Thermal mixing of fractal geometry induced turbulence

Abstract

The present study focuses on exploring numerically the possibility of employing the unique features of fractal geometry to strengthen the thermal mixing performance of a free cooling system that placing within a heating, ventilation and air conditioning system (HVAC). Square duct is used throughout the study with different hydrodynamic inlet conditions, i.e. (i) co-axial flow injection, and (ii) T-duct channelling of working fluid at different temperatures. In general, grid type and orifice-liked inserts are investigated, namely: (i) regular grid (RG), (ii) 2D space filling square fractal grid (SFG), and (iii) orifices. Numerical validations using experimental results from Mazellier and Vassilicos (2010), Morrison, Deotte, Nail, and Panak (1993), Nicolleau, Salim, and Nowakowski (2011), as well as the in-house wind tunnel experiments are conducted. Clearly, all the numerical predictions are in good agreement with the recorded data. Three physical scenarios are then revealed in phases. The first phase aims to evaluate the thermal mixing performance between grid inserts (RG and SFG) and circular orifice (CO) of different thicknesses at δ = 5mm and 40mm. It is found that CO outperforms the rest in thermal mixing, where wider in range and higher in value of turbulence kinetic energy is generated leeward from the orifice. In phase two, the effects of inserts tilting at β = 0°, ±45° are performed using positive square fractal grid (PSFG), negative square fractal grid (NSFG), and CO. It is observed that tilted inserts thermal mixing are significantly better than the non-tilted cases. This is due to the increase in insert surface area in producing larger scales of flow fluctuation. Hence, β = +45° tilted CO thermal mixing performance is about 1382% and 374% higher than PSFG and NSFG, respectively, at x / H = 4.2. Lastly, the implementation of fractal characteristics around the perimeter of an orifice is carried out to further improve thermal mixing. The selected geometries include (i) CO, (ii) square orifice (SO), (iii) square fractal orifice (SFO), and (iv) Koch’s fractal orifice (KSFO). The result show that KSFO generated area-averaged turbulence kinetic energy of about 37%, 48%, 371%, and 1454% higher than those of CO, SO, SFO. Overall, KSFO forms a good balance between the pressure coefficient and the thermal mixing at a Re_H of 1.94×10^4.