Replies

Well, when I was studying EE I was told abruptly taking out (grounding) one phase *could* cause a 3ø generator to pound itself to pieces before it could be stopped. I didn't have the guts to try it.

I do know that a determined foe can figure out an unexpected weakness in any system and exploit it.

I do know that power grids can fail accidently, as has been demonstrated in a surprisingly large number of systems across the world in the past few months. Notably most of these outages affected countries that supported our efforts in Iraq. Perhaps this is only because everyone else doesn't have a grid to down?

I suspect a timed sequential takedown of key links could crowd load onto a known weak link that could then be cut, giving the operators a load sheding problem they may not have enough time to gracefully recover from. IIRC something like this happened accidently causing the '73 blackout.

The human in the loop is only as good as the information s/he has. Cutting communications, or more subtly, presenting normal expected data while the crisis builds renders this the weakest link - bu bye!

That the controls of a reactor are prePC does make the system more robust, but one does not need to take down any single generator to fragment the grid.

If people can design and build it, other people can figure it out and destroy it.

BTW, Do you know the differnce between a mechanical engineer and a civil engineer?

If people can design and build it, other people can figure it out and destroy it.

Today's large, interconnected, redundant, supervised systems ususally fail when a number of factors come together in/at the MOST unfortunate time; it takes/it took the involvement of human error in the BIGGEST events we've seen, sometimes spurred by an induced failure or two and three PLUS the failure to have certain 'backup' systems or procedures in place or 'practiced' by operations to REALLY screw up these LARGE systems; a few lines, a few dozen lines simply 'felled' by persons with devious intent very most likely WOULD NOT result in a widesprad blackout per se; localized outages (due to proper load shedding) due to both intended operator action and load shedding relays would save the system in islands (sine 'lines' also called 'tie lines') between systems would be 'missing' ...

It still looks to me like you're 'subsetting' (looking at isolated subsets on this issue) rather than getting a bigger, more realistic, picture. I moved on from simple 'boxes' (black box design) and onto 'systems' issues long ago; MANY concepts applicable to single 'black box' failure analysis need to be discarded in favor of 'models' that allow some measure of 'self-healing' and recovery or degraded operation when analyzing something as complicated, diverse, widespread and dynamic as 'the grid' ...