No abstract

No abstract

This research investigates the applicability of an ANN and genetic algorithms for modeling and multiobjective optimization of the thermal conductivity and viscosity of water-based spinel-type MnFe2O4 nanofluid. Levenberg-Marquardt, quasi-Newton, and resilient backpropagation methods are employed to train the ANN. The support vector machine (SVM) method is also presented for comparative purposes. Experimental results demonstrate the efficacy of the developed ANN with the LM-BR training algorithm and the 3-10-10-2 structure for the prediction of the thermophysical properties of nanofluids in terms of the significantly superior accuracy compared to developing the correlation and employing SVM regression. Moreover, the genetic algorithm is implemented to determine the optimal conditions, i.e., maximum thermal conductivity and minimum nanofluid viscosity, based on the developed ANN.

No abstract

A schematic diagram of the horizontal pulsed plate column has been used.