Unfortunately for PG&E, but maybe fortunately for the power industry, two attacks in short order on Metcalf substation near Silicon Valley woke us all up.
We’ve all watched a growing concern over cyber-attacks on the grid (see Power System Cyber Threat is the Real Deal) but the PG&E attacks were physical – less exotic and maybe even more difficult to prevent. So, after at least 30 years since the first substation bombings, the heavy wheels of regulatory change are beginning to turn, albeit slowly, and we may actually see something done to protect major substations from terrorists, thieves and just plain vandals.
Recall that the Federal Energy Regulatory Commission (FERC) concluded earlier this year that the entire US power system could collapse if just nine critical substations were suddenly shut down during summer peak load. (U.S. Risks National Blackout From Small-Scale Attack )
So now, as the Nov. 20, 2014 Wall Street Journal reported:
“Federal energy regulators ordered the nation’s utilities to protect important electrical equipment from potential attackers trying to disrupt the U.S. electric grid.
Under a new rule adopted by the Federal Energy Regulatory Commission Thursday, power companies nationwide must identify key transmission substations and other hubs that, if knocked out of service, could cause blackouts or other major problems. The utilities must then put defenses in place, and unaffiliated experts must review their security plans.”
Whoo – hoo! It’s about time. Attacks on the grid are nothing new but until recently utilities and their regulators have pretty much been in denial. Either that or maybe we just don’t even know how to begin to protect the system against physical attack.
I don’t know how PG&E tried to secure the substation after the first attack – and if I did we wouldn’t report it. But the company did conclude publically that security guards had been lax in their duties.
No surprise there. I recall several occasions when my research team at a major utility, working at control centers during the wee hours of the morning, found a guard sleeping at his post.
Then, at the same utility, there was the sadly funny incident of a bored and armed substation guard practicing quick draw and shooting himself in the leg.
Humans just don’t do well with tasks that are glacially tedious and routine while simultaneously keeping an eye out for unexpected events requiring fast action. That’s where automation comes in.
And, speaking of boring, hopefully the process of coming up with an executable plan to protect the grid won’t devolve into some tedious and molasses-like standards process. In fact – we may not want ‘standards’ as such because having a uniform system of protection means that once you know how to successfully defeat the security at one substation, you have the key to them all.
California and Arizona, and maybe several other states are coming up with new rules and legislation as a start to protecting key grid assets. But we haven’t seen an overall national game plan.
So the question remains - can we adequately protect grid assets? And if we can, can we do it in time following a turtle-pace regulatory process with bureaucratic gate-keeping at each review step?
As the WSJ article goes on to say, even the new FERC Commissioner sees a need for more action;
“The rule ‘probably does not go as far as some would like, but it’s a step in the right direction,’ said FERC Commissioner Norman Bay.”
But Bay’s statement begs the question – why not “go far enough”?
Depending on the criticality of the asset being protected, are we talking specially trained security personal, drone surveillance, extensive sensor networks, the National Guard?
Or maybe, as some of my colleagues suggest, let’s don’t do much until and unless we start getting major attacks. After all, making a big deal out of this could attract terrorist attention like flies to honey.
What does “far enough” mean? What do you think?