Magnetic specialists,
I'm curious if anyone has tried to simulate 3-phase transformer inrush in the Finite Element world. I'm currently learning to use the software on the fly and am finding there are a lot of tricks to be learned in order to see the simulation work (and make sense).
If anyone has attempted this I have a question: Can you simulate the inrush event and run the simulation long enough to see the complete transient event?
Here is my particular issue: The saturation curve of the transformer has to be simplified to always have a crossing at 0,0 (no hysteresis). Also, when the core is saturated, you have to extend the saturation curve to be the equivalent of air permeability. As a result of this I'm forced to trick the core into saturation (no remnant flux can be defined). This seems to be the norm among the software. Saturation is accomplished by changing the source voltage(s) when the core has peak flux density.
By doing the above no DC offset occurs in the waveforms. Also, the currents maintain inrush current characteristics once the core flux density is back into a normal operating condition.
I feel like I’m missing something really basic. Either my assumptions and tricks within the software have caused the system response to be unrealistic OR my hopes for the software have been unrealistic.
Any comments are appreciated.
Thank you for any assistance,
Jim
I'm curious if anyone has tried to simulate 3-phase transformer inrush in the Finite Element world. I'm currently learning to use the software on the fly and am finding there are a lot of tricks to be learned in order to see the simulation work (and make sense).
If anyone has attempted this I have a question: Can you simulate the inrush event and run the simulation long enough to see the complete transient event?
Here is my particular issue: The saturation curve of the transformer has to be simplified to always have a crossing at 0,0 (no hysteresis). Also, when the core is saturated, you have to extend the saturation curve to be the equivalent of air permeability. As a result of this I'm forced to trick the core into saturation (no remnant flux can be defined). This seems to be the norm among the software. Saturation is accomplished by changing the source voltage(s) when the core has peak flux density.
By doing the above no DC offset occurs in the waveforms. Also, the currents maintain inrush current characteristics once the core flux density is back into a normal operating condition.
I feel like I’m missing something really basic. Either my assumptions and tricks within the software have caused the system response to be unrealistic OR my hopes for the software have been unrealistic.
Any comments are appreciated.
Thank you for any assistance,
Jim
