Hello
I am looking to find a way to calculate the per unit Y matrix of a thee phase transformer. I can get the Y matrix of the transformer using the ActiveCKTElement object, i have also followed the paper "An Example of 3-phaseTransformer Modeling for Distribution System Analysis" and replicated the MATLAB code to do the Y matrix calculation, and my values match with the result using MATLAB or using the Export Y Prims command.
Being a three phase Delta Wye Transformer, it creates an 8x8 matrix, where I am planning to ignore thr 4th and 8th row; 4th and 8th column. How can i convert the remaining values to corresponding per unti values? Does the current transformer modeling of OpenDSS allow to do these type of calculation?
Thanks
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
There was another paper I published earlier in the same year (2003) at the IEEE PES General meeting titled:
A Perspective On Transformer Modeling for Distribution System Analysis
You might try to get it from IEEE Xplore. It will explain why I don't use per unit values inside OpenDSS for transformer modeling.
I can't really tell you how to take the Yprim back to per unit values, because I've never done it. I don't see the need for it and I have yet to meet a computer that cares whether the numbers are in per units or actual values. Many analyses are easier in actual values. Do you think the values have to be in per unit?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
If we are trying to solve a power flow using newton raphson method in python, and we want normalized values to ensure consistency, isn't it necessary to convert the Y matrix into per-unit alongwith the voltages and powers. Isn't the same procedure used internally by opendss to solve the power flow as we need to set and calculate the voltage bases to solve a power flow?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The Newton-Raphson method should work in actual quantities as well as per units, although the Jacobian may be more difficult to determine. In some cases, I have used the Y matrix as a stand-in for the Jacobian with some success. But it all depends ....
As for converting the transformer Yprim back to per units, it may be possible, but I have never done it so I don't know what to recommend. As OpenDSS develops the Yprim for a transformer, it comes through a step where it is essentially per unit (or percent) just before the actual turns ratios are applied to convert to actual values. I can imagine if it were a simple two-winding transformer or a YY 3-phase transformer, it might be straightforward. Adding a delta winding or another winding, or two, complicates things and I have never tried to do it. All the analysis types we do in OpenDSS work in actual quantities.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Sir,
I totally agree with you that for multi phase case, it is much easier to deal with actual values, rather than doing in per-unit. After going through your paper and formulation, it was much easier to understand and develop the Y matrix.
We are currently working on a multi-phase extension of a linear power flow formulation for radial systems. We were using OpenDSS to get the R matrix and X matrix for all the lines and then we converted them to per unit. I was hoping to do the same for transformers/ regulators, but looks like it is either impossible or it will probably be easier to change the our formulation to deal with actual values.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello
I am looking to find a way to calculate the per unit Y matrix of a thee phase transformer. I can get the Y matrix of the transformer using the ActiveCKTElement object, i have also followed the paper "An Example of 3-phaseTransformer Modeling for Distribution System Analysis" and replicated the MATLAB code to do the Y matrix calculation, and my values match with the result using MATLAB or using the Export Y Prims command.
Being a three phase Delta Wye Transformer, it creates an 8x8 matrix, where I am planning to ignore thr 4th and 8th row; 4th and 8th column. How can i convert the remaining values to corresponding per unti values? Does the current transformer modeling of OpenDSS allow to do these type of calculation?
Thanks
There was another paper I published earlier in the same year (2003) at the IEEE PES General meeting titled:
A Perspective On Transformer Modeling for Distribution System Analysis
You might try to get it from IEEE Xplore. It will explain why I don't use per unit values inside OpenDSS for transformer modeling.
I can't really tell you how to take the Yprim back to per unit values, because I've never done it. I don't see the need for it and I have yet to meet a computer that cares whether the numbers are in per units or actual values. Many analyses are easier in actual values. Do you think the values have to be in per unit?
Hello Roger,
If we are trying to solve a power flow using newton raphson method in python, and we want normalized values to ensure consistency, isn't it necessary to convert the Y matrix into per-unit alongwith the voltages and powers. Isn't the same procedure used internally by opendss to solve the power flow as we need to set and calculate the voltage bases to solve a power flow?
The Newton-Raphson method should work in actual quantities as well as per units, although the Jacobian may be more difficult to determine. In some cases, I have used the Y matrix as a stand-in for the Jacobian with some success. But it all depends ....
As for converting the transformer Yprim back to per units, it may be possible, but I have never done it so I don't know what to recommend. As OpenDSS develops the Yprim for a transformer, it comes through a step where it is essentially per unit (or percent) just before the actual turns ratios are applied to convert to actual values. I can imagine if it were a simple two-winding transformer or a YY 3-phase transformer, it might be straightforward. Adding a delta winding or another winding, or two, complicates things and I have never tried to do it. All the analysis types we do in OpenDSS work in actual quantities.
Sir,
I totally agree with you that for multi phase case, it is much easier to deal with actual values, rather than doing in per-unit. After going through your paper and formulation, it was much easier to understand and develop the Y matrix.
We are currently working on a multi-phase extension of a linear power flow formulation for radial systems. We were using OpenDSS to get the R matrix and X matrix for all the lines and then we converted them to per unit. I was hoping to do the same for transformers/ regulators, but looks like it is either impossible or it will probably be easier to change the our formulation to deal with actual values.