Wildkatze

Unleash the power of CFD by advanced customization through user algorithms

Wildkatze is a general purpose three-dimensional CFD software package with robust Finite Volume and Finite Difference solvers, preprocessing module, and Multiphysics models for a wide range of industrial problems. Flexibility and advanced solver customization through user algorithms have been given paramount importance in the development of Wildkatze. Solver provides access to almost all the variables and internal settings through object oriented C++ code, empowering the user to take control and customize the solver for specific requirements. The flexibility to set up different physics models, utilizing different schemes, in different regions of flow enhances the reproducibility of physical phenomena and quality of analysis.

Why Wildkatze – Prime capabilities

Innovative approach in setting up physics models

Wildkatze package features a wide range of physics models. In general, the flow field is composed of various physical phenomena and it is difficult to capture all of them with one physics model. Wildkatze has the option to divide the analysis region and set up different physics models. For example, it is possible to set up LES model in one region and k-omega turbulence model in the other region.

Flexibility in the selection of CFD schemes

When user selects a time stepping or gradient discretization scheme, generally in conventional solvers, the same scheme would be set for the entire simulation. In contrast, Wildkatze allows the user to select different time stepping and gradient discretization schemes for each physics model.

Customize the Finite Volume solver

Wildkatze provides the user access to almost all variables and internal solver settings through C++ coding. This flexible customization option empower the user to take control of the solver and customers can incorporate their own physics models into Wildkatze for conducting advanced research. User can also enhance the capabilities of existing physics models through Add Feature option in Wildkatze.

Immersed Boundary Method

Immersed Boundary model has been implemented in Wildkatze for solving transient and steady fluid-structure interaction problems like rotation, movement etc. The Immersed Region does not need to be moving, it could be stationary and marked once. This makes adding or removing objects from the simulation easy and also support various types of motions, which are difficult to simulate with other mesh motion techniques, like morphing or chimera grid methods. Fast and robust solid marking algorithm in Wildkatze allows the user to use this method without much penalty on simulation time.

Validation

Flow through the rotating blades is analyzed using MRF (Moving Reference Frame) model. MRF approach is a method used to define and calculate different coordinate systems for several regions of fluid flow that move differently, especially in rotating machinery. Unsteady problem in inertial frame of reference can be analyzed as steady problem with respect to moving frame. The result from steady-state analysis at rotational speed of 10,000 RPM shows that the relationship between the mass flow rate and iterations required for obtaining a converged solution is quite comparable to other commercial solvers.

Example of Solver customization

Application of shallow water equations

The solver has been customized for two-dimensional shallow water equations and conducted Tsunami simulation (Wave propagation model for water and other incompressible fluids). Shallow water equations are suitable to model waves with larger length scales compared to the water depth, where the horizontal flow velocity does not depend on water depth.

Analysis result

Water column with height of 1 m and width of 0.2 m from the water surface is taken as the initial interface, and the shape of the wave generated thereafter has been analyzed. Results over time shows that the simulation has reproduced the state of wave reflection from the boundary wall.

\begin{align} \frac{\partial h}{\partial t} + \frac{\partial (uh)}{\partial x} + \frac{\partial (vh)}{\partial y} = 0 \end{align} \begin{align} \frac{\partial (uh)}{\partial t} + \frac{\partial (u^2h + \frac{1}{2}gh^2)}{\partial x} + \frac{\partial (uvh)}{\partial y} = 0 \end{align} \begin{align} \frac{\partial (vh)}{\partial t} + \frac{\partial (uvh)}{\partial x} + \frac{\partial (v^2h + \frac{1}{2}gh^2)}{\partial y} = 0 \end{align}

t: Time x,y: 2D Coordinates h: Water level or depth u,v: Horizontal velocity components g: Gravitational acceleration

Case studies

Airflow around a bike rider

Simulated the airflow around the bike traveling at 30 km/h

Model: Immersed Boundary Method (Rotating tire)

Compressible flow around RAE 2822 airfoil

Conducted two-dimensional analysis of compressible flow around RAE 2822 airfoil.

Model: RANS Turbulence model – Spalart Allmaras

Freestream Mach number: 0.73

Angle of Attack: 2.79 degrees

Dam break simulation

Simulation of dam break situation has been carried out. Water at rest initially would start flowing down as the simulation progress and splash hitting the obstacle. Wildkatze has captured the splash quite well as shown in the image.

Model: Explicit VOF method

Courant number: 0.1

Physics models implemented in Wildkatze

Turbulence model Multiphase model RANS model k-epsilon k-omega Spalart-Allmaras LES model Smagorinsky Dynamic Smagorinsky Kinetic energy transport equation VOF model Upwind scheme CICSAM scheme HRIC scheme HRIC-U2 scheme HRIC-THINC scheme Algebraic drift flux model Boussinesq approximation CSS model CSF model ○ ○ ○
Pressure based flow model ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ Implicit method / Explicit method Coefficient schemes Body-force-weighted scheme First order upwind scheme Second order upwind scheme CUBISTA (Third order accuracy) Bounded central scheme ○

Comparison of VOF schemes

Bubble rise simulation has been carried out to validate the VOF schemes. Results from each scheme are compared on the basis of volume fraction. HRIC (High Resolution Interface Capturing) scheme combines downwind and upwind schemes, thereby possessing the compressibility characteristics of downwind scheme and stability of upwind scheme. HRIC-U2 and HRIC-THINC are modified HRICs, which captured air-liquid interface sharply.

Volume fraction Upwind CICSAM HRIC HRIC-U2 HRIC-THINC
0-1
0-0.1
0-0.01

Operating System requirements

OS: Red Hat Enterprise Linux, CentOS, Ubuntu
Install following prerequisites- GCC7.3, Open MPI1.10.3, Vtk7.1.1

Get Wildkatze

Customers who are interested in obtaining Wildkatze solver can send us a message by selecting the “Inquiry” option from drop down list. Please mention “Interested in Wildkatze” in title box. Company representative will send the application form and guide you through the process.