When windmill generators will stop generating?

[lz5pl]

Our test engineer has the following case: Windfarm is connected to the grid by one 110 kV line. There is a requirement by the utility to trip 110 kV side CB when the line is switched-off from the remote end. Criterion is absence of all three currents and three voltages after predefined time delay. Point is how long this time delay should be. Windfarm consists of asynchronous generators, rated up to 1000 kW each. After how long generators will stop generating voltage after excitation from the grid disappear? Frankly speaking I have no experience with such generators, so any advice will be helpful.
By the way, setting values are not our responsibility, but I would like to clear this matter for myself.

------------------------
It may be like this in theory and practice, but in real life it is completely different.
The favourite sentence of my army sergeant

 

[slavag]

Plamen.
Sorry, but I not understand what do you meant
"Criterion is absence of all three currents and three voltages after predefined time delay. "
absence current is OK, but voltage you have all time.
You need some decoupling system: ROCOF, maybe low forward power, U1<.
And goodest option is some connection between remote end and generatoe end ( but I assume, isn't possible).
Best Regards.
Slava

 

[slavag]

Of bad. It's wind asynchro generator, sorry.
Need think, Are you have undervoltage protection for generator? what is a setting? What is protection used for the 110kV, what is a setting?

 

[itsmoked]

A few seconds are all you need to drop the generating of induction motors. UNLESS you have capacitors in the picture.

So unless you need something better than "a few seconds" that should do.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[slavag]

Keith, few seconds is not so good seeting. For example if this line have autorecloser Please pay attention all 3 voltages and currents. That mean, Plamen, you need combination of undercurrent protection with logical three pfase undervoltage protection. I think in all cases ROCOF with undervoltage provide good option for this protection, without current criterion.
Regards.
slava

 

[lz5pl]

Sorry slavaq, may be not good explanation from my side.
We are asked by the utility to create CFC (7SA6-relays) with the following logic: all three currents below some value, all three phase voltages below some value, and if all these conditions lasts more then 0,3 s then to trip CB. The idea is this way to recognize tripping of the line from remote end. When remote end trip happens, voltage from the grid infeed will disappear and generators should loose excitation and stop generating. But how long time they will need? And if they are many and in parallel, may be they will excite each other?
According to me the proposed criterion is not the best one. You are right, low forward power will be better, but I should discuss it with the customer. We did CFC as they require, but are not sure if this logic will work.

"Goodest option" you suggest will be available after some time - OPGW on the line. But the customer needs the line now and some temporary solution is necessary to ensure no voltage supply from windmill side during the dead time of AR.

------------------------
It may be like this in theory and practice, but in real life it is completely different.
The favourite sentence of my army sergeant

 

[slavag]

OK, Plamen, I understand you.
But you need undercurrent function, not some meas level of current, it's serious difference. With undervoltage for example 0.8Un for 0.3 sec.
Hmm, I think it's work without problem. But you need check it on the site. Very important, what is a dead time of AR, 0.6-0.8sec?
I think again about it.
Best Regards.
Slava

 

[lz5pl]

Slava, when do you sleep ! For Keith it is clear, it should be morning in CA, but I have an impression about you that you are in Europe !
Seriously speaking: Thank you for quick replies. Unfortunately I have no complete picture for the site. It is not our project, we were called to just test relay protections of the step-up transformer and 110 kV line. We did everything according to the given values, but we have some doubts how well that logic will work. And from your posts I see that these doubts have some reason. I will contact colleagues from the utility tomorrow (sorry, today!) and we will see what to do. Probably one more trip to the site (500 km, sorry) will be necessary.
Good night for Bulgaria.

------------------------
It may be like this in theory and practice, but in real life it is completely different.
The favourite sentence of my army sergeant

 

[slavag]

Plamen, sorry, forgot something.
Are line is dedicated or you have additional load on the generatos, on LV side or on 110kV side?

 

[lz5pl]

It is existing line between two substations. Windmill substation is connected as T-junction. As far as I know, there is no additional load on 20 kV side of windmill's switchgear.

According to me this logic is not so much necessary. They need to be sure that AR from remote end will not "meet" voltage from the windmill side. But in case of failure on the line our CB also will trip without any additional logics.

------------------------
It may be like this in theory and practice, but in real life it is completely different.
The favourite sentence of my army sergeant

 

[waross]

You may wish to consider monitoring the frequency. If the wind generators lose their load, the frequency will rise.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[lz5pl]

Oooops, that 0,3 s time delay is related to our site, important is dead time of the AR. Slava pointed it out p, but I missed it. When I wake up this morning my first thought was this. That's why people say: The morning is more clever than the night!


------------------------
It may be like this in theory and practice, but in real life it is completely different.
The favourite sentence of my army sergeant

 

[Marmite]

Normally the power factor correction capacitors are located in the base of the turbine, which leaves open the possibility of self excitation. I would expect to see overfrequency and over/under voltage protection, also ROCOF or voltage vector shift relays to prevent islanding and/or out of phase autoreclosure. I can't see anything that relies on current or power working correctly as wind turbines only generate 30% of the time and are not a reliable source of current.
Regards
Marmite

 

[slavag]

Yes. Marmite, you are right. In this case only ROCOF and overfrequency ( I mean df/dt> and f> logic). 0.3sec is short time for the another criterions.
For other situation possible build logic on the voltage/current criterions.
Voltage shift, I don't know, I don't like this function, but many companies used it w/o any problem.
Plamen, I strongly recommend check this on the site, in two situation, full load and minimum load or zero load, may be
you will need use two settings of protection and change them according to load.
If will possible, could you please send us DR from relay.
Best Regards.
Slava

 

[davidbeach]

Why not use a low forward power relay - there is very little power in the line charging current - and trip for low forward power. Once tripped, block reclose for dead line. When reclosing into a hot line, block the low forward power relay until after it has dropped out (enough forward power to indicate that the windmills have connected) and then it is ready to go the next time.

 

[slavag]

David, my answer is very simple. I don't know.
I don't have any expirience with windfarm.
I don't know, maybe low forward power is standard situation.
For single steam turbine we used low forward, it's work fine, for group of windgenerator is ???.
ROCOF is always good solution, except few options.
Regards.
Slava

 

[bacon4life]

Are there line side PTs at the utility end that could block reclosing until the voltage had decayed?

 

[slavag]

Good point Bacon.
But usually, its operational problem.
This windfarm connected as T connection, that means utilities must back line to operation with AR wo synchrocheck. Utilities request from the privet power plants
in this case trip wo any conditons from utilities side.
Please see situation:
Fault on the 110kV line, CB utilities opened and AR try back this line to operation, windfarm not disconnected and AR not back line to operation becouse synchrocheck, now you need manually dissconnect windfarm and back 110kV line to operation. Utilities never agree with this option.
Regards.
Slava

 

[cranky108]

Low forward power probally won't work as some wind farms may remain energized whild having no wind.
I would think frequency is a good indication, except the ride through on some wind units is as low as 0.1 pu voltage which is much lower than the normaly allowed frequency relay responce.
Sence the output of a wind unit into a fault can be the same as into a load, and the voltage should be low as the wind units would have to support the stepup transformers, it might be worth consitering tripping on var export on each wind unit. Or maybe just a simple distance rely.