PIV and dynamic LES of the turbulent stream and mixing induced by a V-grooved blade axial agitator

Abstract

The hydrodynamic behavior of a turbulent flow and the mixing characteristics generated by a V-grooved axial impeller inside an agitated tank reactor were investigated both experimentally and numerically. Angle resolved Particle Image Velocimetry (PIV) techniques with an angular displacement Δθ=5° have been applied and two aerodynamic planes along the blades were considered. PIV-based results were compared to those obtained by Large Eddy Simulation (LES), used with the dynamic Smagorinsky-Lilly sub-grid scale (SGS) model. Results showed the existence of distinctive recirculation zones in the aerodynamic planes, and new additional frequencies in the impeller stream, induced by the grooves. A decrease of mixing time of about 11 % was obtained experimentally, consequence of the better suction induced by the grooved blades in the early stages of mixing. Mean velocities, vorticity, TKE obtained from LES showed a good agreement with the PIV-based results. The distributions of turbulence dissipation rate ε were similar to those obtained from PIV, however showing high under-predicted magnitudes.

Validation of large eddy simulation of flow behind a circular cylinder

“…Despite the quantitative discrepancies in the data, all the qualitative trends in the timeaveraged variables are in agreement. Furthermore, the quantitative discrepancies between the experiment and LES are similar or lower than the ones reported for example in [2,8] or in [19]. The initial comparison of measured and computed flow dynamics is performed utilizing the spectra of streamwise and spanwise velocity components evaluated in the points p 1 and p 2 depicted in the left hand side of Figure 4 and located at (4, 0, 0) and (4.53, 1.15, 0), respectively.…”

“…[2,[4][5][6]. Furthermore, if any analysis of the system dynamics is performed, it is usually limited to the comparison of turbulence spectra as in [7,8]. Such a situation follows from the fact that most existing verification and validation methods for fluid flow simulations are derived for the computationally cheaper and better established Reynolds-averaged Navier-Stokes equations (RANS) [9].…”

“…The selected modeling approach provides high-resolution data in the zone of interest corresponding to the measured region. Both experimental and numerical data are, similarly to [7,8] analyzed with respect to (i) time-averaged quantities, and (ii) spectra of velocity components at given locations. Furthermore, the measured and modeled flow fields are studied via the means of proper orthogonal decomposition (POD) [13][14][15][16], where we focus on the most energetic coherent flow structures.…”

Hydrodynamic Behavior of Particles in a 3D Integral Multi-Jet Spout-Fluidized Bed

“…Turbulence kinetic energy is an important index to measure the turbulence degree of a fluid; the fluctuating velocities for the turbulent kinetic energy (TKE 39 or k ) were calculated by considering u ′ = u ″, that is, considering the processed fluctuations free from the influence of the periodic motion. TKE is expressed as where and are the average values of the square of the axial and radial pulsation velocities, respectively, and K a represents the average of TKE over the three cross sections.…”

Thriving artificial underwater drag-reduction materials inspired from aquatic animals: progresses and challenges

“…In addition, for motor blades, inspired by bionics, 160 Zhu and Gao utilized a winged propeller to suppress the generation of tip vortex cavitation (TVC). 45 Also, the simulation results showed that the winglet propeller significantly weakened the strength of the blade tip vortex wake and had a significant effect.…”