Identification of forced convection in pulsating flow at a backward facing step with a stationary cylinder subjected to nanofluid

In the present study, the application of the system identification method for forecasting the thermal performance of forced pulsating flow at a backward facing step with a stationary cylinder subjected to nanofluid is presented. The governing equations are solved with a finite volume based code. The effects of various parameter frequencies (0.25 Hz-8 Hz), Reynolds number (50-200), nanoparticle volume fraction (0.00-0.06) on the fluid flow and heat transfer characteristics are numerically studied. Nonlinear system identification toolbox of Matlab is utilized to obtain nonlinear dynamic models of data sets corresponding to different nanoparticle volume fractions at frequencies of 1, 4 and 8 Hz. It is observed that heat transfer is enhanced with increasing the frequency of the oscillation, nanoparticle volume fraction and Reynolds number. The level of the nonlinearity (distortion from a pure sinusoid) decreases with increasing ϕ and with decreasing Reynolds number. It is also shown that nonlinear dynamic models obtained from system identification toolbox could produce thermal output (length averaged Nusselt number) as close to as output from a high fidelity CFD simulation.