4 ms·
It will never be economically feasible to secure the grid from physical attack. It is too expensive to bury the distribution network, and while isolated attack
by kogepathic 10y ago
It will never be economically feasible to secure the grid from physical attack.
It is too expensive to bury the distribution network, and while isolated attacks like this are extremely unfortunate, the reality is that they are infrequent enough that physically securing the equipment will cost more than the economic losses to the energy company.
The state of industrial security is laughable, and electronic attacks are much easier to accomplish than physical attacks, simply because you can target geographically distributed locations simultaneously.
It's unfortunate that someone thought it was fun to shoot up some transformers, but just look at the article: up to $1 million in costs for the operator in this incident, versus an estimate of hundreds of billions in damages due to a cyber attack.
- Eridrus 10y agoPreventative measures may be costly, but maybe we should have something like shotspotter + video drones deployed so that these people can at least be caught and hopefully it will act as a deterrent.
- DanielBMarkham 10y agoI imagine that this is exactly what we'll see developing over the next decade or so: electronic sensors to identify the attacker, and/or provide a little bit of early warning when an attack is imminent. When for only a few thousand dollars you can permanently equip against millions in potential losses? It's a no-brainer.
- DougN7 10y agoWhat you describe will require increased surveillance. I don't disagree with you, but this is exactly how Big Brother and the surveillance state comes about.
- LyndsySimon 10y agoThat would be ineffective for a number of reasons - chief among them being that shotspotter is largely ineffective, and that substations are often located in rural areas where gunshots are both common and expected occurrences.
- Eridrus 10y agoI had a little bit of a google for shotspotter effectiveness info. It seems like it has a false positive problem (which you allude to) and a response time problem. The first problem seems almost irrelevant if the response cost is low (because you have drones), the latter problem may be more of an issue. Recent drones can get 45 mph, so depending on what kind of range you expect snipers to have, it's probably possible to have a ring of them around the area such that their 2 min range only just overlaps and be able to cover ~3-4 miles around an installation. Is 2 minutes too slow of a response time for these facilities? I don't know. Rural areas will be much easier to secure than urban areas though.