Two-Tank example.
More...
Two-Tank example.
README
TWO TANKS EXAMPLE
The discharge of air from one tank into another through a nozzle. The problem only makes sense, if you think of the flow through the nozzle as a much more complex problem which you have to solve with - let's say 3D CFD. So the two tanks are the macroscopic and the nozzle is the microscopic problem. This is also a backward mapping problem if you assume that the range of inputs is unknown a-priori. twoTanksMacroscopicProblem uses the MoDeNa interface library to embed an even simpler model for the flow rate. While twoTanksFullProblem implements it fully integrated.TODO: The example specific sources should be moved into the example, but this requires building (and finding) the execuables here.
How to run?
Make sure PYTHONPATH and LD_LIBRARY_PATH are set
TODO:
Make this easier to use
export PKG_CONFIG_PATH=${PKG_CONFIG_PATH:-}:${HOME}/lib/pkgconfig:/usr/local/lib/pkgconfig
export PYTHONPATH=${PYTHONPATH:-}:${HOME}/lib/python2.7/site-packages
export LD_LIBRARY_PATH=${LD_LIBRARY_PATH:-}:${HOME}/lib/python2.7/site-packages:${HOME}/lib:/usr/local/lib
Compile project specific sources, i.e. "models":
flowRate="../models/flowRate/src" twoTank="../models/twoTank/src" cmake -H${flowRate} -B${flowRate} && make –directory=${flowRate} cmake -H${twoTank} -B${twoTank} && make –directory${twoTank}
Initialise the model in the database
./initModel
Start the workflow
./workflow
Run again to see that no fitting is done on the second start
./workflow
README
TWO TANKS EXAMPLE:
The discharge of air from one tank into another through a nozzle. The problem only makes sense, if you think of the flow through the nozzle as a much more complex problem which you have to solve with - let's say 3D CFD. So the two tanks are the macroscopic and the nozzle is the microscopic problem. This is also a backward mapping problem if you assume that the range of inputs is unknown a-priori. twoTanksMacroscopicProblem uses the MoDeNa interface library to embed an even simpler model for the flow rate. While twoTanksFullProblem implements it fully integrated.TODO: The example specific sources should be moved into the example, but this requires building (and finding) the execuables here.
How to run?
Make sure PYTHONPATH
and LD_LIBRARY_PATH
are set
TODO:
Make this easier to use
export PKG_CONFIG_PATH=${PKG_CONFIG_PATH:-}:${HOME}/lib/pkgconfig:/usr/local/lib/pkgconfig
export PYTHONPATH=${PYTHONPATH:-}:${HOME}/lib/python2.7/site-packages
export LD_LIBRARY_PATH=${LD_LIBRARY_PATH:-}:${HOME}/lib/python2.7/site-packages:${HOME}/lib:/usr/local/lib
Compile project specific sources, i.e. "models":
flowRate="../models/flowRate/src"
twoTank="../models/twoTank/src"
cmake -H${flowRate} -B${flowRate} && make --directory=${flowRate}
cmake -H${twoTank} -B${twoTank} && make --directory${twoTank}
Initialise the model in the database
./initModel
Start the workflow
./workflow
Run again to see that no fitting is done on the second start
./workflow
README
TWO TANKS EXAMPLE:
The discharge of air from one tank into another through a nozzle. The problem only makes sense, if you think of the flow through the nozzle as a much more complex problem which you have to solve with - let's say 3D CFD. So the two tanks are the macroscopic and the nozzle is the microscopic problem. This is also a backward mapping problem if you assume that the range of inputs is unknown a-priori. twoTanksMacroscopicProblem uses the MoDeNa interface library to embed an even simpler model for the flow rate. While twoTanksFullProblem implements it fully integrated.TODO: The example specific sources should be moved into the example, but this requires building (and finding) the execuables here.
How to run?
Make sure PYTHONPATH
and LD_LIBRARY_PATH
are set
TODO:
Make this easier to use
export PKG_CONFIG_PATH=${PKG_CONFIG_PATH:-}:${HOME}/lib/pkgconfig:/usr/local/lib/pkgconfig
export PYTHONPATH=${PYTHONPATH:-}:${HOME}/lib/python2.7/site-packages
export LD_LIBRARY_PATH=${LD_LIBRARY_PATH:-}:${HOME}/lib/python2.7/site-packages:${HOME}/lib:/usr/local/lib
Compile project specific sources, i.e. "models":
flowRate="../models/flowRate_idealGas/src"
twoTank="../models/twoTank/src"
cmake -H${flowRate} -B${flowRate} && make --directory=${flowRate}
cmake -H${twoTank} -B${twoTank} && make --directory${twoTank}
Append path to the models directory to "PYTHONPATH":
export PYTHONPATH="${PYTHONPATH}:../models"
Initialise the model in the database
./initModel
Start the workflow
./workflow
Run again to see that no fitting is done on the second start
./workflow
README
TWO TANKS EXAMPLE:
The discharge of air from one tank into another through a nozzle. The problem only makes sense, if you think of the flow through the nozzle as a much more complex problem which you have to solve with - let's say 3D CFD. So the two tanks are the macroscopic and the nozzle is the microscopic problem. This is also a backward mapping problem if you assume that the range of inputs is unknown a-priori. twoTanksMacroscopicProblem uses the MoDeNa interface library to embed an even simpler model for the flow rate. While twoTanksFullProblem implements it fully integrated.TODO: The example specific sources should be moved into the example, but this requires building (and finding) the execuables here.
How to run?
Make sure PYTHONPATH and LD_LIBRARY_PATH are set
TODO:
Make this easier to use
export PKG_CONFIG_PATH=${PKG_CONFIG_PATH:-}:${HOME}/lib/pkgconfig:/usr/local/lib/pkgconfig
export PYTHONPATH=${PYTHONPATH:-}:${HOME}/lib/python2.7/site-packages
export LD_LIBRARY_PATH=${LD_LIBRARY_PATH:-}:${HOME}/lib/python2.7/site-packages:${HOME}/lib:/usr/local/lib
Compile project specific sources, i.e. "models":
flowRate="../models/flowRate/src" twoTank="../models/twoTank/src" cmake -H${flowRate} -B${flowRate} && make –directory=${flowRate} cmake -H${twoTank} -B${twoTank} && make –directory${twoTank}
Initialise the model in the database
./initModel
Start the workflow
./workflow
Run again to see that no fitting is done on the second start
./workflow
README
TWO TANKS EXAMPLE:
The discharge of air from one tank into another through a nozzle. The problem only makes sense, if you think of the flow through the nozzle as a much more complex problem which you have to solve with - let's say 3D CFD. So the two tanks are the macroscopic and the nozzle is the microscopic problem. This is also a backward mapping problem if you assume that the range of inputs is unknown a-priori. twoTanksMacroscopicProblem uses the MoDeNa interface library to embed an even simpler model for the flow rate. While twoTanksFullProblem implements it fully integrated.TODO: The example specific sources should be moved into the example, but this requires building (and finding) the execuables here.How to run?
Make sure PYTHONPATH
and LD_LIBRARY_PATH
are set
TODO:
Make this easier to use
export PKG_CONFIG_PATH=${PKG_CONFIG_PATH:-}:${HOME}/lib/pkgconfig:/usr/local/lib/pkgconfig export PYTHONPATH=${PYTHONPATH:-}:${HOME}/lib/python2.7/site-packages export LD_LIBRARY_PATH=${LD_LIBRARY_PATH:-}:${HOME}/lib/python2.7/site-packages:${HOME}/lib:/usr/local/libCompile project specific sources, i.e. "models":
flowRate="../models/flowRate/src" twoTank="../models/twoTank/src" cmake -H${flowRate} -B${flowRate} && make –directory=${flowRate} cmake -H${twoTank} -B${twoTank} && make –directory=${twoTank}Compile project specific sources, i.e. "models":
flowRate="../models/flowRate/src" twoTank="../models/twoTankFortran/src" cmake -H${flowRate} -B${flowRate} && make –directory=${flowRate} cmake -H${twoTank} -B${twoTank} && make –directory${twoTank}Initialise the model in the database
./initModelStart the workflow
./workflowRun again to see that no fitting is done on the second start
./workflow