PDE-ODEsolver

A numerical solver for coupled PDE-ODE systems modelling biofilm growth

TO RUN

bash twoDimension_runall.sh FILENAME.txt

Note, that FILENAME.txt is the parameter file you use. You can copy the original parameter.txt and use that. The order of variables needs to be preserved.

OUTPUTTED

Running the bash script will print the parameters to the screen and then it will print the following after every nOuts timesteps:

time     avgIters     maxIter     avgNit     maxNit     avgM     avgC

Once fortran is done, the script will move all the output files into another folder called 'FILNAMEOutput'. This folder will contain the following:

| 2D_outXX.dat      | The solution at outputted time XX.        |
|                   | Format: x, M(x), C(x)                     |
|-------------------|-------------------------------------------|
| biomassGraphs.pdf | All of the 2D_outXX.dat's in a pdf.       |
|-------------------|-------------------------------------------|
| out.dat           | All of 2D_outXX.dat's in a single file    |
|                   | Format: x, M(x), C(x)                     |
|-------------------|-------------------------------------------|
| FILENAME.txt      | The parameter file used in simulation     |
|-------------------|-------------------------------------------|
| peakInfo.dat      | Data on the Max M value (x,y) and         |
|                   | interface location (furthest x where      |
|                   | M > 0.1)                                  |
|                   | Frm.: t, x at MaxM(x), Max M(x), interface|
|-------------------|-------------------------------------------|
| statReport.dat    | A file reporting computation time and     |
|                   | avg/max iters for lin solv and between    |
|                   | solution iterations                       |
|-------------------|-------------------------------------------|
| total.dat         | The average M and C at each time....      |
|                   | Formate: t, avg M(x), avg C(x)            | 

NOTES

Some parameter values make the problem too stiff to solve and computations begin to take a VERY long time.

Also there is not any measures for when the iterations between two solutions M and C fails to converge. So if the simulator gets stuck and doesn't move, its probably because the parameter set chosen results in a failed convergence (never documented this, so please do if it happens)