Suppose I have an industrial plant, where a switchboard is fed through a transformer. The switchboard have only "passive" loads, i.e. they don't contribute with short circuit currents.
If there is a short circuit on the switchboard feeding that transformer (on the grid side), my intuition tells me that the transformer would be demagnetized (disregarding remanent flux), thus energy must go from the transformer to either the secondary or primary side. Since current can't immediately change in an inductor, I think this energy should go to the secondary side.
My simulations show this behavior. As the image below show, there is current/power flowing for about 2 ms (the results on the primary side is identical, so I haven't included it). However, I'm not sure if this is the correct physical representation or a result of inaccurate mathematics. The minimum integration step I can select is 1 ms, so it might be a result of that (I think).
The reason why I doubt the results is because the exact same current flows from the shorted bus and in to the transformer on the primary side. Investigating the results on both sides of the transformer, I see that the exact same amount of energy goes in and out after the primary side is shorted. This tells me the transformer is not demagnetized.
Now, as we all learned in kinder garden, current in must equal current out, so how could I expect something different? Well, it's not really true in all situations (you can for instance charge a capacitor). I know the energy store in a transformer is magnetic energy, not electrical.
Still, there is something here that doesn't seem right to me. I appreciate any input!
It is possible that my explanation has not been good enough. The short circuit I'm talking about is a fault that occurs after a certain time, 3-phase, line-line, line-line-ground or line-ground.
Here's a part of the single line diagram. I don't have any "justification" for why the delta-star connection is relevant for the analysis (I don't think it is, it's simply how the system is designed). There can be many reasons, elimination of zero sequence currents for one, but this shouldn't be relevant for this quesiton.
The diagram below is how the transformer is represented in my simulation tool. I can't "show" how the circuit is shorted. The way it is done is I add an "event": "Create a (3 phase / Line-Line / ...) fault after 3 seconds. Then I can see the system response as a result of this event.
I would think the general behavior of the transformer is independent of the topology, vector group etc.