Ask Your Question

Dynamic models of the generating units far away from the area of interest

asked 2020-10-04 13:02:46 -0500

Der Momirmeister gravatar image

updated 2020-10-04 13:03:47 -0500

Hi guys,

I have a raw file which contains over 20000 synchronous generating units and has over 100 000 buses. The raw file is like really really big (like half of USA or Europe).

I should run dynamic simulations (tripping lines & generating units as well as bus/line faults, thus analyzing the voltage and rotor angle stability as well as measuring like 15 critical clearing times) in an power system area which is only a small part of the wide-area system and contains approximately 100 gen. units for which I have a good data for a fine tuning of the GENROU/GENSAL, HYGOV/TGOV1/GAST, SEXS/SCRX models of generators, turbine governors and excitation systems, respectively. The dynamic simulations and the units of interest are only limited to the area I described above.

My questions are:

  1. What data in the dyr should I provide for the generators that are not a topic of interest and will not be analysed? (I don't even have enough data for them to be properly modelled) Maybe GENCLS with a classical H=3, D=0, or maybe, model them as infinity bus by setting H=0, D=0.

  2. Should I just leave such units without any dynamic models? Does that automatically mean that PSS-E will treat them as a negative load (of course represented by I-Y Norton equvivalent).

Thank you very much!

Kind regards,


edit retag flag offensive close merge delete


For item 1, you may want to create an equivalent system outside your study area using EEQV activity and from there, you can represent the equivalent source as a classical generator, GENCLS.

ffl gravatar imageffl ( 2020-10-04 20:19:05 -0500 )edit

For item 2, if no generator dynamic models are provided for units located outside of your area of interest, use activity GNET to convert those units as negative loads.

ffl gravatar imageffl ( 2020-10-04 20:19:57 -0500 )edit

Thank you very much! Helped a lot!

Der Momirmeister gravatar imageDer Momirmeister ( 2020-10-05 04:15:34 -0500 )edit

1 answer

Sort by ยป oldest newest most voted

answered 2020-10-06 10:08:53 -0500

perolofl gravatar image

Your alternative 1:

Model GENCLS is not intended to be used for physical generators. It can be used to model the short Circuit Power and inertia of an equivalent generator, representing the rest of the system (Outer system).

Here the rest of the system has more than 19900 generators and the Study System only have 100 gens. The damper Windings are not represented in GENCLS so it is better to use a real generator model, like GENROU, with standard values, for all generators in the Outer System. You can also add a simple exciter if you want, e.g. SEXS.

Alternative 2:

GNETting all unknown generators is possible. The problem is that 99.5% of the system will be a passive network without any voltage source. It is only the voltage dependency of the loads that stabilises the voltage! There will thereby be no in-feed of short Circuit Power from the Outer System.

Hence, I would recommend to go for alternative 1 and use GENROU instead of GENCLS, especially if bad damping is detected when GENCLS is used.

edit flag offensive delete link more


Thank you very much sir!

Der Momirmeister gravatar imageDer Momirmeister ( 2020-10-09 07:23:14 -0500 )edit

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

[hide preview]

Question Tools

1 follower


Asked: 2020-10-04 13:02:46 -0500

Seen: 68 times

Last updated: Oct 06