Deep Learning-Based Heat Transfer Analysis of Titanox and Silver/H2O Hybrid Nanofluid Flow Over a Curved Stretching Surface with Bioconvection

• MHD flow of TiO 2 /H 2 O and TiO 2 –Ag/ H 2 O hybrid nanofluids were analysed. • Hall current and thermal radiation effects influenced the flow behaviour significantly. • TiO 2 /H 2 O performs better than TiO 2 –Ag/H 2 O in velocity, thermal, concentration and microorganism. • An ANN-based predictive model was developed for flow and thermal fields. • Minimum MSE value of 6.40 × 10 − 8 for Nu is obtained at epoch 172. This research focuses on the performance of Titanium dioxide (TiO 2 ) and Silver (Ag) nano sized particles that are suspended in water (H 2 O) under magnetohydrodynamics and bio-convection effects over a curved stretching sheet using machine learning approach. Key factors are the existence of viscous dissipation, heat radiation, chemical reaction, Arrhenius activation energy and gyrotactic microorganisms with the no slip condition. Given these elements, this study intends to offer ideas into productive applications in which energy efficiency and thermal managements are top priorities. By using the proper exponential similarity transformation,the highly non-linear partial differential equations are reverted into non-linear ordinary differential form. The numerical method Bvp4c is applied to solve the non-linear ODEs. In conjunction with the numerical approach, artificial neural networks were employed to accurately capture complex patterns, achieving a mean squared error and fitness values ranging from 10 − 4 to 10 − 8 , thereby reinforcing the reliability of the numerical results. The ANN model allocates 70% of the data for training, 15% for testing, and 15% for validation, facilitating rapid learning and performance evaluation. Additionally, the heat transfer and mass transfer analysis are assessed, the velocity profile lessens with increasing magnetic number values. The Nusselt number diminishes with increasing magnetic number and radiation parameter. The concentration raises with activation energy variables while diminishing with stronger chemical reaction and higher schmidt number. The microorganisms density diminishes with an increased function of bioconvection Lewis and bioconvection Peclet numbers. This two-pronged strategy emphasises the robustness and versatility of the numerical techniques.