Dirichlet boundary condition within a time interval #2272
Conversation
wenqing
force-pushed
the
time_dependent_D_BC
branch
from
1c12b93 to
7fc0b17
Compare
NumLib/TimeStepping/TimeInterval.h
Outdated
| * Class for a time interval, which has a member to check whether the given time | ||
| * is in this time interval. | ||
| */ | ||
| class TimeInterval |
NumLib/TimeStepping/TimeInterval.h
Outdated
| { | ||
| return (current_time >= _start_time && current_time <= _end_time) | ||
| ? true | ||
| : false; |
There was a problem hiding this comment.
The result of comparison is already a boolean, I'd drop the true:false
NumLib/TimeStepping/TimeInterval.h
Outdated
| { | ||
| } | ||
|
|
||
| bool isInThisTimeInterval(const double current_time) const |
There was a problem hiding this comment.
Maybe isActive, s.t.
TimeInterval time_interval{start, end};
...
if (time_interval.isActive(t)) { ... }There was a problem hiding this comment.
changed as if (time_interval.contains(t)).
| { | ||
| if (_time_interval->isInThisTimeInterval(t)) | ||
| { | ||
| return getEssentialBCValuesLocal(t, x, bc_values); |
There was a problem hiding this comment.
The return type of the function is void...
|
|
||
| const double end_time = | ||
| //! \ogs_file_param{prj__process_variables__process_variable__boundary_conditions__boundary_condition__DirichletWithinTimeInterval__end_time} | ||
| config.getConfigParameter<double>("end_time"); |
There was a problem hiding this comment.
Just a suggestion: I'd rather extract the config parsing into createTimeInterval() and have the <time_interval> tag in the project file:
...
<time_interval>
<start>0</start>
<end>1</end>
</time_interval>Because parsing of the time interval is not a task for a boundary condition, in my opinion.
NumLib/TimeStepping/TimeInterval.h
Outdated
| */ | ||
| #pragma once | ||
|
|
||
| namespace NumLib |
There was a problem hiding this comment.
Why is it in NumLib? I would put it in BaseLib.
There was a problem hiding this comment.
At beginning of my implementation, I had put it in BaseLib. Later on, I thought that this class is related to the time stepping only. Therefore I moved it to NumLib/TimeStepping. If you still think BaseLib is the best choice for it after reading this reply, I will move it BaseLib again.
There was a problem hiding this comment.
Could be BaseLib because it's rather general, or ProcessLib/BC/ because this is the only place where it is used now.
There was a problem hiding this comment.
Since both of you prefer to move to BaseLib, I have moved it to there but kept the path of parameter documentation.
|
@endJunction Introduced two functions to avoid class inheritance while to avoid source code duplication. |
…r reducing source code duplication.
wenqing
force-pushed
the
time_dependent_D_BC
branch
from
9a4459e to
deac805
Compare
wenqing
force-pushed
the
time_dependent_D_BC
branch
from
035110b to
96ded07
Compare
|
OpenGeoSys development has been moved to GitLab. |
This PR presents a Dirichlet boundary condition within a time interval.
The benchmark is about a square domain under liquid pressure on its top and bottom. On the top boundary, pressure is always 1.0e+5, while on the bottom boundary, pressure is 1.0e+6 within a time interval of [0, 10] (see the following figure):
