Artificial neural network analysis of bioconvective Casson hybrid nanofluid flow over Howarth's wavy cylinder with viscous dissipation and pollutant concentration

This study examines the artificial neural network (ANN) analysis of bioconvective flow of a Casson hybrid nanofluid involving C u and Al 2 O 3 nanoparticles immersed into the sodium alginate (base fluid) over Howarth's wavy cylinder. This study is significant for promoting environmental sustainability and improving environmental engineering through ANNs. The mathematical model assumes three-dimensional, steady, incompressible CHNF flow over a wavy cylinder under stagnation-point conditions. A special streamline parameter has been introduced to represent the relative strength of the free-stream velocity components. The cylinder's sinusoidal geometry creates stagnation points and the resulting flow including nodal and saddle points, is analysed to reveal velocity patterns. The results reveal that increasing the Hartmann number enhances velocity for nodal and saddle points, while higher special streamline and Casson parameters reduce temperature and concentration distributions. Increasing the Casson parameter increases the Nusselt number by 380.6% (nodal point) and 76.2% (saddle point), while increasing magnetic parameter decreases the Nusselt number by 3.76% (nodal point) and increases it by 0.62% (saddle point). Moreover, increasing chemical reaction parameter increases the Sherwood number by 63.4% (nodal point) and 257.8% (saddle point), while increasing the parameter related to external pollutant source variation decreases it by 0.19% and 0.77%, respectively.