CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   SU2 (https://www.cfd-online.com/Forums/su2/)
-   -   SU2 Divergence Problem (https://www.cfd-online.com/Forums/su2/215527-su2-divergence-problem.html)

Gui_AP March 7, 2019 12:56

SU2 Divergence Problem
 
Hello there!

I want to start to use SU2 for my CFD projects.
I've intalled and ran the test cases without problem, but when I try to run my own case (I designed a 2D cylinder and meshed using Gmsh), with an adapted configuration file, the code is diverging.

I tried to reduce the CFL number, opting to use 2nd equations and etc. but none of these resolved.


This is the error that I got:

-------------------------------------------------------------------------
| ___ _ _ ___ |
| / __| | | |_ ) Release 6.2.0 "Falcon" |
| \__ \ |_| |/ / |
| |___/\___//___| Suite (Computational Fluid Dynamics Code) |
| |
-------------------------------------------------------------------------
| The current SU2 release has been coordinated by the |
| SU2 International Developers Society <www.su2devsociety.org> |
| with selected contributions from the open-source community. |
-------------------------------------------------------------------------
| The main research teams contributing to the current release are: |
| - Prof. Juan J. Alonso's group at Stanford University. |
| - Prof. Piero Colonna's group at Delft University of Technology. |
| - Prof. Nicolas R. Gauger's group at Kaiserslautern U. of Technology. |
| - Prof. Alberto Guardone's group at Polytechnic University of Milan. |
| - Prof. Rafael Palacios' group at Imperial College London. |
| - Prof. Vincent Terrapon's group at the University of Liege. |
| - Prof. Edwin van der Weide's group at the University of Twente. |
| - Lab. of New Concepts in Aeronautics at Tech. Inst. of Aeronautics. |
-------------------------------------------------------------------------
| Copyright 2012-2019, Francisco D. Palacios, Thomas D. Economon, |
| Tim Albring, and the SU2 contributors. |
| |
| SU2 is free software; you can redistribute it and/or |
| modify it under the terms of the GNU Lesser General Public |
| License as published by the Free Software Foundation; either |
| version 2.1 of the License, or (at your option) any later version. |
| |
| SU2 is distributed in the hope that it will be useful, |
| but WITHOUT ANY WARRANTY; without even the implied warranty of |
| MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| Lesser General Public License for more details. |
| |
| You should have received a copy of the GNU Lesser General Public |
| License along with SU2. If not, see <http://www.gnu.org/licenses/>. |
-------------------------------------------------------------------------

------------------------ Physical Case Definition -----------------------
Incompressible Laminar Navier-Stokes' equations.
No restart solution, use the values at infinity (freestream).
Dimensional simulation.
The reference area is 1 m^2.
The semi-span will be computed using the max y(3D) value.
The reference length is 1 m.
Reference origin for moment evaluation is (0.25, 0, 0).
Surface(s) where the force coefficients are evaluated: cylinder.

Surface(s) plotted in the output file: cylinder.
Input mesh file name: mesh_cylinder_lam.su2

---------------------- Space Numerical Integration ----------------------
Flux difference splitting (FDS) upwind scheme for the flow inviscid terms.
Second order integration in space, with slope limiter.
No slope-limiting method.
Average of gradients with correction (viscous flow terms).
Gradient computation using Green-Gauss theorem.

---------------------- Time Numerical Integration -----------------------
Local time stepping (steady state simulation).
Euler implicit method for the flow equations.
FGMRES is used for solving the linear system.
Using a ILU(0) preconditioning.
Convergence criteria of the linear solver: 1e-15.
Max number of linear iterations: 10.
No CFL adaptation.
Courant-Friedrichs-Lewy number: 1e+03

------------------------- Convergence Criteria --------------------------
Maximum number of iterations: 1000.
Reduce the density residual 10 orders of magnitude.
The minimum bound for the density residual is 10^(-14).
Start convergence criteria at iteration 10.

-------------------------- Output Information ---------------------------
Writing a solution file every 250 iterations.
Writing the convergence history every 1 iterations.
The output file format is Paraview ASCII legacy (.vtk).
Convergence history file name: history.
Forces breakdown file name: forces_breakdown.dat.
Surface flow coefficients file name: surface_flow.
Flow variables file name: flow.
Restart flow file name: restart_flow.dat.

------------------- Config File Boundary Information --------------------
+-----------------------------------------+
| Marker Type| Marker Name|
+-----------------------------------------+
| Far-field| farfield|
+-----------------------------------------+
| Isothermal wall| cylinder|
+-----------------------------------------+

---------------------- Read Grid File Information -----------------------
Two dimensional problem.
267712 points.
2 surface markers.
+------------------------------------+
| Index| Marker| Elements|
+------------------------------------+
| 0| farfield| 768|
| 1| cylinder| 128|
+------------------------------------+
534528 triangles.

------------------------- Geometry Preprocessing ------------------------
Setting point connectivity.
Renumbering points (Reverse Cuthill McKee Ordering).
Recomputing point connectivity.
Setting element connectivity.
Checking the numerical grid orientation.
Identifying edges and vertices.
Computing centers of gravity.
Setting the control volume structure.
Area of the computational grid: 0.
Finding max control volume width.
Searching for the closest normal neighbors to the surfaces.
Storing a mapping from global to local point index.
Compute the surface curvature.
Max K: inf. Mean K: inf. Standard deviation K: -nan.
Checking for periodicity.
Wetted area = 0.4 m.
Area projection in the x-plane = 0 m, y-plane = 0.2 m.
Max. coordinate in the x-direction = 0.1 m, y-direction = 0 m.
Min coordinate in the x-direction = -0.1 m, y-direction = 0 m.

------------------------- Driver information --------------------------
A Fluid driver has been instantiated.

------------------------ Iteration Preprocessing ------------------------
Zone 1: Euler/Navier-Stokes/RANS fluid iteration.

------------------------- Solver Preprocessing --------------------------
Incompressible flow: rho_ref, vel_ref, temp_ref, p_ref
are set to 1.0 in order to perform a dimensional calculation.
Force coefficients computed using initial values.
The reference area for force coeffs. is 1 m^2.
The reference length for force coeffs. is 1 m.
The pressure is decomposed into thermodynamic and dynamic components.
The initial value of the dynamic pressure is 0.
Mach number: 0, computed using fluid speed of sound.
For external flows, the initial state is imposed at the far-field.
Angle of attack (deg): 0, computed using the initial velocity.
Side slip angle (deg): 0, computed using the initial velocity.
Reynolds number per meter: 34.4518, computed using initial values.
Reynolds number is a byproduct of inputs only (not used internally).
SI units only. The grid should be dimensional (meters).
Energy equation is active and coupled for variable density.

-- Models:
+------------------------------------------------------------------------------+
| Viscosity Model| Conductivity Model| Fluid Model|
+------------------------------------------------------------------------------+
| POLYNOMIAL_VISCOSITY| POLYNOMIAL_CONDUCTIVITY| INC_IDEAL_GAS_POLY|
+------------------------------------------------------------------------------+
-- Fluid properties:
+------------------------------------------------------------------------------+
| Name| Dim. value| Ref. value| Unit|Non-dim. value|
+------------------------------------------------------------------------------+
| Mu(T) Poly. Coeff. 0| 1.7893e-05| 1| -| 1.7893e-05|
| Mu(T) Poly. Coeff. 1| 1e-08| 1| -| 1e-08|
+------------------------------------------------------------------------------+
| Kt(T) Poly. Coeff. 0| 0.0257| 1| -| 0.0257|
| Kt(T) Poly. Coeff. 1| 1e-05| 1| -| 1e-05|
+------------------------------------------------------------------------------+
| Molecular weight| 28.96| 1| g/mol| 28.96|
| Gas Constant| 287.102| 1| N.m/kg.K| 287.102|
| Therm. Pressure| 17.3996| 1| Pa| 17.3996|
| Cp(T) Poly. Coeff. 0| 1004.7| 1| -| 1004.7|
| Cp(T) Poly. Coeff. 1| 0.1| 1| -| 0.1|
+------------------------------------------------------------------------------+
-- Initial and free-stream conditions:
+------------------------------------------------------------------------------+
| Name| Dim. value| Ref. value| Unit|Non-dim. value|
+------------------------------------------------------------------------------+
| Dynamic Pressure| 0| 1| Pa| 0|
| Total Pressure| 0.00121779| 1| Pa| 0.00121779|
| Density| 0.000210322| 1| kg/m^3| 0.000210322|
| Temperature| 288.15| 1| K| 288.15|
| Velocity-X| 3.40297| 1| m/s| 3.40297|
| Velocity-Y| 0| 1| m/s| 0|
| Velocity Magnitude| 3.40297| 1| m/s| 3.40297|
+------------------------------------------------------------------------------+
| Viscosity| 2.07745e-05| 1| N.s/m^2| 2.07745e-05|
| Conductivity| -| 1| W/m^2.K| -|
+------------------------------------------------------------------------------+
| Mach Number| -| -| -| 0|
| Reynolds Number| -| -| -| 34.4518|
+------------------------------------------------------------------------------+
Initialize Jacobian structure (Navier-Stokes). MG level: 0.

----------------- Integration and Numerics Preprocessing ----------------
Integration Preprocessing.
Numerics Preprocessing.

---------------------- Python Interface Preprocessing ---------------------
Setting customized boundary conditions for zone 0

------------------------------ Begin Solver -----------------------------


Error in "void CSysSolve::ModGramSchmidt(int, std::vector<std::vector<double> >&, std::vector<CSysVector>&)":
-------------------------------------------------------------------------
SU2 has diverged.
------------------------------ Error Exit -------------------------------

wallym May 10, 2019 11:04

Hi,

Thank you for your interest in SU2 and reaching out. I am sorry you are having divergence issues. I am not super familiar with the incompressible solver, however, it seems to me that something is not being properly initialized from the config file (not sure which one). Unfortunately, I cant answer immediately.

If you could post your config file and mesh, I could take a look at it.

angupraveen September 17, 2019 04:19

have you found a solution yet?

MrCrow October 3, 2019 19:25

Did you found a solution for this problem? I have the same error.

akoodlyr October 7, 2019 04:26

Hey,

As Wally suggested it would help if you post the config files or the mesh files here.

As a general suggestion, one option would be to check if MGLEVELS is set to any value other than 0 and set it to 0 if it is. Otherwise perhaps try to reduce your CFL numbers or use a first order scheme(by setting MUSCL_FLOW=NO) and see if you get any results at all. But if you can upload you config file and mesh file, it would be easier to debug.

Cheers,
Akshay

jomunkas February 21, 2020 12:10

If I increase the number of multi grid levels MGLEVELS, what does it change in the calculation?


All times are GMT -4. The time now is 07:25.