Recloser control system

[johndaniel]

Hello guys,

I am working on distribution system where I have to develop a recloser control model in PSCAD. Although I have basic knowledge of recloser system but I couldnt find the control and operation of recloser system which I can emulate in PSCAD.

Any suggestions would be helpful to me.

 

[Mbrooke]

Basically in a recloser loop scheme with simple non communicating controllers you have 2 elements to worry about: Time current curves and time-phase loss logic. That means in any reclsoer loop scheme all reclosers have voltage PT inputs in addition to your normal CT inputs.


First you coordinate time current curves. Lets assume load is even on all portions of the feeder (we will say 4,000kva peak {145 amps} per portion at 16kv to keep it simple), cold and hot load inrush are about the same as your peak load in this example (normal inverse time current curves will be ok with it), substation neutral is solidly grounded and you wont be coordinating with any lateral devices. Largest transformer fuse is a 20T. Your first recloser, the source recloser may have a phase pickup value of say 800amps. Ground values are typically 1/3 to 2/3 that for a multi grounded neutral system having phase to neutral loads. If everything is connected phase to phase your values will be much lower {sometimes just a few amps if coordinating with nothing} (very good idea in the event of a high impedance fault such as a wire on dry soil). In this system we will assume all delta loads, however, because of our 20T fuse, we would like that to blow before any recloser during an internal bolted fault. So we will have a ground value of 1/4 the phase value on each recloser and that curve will be the same as the phase current to keep it simple in this example, as well as the fact the phase curve is typically far 'slower' (to accommodate load inrush) than the fuse which allows for coordination at all times, even on the very last recloser.

So 800amps phase, 200amps ground on your source recloser. A cooper 142 trip curve might be selected here for both, adjusted as needed.

Your midpoint recloser will have another set of trip curves, one set triggered only by a "voltage lock out". Here we will say 650 amps phase 200 amps ground for the normal set. The alternate trip curve set will be 250 amps phase 65 amps ground. Because no other reclosers will coordinate with this alternate pair, only load, the trip curve may even resemble that of a fuse such as a 150T or as low as load permits. A cooper 163 trip curve comes close to a T or K fuse link for example. Of course ground trip here must still coordinate with your largest transformer fuse, a 20T.

Your tie recloser can have a pick up of 400 amps, and a ground of 120 amps in both directions. This point will be open, only triggered by a loss of voltage on all 3 PTs. I recommended the control be configured for all 3 PTs, as the Cooper manual recommends since you only want to be closing on a truly de-energized line. If one had different phase angles of even frequency variants, a blown fuse on one VT will force closing which wont look to good.


Also note that fault current drops off considerably after the substation. Since reclosers tend to be blocks apart each one will see less short circuit current than its prior. This can help because if say time current curves loose coordination after a certain range (which is normal especially for fuses) you know that current wont exceed during a worse case senerio you can ignore that part and call it good. Example, if trip curves on the source and midpoint recloser loose sequence after 6000 amps, and if we know line impedance drops that current to 4000 amps in front of the midpoint recloser we can ignore what happens after 4000 amps on both.

This is more close to a typical text book case, actual scenarios will vary considerably, but its good starting point.

2nd part is timming of voltage which is easy. The maximum amount of time each device will trip-close-lockout on current will determine what the next device is set at voltage lockout wise. In this case we will assume the substation breaker will complete a full trip-close-trip-close-lockout on a hard fault within 45 seconds. We will then set the source recloser to lock out (open) on no voltage after 60 seconds. Assuming the source recloser takes about the same time for a current fault sequence, 75 seconds time is good for the midpoint to reverse its trip curve. Assuming the midpoint has a similar time (will say 30 seconds to lock out on over current) we will give the tie recloser a 90 second time period. This will work well in that it will allow all devices to complete a fault clearing sequence. Ie, if the source recloser opened before the substation breaker closed into a temporary fault, loop restoration would be enabled without need. If the tie recloser closed before the midpoint or source had a chance to open on a permanent fault or under voltage, the current may back feed into the faulted segment tripping the tie, defeating outage restoration on healthy lines.

Since the scheme is none communicating keep in mind that the tie and mid point should not have a reclose function enabled 1 to 2 minutes after closing or the midpoint reversing trip curves. In the event the fault is between the source and midpoint you would want the midpoint to open and stay open. Similarly if the fault is between the tie and the midpoint you would want the tie to open after closing into the faulted segment.


This of course is typical for non communicating reclosers which is rarely done to day. Years ago this method helped reduce outages but it had many flaws since devices are unaware of why they are responding to an under volatage or over current. SCADA however allows reclosers to talk to one another. Many reclosers can be parelled into multiple loops or meshes, under any condition without loosing sequence. Thermal stress is drastically reduced since if a midpoint recloser opens it will send a signal to a tie not to reclose; similarly if a source recloser locks out on an overcurrent it will send a signal out so the midpoint opens before the tie closes since its now known you will be energizing a downed line. SCADA also alerts dispatch an outage has taken place rather than customers having to call in. Once the repair is made the loop can be put back to normal within seconds rather long outages having to go around and resetting each controller.

Hope this helps.