Numerical simulation study on the structure optimization of liquid separation device in supersonic separator

ABSTRACT

The effect of the supersonic separator on the condensation characteristics and fluid separation efficiency is not clear, and the structure of the liquid separation device still has room for optimization. Therefore, based on the rear-swirl type supersonic separator, three kinds of drainage ports and three kinds of liquid collection tank structures were proposed. The supersonic flow and separation process was studied numerically under nine different combination structures. The influence of the drainage port’s length and width and the inclination of the liquid collection tank on the liquefaction and separation process were analyzed. The separation trajectory of condensation droplets and the influence of particle size on the separation efficiency were clarified. The results show that the liquid separation device structure of the Em-Attached wall type can take condensation and separation efficiency into account, and its separation effect is the best. The optimal size of the key components was obtained: the length of the drainage port is 150 mm, the width of the drainage port is 3 mm, and the inclination of the liquid collection tank is 14 °. The particle size of the condensation droplets is in the range of 2 μm to 6 μm.

KEYWORDS:

• Natural gas separation
• supersonic flow
• liquid separation device
• flow field analysis

Acknowledgments

This work was supported by the National Natural Science Foundation of China (Grant No. 51904326), Natural Science Foundation of Shandong Province (Grant No. ZR2019MEE105 and No. ZR2019MEE011), the Science and Technology Plan Projects of Qing Dao (No.19-6-1-87-nsh and No. 17-1-1-88-jch). C.Y. Han would like to express sincere thanks to Mr M.W. Jiang for her continuous encouragement in his scientific research.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Nomenclature

$\rho$	=	fluid density (kg/m3)
u	=	fluid velocity (m/s)
P	=	fluid pressure (Pa)
$\delta$	=	Kronecker delta (-)
$k_{eff}$	=	effective thermal conductivity (-)
${\tau }_{eff}$	=	effective stress tensor (-)
$\tau$	=	viscous stress tensor (-)
$E$	=	total energy (J/kg)
$q$	=	heat flux (W/m2)
$\epsilon$	=	turbulent dissipation rate (%)
$G_k$	=	turbulent kinetic energy item caused by the average velocity gradient (-)
$G_b$	=	turbulent kinetic energy item caused by buoyancy (-)
$k$	=	kinetic energy (J/kg)
${\mu }_{eff}$	=	diffusion Coefficient (m2/s)
$a_k$	=	constants related to $\mathrm{k}$ (-)
$a_{\epsilon }$	=	constants related to $\epsilon$ (-)
$Y_M$	=	effect of compressible turbulent pulsation expansion on dissipation rate, (-)
$m_l$	=	droplet mass (kg)
$u_l$	=	droplet velocity (m/s)
${\rho }_g$	=	gas phase density (kg/m3)
$V_l$	=	droplet volume (m3)
$F$	=	interaction force between the droplet and the surrounding gas (N)
$F_D$	=	drag force of discrete particles (N)
$F_O$	=	other forces (N)
${\mu }_g$	=	gas-phase dynamic viscosity (Pa·s)
$Re$	=	relative Reynolds number (-)
${\rho }_l$	=	droplet density (kg/m3)
$d_l$	=	droplet diameter (m)
$C_D$	=	drag coefficient (-)
$a_1$	=	empirical constants (-)
$a_2$	=	empirical constants (-)
$a_3$	=	empirical constants (-)
$F_T$	=	thermophoretic force (N)
$K_K$	=	empirical constant, ${\mathrm{K}}_{\mathrm{K}}$ = 2.584 (-)
$D_T$	=	thermophoretic coefficient (-)
$F_S$	=	Suffman’s lift (N)
$d_{ij}$	=	strain rate tensor (-)

CrediT authorship contribution statement

Chenyu Han: Writing – original draft. Wenming Jiang: Supervision. Yang Liu: Supervision. Zhanzhao Hu: Software. ZhuoYing Dou: Grammar proofreading.

Additional information

Funding

The work was supported by the National Natural Science Foundation of China [51904326]; Natural Science Foundation of Shandong Province [ZR2019MEE011, ZR2019MEE105].