We are dependent on electricity. Not even water would come from the tap if there is no power on the mains, not to speak of refridgerator, heating, or television - my cellphone I wouldn't miss ;-)
Now, we have a PV installation on our roof and feed more "generated" electricity into the grid than we use ourselves. The issue: If there would be a blackout then also grid feeding would instantly stop - for safety reasons (don't fry your utility electrician), but also due to the fact that the PV inverter supplies only when it sees a grid frequency to synchronize with.
What would happen in the case of a longer blackout? Could we at least locally make the available renewable energy usable? What would be the minimal requirement to be fulfilled for making a PV feed in again (no island installation!)?
Wouldn't this be worth a bachelor thesis or even a master thesis (providing a solution)?
I wrote this also up in https://energyhacks.wordpress.com/2016/09/03/what-if/
Any comments?
Negawatts are not only those not spent but also those available "for free"...
I think the Tesla Powerwall allows you to do that, switch to being an energy island when grid drops
Sure, Petur, you can buy a storage unit for a lot of money as you can buy island inverters; but what are the minimal requirements to overcome at least locally a blackout scenario and make your PV feed on?
I guess at a minimum you need a bit of storage for the transition from local to net to give the PLL time to phase-match the net?
But that is only when your local production is > usage, so at night or dusk/dawn or even bad weather this will not work. So more storage is not the minimum but more practical.
Something like that? ;)
Without battery storage, the PV system wouldn't be much use during a blackout, even if it could supply the house during that time. Just think about fluctuating load and fluctuating production. You'd easily end up with the power turning on and off every few tens of seconds on a day with patchy clouds. There is also the issue of, when power can be turned back on during the blackout, how to deal with the load spike when all the devices in the house start drawing current simultaneously.
As far as I can see, doing this isn't feasible without a battery as a buffer. And, with a battery, there are several solutions, including Tesla, Samsung, LG, Enphase, and others.
Michi.
This is all clear, Michi. What I am heading for is something different: Think beyond a single installation and just PV and on the opposite a nation or continent wide power supply network; there are wind turbines, co-generation units, biogas units that in case of a broader blackout are turned down and you cannot "just" use them in a smaller environment. Of course, the grid control wants to regain online status as fast as possible. But to power up thousands of kilometers of grid lines is something different than powering up a smaller section. So, again and maybe in different words: What is the minimal requirement to set up a micro grid? Is there an "easy" way that would enable that? An easy way to do, if the "infrastructure we know" is not available for a month or longer?
I am not arguing against the powerwalls, but this is an individual issue - I am thinking of quickly regaining "some" power for, say, a quarter, a village, a suburb, not a single house and not a county, something in-between.
Ah, OK, I get the idea. There are people thinking about this sort of thing, where all devices on the grid are actively participating in balancing load. For example, when my car is plugged in for recharging, it would deliver charge back to the grid if there are load spikes to be dealt with.
I think we'll eventually end up with that kind of intelligent grid, where "blackouts" don't happen any more because the grid will locally be able to deliver energy from buffered capacity.
I don't know what it would take to minimally enable that for a suburb or village. At the very least, I suspect we'd have to have lots of batteries all over the place, with a universal comms protocol so the devices can talk to each other and be aware of where load is needed or excess production is available. The economic model for all of this is missing too at this point :(
Michi.
Michi, you "hit the nail on its head"; the "missing economic model" is the cause why there seems no real investment in this kind of "commodity", yet there are some examples where it already works, on scientific level "where money plays no role, you know...".
I must admit that I do not buy the necessity of an energy storage yet. So, still the question "what is the set of minimal requirements to establish a microgrid within an infrastructure of existing renewable energy sources that are intended to "only" function in a stable large scale grid?"
there must be some kind of protocol in place because my SMA Sunnyboy has a setting dealing with grid requests for production limiting. Haven't checked further on this because in our crappy country (BE) it is disabled of course. But I read that Italy uses it already.
In Germany this may (!) also be used corresponding to the currently valid legal regulations on supplying into the grid (alternative is a fixed limiting to 70% of the supply power) It is called "Rundsteuertechnik" (ripple control), that exists for 100 years by now and may be used for a variety of things: tariff control, supply switching, light switching etc. This I claim to be no necessity in the sense of my initial question (my SB does not provide this feature resp. for generators below 5kWp it is not requested).
Well, so if we assume no local storage anywhere, you'd allow producers to feed into the grid even when the grid is dead. With some protocol to tell them to stop doing that so people can fix things in the grid without getting killed.
The main problem I see here is that you can't send mains voltage very far without big losses. A few hundred meters at most, I suspect. There are also the issues about phase matching/balancing. The protocol would have to ensure that, within a certain area, the phases remain balanced, even though most producers will be single-phase feed-in. I have no idea whether that's even feasible with my limited electrical engineering knowledge.
Presumably, it would be necessary to (again, under remote control) partition the grid into a number of separate cells, each of which would try to be autonomous during an outage. If some providers also have batteries and can feed from that, that would be a bonus.
I strongly suspect that this is anything but a simple problem…
Michi.
I see - http://energy.gov/sites/prod/files/2012%20Microgrid%20Workshop%20Report%... provides a well-shaped overview of all the issues that would need to be answered on setting up a microgrid; now I know why in Africa there are so many places unprovided, but I also argue that we will "give a sh*t" on all these issues in case we "just really" need some power, so, still - what are the minimal requirements?
What comes from the discussion so far:
REQ-MCRGRD-0 Safety is key; thou shalt not fry anyone. Note: This may be "easily" established by relying on isolated cables, plugs and boxes "the normal DAU" cannot touch except willingly.
REQ-MCRGRD-1 On setting up an autonomous small net of electrical providers and consumers there must be some kind of balancing provisioning and consumption. Note: This implies some kind of control on the grid (that actually was established using the "control frequency" of 50/60Hz, simulating the response of an inductive generator)
All this can of course be done with load sharing, voltage regulation, etc etc.
Only problem I see is the Global issue of corporate profits.
It will need to use existing infrastructure, cable, transformers etc.
I doubt unless it as a new setup owned by a local cooperative it just wont happen...sadly.