perolofl's profile - activity

Missing sequence data shouldn't cause PSSE to crash. It must be a fatal error leading to division with 0.

See post "What is the correct way to finish a simulation"

Google swing bus or slack bus.

We are not dealing with a real generator, it is a swing bus! A swing bus has no boundaries.

Select the line, Right-click, Item Properties > Line Style / Color

Check that there is no generator with Mbase=0. That will crash PSSE when performing ASCC.

Forget your "solution".

Use this to define the variables for the default values:

import psspy
_i=psspy.getdefaultint()
_f=psspy.getdefaultreal()
_s=psspy.getdefaultchar()
psspy.branch_chng(frombusNum,tobusNum,r"""cktID""",[_i,_i,_i,_i,_i,_i],[_f,_f,_f, newRatingAfloat,newRatingBfloat,newRatingCfloat,_f,_f,_f,_f,_f,_f,_f,_f,_f])

The active power generation at the swing bus becomes negative if the total generation of type 2 generator buses is larger than the total load + losses.

You have to decrease the generation at some other generators.

This is really basics for load flow calculation... I recommend to study the theory of load flow calculation prior to use PSSE.

Right-click in the grey background in the toolbar section and activate toolbar "Diagram Primitives".

The syntax for the DYR-file is described for each model in manual "Model Library"

Use tool "Locate bus" in toolbar.

Enter the bus number (or Select and Filter to select a bus name)

PSSE will zoom into the selected bus if it is present in the diagram.

Is PGEN at bus 7 changed? Is it the same PGEN and QGEN at bus 13 after changing to wind generator?

It seems like the change to wind generator changed the load flow and other generators became overloaded in reactive power.

Is bus 7 swing bus? Bus 7 has a PGEN outside the power limits in model IEESGO.

Did you replace generator at bus 13? It seems like the generator is overloaded.

There may be warnings even for proper parameters, so the warnings are only an indication.

The limits for warnings are described in Program Application Guide, Vol II, Chapter 26.5. DOCU,CHECK gives the following report:

 Model SEXS Bus 3011 [MINE_G      13.800] Machine "1 " :
 TB = 4.0, outside of range > 5.0 and < 20.0
 K = 200.0, outside of range > 20.0 and <= 100.0
 TA/TB=     0.2500   K=       200.0000   TRANSIENT GAIN=    50.0000
 EMIN = -4.6, not = 0.0

                    C O N S       S T A T E S
                     133-138          50-51

                TA/TB     TB      K       TE     EMIN   EMAX
                0.250   4.000   200.0   0.020   -4.60   6.00

The used exciter has a very short exciter time constant (20 ms), which means that the block is ignored in a 50 Hz system with simulation step 10 ms.

The transient gain is 50 which is high, resulting in a fast response in field voltage.

See posts "How to model multiple infinite buses?" and "External equivalent (impedance source)".

The inertia constant has no effect in load flow calculation.

Normally you select a large generator as swing bus. However, any bus can be selected as swing, there is no need for a generator at the swing bus.

The most important is that the swing bus is electrically strong and is able to absorb any active power imbalance created when tripping generators, loads, HVDC lines, etc or after scaling total load/generation in the whole network or a subsystem.

A swing bus at high voltage level (e.g. 400 or 500 kV) is stronger than a swing bus at generator level (15-25 kV).

When performing voltage stability studies it is also good to locate the swing bus outside of the studied area, since the swing bus has no reactive power limits.

Slider diagrams are not available when running PSSE from python, i.e. when PSSE is started by a python script.

However, if you start PSSE manually and run a script from the GUI ("PSSE runs python"), then sld can be opened and exported.

Normal time step is 0.01 s for 50 Hz systems and 0.008333 s for 60 Hz systems. You may have a problem with too short time constants in some dynamic models.

In PSSPLT it is possible to save an outfile in text format with activity RAWC (in File menu) and also to create a binary outfile from a text file!

• Enter the binary file (outfile) and the text file in RAWC and select option "Read Binary Channel File and Create Source File".
• Edit the text file with channel data.
• Enter the edited text file and a new name for the outfile and select option "Read Source Channel File and Create Binary File".

The edited outfile can then be plotted with PSSPLT, Plotbook or with matplotlib. This outfile editing doesn't need any coding, just simple usage of PSSPLT activities and a text editor.

Use this in your script:

import psspy
_i = psspy.getdefaultint()
_f = psspy.getdefaultreal()
_s = psspy.getdefaultchar()

It depends on the dynamic model used. Most modern models have all parameters in pu of Mbase, so the dynamic model will work for any MW size of the farm.

Some vendor specific models may have the rating as a CON-parameter. You have to the check the model's datasheet if the rating is a parameter in the model.

The problem is that DYRE don't clear the channel data when a dynamic raw data file is read into PSSE. This means that any previously defined channels are kept after redoing DYRE.

In your case every simulation starts with loading the case, reading DYR-fil and adding channels. You have to clear the channel memory before adding channels. This is done with API:

psspy.delete_all_plot_channels()

Yes, it is correct. The fault is represented with a shunt of 62 MW and -268 Mvar. Plot the bus voltage and you will see the voltage drop during the short circuit.

REGCAU1 is called as a user written model in PSSE so you have to add the first instance of the model with DYRE,ADD.

See the model manual for the DYRE record for the model.

After the first model has been added in the snapshot, the model will appear when doing "Add/Replace Model" in Wind Machine spreadsheet.

It is just to change the load in "Load" spreadsheet.

Why not just skip those lines when reading the file?

Split the faulted bus in saved case and add a MVA channel for the new ZIL between the buses. Calculate and plot the fault current with method 1 described in post "Monitor branch current in dynamic study".

Do you mean to plot the short circuit current in a dynamic simulation?

The SCR is not calculated, but activity ASCC reports the short circuit power at the bus for three-phase faults. From that you can calculate SCR.

The shunt is removed at fault clearing!!! Check the content of the progress window.

You can load a dll into PSSE with Load Model library.

rev 33: Tools > Load Model library...

rev 35: Dynamics > Model Library > Load Model library...

In python:

psspy.addmodellibrary(dll-file)

does help(dyntools) work for you?

Use Bus Subsystem Selector (Ctrl+Shift+U) in the GUI to see how a bus subsystem is defined. A bus subsystem may be a combination of area, zone, owner, kV and buses.

API subs defines a subsystem and assigns it to a number, SID, between 0 and 11.

Use SID=-1 to convert all loads in the system. Use subs if the load conversion is not the same for all buses in the network.

The shunt represents the unbalanced faults. Just continue the simulation!

Try to enter load channels for motor load and save a temporary snapshot after simulation.

Which R and X do not have values? Resistance and reactance of L-G fault impedance shall be 0 for solid faults.Is a shunt added at the faulted bus when you click OK?

Use module dyntools to convert the file to CSV-format.

import dyntools
output_obj = dyntools.CHNF('outfile.out')
output_obj.csvout(channels=[])

This will export all channels to CSV-file. Select specific channels with argument channels. e.g. channel 4, 5 and 6 with channels = [4,5,6]. See also help documentation for dyntools.

Is the initialisation (STRT) ok without suspect STATES? There areany messages about missing models? You write that you cannot load .dlls.

This is not an error!

It is just a message that the DC rectifier hits the minimum alpha angle limit and cannot maintain the power order.

The reason is that the simulated disturbance leads to decreased AC voltage at the rectifier side of the DC line. The model is behaving correctly.

The best way to find the API for data changing in PSSE is to do the following.

1. Start recording of a python script.

2. Apply the change manually via PSSE GUI.

3. Stop recording.

4. Open the created python script in a text editor and check the content. You will then see the API with its arguments.

How is the SVC modelled in load flow?

You can save the case at any time when performing a dynamic simulation. Just remember to use a new file name so you don’t destroy the original saved case.

This temporary case cannot be used as a start of a new simulation, i.e. don’t initialise a new simulation from that case.

There is no critical fault clearing time if the generators maintain synchronism during and after the fault.

Exactly, this shows that there are fault locations where no generator will go out of step. It is best seen in the rotor angles but also in speed. Nothing is wrong with the results.

Yes, there is!

psspy.getmodellibraries()

returns a list with all loaded libraries.

Maybe you are isolating the generator bus from the rest of the system when you trip a line (or transformer?) at fault clearing? Or you are splitting the system into two separate systems?

EDIT: Maybe you are not doing anything wrong. It is a small test system and it might be that all generators are accelerating similarly for some fault locations and synchronism is maintained even for long fault clearing times. Plot the speed deviation for all generators and see if they are following each other.

What kind of dynamic models are you using? Usage of CONEC and CONET models is discouraged.

PSSPLT is a part of PSSE package so you should find it in the installation folder.