Intelligent levenberg-marquardt neural network solution to flow of carbon nanotubes in a nozzle of liquid rocket engine.

We have analyzed velocity and temperature profiles for the flow of a nanofluid composed of carbon nanotubes (CNTs) and kerosene oil in a regenerative cooling channel of a rocket engine. After developing the flow model using conservation laws, we have employed similarity transformations to get a system of dimensionless nonlinear ordinary differential equations. The velocity and temperature profiles are approximated by employing artificial intelligence. The use of Artificial Neural Networks based on Levenberg-Marquardt algorithm (ANN-LMA) has been made for this purpose. Both single and multi-walled carbon nanotubes (SWCNTs and MWCNTs) have been considered. The reference data to train test and validate the feedforward neural network was created using built-in subroutines of mathematical software MATLAB. Different scenarios involving various sets of values of significant physical parameters were considered and corresponding to each scenario a network was trained. The accuracy of the network was also ensured by comparing the network values with numerical outputs of reference data. We validated the performance of the network through regression analysis, histogram studies, and the calculation of the mean square error (MSE). The parametric study conducted in this investigation will assist aerospace engineers in designing regenerative equipment in an efficient manner.&#xD.