PISO - pressure-implicit split-operator

SIMPLE - semi-implicit method for pressure-linked equations.

PISO being used for transient problems and SIMPLE for steady-state.

Both algorithms are based on evaluating some initial solutions and then correcting them.

SIMPLE only makes 1 correction whereas PISO requires more than 1, but typically not more than
4.
The user must therefore specify the number of correctors in the PISO dictionary by the nCorrectors
(nCorrPISO).

An additional correction to account for mesh non-orthogonality is available in both SIMPLE and
PISO in the standard
OpenFOAM solver applications.
The number of non-orthogonal correctors is specified by nNonOrthogonalCorrectors.
The number of non-orthogonal correctors should correspond to the mesh for the case being solved,
i.e. 0 for an orthogonal
mesh and increasing with the degree of non-orthogonality up to, say, 20 for the most non-
orthogonal meshes.

/*..................................................................*/
PIMPLE Algorithm in OpenFOAM

nCorrPIMPLE = No of pressure-momentum coupling per time step.

nCorrPISO = No of pressure corrections per pressure-momentum coupling

1 time step = nCorrPIMPLE loops.

if nCorrPIMPLE <= 1 PIMPLE: Operating solver in PISO mode

1 nCorrPIMPLE loop = nCorrPISO pressure corrections.

Default values for pimple control

/*..................................................................*/
nOuterCorrectors (nCorrPIMPLE) is set by default to 1
nCorrectors (nCorrPISO) is set by default to 1
nNonOrthogonalCorrectors (corrPISO) is set by default to 0 (not shown here)
turbOnFinalIterOnly(turbulence is calculated at the end of every pimple iteration) is set by default
to true
Residual control is set to false.

We introduced the relaxation factors to obtain a smooth convergence.

The problem now is that we use the relaxation factors in the first pimple loop but after we reach the
last one
(nCorrPIMPLE) the relaxation factor will be 1 again to ensure time consistence.
Therefor nCorrPIMPLE-1 times with relaxation (like SIMPLE) and the last time without relaxation.

Observations
/*..................................................................*/
deltaT(could be set to very high) in controlDict has no impact on the simulations because it is
governed by maxCo when adjustableRunTime.

Set nCorrPIMPLE up to a very high number (~ 50 to 200) and control your pimple loop with the
residual control.
At the beginning of an transient simulation it could happen that after 100 pimple loops the solution
is still
not converged but this will change during the simulation time

Increasing the maxCo increases the no of nCorrPIMPLE.

Increasing nCorrPISO(1 to 3 is the best) increases the error in k and epsilon(For high turbulent flow
you should
add turbOnFinalIterOnly = false to your PIMPLE control)
If you set nCorrPISO higher it could be possible to leave the pimple loop earlier and prevent extra
calculation
but could also do the opposite.

The optimum relaxation factors depend on the simulation case you are running.
The converged solution is independed of the relaxation factors which are applied.

Due to the fact that you have bigger time steps it could happen that some phsics are not resolved
(extreme fast pressure drop etc.)

/***********************************************************************/
PIMPLE
{
momentumPredictor no;
nOuterCorrectors 50;
nCorrectors 1;
nNonOrthogonalCorrectors 0;

residualControl
{
U
{
tolerance 1e-5;
relTol 0;
}
p
{
tolerance 5e-4;
relTol 0;
}
}
}

relaxationFactors
{
fields
{
p 0.3;
pFinal 1;
 }
equations
{
"U|k|epsilon" 0.3;
"(U|k|epsilon)Final" 1;
"h|Y" 0.3;
"(h|Y)Final" 1;
"(T|R)" 0.3;
"(T|R)Final" 1;
}
}

Example
nNonOrthogonalCorrectors 2;
nCorrectors 4;
nOuterCorrectors 200;
/*..................................................................*/
PIMPLE: iteration 1
.
.
.
PIMPLE: iteration 3
smoothSolver: Solving for Ux, Initial residual = 8.25413e-06, Final residual = 8.25413e-06, No
Iterations 0
smoothSolver: Solving for Uy, Initial residual = 0.000140663, Final residual = 1.37053e-06, No
Iterations 1
GAMG: Solving for p, Initial residual = 0.149538, Final residual = 0.000555775, No Iterations 2
GAMG: Solving for p, Initial residual = 0.00122447, Final residual = 1.1619e-05, No Iterations 6
GAMG: Solving for p, Initial residual = 0.000120136, Final residual = 9.26922e-07, No Iterations
4
time step continuity errors : sum local = 3.08687e-09, global = -3.15106e-10, cumulative = -
9.93989e-09
GAMG: Solving for p, Initial residual = 0.100183, Final residual = 0.00036583, No Iterations 2
GAMG: Solving for p, Initial residual = 0.000843761, Final residual = 7.76257e-06, No Iterations
6
GAMG: Solving for p, Initial residual = 8.25385e-05, Final residual = 6.08004e-07, No Iterations 4
time step continuity errors : sum local = 2.02234e-09, global = -2.05507e-10, cumulative = -
1.01454e-08
GAMG: Solving for p, Initial residual = 0.100281, Final residual = 0.000367295, No Iterations 2
GAMG: Solving for p, Initial residual = 0.000846467, Final residual = 7.79512e-06, No Iterations
6
GAMG: Solving for p, Initial residual = 8.27594e-05, Final residual = 6.10367e-07, No Iterations 4
time step continuity errors : sum local = 2.02982e-09, global = -2.05435e-10, cumulative = -
1.03508e-08
GAMG: Solving for p, Initial residual = 0.100286, Final residual = 0.000367476, No Iterations 2
GAMG: Solving for p, Initial residual = 0.000846685, Final residual = 7.80175e-06, No Iterations
6
GAMG: Solving for p, Initial residual = 8.27854e-05, Final residual = 6.10995e-07, No Iterations 4
time step continuity errors : sum local = 2.03182e-09, global = -2.05206e-10, cumulative = -
1.0556e-08
PIMPLE: iteration 23
.
.
.
GAMG:
time step continuity errors :
smoothSolver: Solving for epsilon, Initial residual = 7.79974e-05, Final residual = 4.06576e-06,
No Iterations 2
smoothSolver: Solving for k, Initial residual = 3.58729e-05, Final residual = 6.01955e-06, No
Iterations 2
PIMPLE: converged in 23 iterations
ExecutionTime = 2.36 s ClockTime = 2 s

/***********************************************************************/
SIMPLE
{
nNonOrthogonalCorrectors 0;
rhoMin rhoMin [ 1 -3 0 0 0 ] 0.2;
rhoMax rhoMax [ 1 -3 0 0 0 ] 1.5;

/*
residualControl
{
p 1e-4;
U 1e-4;
h 1e-5;
"(k|epsilon|omega)" 1e-4;
Y 1e-5;
G 1e-5;
}
*/
}

relaxationFactors
{
fields
{
p 0.3;
rho 0.05;
}
equations
{
U 0.7;
"(k|epsilon)" 0.7;
h 0.5;
Y 0.5;
T 0.5;
"ILambda.*" 0.7;
G 0.7;
}
}