[bug] `pandapower.converter.{pypower, matpower}` broken

[determ1ne]

Bug report checklis

• Searched the issues page for similar reports

• Read the relevant sections of the documentation

• Browse the tutorials and tests for usefull code snippets and examples of use

• Reproduced the issue after updating with pip install --upgrade pandapower (or git pull)

• Tried basic troubleshooting (if a bug/error) like restarting the interpreter and checking the pythonpath

Reproducible Example

"""
% export a mpc from matpower first
mpc = case9241pegase;
save('mpc.mat');
"""
from pandapower.converter.matpower import from_mpc, to_mpc
net = from_mpc('mpc.mat')
to_mpc(net, 'mpc_out.mat', init="flat")

# the matpower will fail to run OPF with the new mat
"""
load("mpc_out.mat");
runopf(mpc);
% Did not converge in 2 iterations
"""

Issue Description and Traceback

This issue may be similar or related to #2620 .

1. Currently from_mpc and to_mpc combination destroys the data.

from pandapower.converter.matpower import from_mpc, to_mpc

net = from_mpc('case9241.mat')
to_mpc(net, 'case9241_out.mat', init="flat")

output:

There are 22 branches which are considered as trafos - due to ratio unequal 0 or 1 - but connect same voltage levels.

2. from_ppc get the same output, seems broken too.

Expected Behavior

MATPOWER OPF should converge with the exported case file.

Installed Versions

• pandapower version: develop branch, commit f91a669
• python version: 3.13.12
• pandas version: 2.3.3
• networkx version: 3.6.1
• scipy version: 1.16.3
• numpy version: 2.3.5
• packaging version: 25.0
• tqdm version: 4.67.3
• deepdiff version: 8.6.2
• Operating System name/version: Microsoft Windows Version 10.0.26200.7628

Label

• Relevant labels are selected

[KS-HTK]

So testing this it seems to work with case118 and only an issue with the case9241pegase.

When trying to load this directly from the .m file (as I do not have matlab) I stumbled upon a bug in matpowercaseframes that I have reported there. I guess that this might be related.

[determ1ne]

So testing this it seems to work with case118 and only an issue with the case9241pegase.

When trying to load this directly from the .m file (as I do not have matlab) I stumbled upon a bug in matpowercaseframes that I have reported there. I guess that this might be related.

I think pandapower embedded case9241pegase JSON also has the problem:

import pandapower as pp
from pandapower.networks.power_system_test_cases import case9241pegase
from pandapower.converter.matpower import from_mpc, to_mpc

net = case9241pegase()
pp.runpp(net)
to_mpc(net, 'mycase.mat')
# Both MATPOWER and pp.runopp failed the OPF