Reverse power protection with two separate feeders 4

[123MB]

Hi All, hope you’re all well!

I am working on a customer site with lots of behind the meter generation. Power is from the supply authority with 2-off transformers, each feeding a separate main switchboard and sub boards.

The supply authority requires us to limit the total export real power in kW to a set figure. How would you accomplish this? Because the site has two separate supplies, the active power for each needs to be cal calculated and summed, and the result used in some logic to trip the breaker. Key requirement is that the trip is ‘protection grade’ and has a 400ms response time.

The options I have though of:

1) a PLC which reads the power data out of the power meters with Modbus TCP and sends a trip command via digital output. But would this be fast enough?

1) somehow providing IEC 61850 IEDs to measure the power (ie 61850 power meters) and a 61850 programmable IED ie an RTU to do the trip logic and issue a command via digital output

Thoughts?

 

[stevenal]

Modbus would be too slow I believe.
Mirror bits from SEL work at protection speed.

 

[catserveng]

Not sure where you are at, but in the areas I work in with Rule 21 interconect agreements PLC's are specifically NOT allowed for that kind of protection.

I'll second the SEL mirrored bits comms and properly configured logic, I work on a couple sites with similar issues and the installed solution is SEL 751 relays with mirrored bits comms between the relays. If you're needing to manage more than two relays in your system to make it work properly an SEL 2100 Logic processor can be used and still supply very fast response times.

Hope that helps, MikeL

 

[123MB]

Thanks all for fast responses!

I have edited the above from Modbus RTU to Modbus TCP if that makes any difference.

Is it a general rule that Modbus comm’s aren’t used for protection level trip signals? I don’t have a lot of experience in this area.

Thanks!

 

[FreddyNurk]

Modbus not only will be too slow, but is non-deterministic and unable to guarantee that any sort of signals will be received when required, hence its lack of use for such applications.
There may be scope to approach the utility and find out if there's a set export limit on either supply, which would reduce the need for summation. The other thing to consider is whether the utility allows for tripping the onsite generation, or whether exceeding the export limit needs to drop the entire site.
Otherwise as suggested the SEL equipment would do the job.
ComAp may have something that has IEC 60255 or equivalent protection approvals and also allows for regulation of the generator plant so it doesn't trigger a trip, but your integration task may be exponentially harder depending on your generation.


EDMS Australia

 

[waross]

A suggestion for a kW back feed limiting system:
Install a kW transducer or transmitter.
Install CTs on both incoming lines and run them in parallel.
Caution: Be sure that the current input of the transducer/transmitter is connected to both sets of CTs before energizing the system.
Use the output of the kW transmitter to instruct the load controller to reduce the kW output of the generators.
I have used more than one CT to energize meter elements.
I was skeptical at first and used a single phase meter on one phase to check for gross errors.
It works.
The current element of a relay or meter has a very low impedance compared to the effective impedance of an energized CT.
Faced with a choice of sending current through a meter element or forcing the current through another CT, the line of least resistance is to send the current through the metering element.

There is an accepted metering scheme that uses CT forcing current through the metering elements of two other CTs to meter the third phase of a four wire circuit using a two element meter.
The accuracy is claimed to be suitable for revenue service.
The circuit makes the assumption that the voltage on the unmetered phase is equal to the average of the other two phases.

Metering the outgoing kW directly is so much easier, quicker and dependable than trying to sum the kW outputs with a computer.

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

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

 

[123MB]

Thanks all

I forgot to mention the system needs to be IEC 60255 compliant. Unfortunately I don’t have an appreciation of this. I’ve read as much of it as I can find but I don’t have a full appreciation.

Waross, thanks for your response. I considered something similar with summation CTs, but I didn’t think this would work since any protection relay requires the phase voltages in order to calculate the unbalanced three phase power. Your solution would work for a current limit but would it work with power?

 

[waross]

The limitation would be that one system would meter properly.
The second system would assume that the voltage was the same as the voltage of the other system.
Why do you want to trip the breakers instead of reducing the throttle settings?
How much power are you allowed to export?
You may be able to use reverse power relays to trip the sets off if the export power limit is fairly low.
We have to know if the limit is a sum or if it is an individual limit for each service.

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

 

[123MB]

Thanks waross

The export limit is 550kW. So the total real power across the two feeds needs to be less than -550kW. So it is a sum.

The generators aren’t conventional, they can’t be throttled.

For the export limit to be exceeded, other things would have to go wrong. This is a form of backup protection, so on the balance of probability it will happen we are happy to trip the breaker.

I looked into the SEL-751 and since it can’t do analog readings via MIRROREDBITS, the only option would be to send the power value from one feeder relay to the other with binary coding and do logic to decode. Messy and I don’t know if it will be possible as I’ve never used these devices.

I’d prefer not to rely on the voltage and phase angles at the two transformers being identical at this point. They’re identical transformers but they have very different loads and they aren’t supplied from the same HV feeder.

 

[FreddyNurk]

IEC-60255 is a suite of protection equipment standards, including input withstand, shock testing and so on.
The SEL products will very likely be 60255 compliant. 60255 compliance will also eliminate any PLC based scheme that you may come up with.

Effectively it means you can only use equipment that is designed and tested to comply with relevant standards for equipment used for network protection, which in my opinion is not a bad thing.
Other stuff to be considerate of if the utility has to approve your scheme will include the interface to the relevant breakers to trip, and whether or not compliant devices are used for any interposing relays or trip coil interfaces.

If the utility permits it I'd trip the generation sources rather than incoming breakers. SEL also has a great product that allows for remote tripping of breakers using a fibre optic interposing link, and also 60255 compliant, which may permit tripping the generation sources.

SEL is not your only option, although their logic processor platform is likely to be able to do what you want, perhaps with a little OEM support. Generator control OEMs (ComAp, Woodward, Deif) will likely have equipment that may do what you want, including relevant 60255 compliance, but their approach is slightly different to a relay OEM.


EDMS Australia

 

[catserveng]

Freddy,

I think you pretty much nailed it, and distributed generation, mostly with RICE units is a major part of my living.

I will add that I have, in working thru a number of utility interconnected projects that generation controllers, like Woodward MSLC2/DSLC2 systems, even though they would do what was wanted could not get acceptance by the utility protection engineers, in most cases we resorted to using an SEL relay based solution, but have also worked projects using other protection system products, such as ABB. Recently had a similar situation with a ComAp based system, that while it worked, utility wanted separate "approved" protection relaying to manage the protection functions.

To the OP, you have asked about a pretty specific problem without a lot of details to get you a really good answer, like where you are at, what kind of generation are you dealing with, who your utility is, etc.

It sounds like you are possibly dealing with a renewable or possible fuel cell generation, based on your comment that they can't be "throttled". But that is just assumption on my part.

I wouldn't call the SEL programming required "messy", but what you are trying to do in the time required by the standard is a bit complicated, however you're certainly not the first engineer faced with this problem, so you may need to find someone who as actually done it before.

MikeL

 

[waross]

A reverse power relay doesn't much care about phase angles or PF.
A reverse power relay will sum the kW and ignore the PF.
Are both feeds the same voltage?
Are any variations in voltage with an acceptable window of error?
Are both feeds from the same source?
Is your back feed limit kW or KVA?
Using a reverse power relay to trip a generator is old and proven technology.
I haven't installed a reverse power relay for about 20 years and haven't had occasion to look at the specs since then.
I don't remember the setting range.
If you need more range you can fool the relay by using oversized CTs.

Here is a relay with a 100 ms response time.
Single phase relays are generally adequate for generator protection.
You may use three relays, one per phase, if you may experience unbalanced reverse power.

Link to Web Page


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

 

[123MB]

Thanks all

Apologies for not giving some information, some of it is proprietary info.

The project is in Australia. The generators are essentially capacitor banks that are being discharged as part of manufacturing testing.

Hi waross, the export limit is in kW. The voltages are probably similar but not the same. They are two identical trafos on different HV feeders and with different loads. Both feeds are not from the same source.

The issue with applying an off the shelf relay is that it won’t be able to calculate power correctly as the voltages and PF will be different. At this stage I don’t want to make the supply authority approval an argument about how different they might be and how that might affect the kW calculation.

 

[123MB]

Can anyone weigh in on the idea of sending the active power from one SEL relay to the other using the 8 MIRROREDBITS available in the SEL-751.

I don’t really see how it could work as even if the logic for encoding and decoding could be done in the relay, there’s only 8 bits each way which is only enough for +\- 127 kW to be sent each way.

 

[waross]

I don't think that you understand the relationship between kw and PF.
There isn't any.
A reverse power relay measures kW. It ignores PF.
If the two transformers are fed from different sources, then each must have individual revenue metering.
I very much doubt that the metering in place is able to determine the sum of the reverse power from two separate systems.
I suspect that someone in the chain of information looked at a utility requirement of maximum 275 kW per system and took a short cut, summing the limits and passing the information on as 550 kW total.
An independent reverse power relay set at 275 kW on each system will be simple and relatively economical.
It will meet the utility requirements.
Putting a total limit as the sum of the backfeed of two separate systems does not make sense.

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

 

[catserveng]

Mirrored bits communication is just what is says, a fast secure means to move up to 8 bits of data per channel used in protection systems. If you feel you need to use analog values and still maintain compliance with IEC61850, then you may need to go the route of a protection grade IEC61850 compliant processor using GOOSE messaging. Not as fast as mirrored bits, but still considered by many as suitable for protection applications. I am only familiar with systems by NovaTech and SEL, although I know others exist. We are seeing more IEC61850 compliant automation processors in more applications, mostly the SEL RTAC family of processors and I/O. Also, as far as I know most of this hardware also complies with parts of the IEC 60255 standards.

Here is a paper discussing the protocol,

https://cdn.selinc.com/assets/Liter...nSpeeds_NS_20080221_Web.pdf?v=20191014-173540

And current SEL IEC 61850 family of controls,

https://cdn.selinc.com/assets/Liter.../IEC_61850_Family_Flyer.pdf?v=20190905-155334

Here is some info on NovaTech although I'm not sure it is what you need,

https://www.novatechweb.com/wp-content/uploads/2016/09/AN_IEC-61850_091616-1.pdf

Lately a lot of systems we're working on that require complying with local utility interconnect agreements are based on the SEL RTAC style controllers, for smaller systems the Axiom product line does a lot, and SEL is usually pretty helpful in getting the engineers on a path to being successful. But we also still do a lot with Mirrored bits as well.

Since you can't provide all the required information due to it being proprietary then you'll likely need to engage a local engineer with knowledge of both the applicable standards, and how the local utility (supply authority in your case) applies those standards to their system.

Hope that helps, MikeL.

 

[FreddyNurk]

Well, at least where I am, the utilities care less about the type of generation, and more about maintaining the protection for anything that might backfeed their network.
Some of them post lists of approved relays (and whilst waross' suggestion is an excellent one, it generally won't fly as there are other protection elements also required), but there's also the possibility of making the case of why a different relay can be used, although its much harder to get through the utilities acceptance process.
At least in Queensland, there are published standards from the utility that are likely to cover your situation, although if you have 2 supplies each with separate metering its going to be a pain no matter what.

Lastly, as export is generally in Watts as waross noted, just how much real power are you going to export from discharging a cap? I would have thought this situation is fundamentally different to the context of behind the meter generation, whether its synchronous kit, or inverter based battery or solar, as there's unlikely to be the same level of real power output.

EDMS Australia