Lian, YongSheng*

About our group
The computational thermo-fluids group at the University of Louisville was established in 2008. It currently hosts two post-docs and six doctoral students. Our group engages cutting edge research in the broad areas of aerodynamics, multiphase flow, and heat transfer. Our work has been funded by U.S. Air Force, NASA, NSF, DOE, General Electric, Cummins, and Hitachi. Since 2008 we have graduated more than 10 master students and 1 doctoral student. We have published more than 30 journal papers and conference papers since 2009.

Project 1

Simulation of Supercooled large droplet for aviation safety

This project investigates the fundamental fluid mechanics phenomena associated with the so called supercooled large droplet. Supercooled large droplets (SLDs) are droplets of radius greater than 40 µm. They exist in liquid form in freezing drizzle and rain because of the absence of ice nuclei in the environment. Unlike small droplets which freeze immediately after striking aircraft surface, SLDs can splash into smaller droplets and/or bounce off from surface and spread into unprotected areas to form clear ice which is heavy and difficult to remove. The accretion of ice in unprotected areas can cause a drastic decrease in lift and an increase in drag. SLDs are considered as a significant aviation hazard even to aircraft certified to operate in icing conditions and happen more frequently than previously accepted. We solve the three-dimensional Navier-Stokes equations using adaptive mesh refinement method. The gas-liquid-solid interface will be captured with a novel moment of fluid method.

Students

Andrew Work (Ph.D student), UofL
Yisen Guo (Ph.D student), UofL
Yongsheng Lian (faculty), UofL

Project 2

Simulation of flat fan nozzle

This project focuses on the numerical simulation of two-phase flow near the outlet of flat fan nozzles under low operating pressure. The moment-of-fluid (MOF) method is used for the representation of the liquid gas interface and the directional split method is used for the advection of the interface. A variable density pressure projection algorithm is used for the fluid solver and a block structured adaptive mesh refinement (AMR) method is used to locally increase the resolution near interface. The internal geometry of the nozzle is defined by three parameters: nozzle inlet diameter D, V-cut or U-cut width W and V-cut or U-cut offset H. The effects of these three parameters on the fan exit angle and pressure loss through the nozzles are studied.

Students

Guibo Li (post-doc), UofL
Yongsheng Lian (faculty)

Project 3

Transitional flow simulation of dynamic stall for helicopters

This project investigates the impact of transitional flow on dynamic stall. Dynamic stall is an important unsteady phenomenon frequently encountered in helicopter, turbo-machinery, wind turbine, and military fighters. The problem has been widely studied but it is still not well understood due to the complex multiscale fluid mechanics phenomena involved. To accurately capture dynamic stall in a computational environment, considerable emphasis has to be placed on the grid dependence, temporal convergence, and influence of turbulence. Most grid dependence and convergence studies are typically performed as an after thought and they are computational expensive and difficult to verify simulation accuracy. We propose to study dynamic stall using a feature based grid adaption method. The feature based method automatically increases the grid resolution in regions of importance and coarsen the grid resolution in other regions. The method can accelerate spatial convergence to reduce the computational cost. Initial results suggest feature based adaptation has potential in refining the mesh in the wake of the airfoil, allowing vorticity to be carried out a several chords behind the airfoil without dissipation.

Students

Kyle Hord (doctoral student), UofL
Yongsheng Lian (faculty)

Collaborators

Bailey, Sean (Faculty at UKY)

Computational methods

For project 1 and 2 we use an in-house code. The code solves the Navier-Stokes equations using a projection method.
For project 3 we use an in-house code. The code solves the compressible Navier-Stokes equations on the composite overlapping grid.

Software

Using in-house codes

UK participation - collaborators

We are preparing a joint proposal with Prof. Saito Kozo at UK Mechanical Engineering department. His group will involve in project 1-2 if the proposal is funded.
We collaborated with Prof. Sean Bailey at the UK Mechanical Engineering department on a topic similar to that in project 3. We intend to continue to work with him for future funding opportunities.