You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Both stochastic implicit depletion integrator: SICELI & SILEQI did not create statepoint file named openmc_simulation_n.h5 on each depletion step as usual (other time integrators). Here I show you my log file when using SICELI in comparison to CE/LI.
In general, SICELI and SILEQI only produce the first statepoint file at t=0 (n0), and after that, they didn't produce any statepoint/openmc_simulation_n file on the following depletion steps until they generated a statepoint file for the last burnup step.
one might benefit from these state point files when tallying something for their problem, i.e: neutron flux spectrum, reaction rate, photon flux for dose and fluence calculation, ...
Steps to Reproduce
since I am just comparing each depletion method provided by the openmc, then I only use a standard integrator command in my script.
I think it would be better to use jupyter notebook to show this phenomenon
cell 14:
PIintegrator = openmc.deplete.PredictorIntegrator(operator, time_steps, power, timestep_units='d')
SICELIintegrator = openmc.deplete.SICELIIntegrator(operator, time_steps, power, timestep_units='d', solver= 'cram48', n_steps=1)
SILEQIintegrator = openmc.deplete.SILEQIIntegrator(operator, time_steps, power, timestep_units='d', solver= 'cram48', n_steps=1)
CELIintegrator = openmc.deplete.CELIIntegrator(operator, time_steps, power, timestep_units='d', solver= 'cram48')
LEQIintegrator = openmc.deplete.LEQIIntegrator(operator, time_steps, power, timestep_units='d', solver= 'cram48')
for SICELI and SILEQI, I use n_steps=1 since I have tried the default 10 value and it is a long-run simulation.
the jupyter notebook and its results can be seen here depletion.zip
I also read a paper from 2023 that indicated a similar trend, as I quote Tanash (2023) - Selecting Burnup Algorithms in OpenMC Using the Calculated Benchmark of LEU Assembly and MOX Fuel S1063778823110431.pdf
"Burnup modeling was performed using only six integrators. The SI-CE/LI and SI-CE/QI integrators were not used because of the extensive time of computation (requiring more than 50 computational steps per burnup step with no results obtained)."
I think the no results obtained they mean is the statepoint_n files on each burnup step.
Thank you
Environment
Openmc 0.14.1-dev
The text was updated successfully, but these errors were encountered:
wahidluthfi
changed the title
depletion time integrator SICELI & SILEQI did not create statepoint file (openmc_simulation_n.h5) on each step
depletion time integrator SICELI & SILEQI did not create statepoint file (openmc_simulation_n.h5) on each depletion step
Jun 17, 2024
Bug Description
Both stochastic implicit depletion integrator: SICELI & SILEQI did not create statepoint file named openmc_simulation_n.h5 on each depletion step as usual (other time integrators). Here I show you my log file when using SICELI in comparison to CE/LI.
In general, SICELI and SILEQI only produce the first statepoint file at t=0 (n0), and after that, they didn't produce any statepoint/openmc_simulation_n file on the following depletion steps until they generated a statepoint file for the last burnup step.
one might benefit from these state point files when tallying something for their problem, i.e: neutron flux spectrum, reaction rate, photon flux for dose and fluence calculation, ...
Steps to Reproduce
since I am just comparing each depletion method provided by the openmc, then I only use a standard integrator command in my script.
I think it would be better to use jupyter notebook to show this phenomenon
cell 14:
PIintegrator = openmc.deplete.PredictorIntegrator(operator, time_steps, power, timestep_units='d')
SICELIintegrator = openmc.deplete.SICELIIntegrator(operator, time_steps, power, timestep_units='d', solver= 'cram48', n_steps=1)
SILEQIintegrator = openmc.deplete.SILEQIIntegrator(operator, time_steps, power, timestep_units='d', solver= 'cram48', n_steps=1)
CELIintegrator = openmc.deplete.CELIIntegrator(operator, time_steps, power, timestep_units='d', solver= 'cram48')
LEQIintegrator = openmc.deplete.LEQIIntegrator(operator, time_steps, power, timestep_units='d', solver= 'cram48')
for SICELI and SILEQI, I use n_steps=1 since I have tried the default 10 value and it is a long-run simulation.
the jupyter notebook and its results can be seen here
depletion.zip
I also read a paper from 2023 that indicated a similar trend, as I quote
Tanash (2023) - Selecting Burnup Algorithms in OpenMC Using the Calculated Benchmark of LEU Assembly and MOX Fuel S1063778823110431.pdf
"Burnup modeling was performed using only six integrators. The SI-CE/LI and SI-CE/QI integrators were not used because of the extensive time of computation (requiring more than 50 computational steps per burnup step with no results obtained)."
I think the no results obtained they mean is the statepoint_n files on each burnup step.
Thank you
Environment
Openmc 0.14.1-dev
The text was updated successfully, but these errors were encountered: