add stratified heat buffer #225

octavianor · 2024-12-09T15:45:59Z

now we implement not ideal buffer, it means that the temperature of return and supply temperature depends on the states of the storage.

this makes difficult for controller to do balancing based on power. Example: controller asking 1 MW from storage. Storage will calculate mass flowrate based on dT. The consequence: network temperature becoming 75 degree at supply side, and it will go down over time when you ask more power from storage.

in Warming Up 1, we assumed ideal buffer with fixed dT so we have no problem

src/omotes_simulator_core/entities/assets/heat_buffer.py

samvanderzwan · 2025-01-08T08:18:27Z

src/omotes_simulator_core/entities/assets/heat_buffer.py

+        self.timestep = 3600  # seconds
+        self.solver_asset = ProductionAsset(name=self.name, _id=self.asset_id)
+
+        # Output list


remove redundant comment

src/omotes_simulator_core/entities/assets/heat_buffer.py

samvanderzwan · 2025-01-08T08:23:17Z

src/omotes_simulator_core/entities/assets/heat_buffer.py

+        )
+
+    def _calculate_fill_level(self) -> None:
+        """Calculate fill level of the storage."""


You also now the maximum volume, so should we not also calculate the volume stored?

indeed we can have both output

samvanderzwan · 2025-01-08T08:24:40Z

src/omotes_simulator_core/entities/assets/heat_buffer.py

+            raise ValueError(
+                f"The new fill level is {new_fill_level}. It should be between 0 and 1."
+            )
+


You only calculate the volume stored, but you should also track the temperature or the energy stored in the buffer, since now you assume that you can always supply at the temperature which is set, but you will supply at the temperature which is in the buffer.

currently I implement ideal buffer, only tracking volume without modeling temperature losses. Thus temperature of storage always at setpoint supply and return

But still without temperature loss, you can have different temperatures as inflow and then you just want to calculate mixing temp of fluid in buffer.

for ideal water buffer without thermal losses, i am not interested with mixing temperature in the middle layer. The model is only temperature on top layer and bottom layer which is remains constant (hot and cold)

The above image is not what is being modelled based on the provided code. You have a storage with a 2-node component now.

In addition, you mention hot and cold remain constant but that is not always the case right? These temperatures can vary or are you now implying that the whole simulation has a fixed supply and return temperature.

Ideal Buffer model CHESS.pptx

Buffertank model.docx

we conclude that for simplicity , the ideal buffer is tracking energy with a fixed deltaT. Since Energy = Power * dt and Power = deltaT * Flow * HeatCapacity. Thus tracking energy is similar with tracking volume V = V + Flow * dt

@samvanderzwan @vanmeerkerk but you are correct, now the implementation is driven by temperature supplied by controller, it should be using temperature from the object itself defined from ESDL carrier.

let's discuss first if this model is enough for our simulator or we want to have more detailed model

src/omotes_simulator_core/entities/assets/heat_buffer.py

…-buffer-asset

vanmeerkerk

Hey Ryvo I tried to understand the component but could not yet figure it out. Could you write a manual page describing the equations that you would like to solve?

vanmeerkerk · 2025-02-06T10:01:13Z

src/omotes_simulator_core/adapter/transforms/esdl_asset_mappers/heat_buffer_mapper.py

+from omotes_simulator_core.entities.assets.asset_defaults import HEAT_BUFFER_DEFAULTS
+
+
+class EsdlAssetAtesMapper(EsdlMapperAbstract):


If it is a mapper to a heatbuffer class why do I see the Ates referecne? Or do we no assume the Ates is the idealized heat buffer?

i copy paste from ates, indeed i should change to buffermapper

vanmeerkerk · 2025-02-06T11:46:21Z

src/omotes_simulator_core/entities/assets/heat_buffer.py

+        self.thermal_power_allocation = 0
+        self.mass_flowrate = 0
+        self.maximum_volume = maximum_volume
+        self.fill_level = fill_level


Personally I would rename fill_level to percentual_fill_level to make it more clear that we are using it as a fraction (e.g., value between 0 - 1)

vanmeerkerk · 2025-02-06T11:47:00Z

src/omotes_simulator_core/entities/assets/heat_buffer.py

+        self.output: list = []
+
+    def set_setpoints(self, setpoints: Dict) -> None:
+        """Placeholder to set the setpoints of an asset prior to a simulation.


It is no longer a placeholder, right?

vanmeerkerk · 2025-02-06T11:50:03Z

src/omotes_simulator_core/entities/assets/heat_buffer.py

+        else:
+            raise ValueError(
+                f"The new fill level is {new_fill_level}. It should be between 0 and 1."
+            )


This error should be handeld by the controller; a fill level of 0 means no discharge from the storage and 1 no more supply to the storage.

I am still implementing this error to make sure the flow that you set from controller does not hit buffer volume constraint. It could be the controller is not taken account this

vanmeerkerk · 2025-02-06T11:50:48Z

src/omotes_simulator_core/entities/assets/heat_buffer.py

+        self.maximum_volume = maximum_volume
+        self.fill_level = fill_level
+        self.current_volume = fill_level * maximum_volume
+        self.timestep = 3600


Timestep should not be a fixed value but depend on the simulation dt; now it cannot be adjusted as the input is not there!

this is similar with ates discussion, currently it is also fixed there until we have solution to include timestep from controller or esdl

vanmeerkerk · 2025-02-06T11:55:40Z

src/omotes_simulator_core/entities/assets/heat_buffer.py

+            raise ValueError(
+                f"The new fill level is {new_fill_level}. It should be between 0 and 1."
+            )
+


The above image is not what is being modelled based on the provided code. You have a storage with a 2-node component now.

In addition, you mention hot and cold remain constant but that is not always the case right? These temperatures can vary or are you now implying that the whole simulation has a fixed supply and return temperature.

vanmeerkerk · 2025-02-06T11:58:53Z

src/omotes_simulator_core/entities/assets/heat_buffer.py

+        :param str asset_id: The unique identifier of the asset.
+        """
+        super().__init__(asset_name=asset_name, asset_id=asset_id, connected_ports=port_ids)
+        self.temperature_supply = DEFAULT_TEMPERATURE


Temperature_supply is the 'hot storage temperature' and Temperature_return is the 'cold storage temperature'.

yes indeed, i can rename that also

vanmeerkerk · 2025-02-06T12:01:36Z

src/omotes_simulator_core/entities/assets/heat_buffer.py

+            PROPERTY_TEMPERATURE_SUPPLY,
+            PROPERTY_TEMPERATURE_RETURN,


The controller is not able to set these values at least not with my understanding of the current implementation. I do not see how it can continously change the hot and cold storage temperatues. You should get some sort of value for the energy it is supplying to the vessel and calculate the new fluid level associated with the mixed hot/cold temperature level, right?

…rom carrier esdl during mapping

…ct-OMOTES/simulator-core into 221-day-night-buffer-asset # Conflicts: # src/omotes_simulator_core/entities/assets/asset_defaults.py

…-buffer-asset

octavianor · 2025-08-11T08:54:20Z

i did major re-implementation of water buffer tank. based on 1-D stratified buffer with 5 layers

reimplement as 1D stratified buffer tank

octavianor · 2025-08-12T07:00:16Z

reference: https://www.sciencedirect.com/science/article/pii/S0306261919307901?via%3Dihub

but i dont include buoyancy and thermal losses to outside, only direct heating (yellow part)

vanmeerkerk

Some minor comments should be fine after a final fix

vanmeerkerk · 2025-08-28T10:59:29Z

ci/win32/lint.cmd

 cd /D "%~dp0"
 cd ..\..\
-flake8 .\src\omotes_simulator_core
+flake8 .\src\omotes_simulator_core .\unit_test


Can you either remove this statement or explain why it is needed to also run flake8 on the unit test separately?

because the linux script also include unit test, so i just want to make sure it is similar

vanmeerkerk · 2025-08-28T11:00:14Z

ci/win32/typecheck.cmd

 call .\venv\Scripts\activate
 set PYTHONPATH=.\src\;%$PYTHONPATH%
-python -m mypy ./src/
+python -m mypy ./src/omotes_simulator_core ./unit_test/


Same comment as with the linting. Why is it necessary also call mypy on the unit test? Shouldn't the pyproject.toml handle this?

because the linux script also include unit test, so i just want to make sure it is similar

vanmeerkerk · 2025-08-28T11:02:43Z

src/omotes_simulator_core/adapter/transforms/esdl_asset_mappers/heat_buffer_mapper.py

+            asset_name=esdl_asset.esdl_asset.name,
+            asset_id=esdl_asset.esdl_asset.id,
+            port_ids=esdl_asset.get_port_ids(),
+            volume=esdl_asset.get_property(


You won't need to call the EsdlAssetObject.get_property()[0] with the zero at the end; the method returns a float so it should work without.

vanmeerkerk · 2025-08-28T11:04:54Z

src/omotes_simulator_core/entities/assets/controller/controller_storage.py

+        # TODO: The loop is not complete as the asset also has a fill-level that should not
+        #  surpass the maximum fill-level.


That is an implementation I made that can be merged/changed after we have completed the current PR.

vanmeerkerk · 2025-08-28T11:08:44Z

unit_test/entities/test_heat_buffer.py

+#  You should have received a copy of the GNU General Public License
+#  along with this program.  If not, see <https://www.gnu.org/licenses/>.
+
+"""Test Ates Cluster entities."""


Rename to test the HeatBuffer or descriptive name of the asset you are testing.

vanmeerkerk · 2025-08-28T11:18:40Z

src/omotes_simulator_core/entities/assets/heat_buffer.py

+
+    def _set_solver_asset_setpoint(self) -> None:
+        """Set the setpoint of solver asset."""
+        if self.mass_flowrate > 0:


This should be the zero flow limit; >= MASSFLOW_ZERO_LIMIT and probably the lowerbound should be <= - MASSFLOW_ZERO_LIMIT. Question, what happens when m_dot = 0 or between those bounds?

vanmeerkerk · 2025-08-28T11:20:33Z

src/omotes_simulator_core/entities/assets/heat_buffer.py

+            self.mass_flowrate = 0
+
+    def _calculate_new_temperature(self) -> None:
+        """Calculate new temperature of the tank storage."""


Can you include more details in the docstring what you are exactly solving?

vanmeerkerk · 2025-08-28T11:20:56Z

src/omotes_simulator_core/entities/assets/heat_buffer.py

+
+
+class HeatBuffer(AssetAbstract):
+    """A HeatBuffer represents an asset that stores heat. Thus, it has the possibility to supply \


Maybe include a reference or doi to the paper you are using to model this system.

The temperature at the in and out port should be possible to define right? Or at least some sort of initial storage temperature, now you will always start at 300 k

The layer mass now remains constant over the calculation; shouldn't that be a parameter defined by the local temperature difference over the layer? So that layer mass is a variable over the height of the storage tank?

This off course depends on the structure of the tank

vanmeerkerk · 2025-08-28T12:23:12Z

src/omotes_simulator_core/entities/assets/controller/controller_storage.py

+        # TODO: Current implementation loops over the entire profile; should be improved!
+        # TODO: Unclear why there is a timestep of 1 hour in the profile.
        for index in range(self.start_index, len(self.profile)):
            if abs((self.profile["date"][index].to_pydatetime() - time).total_seconds()) < 3600:


This might not work as expected; what is the intention? Maybe we should make an seperate issue to review this part.

vanmeerkerk · 2025-08-28T12:24:46Z

src/omotes_simulator_core/entities/network_controller.py

        """Method to get the total storage discharge power of the network.

        :return float: Total heat discharge of all storages.
        """


This is still something that needs to be implemented. Revert back the todo.

vanmeerkerk · 2025-08-28T13:30:10Z

I checked some more of the implementation, when I create a system with one layer the stored power is approximately 36 MJ after one step (1H), but this increases to 97 MJ for 5 layers, even though the input doesn't change. There might be something wrong with the implementation @octavianor

--- Addition ---

I tried a few things - please correct me if I am wrong - but do you now compute the stored power as:

`
def _get_stored_power(self) -> float:
"""Get the stored power in the heat buffer.

    Compute the power (J) stored in the heat buffer, as

        Q_i = m * c_{p,i} * (T_{i} - T_{i+1})

    with i as the layer index.

    :return: The stored power in the heat buffer (J).
    :rtype: float
    """
    # Average temperature per layer
    average_layer_temperature = (self.layer_temperature[:-1] + self.layer_temperature[1:]) / 2
    # Temperature difference per layer
    delta_layer_temperature = np.diff(self.layer_temperature[::-1])
    # Layered heat capacity
    heat_capacity_layered = np.array(
        [
            fluid_props.get_heat_capacity(average_layer_temp)
            for average_layer_temp in average_layer_temperature
        ]
    )
    # Stored heat
    stored_heat = np.sum(heat_capacity_layered * delta_layer_temperature * self.layer_mass)

    return stored_heat

`

octavianor · 2025-09-11T12:38:01Z

add energy calculation

octavianor · 2025-09-11T14:05:36Z

i put this into draft because i confuse with convension positive and negative flowrate

…-buffer-asset

octavianor · 2025-09-17T12:35:13Z

implement ODE solver, has been tested with esdl file

octavianor · 2025-09-19T07:34:44Z

@vanmeerkerk ODE takes 30 iteration for 5 layers (with solving time 1 ms)

add ideal heat buffer tracking fill level

b2875c9

octavianor linked an issue Dec 9, 2024 that may be closed by this pull request

Day-night buffer asset #221

Open

octavianor requested review from samvanderzwan and vanmeerkerk December 9, 2024 15:46

octavianor self-assigned this Dec 9, 2024

samvanderzwan previously requested changes Jan 8, 2025

View reviewed changes

octavianor added 2 commits January 17, 2025 10:35

Merge remote-tracking branch 'remotes/origin/main' into 221-day-night…

ff053a8

…-buffer-asset

add volume calculation and update the comments

83ff1b7

octavianor requested a review from samvanderzwan January 17, 2025 09:43

octavianor added 3 commits January 17, 2025 11:04

add heat buffer mapper to remove add physical data

71691a1

add new parameters for unit test

1ede1fb

fix current volume calculation

d511e89

vanmeerkerk previously requested changes Feb 6, 2025

View reviewed changes

octavianor and others added 17 commits February 7, 2025 09:42

rename mapper

f564c58

map esdl water buffer as heat buffer

703f0e5

remove supply and return temperature from controller, it is now set f…

c609e4a

…rom carrier esdl during mapping

add ideal heat buffer tracking fill level

4f3a415

add volume calculation and update the comments

198cb9e

add heat buffer mapper to remove add physical data

7612d81

add new parameters for unit test

86a88db

fix current volume calculation

39cbbdf

rename mapper

c0b64b6

map esdl water buffer as heat buffer

51310f5

remove supply and return temperature from controller, it is now set f…

77cc69c

…rom carrier esdl during mapping

Merge branch '221-day-night-buffer-asset' of https://github.com/Proje…

623b700

…ct-OMOTES/simulator-core into 221-day-night-buffer-asset # Conflicts: # src/omotes_simulator_core/entities/assets/asset_defaults.py

Minor code changes

4dc972e

Merge remote-tracking branch 'remotes/origin/main' into 221-day-night…

1371134

…-buffer-asset

run formatting

1ca1295

reformat heat_buffer_mapper.py

d47c6fd

implement stratified buffer tank with 5 layers

73bab55

fix unit test

45b2f92

octavianor requested a review from vanmeerkerk August 11, 2025 08:54

remove fault handler

05f0321

vanmeerkerk requested changes Aug 28, 2025

View reviewed changes

octavianor added 4 commits August 29, 2025 10:28

fix typo layer volume

4d63fef

fix test heat buffer after correction layer_mass

e357c8e

change comparison value

99780f0

add energy calculation

f7b33bd

add stored energy calculation

0685f57

octavianor marked this pull request as draft September 11, 2025 14:05

octavianor added 5 commits September 13, 2025 09:49

change to ODE solver

ac27693

add time in asset, integration test with ESDL

b491dce

Merge remote-tracking branch 'remotes/origin/main' into 221-day-night…

d961895

…-buffer-asset

fix linting, formatting, typecheck

3ab1f8c

fix formatting and unit test buffer with time

721516d

octavianor requested a review from vanmeerkerk September 17, 2025 12:33

octavianor marked this pull request as ready for review September 17, 2025 12:34

octavianor marked this pull request as draft September 17, 2025 12:35

octavianor marked this pull request as ready for review September 17, 2025 12:36

octavianor changed the title ~~add ideal heat buffer tracking fill level~~ add stratified heat buffer Sep 19, 2025

update typecheck for numpy

edf3df5

		from omotes_simulator_core.entities.assets.asset_defaults import HEAT_BUFFER_DEFAULTS


		class EsdlAssetAtesMapper(EsdlMapperAbstract):

		# TODO: The loop is not complete as the asset also has a fill-level that should not
		# surpass the maximum fill-level.



		class HeatBuffer(AssetAbstract):
		"""A HeatBuffer represents an asset that stores heat. Thus, it has the possibility to supply \

add stratified heat buffer #225

Are you sure you want to change the base?

add stratified heat buffer #225

Uh oh!

Conversation

octavianor commented Dec 9, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

octavianor Feb 7, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

octavianor Feb 7, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

vanmeerkerk left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

octavianor commented Aug 11, 2025

Uh oh!

octavianor commented Aug 12, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

vanmeerkerk left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

octavianor commented Dec 9, 2024 •

edited

Loading

octavianor Feb 7, 2025 •

edited

Loading

octavianor Feb 7, 2025 •

edited

Loading

octavianor commented Aug 12, 2025 •

edited

Loading

vanmeerkerk commented Aug 28, 2025 •

edited

Loading