through a peristaltic channel in the presence of magnetic field, motile micro‐organisms, and slip effects. The problem of the governing nonlinear partial differential equations is reduced to a set of ordinary differential equations by using the lubrication approximation along with the Debye–Hueckel transformation with suitable nondimensional variables. The resultant dimensionless ordinary differential equations are solved numerically by using the NDSolve routine to obtain the accurate numerical profiles of velocity, temperature, concentration, and bioconvection fields. These numerically computed solutions are used to create the datasets, which are used to train the ANN model. The ANN is implemented in Python with the ready‐made library TensorFlow 2.17.0, the network is built with one input layer, two hidden layers (each layer consisted of 64 neurons), and one output layer. A sigmoid activation function is used in the hidden layer, and an Adam optimizer is used for training models. Performance indicators such as the mean square error (MSE), root mean square error (RMSE), regression coefficient (), error histogram, gradient, and relative error are examined in order to analyze the prediction ability of the network. The results show that the ANN gives high accuracy in the learning and prediction of the numerically obtained velocity, thermal, concentration, and bioconvection profiles. Furthermore, the results shown in the analysis indicate that the magnetic field parameter and Eckert number have a significant effect on the thermal profile while the velocity profile is highly affected by the magnetic field. The results of this work have implications for biomedical engineering applications involving transport in microfluidic systems and targeted drug delivery, and also hold potential relevance for processes in environmental engineering.
Ishaq et al. (Fri,) studied this question.