diff --git a/Usecases/README.md b/Usecases/README.md index 694e224245b2f2b979dbecd249d972890cc9d45e..45cb50400df26476cd2e2cf474c27d338fc11ac0 100644 --- a/Usecases/README.md +++ b/Usecases/README.md @@ -133,11 +133,44 @@ In detail: might be an issue depending on your machine. >>> -- `starttimes = {0: 0.0}` starttimes gives a list of starttimes to run the simulation from. - The list is looped over and a simulation is run with t_0 as initial time - for each element t_0 in starttimes -- timestep_size = 0.001 -- number_of_timesteps = 20 + - `starttimes = {0: 0.0}`: `starttimes` is a dictionary containing pairs + `t0_index: t0` specifying starttimes `t0` along with the number this start time should be given `t0_index`. Usually you will want to have `t0_index` + set to zero because `t0` is the intial timestep. + However, if the simulation stopped for some reason and part of the data is + valid, a later time can be specified along with the index of the timestep + it had in the first attempt of the simulation. + >>> + **Example:** Say you wanted start + the simulation at `t0 = 0.5` and let that be the 87th timestep, you would set + `starttimes = {87: 0.5}`. + >>> + + In case more than one element is given parallel simulations are started starting from the specified starttimes. + >>> + **Example:** + Assume `timestep_size = 0.01` and `number_of_timesteps = 50` + `starttimes = { + 0: 0.0, + 50: 0.5 + }` + yields a simulation over [0.0,1.0] but split into to processes and saved into + two different folders. + >>> + This is also usefull to test parameters if `number_of_timesteps = 1` and + the behaviour of the solver want to be tested at various timesteps one + could set up an example as + ~~~python + starttimes = { + 0: 0.0, + 1: 0.5, + 2: 1.0 + 3: 1.5 + } + ~~~ + to test the behaviour of the solver at different times. + +- `timestep_size`: Set size of the timestep. +- `number_of_timesteps`: Set numbers of timesteps to calculate. ### LDD SCHEME PARAMATERS ~~~python