Island Utility HV Capacitor Reactive Compensation

[fb599]

We have an island utility system of nominal load of 45MW. We are also connected via a 100km 90 kV AC cable link. During the summer months we do not run any generation, therefore we are hanging on this long length of cable. We use shunt/series reactors so that we balance out the capacitive reactance of the link. However, during light load periods (ie night time) we have to switch in 11 kV capacitors to balance reactance and nominal system voltage back to 11 kV. My Question is this, when we switch in the cap's, which are 3.15 MVAR and which in addition are connected in series with inrush reactors of 0.05mH the 11kV network suffers from large voltage transients. My phone hasn't stopped ringing since we went solo on the link ! We have tried to reduce sys. volts before closure and raise volts prior to opening the Cap. circuits but the effect obviously is still there for all to see. Is there a way of calculating the size of inrush reactor to reduce this voltage transient or do I need to rethink the whole VAR compensation issue with regard to HV Capacitors. FYI we have 6 of these units based at the power plant. Any help gratefully received. Otherwise I can't wait till winter when we start generating in parrallel again !.......

 

[avishai]

Your thinking is not wrong your swiching is .If instead of conventional swiching you use HV Solid State Swiches meaning you swich the capacitors at zero crossing then you solve the problem .If the capacitors are stayng swiched on for long time then you should also think of a parallel realy for after swiching connection . A.Rash BScEE & PE
a_rash@bigfoot.com

 

[NormGA]

Are you talking about a temporary transient such as harmonics that you often get when switching capacitors or a jump in RMS voltage that is unacceptable? If the RMS voltage is going too high the size of the cap bank is too great for the size of the transformer and other system variables. It has been my experience that the voltage should not increase more than about one or two percent, preferrably one, when switching capacitors on line. More and you will have problems. Years ago we had some 12kvcapacitor banks that were switched in 4 stages to avoid the jump in voltage. Each stage was limited to about 1200 kvar.

The previous poster has a good idea, but my guess is the price is steep. A static var compensator would work also, but again, very pricey. You could use smaller banks on your distribution feeders to switch in smaller chunks, but control could be a challenge.

You might get more input if you can give more info, like what size transformer is this, system load when switching caps on, method of control, etc.

If you are talking temporary transients you might need a different type of switch. How are you switching the caps by the way? We have had problems with 115kv caps causing some very damaging transients when the switching power circuit breaker had defective pre-insertion resistors.

 

[fb599]

Hi NormGA

Interesting thoughts, I like your idea regarding installation of smaller Caps on the distribution network, as the CB's I'm using on the main 11kV system at the Power Station are a little pricey. You are also right in the assumtion of the RMS increase in voltage. It is very noticable when you are at home and the station switches a 3MVA Capacitor in. We have 6 of these units around the site. The increase in volts is around 4-8% on closure of each CB but the capacitive reactance is necessary to balance the system when we are lightly loaded, at say 25MW. The problem here I think is that in theory we are connected to an infinite bus but the reality is it is 100km till we get to an infinite system....Ideally it would be beneficial to split the cap banks into smaller lumps but the cost would be very high. I wondered if the sizing of the inrush reactors could be changed to "tune" each bank ?

Thanks also to Avishai, that's a new one on me I wonder where you can buy Solid State switches in EU ?

Regards fb 599

 

[NormGA]

If your voltage increases as much as 4 to 8 percent I would think you are putting on too many capacitors at one time. I don't think the inrush inductors have much to do with it, since their purpose is to limit the current inrush on energizing. As a rule, anything over a 5% increase in voltage is noticeable. Capacitors on the end of a long line generally lead to high voltage. When lightly loaded at 25 MW, what is your reactive load?

Take a look at the following link and see if this sounds like what you are looking for. I have some field experience with an ABB static var compensator and I believe it might solve your problem. It divides a capacitor bank into many small groups and switches them on one group at a time at extremely high speed. The unit I am familiar with operates on the 600 volt secondary of an appropriately sized 46/.6 kv transformer. Capacitor switching is done by hockey puck size SCR's. Control of the unit is either current or voltage. In your case, voltage control would tend to stabilize the bus voltage. You might try using the

http://www.google.com

search engine and look for static var compensator for more info. There may be other solutions worthy of consideration.

http://www.abb.com/global/abbzh/abb...d&e=us&m=V&c=E85D0745A26CE1A1C12568780037DD00

 

[fb599]

Hi again NormGA

I've checked up on last nights activities and the capacitors are generally switched in when our load is light ie 30ish MW as the load is increasing with the morning peak. We generally switch in at say approx 30 MW and around 4.5MVA lagging VARs this is the general approach. Very late at night as the load drops off the Caps are opened. This is typically when we have around .6 - 1.0 MVAR leading. I hope i did not mislead in the start of my thread ! There are shunt reactors on the 90 kV link system which balances out the VAR situation as best as possible. Thanks for your guidance on the links, I'll certainly have a look at those.
Regards fb 599