South Australia statewide electricity blackout. 10

[Trenno]

SA may be more worried now that the Hazelwood Coal Power Station has closed.

 

[byrdj]

it mentions that the inverter storage software can be configured to provide inertia? is that blowing smoke or possible. 800 tons at 1800 rpms is inertia!

 

[GregLocock]

I proposed exactly that a while back. I imagine some way of broadcasting a nominal clock pulse across the state , and the inverters were configured to be driven by that, rather than the local transmission line clock speed. If only there was a way of transmitting clock pulses over long distances. Once we've solved that, say by about 1900, the rest is easy.

Sure, it would cost about $43 in hardware.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[mgtrp]

it mentions that the inverter storage software can be configured to provide inertia? is that blowing smoke or possible. 800 tons at 1800 rpms is inertia!

Google "synthetic inertia" and wind turbines. The trick in the inverter is for it to be able to take the inertia stored in the wind turbine and pass it through to the electric grid.

I imagine some way of broadcasting a nominal clock pulse across the state , and the inverters were configured to be driven by that, rather than the local transmission line clock speed.

GPS clocks have been around for some time, which allowed synchronised time to be available almost anywhere. But I'm not sure what the intended benefit is in using a synchronised clock pulse to drive the inverters.

 

[GregLocock]

The advantage would be that it would increase their ride through tolerance, or if you like, it would add more inertia to the system.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[davidbeach]

I'm not sure that would actually matter or help. In a system with actual inertia, the frequency would be continually varying as everything found a new stable operating point (or not). Without extremely robust communications I'm not sure that any combination of artificial inertia devices could adequately perform that dance. That isn't to say that it couldn't be less bad than it is, nor that artificial inertia couldn't help up to some point, but I'm not convinced that a wide area system can simply rely on artificial inertia.

It might be possible to design an electrical grid that didn't need inertia, whether real or artificial, but we're far from implementing such a beast.

 

[GregLocock]

I've worked with largeish machines that simulate inertia quite happily, it's mechanics 101.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[cranky108]

Here's another problem with simulated inertia, if you can't model it, and communicate that model to the planning engineers, then it can only be assumed to not exist. Inverter manufacturers have done a poor job of communicating with the system planning engineers, let alone the protection engineers. So to date, most of the utility engineers can do nothing but discount the statements of the inverter manufacturers.

The issue might be that there is a third party involved, called the customer, who has no idea that any of this information is important.

So none of us will really know how the system will respond, until there is an event. And even then, any units not within the event won't be known.

 

[GregLocock]

And why wouldn't you model it?

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[cranky108]

Model what? Hot air? 'if you can't model it, and communicate that model to the planning engineers, then it can only be assumed to not exist.'

Inverter manufacturers have done a poor job of communicating with the system planning engineers, let alone the protection engineers.

 

[GregLocock]

I'll try again. I have worked with large electric machines that specifically simulate inertia (chassis dynos). Therefore it does not seem intrinsically impossible to alter the output of an electric drive system to simulate inertia. To point out the obvious my off grid inverter simulates a high inertia, that is, its output frequency does not change as the load changes.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[byrdj]

Is it that the inverters are allow to operate at max output and thus have nothing to add when needed.
I recall steam turbine plants that were not allow to operate at CVWO because of contract to be able to support grid stability.

 

[GregLocock]

No, the issue is that the output of the inverters is locked (in ways I do not know) to the clock frequency of the transmission line, and, obviously, its voltage is greater. So for example my on-grid inverter tracks the clock frequency of the local line, but does not attempt to drive it. It pumps electricity in by delivering a higher voltage than the transmission line. That is if the TL is operating instantaneously at V.sin(w.t) my inverter delivers (V+dV).sin(w.t). A truly synchronous machine would actually drive the transmission line at (V+dV).sin((w+dw).t) where dw is the difference between line frequency w and the synchronous generators speed.

Any reasonably gifted electrical engineering student can design a motor controller to accomplish that efficiently.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[MikeHalloran]

So, you want to phase lock all the inverters throughout the system to a remote master clock, so if they happen to be disconnected from the grid at some instant in time, and you want to reconnect them to the grid, all you have to do is close the tie breaker, without having to check or adjust the phasing first? Sounds possible technically, perhaps less so politically.

OR, could you implement a persistent local phase lock, that would just keep the local inverters' clocks running pretty much in sync with the grid even through a disconnect of some specified finite time? Sounds like something that could be sold as a product improvement, or maybe mandated for new designs, and rolled out with normal replacements as the supply chain flushes.



Mike Halloran
Pembroke Pines, FL, USA

 

[mgtrp]

A truly synchronous machine does not operate above the grid frequency, otherwise it would be an asynchronous machine. It is much the same way that the turbine operates at the same speed as the generator--it is not turning at a slightly higher speed, although it would speed up if the shaft broke in much the same way that a generator would if its connection to the grid were lost.

With regards to inverters, they do not need to be all synchronised to the same independent clock source to be able to provide synthetic inertia. All they need to do is be capable of providing a brief overload in the event of system frequency dropping. Plenty of wind turbines already have this capability.

 

[GregLocock]

Yes they could do it, but thdy didn't.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[cranky108]

And if they did, there simulated anything could not be used as an input to power system studies, because the inverter manufacturers don't want to provide information to the public.

Utilities would need to install spinning devices because they can not count on inverters with unknown synthetic inertia.

And what master clock are you talking about? There is no master clock, other than the speed of the grid, which ever one you are on. None of them operate at any exact speed, but at a ball park speed.

 

[HamburgerHelper]

With synthetic inertia, I guess it pulls power from the system to motor the turbine if the system is accelerating?

 

[IRstuff]

" None of them operate at any exact speed, but at a ball park speed."

I don't know about now, but, at least in the US, they used to be at an exact average speed, as many AC-powered clocks used the line frequency as a clock reference. This allowed AC clocks to be accurate to better than a second per year, or so.

I'm not really following parts of the discussion here, since "inverters," by definition, are converting DC to AC, and as such, they have to be phase-locked to the line frequency to efficiently transfer power. They are nominally crystal-driven to the 60-Hz line frequency if they are free-running, but if they're connected to the grid, they phase lock to the grid's frequency. Which means. they'll accommodate variations in frequency, up to a point, assuming the changes aren't rapid and random.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Compositepro]

I think this discussion of synthetic inertia is misleading. There is no such thing. What the inertia of a generator provides is energy storage, period (full stop). In an inverter generator this would be provided by batteries, capacitors, or even a flywheel/generator. Diesel or steam generators are just more robust than inverter generators where the weakest link is the solid state electronic components