Computational intelligence approach of Williamson hybrid nanofluid flow over a triangular wedge with nonlinear stratification and convective boundary effects

: Advanced and new technologies are highly depended on the thermal capability and efficient performance of cooling and heating for long standing working management. Nanofluid and hybrid nanofluids play significant role to uplift this performance. In view of such challenging task, this study examines the Falkner–Skan wedge flow of a thermally stratified Williamson hybrid nanofluid composed of nanoparticles dispersed in Methanol. The flow is generated through a nonlinear permeable stretchable wedge in the stagnation region. Nonlinear thermal stratification and viscous dissipation are imposed to elaborate the nature of heat transportation. Convective surface condition and thermal radiation are utilized to visualize the heat flow near the boundary. Features of hybrid nanofluid are analyzed though mixed convection and nanoparticle shape factors. The governing nonlinear equations are reduced via similarity transformations and solved through bvp5c. To enhance predictive capability, Artificial Neural Networks (ANNs) trained with the Levenberg–Marquardt algorithm is employed to estimate surface drag and heat flux under non-stratified scenarios. The ANN simulation represents strong accuracy, with a mean square error (MSE) of and for the Eckert number ( ) case at epochs, while the stratification parameter ( ) case yields and over training, validation, testing and epochs respectively. It is interesting to note that fluid’s velocity is higher in the case of nanofluid in comparison to hybrid nanofluid and viscous fluid for velocity ratio parameter. These findings confirm that Williamson hybrid nanofluids substantially improve heat transfer and flow characteristics, supporting their applications in heat exchangers, solar collectors, cooling systems, and porous media technologies.