How to create restart files

[studdrud]

Hi elmer team,
i want to do some transient computations and i have to use the restart function. Does an example existe, which shows the use of this funktion?
If not, can someone explain me how to use this, please. I'am using ElmerGUI. Is there somewhere an option, with which i can set the restart?

regards
rn

[studdrud]

ok i created restart files with:
Output File = "case.result"
and then restarted from a timestep with:
Restart File = case.result
Restart Position = 2.
I computed a beam with a bc on the left side like this:

Boundary Condition 1
Target Boundaries(1) = 5
Name = "wall"
Displacement 3 = 0
Displacement 2 = 0
Displacement 1 = 0
End

and an initial condition on the right side:

Initial Condition 1
Name = "InitialCondition 1"
Displacement 1 = 0.01
End

I computed the first 10 Timesteps with a dt=0.001 and wrote the restart file. Then i restarted the computation with the result fil, i've written,
at the time of 0.002. I've compared both results: first from the first computation and then from the restart. The values for the displacement were different
in the both cases. What is the fault? Here my .sif:

Header
CHECK KEYWORDS Warn
Mesh DB "." "."
Include Path ""
Results Directory ""
End

Simulation
Max Output Level = 4
Coordinate System = Cartesian
Coordinate Mapping(3) = 1 2 3
Simulation Type = Transient
Steady State Max Iterations = 1
Output Intervals = 1
Timestepping Method = BDF
BDF Order = 2
Timestep intervals = 500000
Timestep Sizes = 0.000001
Solver Input File = case.sif
Post File = case.ep
Output File = "case.result"
!Restart File = case.result
!Restart Position = 1
End

Constants
Gravity(4) = 0 -1 0 9.82
Stefan Boltzmann = 5.67e-08
Permittivity of Vacuum = 8.8542e-12
Boltzmann Constant = 1.3807e-23
Unit Charge = 1.602e-19
End

Body 1
Target Bodies(1) = 1
Name = "Body 1"
Equation = 1
Material = 1
Initial condition = 1
End

Solver 2
Equation = Result Output
Procedure = "ResultOutputSolve" "ResultOutputSolver"
Output File Name = case
Output Format = Vtk
End

Solver 1
Equation = Linear elasticity
Procedure = "StressSolve" "StressSolver"
Variable = -dofs 3 Displacement
Exec Solver = Always
Stabilize = True
Bubbles = False
Lumped Mass Matrix = False
Optimize Bandwidth = True
Steady State Convergence Tolerance = 1.0e-5
Nonlinear System Convergence Tolerance = 1.0e-8
Nonlinear System Max Iterations = 20
Nonlinear System Newton After Iterations = 3
Nonlinear System Newton After Tolerance = 1.0e-3
Nonlinear System Relaxation Factor = 1
Linear System Solver = Iterative
Linear System Iterative Method = BiCGStab
Linear System Max Iterations = 500
Linear System Convergence Tolerance = 1.0e-8
Linear System Preconditioning = ILU0
Linear System ILUT Tolerance = 1.0e-3
Linear System Abort Not Converged = False
Linear System Residual Output = 1
Linear System Precondition Recompute = 1
Time Derivative Order = 2
End

Equation 1
Name = "Equation 1"
Calculate Stresses = True
Active Solvers(2) = 2 1
End

Material 1
Name = "Steel (alloy - generic)"
Heat Conductivity = 37.2
Youngs modulus = 200.0e9
Mesh Poisson ratio = 0.285
Heat Capacity = 976.0
Density = 7850.0
Poisson ratio = 0.285
Sound speed = 5100.0
Heat expansion Coefficient = 12.0e-6
End

Initial Condition 1
Name = "InitialCondition 1"
Displacement 1 = 0.01
End

Boundary Condition 1
Target Boundaries(1) = 5
Name = "wall"
Displacement 3 = 0
Displacement 2 = 0
Displacement 1 = 0
End

regards rn

[raback]

At least one source for a difference could be that the restart includes just the one instance. Hence when using 2nd order timestepping it will force 1st order scheme at restart. Maybe you could test whether you have exact correspondence with BDF order 1.

-Peter

[studdrud]

Ok i tried with BDF Order 1: i computed first the deformation without restarting:
Timestep intervals = 10
Timestep Sizes = 0.001
Output File = "case.result"
!Restart File = case.result
!Restart Position = 2

and then i started a computation with the restart (the same timestep size):

Header
CHECK KEYWORDS Warn
Mesh DB "." "."
Include Path ""
Results Directory ""
End

Simulation
Max Output Level = 4
Coordinate System = Cartesian
Coordinate Mapping(3) = 1 2 3
Simulation Type = Transient
Steady State Max Iterations = 1
Output Intervals = 1
Timestepping Method = BDF
BDF Order = 1
Timestep intervals = 10
Timestep Sizes = 0.001
Solver Input File = case.sif
Post File = case.ep
Output File = "case_1.result"
Restart File = case.result
Restart Position = 2
End

Constants
Gravity(4) = 0 -1 0 9.82
Stefan Boltzmann = 5.67e-08
Permittivity of Vacuum = 8.8542e-12
Boltzmann Constant = 1.3807e-23
Unit Charge = 1.602e-19
End

Body 1
Target Bodies(1) = 1
Name = "Body 1"
Equation = 1
Material = 1
End

Solver 2
Equation = Result Output
Procedure = "ResultOutputSolve" "ResultOutputSolver"
Output File Name = case
Output Format = Vtk
End

Solver 1
Equation = Linear elasticity
Procedure = "StressSolve" "StressSolver"
Variable = -dofs 3 Displacement
Exec Solver = Always
Stabilize = True
Bubbles = False
Lumped Mass Matrix = False
Optimize Bandwidth = True
Steady State Convergence Tolerance = 1.0e-5
Nonlinear System Convergence Tolerance = 1.0e-8
Nonlinear System Max Iterations = 200
Nonlinear System Newton After Iterations = 3
Nonlinear System Newton After Tolerance = 1.0e-3
Nonlinear System Relaxation Factor = 1
Linear System Solver = Iterative
Linear System Iterative Method = BiCGStab
Linear System Max Iterations = 500
Linear System Convergence Tolerance = 1.0e-8
Linear System Preconditioning = ILU0
Linear System ILUT Tolerance = 1.0e-3
Linear System Abort Not Converged = False
Linear System Residual Output = 1
Linear System Precondition Recompute = 1
Time Derivative Order = 2
optimize bandwidth = False
End

Equation 1
Name = "Equation 1"
Calculate Stresses = True
Active Solvers(2) = 2 1
End

Material 1
Name = "Steel (alloy - generic)"
Heat Conductivity = 37.2
Youngs modulus = 200.0e9
Mesh Poisson ratio = 0.285
Heat Capacity = 976.0
Density = 7850.0
Poisson ratio = 0.285
Sound speed = 5100.0
Heat expansion Coefficient = 12.0e-6
End

Boundary Condition 1
Target Boundaries(1) = 3
Name = "force"
Force 2 = 250000
End

Boundary Condition 2
Target Boundaries(1) = 5
Name = "wall"
Displacement 3 = 0
Displacement 2 = 0
Displacement 1 = 0
End

and get this here in the case.result, the first computation:
Time: 3 3 0.300000000000E-002
displacement 1
Perm: 179 179
1 1
2 2
3 3
4 4
5 5
...
177 177
178 178
179 179
-6.28663559146706865E-074
1.54672114805168850E-005
-1.53459693136461363E-005
-6.28663559146706865E-074
-6.28663559146706865E-074
-1.53571297335832839E-005
1.54458313151421980E-005
-6.28663559146706865E-074
-3.44764042025630022E-006
-6.42809396984468832E-006
-9.18980443138580843E-006
-1.11867439297244321E-005
-1.26908776083000292E-005
-1.40778220321594874E-005
-1.48297661527703205E-005
-1.51886249759363248E-005
-1.87908601104894613E-006
-4.89734963734353630E-006
...

and in the case_1.result there is:
Time: 1 1 0.300000000000E-002
displacement 1
Perm: 179 179
1 1
2 2
3 3
4 4
5 5
6 6
...
176 176
177 177
178 178
179 179
0.0000000000000000
8.88951211590773745E-006
-8.80837622027232527E-006
0.0000000000000000
0.0000000000000000
-8.83746455822413260E-006
8.85053852177481628E-006
0.0000000000000000
-1.58022694914263818E-006
-3.05174433867506328E-006
-4.52955368637457124E-006
-5.70158821635011704E-006
-6.70075506307938182E-006
-7.71372170159645116E-006
-8.32876659783153695E-006
-8.65734723576483902E-006
-8.48597368262170670E-007
-2.28120728666105694E-006
-3.81654889245630651E-006
-5.14465071532359973E-006
-6.18709254452092831E-006
...

any ideas?

[studdrud]

ok, i uploaded the sif file and the Gmsh-file. You can try it yourself and tell me then.

regards rn

[mulvrova]

hello,

i have experienced some difficulties with restart as well, but it still remains unresolved.
see the old elmer forum https://postit.csc.fi/sympa/arc/elmerdi ... 00000.html

regards, Martina.

[studdrud]

hello,
i've thought about that problem and maybe i've a solution. Is it possible that the velocity of the structure is forgotten in the restart?
How can i write the velocity in the result file as variable? Maybe then it will compute the right values?

regars rn

[studdrud]

to write out the velocity look at this thread "velocity of nodes in the linear elasticity".
now i will try the restart with this information, if i have time.
regrads rn

[law]

If anyone here is trying to use restart files, I couldn't get it working for elasticity either. However I noticed in the elmer models manual a possible alternative called "reloaddata" on page 160. It says:

This module offers extended features compared to the Restart File option in the Simulation
section. The module reads a new solution step from the solution file on each timestep, whereas the restart
file option reads only the initial state for a simulation

I might try it some time if I can get it working.

[stoykov]

Hello,

I think that for second order problems, like the Linear Elasticity problem, the displacement and the velocity are necessary to continue from existing solution. Thus the vector of velocity has to be written in the Restart File, i.e. one should use "Calculate Velocity = Logical True" in the solver section.

Still there is a problem, that I do not obtain exactly the same solution and there should be a way to do it. I mean that I start the transient response and I save it using Output File = "case.dat". Than I start it, let say from Restart Position = 3. So the first solution has to be equal to the solution saved on position 4, but it is not. Maybe I miss something small. Here is the sif file. Can someone help me?

The problem which I solve is the transient response of a clamped-free beam excited by a periodic force on the free end.

Code: Select all

Header
  CHECK KEYWORDS Warn
  Mesh DB "." "."
  Include Path ""
  Results Directory ""
End

Simulation
  Max Output Level = 4
  Coordinate System = Cartesian
  Coordinate Mapping(3) = 1 2 3
  Simulation Type = Transient
  Steady State Max Iterations = 1
  Output Intervals = 1
  Timestepping Method = BDF
  BDF Order = 1
  Timestep intervals = 10
  Timestep Sizes = 0.001
  Solver Input File = case.sif
  Post File = case.ep
  Output File = "case.dat"

!  Restart File = "case.dat"
!  Restart Position = 3
End

Body 1
  Target Bodies(1) = 1
  Name = "Body 1"
  Equation = 1
  Material = 1
End

Solver 1
  Equation = Linear elasticity
  Variable = -dofs 3 Displacement
  Exec Solver = Always
  Stabilize = True
  Bubbles = False
  Lumped Mass Matrix = False
  Optimize Bandwidth = True
  Steady State Convergence Tolerance = 1.0e-8

  Linear System Solver = Direct
  Linear System Direct Method = Umfpack
  Linear System Convergence Tolerance = 1.0e-8

  Time Derivative Order = 2

  Calculate Velocity = Logical True
End


Equation 1
  Name = "eqs"
  Calculate Stresses = True
  Active Solvers(1) = 1
End

Material 1
  Name = "alm"
  Density = 2778
  Poisson ratio = 0.34
  Youngs modulus = 7e10
End

Boundary Condition 1
  Target Boundaries(1) = 2
  Name = "mass"
  Force 3 = Variable time
	Real MATC "-1.25e5*cos(30.0*tx)"
End

Boundary Condition 2
  Target Boundaries(1) = 4
  Name = "wall"
  Displacement 3 = 0
  Displacement 2 = 0
  Displacement 1 = 0
End

Regards,
Stan