Coal fired power station failed - Callide

[octave]

I think the most likely target of a cyber attack may not be a power station but the grid itself with smart meters and other possible points of entry.  

[Jerry_Atrick]

The switchgear may not be IP addressable, but IP is one protocol of many networking protocols. UDP is another, and there are a few proprietary ones out there as well. Because something does not have an IP address does not mean it can't be contacted. However, I would be surprised if switchghear on older stations is anything but hardwired to controllers on the network associated with the SCADA system.

 

I have read the bigger issue is the older telemetry systems which use wireless, insecure intranets.. these will be dead easy to hack and send fake readings back (e.g. excessive vibrations on a series of feedwater pumps, which will shut the whole thing down. I doubt there would be many plant engineers that would override this until they get to the bottom of why the false readings were being provided across the array of pumps.

 

Systems on an intranet that are "away" from gateways, switches, bridges, routers, and the like have to be on a physically different network. They may deploy internal de-militarised zones (DMZs). Note, these by themselves are very secure anyway, and traditional "hacking", of which the correct term is cracking, is in the lower order of risk these days due to the advances of info-security systems. Of course, it still happens.. but only because of the fallibility of humans, but the cyber-security offerings today are still effective at defending against these sort of attacks where humans stuff it up.

 

Cybersecurity these days is a lot more, mostly geared to the weakest link in the chain - people. Stopping fishing emails getting through, but then ensuring people don't click on the links when they get through is high on the agenda. As is the plethora of social engineering tricks used to get people to disclose passwords, etc. Penetration testers are paid handsomely, but they aren't just geeks running password cracking software; they will try to physically breach the walls and see if they can launch a cyber-attack from within.

 

From memory, internal breaches (accidental clicking on links to maliciously installing malware or disclosing passwords) consisted of around 80% of all cyber attacks..

 

Agree with Octave - the grid will be a better target, but ultimately, criminals will try the least difficult path, so if the grid is well secured, they will go for the power stations (if they want to disrupt power supply, for example). When I worked in the Nuke industry from 97 - 02, the average cost of a single unit being off grid to the power generator was £1m/day.. Most generators had 2 units in the facility, so that would be £2m/day. Knock one out for a week and in an industry that was operating on thin margins, that is a lot of cash to try and blackmail/ransom someone with. There are of course other reasons to target nukes, so they spend a lot on cybersecurity.

[onetrack]

It looks to me very much like a similar event to the East Perth powerhouse explosion I previously outlined - a disintegration of a turbine, quite likely due to poor or delayed maintenance.

 

https://www.abc.net.au/news/2021-06-02/qld-cs-energy-releases-photo-of-damaged-callide-power-station/100186330

[nomadpete]

Gents, smart (customer) meters are not part of the grid. It would hardly be a terrorist act if a couple of houses went black. These are in the distribution network, or 'poles and wires'.

 

The grid is the national bulk supply network, so it involves whole cities and states. At that level there's a higher risk of being noticed if something goes black. Yes, Jerry most SCADA systems are totally stand alone rings in a substation. I didn't wish to bore anyone with the multi layers of internal comms.

In truth there will always be some way to work around any 'secure' electronic communication system. But even if an outsider managed to burrow a path to anything operational,  to cause any malicious damage would require very very detailed knowledge of the individual brands and models of the devices. These devices usually use inhouse protocols and unless you know what a device is, you'd be yelling into a void without answer. 

 

I really think that terrorists are too lazy to commit such difficult crimes.

[octave]

There have certainly been cyber successful attacks overseas against power grids.  There is also a report  by our chief scientist Alan Finkel which I think is available online.

 

 

[octave]

A search of the net shows that the threat to the power system from hackers or foreign governments is being taken seriously.   Whilst it may be difficult to access the controls of a power generator (I have little knowledge of this)  I can think of important areas of infrastructure that have vulnerabilities.    My brother in law is a computer programmer and he works for a company that designs software to control large factories.    He is able to access these software installations remotely in order to tweak them or fix them when necessary.    This is a great thing of course but is a point of vulnerability which I know they take seriously.    My wife used to work at a water treatment engineering company and one of her ex colleagues now works for a large water treatment plant.  He is able to access several different sights form one location or even from home overnight.

 

When it comes to power security, whilst it may be difficult to directly hack into a generator for example, the grid has many layers.    My power retailer can remotely shut off my power if I don't pay my bill.  They could also shut off my power if my solar system is malfunctioning and endangering the grid.   On a larger scale input from the many and varied generators is often "curtailed" when the grid voltage is too high or capacity is brought on line when it is required.    Unless all these entities are hard wired together I don't see how we can rule out cyber mischief.

 

I am definitely not anti smart grid in fact the smart grid is fantastic.   The chance of a cyber attack may well be small but should not be dismissed.

[nomadpete]

I'm not asserting that there isn't any risk to our power grid security.

What I'm saying is that in Qld at least the grid operator/designer/maintainer is acutely aware of the potential risk and has historically gone to great lengths to make it hard for outsiders to sabotage the grid.

 

I spent 17 years commissioning and testing power grid control systems throughout Queensland. Yes it's true that no system is totally, absolutely secure. But what I saw was that even a well trained multi skilled employee with insider knowledge would have a hard time to even begin to do any damage. It isn't something that can be learnt by outsiders. There, I've just exposed a small vulnerability!

I  left that industry a couple of years ago, so I can't comment on recently installed equipment.

But I do know that these organisations do go to great lengths to secure the system.

 

Personally I am more concerned with the possible security risks to drinking water processing plants. But that may be because I don't know anything about their control system vulnerabilities.

Edited June 3, 2021 by nomadpete

[octave]

 

51 minutes ago, nomadpete said:

I'm not asserting that there isn't any risk to our power grid security.

What I'm saying is that in Qld at least the grid operator/designer/maintainer is acutely aware of the potential risk and has historically gone to great lengths to make it hard for outsiders to sabotage the grid.

 

I think we are broadly in agreement.  There have been successful attacks overseas and many attempted attacks.   This does not keep me awake at night. I also don't worry too much about whether the airbus I am travelling on can be hacked into but serious people are working on being one step ahead of anyone who might attempt such a thing.  I get that there are many safe guards and that systems  are separated from the internet but I don't think we should believe that it could never be done.

 

This is from an article which most definitely is not alarmist and points at how difficult it is but at the same time we should not just take it for granted that it will not happen or could not happen (I realize you are not saying i could never happen)

 

 

In theory, OT systems are "air-gapped" from IT systems, with no network connections between the two. But with the exception of nuclear power plants, which strictly regulate their operational systems' disconnection from outside networks, that air-gap is often more permeable than it ought to be, says Galina Antova, a co-founder of the industrial control system security firm Claroty. She says that Claroty has never analyzed an industrial control facility's setup and not found a "trivial" way in to its OT systems. "Just by mapping the network, we can see the pathway from IT to OT," she says. "There are ways of getting in."

 

https://www.wired.com/story/hacking-a-power-grid-in-three-not-so-easy-steps/

 

 

 

 

 

[Yenn]

Nomad it sounds as if you took over from Neville Gersch at Boyne. My time remark was based on what a friend of mine who runs pot line 1 told me. I think he said he could revive after 34 hours, but I never asked him how.

They have just found a 350kg lump of what they call shrapnell in the roof at Callide. that sounds like a big lump of the casing as I doubt that there would be that much in the turbine except the shaft  and I reckon we would have been told if the shaft disappeared.

[nomadpete]

Thanks Yenn. As I said, I was mostly repeating the site induction talk. Maybe they were exaggerating  And the four hours was mentioned by a power link person, who was cautioning us to be careful when testing the intertrip gear.....The four hours was when big financial penalties started biting our bottom line.... When I think about it, even when the arc stops, there must be a considerable thermal store in the pot.

 

On the subject of 'shrapnel' I recall a photo of a portion of turbine drive shaft embedded in a foot thick concrete floor when a (not Aussie) generator failed. The shaft had gone through the brick walls of the equipment room before going into the floor. And that was only a small plant of about 65Mw. 

The Callide turbine must have been close to 50,000hp, so imagine letting all that energy out of the casing

 

Years ago, a close friend was working in a NSW power station equipment room in the turbine hall when it suddenly got quiet. If a Power station goes quiet it's not a good sign. Concerned, he opened the door and saw a flame shooting up from the turbine. He said the roof I beams sixty feet up 'were slowly bending in like spaghetti. Deciding that his job description didn't include this scenario, he left with all haste. It was an oil fire from the turbine bearing. The oil is what keeps the hydrogen in.

 

Such things don't happen very often but they are spectacular.

 

Edited June 3, 2021 by nomadpete