Developing the Approximate Luzon Network Model

Almost two weeks of vacation gave me enough reason to delay the update on the approximate Luzon power system model I have developed in Powerworld. In this post, I will cite the modeling assumptions I have made and some issues I encountered in the approximate model. The aim is to allow discussion to set in to refine the model as to depict as near as possible the real power system performance.

Component Modeling:

Generation:
I modeled the generation of an existing power plant as one lumped model. Data was taken from WESM Luzon single line diagram. For example, if Sual has two machines, the plant is modeled as one unit with the maximum capacity equals the sum of the Pmax of the units. The machines are assumed to operate at 0.95 power factor whether lagging or leading condition. If the dispatch is less than or equal the capacity of one unit, the reactive power must be adjusted too.

Transmission lines:
The line impedances were computed using the table of line parameters from page 68 of Power System Dynamics: Stability and Control book. The distances were determined with Google Earth. Application of plus 5 km. was considered to account for right of way instead of the straight distance measured.

Transformers:
The transformer capacities and voltage level were taken from the WESM single line diagram. The impedances were taken from the appendix of Elements of Power System Analysis book by Stevenson.

Load:
The load allocation at the buses was taken from the document posted at ERC. I added Zapote substation to further zoom into Metro Manila.

The issues I have encountered with this approximate model are:
1. The load power factor is an important parameter for voltage magnitudes. Low power factor will provide low voltages especially at highly loaded buses.

2. The assignment of swing bus brings a concern since two or more plants are lumped at one bus. The pick-up of needed supply or decreasing of output due to excess supply is shared by the plants connected at the swing bus and might cause unrealistic dispatch in real and reactive power.

Further improvement of the model can be the following:

• Validation with the real system performance - I do not know any public document that would give a snapshot of the Luzon grid at any load condition

• Plot the model with the Luzon map

• Include generation cost data

Email me at ebcano@gmail.com if you are interested to have the approximate Luzon network model. You have to download the Powerworld software for this.

Comments on the approximate model are very welcome.

Two things come into mind after I have completed this job - Visayas and Mindanao. ;)