Active and Passive Control of Nanofluid Heat Transfer via Fractional Operators and Neural Networks

The free convective flow of heat transport with nanofluids over a plate must be studied in order to enhance thermal technologies and increase the efficiency of heat transfer in engineering system. The present article introduces the study of Tio2 and Co3O4 nanoparticles with the base fluid of C3H8O2. Heat transfer on Casson with the multiple effects is taken into account. This new research presents an innovative approach to studying how mass transport and heat transfer interact over a plate. The core of the problem involves complex mathematical equations called fractional partial differential equations, which are difficult to solve using traditional methods. Instead of using the standard mathematical techniques, the researchers took an artificial neural network (ANN), a type of Artificial Intelligence, and used it to solve these equations. The ANN was trained by the Levenberg–Marquadrt backpropagation algorithm, which is a sophisticated optimization algorithm. The required training data were created analytically (which means precisely calculated) using the Laplace transform method. The data were divided up with 70% for the training of the primary and 15% for checking the generalization capacity of the model and the other 15% for validation. The most significant achievement of this work is the successful combining of the power of fractional calculus (for the modeling of the transport processes in detail, strictly nonjust behaving) with AIAN (neural network)-based optimization (for efficient and accurate solution). This provides a credible and efficient means of modeling and ultimately optimizing heat transfer in state-of-the-art nanofluid based engineering systems.