2D and 3D numerical modelling of internal flow of Pressure-swirl atomizer

conference paper

en

This paper compares 2D axisymmetric and 3D numerical models used to predict the internal flow of a pressure-swirl atomizer using a commercial software Ansys Fluent 18.1. The computed results are compared with experimental data in terms of spray cone angle (SCA), discharge coefficient (CD), internal air-core dimensions and swirl velocity profile. The swirl velocity was experimentally studied using a Laser Doppler Anemometry in a scaled transparent model of the atomizer. The internal air-core was visualized at high temporal and spatial resolution by a high-speed camera with backlit illumination. The internal flow was numerically treated as transient two-phase flow. The gas-liquid interface was captured with Volume of Fluid scheme. The numerical solver used both laminar and turbulent approach. Turbulence was modelled using k-ε, k-ω, Reynolds Stress model (RSM) and coarse Large Eddy Simulation (LES). The laminar solver was capable to predict all the parameters with an error less than 5% compared with the experimental results in both 2D and 3D simulation. However, it overpredicted the velocity of the discharged liquid sheet. The LES model performed similarly to the laminar solver, but the liquid sheet velocity was 10% lower. The two-equation models k-ε and k-ω overpredicted the turbulence viscosity and the internal air-core was not predicted

This paper compares 2D axisymmetric and 3D numerical models used to predict the internal flow of a pressure-swirl atomizer using a commercial software Ansys Fluent 18.1. The computed results are compared with experimental data in terms of spray cone angle (SCA), discharge coefficient (CD), internal air-core dimensions and swirl velocity profile. The swirl velocity was experimentally studied using a Laser Doppler Anemometry in a scaled transparent model of the atomizer. The internal air-core was visualized at high temporal and spatial resolution by a high-speed camera with backlit illumination. The internal flow was numerically treated as transient two-phase flow. The gas-liquid interface was captured with Volume of Fluid scheme. The numerical solver used both laminar and turbulent approach. Turbulence was modelled using k-ε, k-ω, Reynolds Stress model (RSM) and coarse Large Eddy Simulation (LES). The laminar solver was capable to predict all the parameters with an error less than 5% compared with the experimental results in both 2D and 3D simulation. However, it overpredicted the velocity of the discharged liquid sheet. The LES model performed similarly to the laminar solver, but the liquid sheet velocity was 10% lower. The two-equation models k-ε and k-ω overpredicted the turbulence viscosity and the internal air-core was not predicted

Pressure-swirl, CFD, Internal flow, LDA

28.06.2019

EDP Sciences

2100-014X

EFM18 – Experimental Fluid Mechanics 2018

213

1

1–6

6