Kato Chisachi Laboratory Official Site / Institute of Industrial Science, The University of Tokyo

Top |  Research |  Papers |  Awards and media |  Facilities |  Member |  Access |  Link
Fully Resolved Large Eddy Simulation as an Alternative to Towing Tank Resistance Tests
The demand for towing tank tests has been increasing due to the requirements of the EEDI*1 imposed by IMO*2 as a measure for controlling CO2 emissions from ships. Although numerical simulations by RANS (Reynolds-averaged Navier-Stokes) have been widely used in the early stage of hull design, the prediction accuracy has to be further improved for such simulations to become a complete alternative for towing tank tests. In particular, for resistance tests and self-propulsion tests, the relative error to the experimental data has to be reduced to within 1%. By tuning the turbulence models, RANS simulations may provide such accuracy. But, it seems difficult for RANS simulations to always guarantee such accuracy for different types of ships. With the recent speed-up of high-end computers, fully resolved large eddy simulation (LES), which directly computes the streamwise vortices in the turbulent boundary layer (TBL), is expected to become feasible within a few years. Fully resolved LES provides almost the same accuracy as DNS (Direct Numerical Simulation) and will probably achieve simulations with a relative error of 1% or smaller.
Clarifying rear flow vortical structures of 2box vehicle model by using unsteady large-scale CFD
If we can control the vortex behaviors in boundary layer which affects the flow separation, we will be able to reduce drag coefficient (Cd) without changing the car shape. We analyzed in detail the flow all around a vehicle body by LES (Large Eddy Simulation). We succeed in clarifying the necessary precision of calculation for analyzing the vortical structures in the boundary layer. We also found an essential difference between the vortical structures that resolution high and low high and low aerodynamic drag.
Cavitating flow in a Francis turbine
The operation of a turbine at part load or at full load conditions changes the axial and tangential velocity distributions at the inlet of the draft tube, inducing the development of the well-known vortex rope. The volume of the vortex rope is changing over time resulting in a pressure variation that can be felt throughout the power plant. It is thus primordial to predict accurately the dynamical behavior of the vortex rope in order to avoid that the Eigen frequency of the vortex rope reach the Eigen frequency of the hydraulic system. Therefore, the challenge is to predict accurately the behavior of the flow field, which requires to take into account the cavitation phenomenon.
Mechanism of appearance and disappearance of vortices in a pump sump
Unsteady vortices in a model pump sump are investigated by performing large eddy simulations (LES). The model sump is composed of a 2,500 mm-long water channel of rectangular cross section of 300 mm (channel width) by 100 mm (water height) and a vertical suction pipe of a 100 mm diameter installed at its downstream end. At the upstream end of the channel, a uniform velocity of 0.37 m/s is given. The computational grids for the LES are composed approximately of 2 billion hexahedral elements with 0.255 mm maximum resolution. Although they are not sufficiently fine to capture the viscous cores of the submerged and air-entrained vortices, these grids can resolve the streamwise vortices in the approaching turbulent boundary layers that develop on the channel walls. The primary objectives of this study are two folded: to determine the origin of the suction vortices, and to clarify mechanism of appearance and disappearance of these vortices. By performing several cases of LES with different wall boundary conditions, we have first clarified the origin of the submerged vortices as well as air-entrained vortices. We then performed LES for a long time period of 16 seconds, and successfully captured unsteady events several times when submerged and air-entrained vortices appeared and disappeared. Detailed investigations of these events have revealed mechanism of the appearance and disappearance of the suction vortices.
Professor Kato Chisachi

Research fields
・Micro-energy conversion devices
・Prediction and control of unsteady
 turbulence and aerodynamic noise
・Numerical prediction of unsteady
 turbulence and its engineering application

Academic societies
・Japan Society of Mechanical Engineers
・Gas Turbine Society of Japan
・Turbo machinery Society of Japan
・The Japan Society of Fluid Mechanics
  〇Numerical simulation of unsteady fluid flows
    ・Fully Resolved Large Eddy Simulation as an Alternative to Towing Tank Resistance Tests
    ・Clarifying rear flow vortical structures of 2box vehicle model by using unsteady large-scale CFD
    ・Cabin noise prediction for an automobile
    ・Prediction of performance and aerodynamic noise of a centrifugal fan
    ・Numerical Investigation of Unsteady Flows and Particle Behavior in a Cyclone Separator
  〇Research on energy conversion systems
    ・Cavitating flow in a Francis turbine
    ・Mechanism of appearance and disappearance of vortices in a pump sump
    ・Large-Eddy Simulation of Flow around Magnus Wind Turbine with Spiral Fins
    ・Research of Ultra Micro Gas Turbines

  〇Journal papers
  〇Invited talkes
Awards and media
  〇Low-noise wind tunnel test facilities
  〇40-mm ultra micro gas turbine measuring facilities
  〇4-mm ultra micro gas turbine measuring facilities

  〇Staff, Research Fellow, Other