PSS/E Dynamics Simulation Problem

[boregard]

Hi,
I am new to dynamics. I am currently trying to run a dynamics simulation on a 17 generator system, 165 bus system. Activity STRT has confirmed that that my initial conditions have checked out o.k. However, when I RUN the simulation, the solution diverges (see below). Any help will surely be appreciated..Boregard

TEXT, RUN FOR 20 SECONDS PRINT 0 PLOT 5 CRTPLT 0
RUN
19.99 0 5 0

NETWORK NOT CONVERGED AT TIME = 0.0417
401 10020.453 60 [SX FLL7 115.00] -0.5008E+00 -0.1344E-01

NETWORK NOT CONVERGED AT TIME = 0.0458
401 16097.673 60 [SX FLL7 115.00] -0.8035E+00 0.4676E-01

NETWORK NOT CONVERGED AT TIME = 0.0708
401 65058.355 60 [SX FLL7 115.00] -0.3195E+01 -0.6094E+00

NETWORK NOT CONVERGED AT TIME = 0.0750
401 ********* 60 [SX FLL7 115.00] 0.1832E+02 0.3111E+02

NETWORK NOT CONVERGED AT TIME = 0.0792
401 ********* 64 [SIOXLS 345.00] 0.7690E+01 -0.1035E+02

NETWORK NOT CONVERGED AT TIME = 0.0833
401 99970.383 60 [SX FLL7 115.00] -0.4686E+01 0.1739E+01

NETWORK NOT CONVERGED AT TIME = 0.0875
401 ********* 60 [SX FLL7 115.00] -0.2040E+02 -0.2130E+02

NETWORK NOT CONVERGED AT TIME = 0.0917
401 ********* 64 [SIOXLS 345.00] 0.6048E+01 0.3197E+02

 

[dpc]

Who knows? It could be probably 100 different things. One of the terms is obviously blowing up. First thing to check is the data in the model around this node.

Things to check:

Transformer ratios, size, impedance, etc
Units
Cable/T-line lengths (very short branches are sometimes a problem)

Check all your input data for obvious errors.


David Castor

http://www.cvoes.com

 

[tlrols]

Do you have the PSS/E documentation? Read it and follow the instructions to the letter. From your output it would appear that the case is divergent. Did you run FACT, TYSL, CONG, CONL, etc? You MUST perform those activities prior to running the transient stability program.

Frankly, the PTI transient stability program is a pain to use. It is unnecessarily complex and requires the user to perform numerical housekeeping chores that may have been appropriate for the 1970's, but are most certainly NOT appropriate for the 21st century!

Check that the power flow case converges first. If it does, then this case is good to go. You CONG, CONL, FACT, TYSL, and then you should be able to "flat-line" the dynamic simulation to test for numerical stability.

I think you missed a step in your running of the software.

My honest advice: Use the GE PSLF/PSDS software instead...or wait a few months and use PowerWorld Simulator version 15. Life is too short to waste on clanky software that isn't a mainstream product for the company that sells it. The GE software certainly isn't a GE mainstream product but it clanks less than PSS/E (since it was written by unhappy, former PTI employees).

 

[boregard]

Hi,
Thanks for your input. Just to give you a little background, I started with a solved power flow case and then proceeded with the following activities:
GNET
CONL
CONG
ORDR
FACT
TYSL

The tysl produced a well solved case, so I'm assuming my data is good. Activity STRT gives me good initial conditions prior to the start of my simulation. My original post is from the output of a flat line simulation (no contingency). I know there are better software packages bus I'm not able to use any other software but the PSSE.
Your help is appreciated.
Boregard