Stability of dynamic simulation for the BESS

asked 2023-09-14 05:37:37 -0600

Zhnoob
21 ●9 ●10 ●13

updated 2023-09-15 04:10:13 -0600

Hi, I am modeling an BESS and with one load and one normal generator, using REGCA1 and REECCU model, but after a short disturbance, the system frequency seems not going back to 60Hz, (the non disturbance test will get a flat result), in the output bar there isn't any specific error message, is there any document to check or suggesting in this kind of situation?

Or trying to adjust the parameter according to the transfer function in "MODELS" and the parameter range according to "WECC Type 4 Wind Turbine Generator Model – Phase II" is the right move?

Any suggestion will be much appreciate!

The dynamic parameter is as below:

1,'GENROU',1 6.00 0.05 0.535 0.05 3.33 0.67 1.72 1.66 0.23 0.37 0.21 0.10 1.01 1.02 /

1 'IEEET1' ,1 0.0000 20.000 0.20000 8.535732 -8.535732 1.0000 0.31400 0.63000E-01 0.35000 0.0000 2.8000 0.30338 3.7300 1.2884 /

// governor

1, 'IEESGO', 1, 25.0, 0,1.0,1.0,12.5, 0, 5.0,3.0,0.5, 1.0, 0,/

2,'USRMDL',1,'REGCAU1',101,1,1,14,3,4, 1, 0.02,10.0,0.9,0.5,1.22, 1.2,0.8,0.4,-1.3,0.02,0.7,9999.0,-9999.0,1.0, / Generic RE Converter model

2,'USRMDL',1,'REECCU1',102,0,5,45,7,6, 0,0,1,0,0, -99.0,99.0,0.0,-0.05,0.05,0.0,1.05,-1.05,0.0,0.05, 0.436,-0.436,1.1,0.9,0.0,0.1,0.0,40.0, 0.02,99.0,-99.0,1.0,0.0,1.82, 0.02,0.0,0.0,0.0,0.0,0.0, 0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0, 100.0,1.0,1.0,0.0, /

Electrical Control Model for Utility Scale Battery Energy Storage

Update:

Actually, my goal is to introduce a signal through BESS, and see how the system works, so at first I try to introduce a simple signal like step function, but the system frequency didn't recover to 60Hz as I thought.

So I also tried to do the dynamic simulation with a bus fault at the GENROU side for 0.5 sec then clear the fault, but the system frequency didn't recover also, so I thought the system might be unstable.

Thanks for your reply! really helps me to understand the theory behind.

And another question is, if the frequency reach a stabilise on a new level after primary frequency control, in order to bring the frequency back to 60Hz I need to apply the secondary frequency control by the model like LCFB1 model right?

Link for the figure (google drive):

https://drive.google.com/drive/folders/1lsYtNwQqcZ_bhVABBkAVdDmdp7DIPJjF?usp=sharing (https://drive.google.com/drive/folder...)

edit retag flag offensive close merge delete

Comments

What is the disturbance?

perolofl ( 2023-09-14 06:50:15 -0600 )edit

Did you simulate long enough to get a new steady state after the short circuit?

perolofl ( 2023-09-15 02:07:40 -0600 )edit

In the test with bus fault, the bus tripped at t=2 and cleared fault at t=2.5, run the dyn simulation till t=200s. The link of figure of the test is also posted above, if you are willing to take a look. Another wierd thing is, the active power of the BESS bus will decrease to zero at around t=142s

Zhnoob ( 2023-09-15 04:07:21 -0600 )edit

The BESS goes out of charging at t=142 s! It is stated 100 s battery discharge time. I guess Mbase for BESS is 100 MVA.--> 70 MW output will discharge the battery in 100 / 0.7 = 143 s. Better to plot speed for machine (bus 1)

perolofl ( 2023-09-15 07:04:52 -0600 )edit

See my updated answer!

perolofl ( 2023-09-15 08:31:41 -0600 )edit

add a comment

answered 2023-09-14 07:37:19 -0600

perolofl

3332 ●1 ●8 ●19

updated 2023-09-16 05:01:31 -0600

Most governor models in the library, like IEESGO, are speed droop models, which means a primary frequency control. The purpose of those models is to stop the frequency drop and stabilise the frequency on a new level. The frequency will not be brought back to 50/60 Hz by those models.

A generator trip will lead to permanent underfrequency and vice versa at load trip or increased power injection.

So, there is nothing wrong with your results (I assume you are changing the power injection from BESS during the simulation).

EDIT:

There is a fatal problem with the governor model used. Where did you find it?

With the chosen parameter that model (IEESGO) is unstable when using long simulation times. I recommend to use a stable model, like TGOV1, with the following parameters:

    R         T1       VMAX      VMIN       T2        T3        DT
  0.050     0.500     1.000     0.000     2.000     7.000     0.000

EDIT 2:

I found some typical parameters for IEESGO: (https://www.phasetophase.nl/pdf/Synch...):

    T1    T2     T3    T4    T5    T6    K1    K2    K3   PMAX  PMIN
   0.20  0.00  0.100 0.250  7.00  0.40  20.0  0.40  0.30  0.95  0.00

Here I assume cos-phi 0.95, hence PMAX=0.95.

edit flag offensive delete link

Comments

Really appreciate! ! Now both the scenarios I want to simulate are working fine, would you mind guiding me how to find the fatal problem with the governor model I am using, so that I don't run into the same problem in the future. Thanks! ! (I got these parameters from a IEEE 9 bus example).

Zhnoob ( 2023-09-16 00:39:42 -0600 )edit

The problem is parameter K2 and K3 which are too high, and too long time constants.

perolofl ( 2023-09-16 04:56:34 -0600 )edit

add a comment

Stability of dynamic simulation for the BESS

Comments

1 answer

Comments

Your Answer

Question Tools

Stats

Related questions