Debugging NR non convergence case

[AnkurArohi]

I have a non converging NR and I am trying to find out what could lead to this failure. While running the diagnostic function i could find out two major points:

{'impedance_values_close_to_zero': [{'xward': [5, 31], 'impedance': [2, 9, 10, 23, 26, 27, 41, 43, 49, 67, 68, 73, 81, 82, 84, 92, 99, 100, 104]},

I tried to set the impedance out of service but still the network is not converging. A further analysis showed that the same network has a NR convergence case in INTEGRAL. Any suggestions?

[rbolgaryn]

try increasing the impedance instead of setting it out of service

[AnkurArohi]

i am sorry, i actually increased the impedance, i meant i tried to set xward out of service, as i see that the impedance of xward is not low, still it does not converge

[rbolgaryn]

you set the impedance of these elements higher, and still doesn't converge? You can also try running DC power flow and seeing if any elements are overloaded to a high extent.

[AnkurArohi]

yes still does not converge, DC Power flow/Initial guess converges

[AnkurArohi]

Are extended wards also modelled as PV node or PQ node, as per the description in pandapower docs, I came to the conclusion that they are modelled as PQ nodes, is that right?

[rbolgaryn]

Yes, ward is PQ and xward is PV