Roger Dugan

The KLU solver can handle an asymmetric Y matrices if you create one. You can get one with elements like the VSOURCE or REACTOR by specifying that the positive- and negative-sequence impedances are different. Is that what you are looking for or something else?

Somewhere in the beginning of your DSS script, you need a statment that creates a new circuit. Something like this code borrowed from the IEEE 13-bus test feeder: new circuit.IEEE13Nodeckt ~ basekv=115 pu=1.0001 phases=3 bus1=SourceBus ~ Angle=30 ! advance angle 30 deg so result agree with published angle ~ MVAsc3=20000 MVASC1=21000 ! stiffen the source to approximate inf source When you instantiate a Circuit object you are defining a Thevenin equivalent (short-circuit equivalent) for the beginning...

There are few optimization algorithms built into OpenDSS. This was done intentionally in 1997 when we first built the code. It was understood that it would be difficult to anticipate all the ways power distribution engineers would choose to design optimization algorithms. So we provided a way for users to implement their own optimization code and that was primarily the COM interface, which works on Windows implementations. Few restrictions were placed on the coding. Some users have implemented optimization...

CalcVoltagebases essentially does a zero-load solution for the purpose of establishing the voltage bases. It uses the voltage bases defined with the "Set Voltagebases= ...." list to guess at the appropriate voltage base for each bus. This command is provided for user convenience. You can set each base voltage base separately using the SetkVBase command. For a large circuit, this is tedious so we provided the CalcV command that works nearly all the time. There are a couple of transformer configurations...

There are a few errors in the script. I'll give you some hints and you can try again. You defined the main Vsource with this: "basekv=110, basekv=220". Not sure what you were trying to do, but the 2nd declaration of basekv overwrites the first. So you end up with 220 as the base kV. The kVA ratings of the transformers have too many zeroes. 150 MVA is defined as ** kVA=150000**. Make these changes and try again.

I put your OpenDSS script back into OpenDSS and it seems to solve correctly and quickly with no more than 2 iterations per tap change. (4 tap changes are required). You can find several posts on this Forum from users who have tried something similar to what you are doing with the Y matrix and had similar difficulties. I suspect the Y matrix is not entirely what you think it is. Please review the documentation that is on line (Help>UserManual) for how the fixed-point solution works. I think you'll...

For just a simple transformer with 2 windings (H and L) the short circuit impedance from H to L in percent is %R(H) + %R(L) + j XHL. When there are more than 2 windings, it can seem more complicated. The required impedances are the winding resistances (in percent) and the short circuit reactance (in percent) between each pair of windings. Some power system analysis programs will want you to specify an X for L and and X for H. But there is actually no such thing for this analysis. A short circuit,...

Specifically, what problems are you encountering?