I think that I've got my head around how to run a harmonic flow simulation in OpenDSS, in which harmonic current spectra are applied to connecting loads and a harmonic voltage spectra is applied to the source bus. Thanks to all of the responses to questions in the forums by other users that helped to increase my understanding of this.
As such, my question surrounds the basis by which harmonic currents and voltages are determined in a harmonic flow simulation rather than how to actually implement the code.
In the scenario I introduced above, both harmonic voltage and current distortion are introduced into the network under study. If we refer to IEEE Std. 1459-2010 we obtain the following equations for the flow of rms apparent power (S) under such conditions (See attached images 1 and 2).
In reality, it is evident (from these equations) that some amount of cross-coupling can be expected to occur between the harmonic orders of current and voltage to form distortion powers such as DI, DV and DH (definitions for which are described in IEEE 1459).
I would like to understand how, or if, OpenDSS models these cross-coupling interactions in a harmonic flow program, given that we simuate one harmonic order at a time. For example, are the currents that are reported from such simulations reflective of the cross coupling observed in the formation of what IEEE-1459 terms the "current distortion power", DI, where DI^2 = [V1.I2]^2 +[V1.I3]^2 + [V1.I4]^2 +[V1.I5]^2, ..., etc ?
OpenDSS computes the solution for each frequency and Monitor objects capture the solution for each frequency component. To compute S, Vrms, THD due to multiple frequencies you have to post-process the solution. Fairly easy from the COM interface. I will frequently set up an Excel spreadsheet and load the CSV files into it where it automatically calculates the values.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thank you for getting back to me, and may I take this opportunity to also thank you for all of your efforts in developing and maintaining OpenDSS over the years, as it is a very useful tool.
As you suggest, I already have some post-processing in place using Python to compute the associated THD and rms quantities. This works well.
Unfortunately, I don't think that my initial post was very clear - apologies for this. Ultimately, what I'm trying to learn more about in OpenDSS is how loads that are configured with harmonic current spectra are expected to respond in simulation, under the influence of different applied voltage harmonic spectra and in response to changes in the fundamental voltage magnitude observed at the terminals of the device .
In practice, electronic devices are often observed to emit higher levels of current distortion if their terminal voltage is subject to higher levels of voltage distortion. For reference, an example of this can be found in a paper I wrote: http://doi.org/10.1049/oap-cired.2017.0190
Is there an easy way to adjust the harmonic current spectra defined for a Load object as the software progresses through the different harmonic orders in simulation? Or can you only specify this before the first fundamental solution is performed and not update it again until after all non-fundamental solutions have been completed?
Ultimately, I am interested in incorporating a voltage level dependency effect into the harmonic current emissions of the Load objects in my simulations. I know that ZIP coefficients etc can be used to enact this for voltage led changes in P and Q. I was wondering if a similar approach could be applied to effectively adjust the harmonic current spectra with changes in applied voltage level and/or voltage distortion?
For example, if a set of ZIP coefficients are defined for a Load object, do changes in the voltage observed for the 3rd, 5th etc harmonic orders effect the harmonic currents drawn by the load in the 3rd or 5th order?
Thanks
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
For cases in which the harmonic spectrum changes significantly due to voltage distortion, we will recalc the Spectrum and do the Harmonic mode solution again. We are doing this with our GIC_Harm app in which the transformer current distortion changes as a function of the voltage distortion. So we iterate through the harmonics solution and GIC solution 3 to 7 times for it to converge. Seems to work well, but we are expecting to make refinements in the future. This app is in Python and will eventually be posted open source, but is not public for now.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
I think that I've got my head around how to run a harmonic flow simulation in OpenDSS, in which harmonic current spectra are applied to connecting loads and a harmonic voltage spectra is applied to the source bus. Thanks to all of the responses to questions in the forums by other users that helped to increase my understanding of this.
As such, my question surrounds the basis by which harmonic currents and voltages are determined in a harmonic flow simulation rather than how to actually implement the code.
In the scenario I introduced above, both harmonic voltage and current distortion are introduced into the network under study. If we refer to IEEE Std. 1459-2010 we obtain the following equations for the flow of rms apparent power (S) under such conditions (See attached images 1 and 2).
In reality, it is evident (from these equations) that some amount of cross-coupling can be expected to occur between the harmonic orders of current and voltage to form distortion powers such as DI, DV and DH (definitions for which are described in IEEE 1459).
I would like to understand how, or if, OpenDSS models these cross-coupling interactions in a harmonic flow program, given that we simuate one harmonic order at a time. For example, are the currents that are reported from such simulations reflective of the cross coupling observed in the formation of what IEEE-1459 terms the "current distortion power", DI, where DI^2 = [V1.I2]^2 +[V1.I3]^2 + [V1.I4]^2 +[V1.I5]^2, ..., etc ?
Thanks in advance.
Last edit: Gareth McLorn 2019-01-12
OpenDSS computes the solution for each frequency and Monitor objects capture the solution for each frequency component. To compute S, Vrms, THD due to multiple frequencies you have to post-process the solution. Fairly easy from the COM interface. I will frequently set up an Excel spreadsheet and load the CSV files into it where it automatically calculates the values.
Hi,
Thank you for getting back to me, and may I take this opportunity to also thank you for all of your efforts in developing and maintaining OpenDSS over the years, as it is a very useful tool.
As you suggest, I already have some post-processing in place using Python to compute the associated THD and rms quantities. This works well.
Unfortunately, I don't think that my initial post was very clear - apologies for this. Ultimately, what I'm trying to learn more about in OpenDSS is how loads that are configured with harmonic current spectra are expected to respond in simulation, under the influence of different applied voltage harmonic spectra and in response to changes in the fundamental voltage magnitude observed at the terminals of the device .
In practice, electronic devices are often observed to emit higher levels of current distortion if their terminal voltage is subject to higher levels of voltage distortion. For reference, an example of this can be found in a paper I wrote: http://doi.org/10.1049/oap-cired.2017.0190
Is there an easy way to adjust the harmonic current spectra defined for a Load object as the software progresses through the different harmonic orders in simulation? Or can you only specify this before the first fundamental solution is performed and not update it again until after all non-fundamental solutions have been completed?
Ultimately, I am interested in incorporating a voltage level dependency effect into the harmonic current emissions of the Load objects in my simulations. I know that ZIP coefficients etc can be used to enact this for voltage led changes in P and Q. I was wondering if a similar approach could be applied to effectively adjust the harmonic current spectra with changes in applied voltage level and/or voltage distortion?
For example, if a set of ZIP coefficients are defined for a Load object, do changes in the voltage observed for the 3rd, 5th etc harmonic orders effect the harmonic currents drawn by the load in the 3rd or 5th order?
Thanks
During a Harmonics mode solution a Load model is a linear, time-invariant model (Norton equivalent with a harmonic spectrum). See:
https://sourceforge.net/p/electricdss/code/HEAD/tree/trunk/Version7/Doc/Harmonics%20Load%20Modeling.docx
For cases in which the harmonic spectrum changes significantly due to voltage distortion, we will recalc the Spectrum and do the Harmonic mode solution again. We are doing this with our GIC_Harm app in which the transformer current distortion changes as a function of the voltage distortion. So we iterate through the harmonics solution and GIC solution 3 to 7 times for it to converge. Seems to work well, but we are expecting to make refinements in the future. This app is in Python and will eventually be posted open source, but is not public for now.
Thanks for this.