If we're talking LV systems, then BS 7671 (especially section 551 - generating sets) covers many of the requirements for both stand alone and grid connected generators. In terms of "grid code" G98 and G99 (and perhaps G100) is probably what you need to be looking at.

The general gist though in that in island mode you're required to have no galvanic connection with the grid's live conductors (only PE can remain connected) and you can make no demand on the grid's Earth (you'd need your own local generator electrode(s))  - but then you're pretty much left to your own devices. In grid connected mode there are far more restrictions (mostly to safeguard the grid and people working on it) and often specific agreements with the local DNO are required for larger setups (>16A/phase).

Yes, you can export to the grid - and be paid for doing so. Most suppliers operate some kind of export tariff (the numbers depend on the source of generation - renewables usually get preferential treatment).

   - Andy.

Hi Andy, thanks for replying. My question is related to Type D plants, connecting to the 400kV transmission system.

Times are given in The Grid Code for remaining connected for banded frequency and voltage variations, though I can't find anything definitive about requirements to remain connected and operate in Island Mode (Islanding) should an area become disconnected from the national grid. Any thoughts please? Cheers, Bob.

Hi Andy, thanks for replying. My question is related to Type D plants, connecting to the 400kV transmission system.

Times are given in The Grid Code for remaining connected for banded frequency and voltage variations, though I can't find anything definitive about requirements to remain connected and operate in Island Mode (Islanding) should an area become disconnected from the national grid. Any thoughts please? Cheers, Bob.

connecting to the 400kV transmission system

Ah, that really is big stuff! Ignore what I said - that was for quite literally the opposite end of the spectrum.  I don't know if anyone else here has any experience of things of that scale - I suspect you might have to talk to the national grid people directly (the real national grid people that is, not the DNO that was until recently known as WPD).

I suspect that the bias now is to "stay connected at almost any cost" - after that episode a couple of years back when one power station failure triggered a small departure from normal voltages/frequencies which then triggered some other generators to automatically disconnect in a cascading fashion until we almost had a national blackout (and electric train services were disrupted for long after the power returned).

   - Andy.

Thanks for replying Andy. Sounds like the UK nearly had an "East Coast" blackout scenario!!

If your not familiar, then it is this one Friday Aug 9th 2019

assets.publishing.service.gov.uk/.../20191003_E3C_Interim_Report_into_GB_Power_Disruption.pdf

The upshot was that a few million customers lost power some  for up to 45 minutes.

or in a more newsy way..

https://www.current-news.co.uk/blackout-investigation-what-went-wrong-at-hornsea-one-and-little-barford/

Basically there was a lightning surge that disconnected some kit and the grid frequency fell, and because we use grid frequency as an overload indicator - because on rotating gensets it naturally is- things slow down when overloaded, and on inverters and so on it has to be faked, more things disconnected. And then could not reach the condition to reconnect quick enough so load shedding took place.

The other disappointing discovery was that some of the new Avanti trains don't restart without a lockout being reset after  a low frequency event, and that left some trains in the wrong places until the Sunday, as chaps with laptops and programming leads visited each one in turn.

Mike.

PS the other fun read about power going badly wrong in the UK is  the Lancaster floods - the moral is don't put a substation on a river bank, basically.
www.theblackoutreport.co.uk/.../

Hi Mike, Interesting!! I've done some work in UK using EPRI who were formed in the USA after the notorious "East Coast" blackout. With less "spinning reserve" and a "softer grid" I feel there will be a lot of challenges to "keeping the lights on"!!!

I think we need to either move away from the traditional load/ frequency model of detecting real overloads or emulate it very carefully where we are in effect inverting up to AC from DC. Unlike spinning iron,  there is no real mechanical inertia to borrow energy to tide over 100% overloads for a few cycles in a solar panel, and not really a lot in a wind farm, as the interconnects are all DC.
On the other hand, we wanted to and did it to an agreed standard,  we could easily have self shedding loads that turn themselves down if the mains frequency drops below some threshold - to add such a thing to car chargers and heat pumps would not be especially tricky. Sadly as it does not involve AI, or broadband, I suspect being simple, it won't ever be done.

The other way to look at this is that 5 years on we still consider a 45 minute power cut as the most serious things that happened recently, so perhaps compared to much of the world, really it's not so bad here.

Mike