VFD Model for Harmonic Analysis

[EP007]

For harmonic analysis on industrial power systems, I normally see the VFD's represented by current harmonics, for ac induction motor applications, and pulses of 6,12,18, etc. Now I see a report that represented the VFD's by voltage harmonics. The software is not an issue as it can accept both. My concern is that this may not be the correct method, which then affects decisions. My question is: Does anyone know of any reasons or situations where that approach could be used?

 

[dpc]

The front end of VFDs draws non-linear current. This creates non-linear voltage drops upstream of the VFD. The voltage harmonics are seen by the other shunt loads on the system.

It's hard to give a more specific answer without knowing the software being used and the purpose of the modeling.



David Castor

http://www.cvoes.com

 

[EP007]

I agree with that dpc. Normally we see the VFD represented by it's non-linear current in the form of a harmonic current spectrum. Hence, those currents produce the voltage harmonics in the system.

They have not done that. They have represented the VFD by non-linear voltage in the form of harmonic voltage spectrum. It still results in calculated harmonic currents and voltages in the system. However, I can't see how a VFD can be modeled that way, since the harmonic voltages at the VFD terminal are dependent on the network inpedance and other factors, and to represent them as constant does not seem correct. Also, how can anyone know what the harmonic voltage spectrum of a VFD is?

The purpose of the modeling is to assess the harmonic currents and voltages in the power system network.

 

[LionelHutz]

I'd agree with your assessment. The VFD model should produce current harmonics in the model power system which then generate the voltage harmonic levels.

I also don't see how you can represent the VFD as a fixed voltage harmonic model since the voltage harmonics vary with each power system.

My first guess would be the report you saw bypassed the part where the current harmonics are modelled to produce the voltage harmonics. It almost sounds like whoever generated the report just plugged in some arbitrary or "field proven" voltage harmonic levels and looked at their effects on various components in the power system.

 

[karalahana]

do not VFD's at least up to date ones have harmonic filters and cant we take them as without harmonics in our simulations?

 

[peebee]

karalahana -- I believe you've made two bad assumptions there:

1) not all modern VFDs have filters, and

2) even the ones with filters will still produce some harmonics (just not as badly as if they didn't have the filter).