Implement pressure loss calculation via pandapipes

[jnettels]

I have added a new argument calculation to delta_p() and the functions which call it (v_max_secant(), v_max_bisection()) which allows using a pandapipes simulation to calculate pressure losses in a pipe as an alternative to the existing internal implementation.

• Update code
• Add new test
• Update changelog
• Update documentation

Fixes #162

Context

dhnx includes the precalc_hydraulic module which allows converting the resulting capacity of each pipe into norm diameters.
For this, a pressure loss calculation is needed, for which precalc_hydraulic provides an internal solution.

In my workflows, I design a network with dhnx and then run a simulation of temperature and pressure distribution with pandapipes to check if the network is valid in terms of pressure drops and velocities. In some rare edge cases I noticed differences where dhnx was supposed to only give e.g. a maximum of 100 Pa/m pressure losses in any given pipe segment but pandapipes showed slightly larger values.

Implementing a new function delta_p_pandapipes() in precalc_hydraulic solved the issue because it synchronizes the precalculation and post-processing. It allows calculating the pressure loss with pandapipes instead of using the internal implementation.

Independently, in #162 another user describes issues with the internal implementation. Instead of trying to fix/improve the internal pressure loss equations I think it is the best approach to use external modules that have already put more resources into this problem. (Nevertheless: pandapipes itself offers different friction models, and judging by the issues there, not all of them are perfect, either.)

Considerations / feedback

• My approach gives room to add other calculation methods in the future, which may coexist. I hope that is useful
• Running pandapipes is significantly slower than the internal method. But in the context of a long optimization procedure that is fine
• precalc_hydraulic is briefly mentioned in the documentation, but not actually documented. Thus there was no documentation to update
• I have added matching tests
• In order for the new test to work, I have added pandapipes to the deps in testenv in tox.ini. But I am unsure if this is the best way and I need feedback. I guess I could create a new entry of optional dependencies in pyproject.toml instead, but that feels like overkill to me. If a user knows they want to use pandapipes alongside dhnx, they can just install it.
• pandapipes requires pandapower, which is currently still restricted to pandas<3.0. Thus with this change all of our tests here will run with pandas 2.x, which is not ideal, I think. But I see no way around that. Thoughts?

[github-actions]

Coverage report

Click to see where and how coverage changed

File	Statements	Missing	Coverage	Coverage (new stmts)	Lines missing
src/dhnx/optimization
precalc_hydraulic.py
Project Total

This report was generated by python-coverage-comment-action