Beginner: Connected heat transfer networks

[alfadir]

Hi,

I am new to pandapipes but am using pandas and pint-pandas for physics to work to estimate cooling of machinery.
I have made a simple network,

Where when I run it I get

For component <class 'pandapipes.component_models.circulation_pump_pressure_component.CirculationPumpPressure'> you gave a value for t, although the component is not of 't'-type (type is p), i.e. this value is probably neglected internally.

Traceback (most recent call last):
  File "pipes.py", line 93, in <module>
    pp.pipeflow(net, mode='all')
  File "/usr/lib/python3/dist-packages/pandapipes/pipeflow.py", line 103, in pipeflow
    identify_active_nodes_branches(net, branch_pit, node_pit, False)
  File "/usr/lib/python3/dist-packages/pandapipes/pf/pipeflow_setup.py", line 531, in identify_active_nodes_branches
    raise PipeflowNotConverged(" All nodes are set out of service. Probably they are not supplied."
pandapipes.pf.pipeflow_setup.PipeflowNotConverged:  All nodes are set out of service. Probably they are not supplied. Therefore, the heat transfer pipeflow did not converge.  Have you forgotten to define an external grid?

It is not visible in the image, but the network consists of a few DN150 pipes and that branches into DN50 pipes.
The pipes are well defined in the system but here just added create_pipe_from_parameters.
I would add proper pipes to the CSV or if there is some other way to define Pipes, Pumps and HeatExchangers, create the types as code.

I can extend this to further steps, in the normal case there are 3 circuits where Heat created out of electricity cools down in a Machine network, then a building network, and the last step an external network (to a chiller or other heat reducing technologies) . This image shows the building network with 2 machines.

In addition there was a comment about the units in this discussion
I have been successful in using pint, as it makes it easy to switch between different units.
This is helpful when specification is given in USgpm vs m3/h or pressure in bar vs PSI.
Then the units can be defined using "their" specification, and the calculation can be done in the SI units.
Results can then be converted back if needed.

I can hence also define things like Water at a specific temperature as :

        self._constants["Water"] = {}
        self._constants["Water"]["Density"] = self.Q_("992.22 kg/m**3")
        self._constants["Water"]["HeatCapacity"] = self.Q_("4.0737 kJ/(kg °K)")

and calculation becomes somewhat easier, as it is unit aware

    def flowrate(self, medium, energy, inlet, outlet):
        d = self._constants[medium]["Density"]
        c = self._constants[medium]["HeatCapacity"]
        return (energy / (d * c * (outlet - inlet).to("°K"))).to("m ** 3 / hour")

Yes, Density and Heat Capacity is temperature dependent.. so.. there is room for improvement.
I have still to find a good package to just get the different constants, also for other liquids.

It is also possible to use this notation in pandas, with pint-pandas. (with the disadvantage of a dependency)

So, long post, but basically I am interested in pointers to how I get this closed system to "show" the different pressure drops.
Is it possible to "connect" both sides of the heat exchanger to look at things like thermal resistance and heat transfer ?

If pandapipes is not for this, then it would also be great to get a pointer to something that might be more reasonable.
To calculate the "steady state" of several machines needing to cool say 100kW, moving the heat out of the building to an external network or chiller. To see the effect of pipe selection etc.
It would also be interesting to know how you define different pipe connections. L, T or 45 degree elbows have all different properties in preasure loss and flow.

I have tried to read the documentation, and follow the examples that seemed reasonable, but I guess i need more pointers to the docs.

Thank you