Thoughts on "SmartGrid" Implementation

[tlrols]

Ok, we are doing some "SmartGrid" work...reconductoring, switch installations, communications and control. We have federal funds to spend...

I have been asked to give my thoughts on using a wireless 802.11 network to communicate the control signals to the associated switches, etc. Not surprisingly this project is run by an IT person (long sigh...).

I feel that the use of a unlicencsed network such as this is a bad idea. Keep in mind that this system will allow load to be tripped or transferred, or possible paralleling of the 13.8 kV with the 115 kV.

Does anyone else think this is a bad idea?

It isn't so much the security issue I am concerned about (passwords and encryption would be used) but rather the lack of reliablity associated with using a non utility owned comms system (it is regular old Wi-Fi communications at this point).

Thoughts anyone?

 

[jeebusmn]

These same issues were brought up when I went to the OTC convention in Houston. There is a big push to move to wireless by the vendors for refinery controls. No one in the industry is very hot on replacing something they know works. I can see it being used for simple non-safety applications but for other stuff that requires quick response it is kind of scary.

 

[dpc]

You can get very strongly encrypted communications via WiFi using appropriate radios.

Check the SEL-3022 from Schweitzer as an example. This is supposed to meet NERC security requirements. Of course, it has to be separate from the normal office WAN/LAN system.

Is it 100% secure - no, but nothing else is either.





David Castor

http://www.cvoes.com

 

[rbulsara]

What is 'SmartGrid'? Why is a IT person running a project that decides what a electric grid should trip? Whose loads are you taking about tripping?

This is not a social networking site like LinkedIn, where such topics are often posted and discussed by many who have no clue as to how a grid works or any electrical circuit for that matter.

Narrow down your question to specific technical ones, to get a better response, if any.



Rafiq Bulsara

http://www.srengineersct.com

 

[rcw retired EE]

My thoughts: If there is a radio communicatiosn system that is compatible with IEC 61850 standards for relay communication, it might be possible to put a secure system together. I think that SEL has done it.

I would bet your IT types will balk at the cost and complexity compared to the ease of an open wireless link.

I am not too excited about the idea, but if enough security can be put in place, maybe it will work. Can you try it out on a small project?

Also, make sure some of the "dumb" equipment and protections are still in place to keep the disaster to just lost load and not lost protection.

 

[tlrols]

Thanks for your comments...

I am mostly amused by rbulsara's comments...Social Networking indeed. I am an electrical engineer with 30 years in the industry, much of it driving a certain really large federal system in the NW US--about 18,000 MW worth of transmission sales, etc (Operations/Planning/SCADA/Communications)--I am fairly confident that I have more than a clue about designing/planning/operating an interconnected electrical system.

I now work for a modest sized IOU. The loads that could be tripped, switched, or otherwise manipulated are those native loads that we serve on our distribution system.

We use Ed Schweitzer's products--I was merely wishing to get a relatively unbiased opinion from others. My experience with SCADA and leased PSTN lines makes me prefer comms system that we own/control. And yes, it IS an IT person running things--mark that off to corporate politics...it is what it is.

I think we will scope this project so that if the wireless part turns into a "goat rope" that the project in general will still succeed. I really dislike the WiFi solution really from the standpoint of multiple parties making use of it. However, the cost of deploying a traditional wires/fiber/digital radio (read licensed) is quite high...

 

[PHovnanian]

the lack of reliablity associated with using a non utility owned comms system (it is regular old Wi-Fi communications at this point).

Well, first of all, somebody owns components of the WiFi system. It could be owed/maintained by your own utility, your local telephone company, municipality, etc. Who owns the various components of the wireless (and underlying network backbone) is something your IT person needs to verify. In addition to the ownership, you (or your IT person) is going to have to find the service level that this network infrastructure provides. In my town (I live within spitting distance of Bill Gates) for example, our cellular operators do provide backup in the event of power failures. But only for a matter of hours. Once the base station backup batteries go dead or gensets run out of fuel, cellular service is dropped. We've seen that happen for power outages that last more than a day or two. Cable TV based broadband has even worse backup provisions. And while POTS (plain old telephone service) is generally reliable as required by telecom regulations, the newer fiber to the customer services are not subject to the same regulations. Unless you have explicit service agreements negotiated for each link to your systems, these may fare poorly when the lights go out.

Your IT person needs to identify every piece of networking infrastructure needed to provide the SCADA services you require. For example, people often overlook things like DNS services and while the links between your systems have proper backup power, if they can't resolve the host names they need, your network is worthless.

 

[pwrtran]

A part of our SCADA communication uses unlicensed radios, believe me it is a nightmare!

 

[Skogsgurra]

It is interesting to read about today's developments and what was said about it back in 2003. A couple of threads are available here: thread238-67677 and just below it the sequel can be found.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[LionelHutz]

This seems to go back to that old problem that you have to let someone without experience prove to themselves that their wonderful new cheap money savings ideas don't work.

I'm dealing with one of those problems today. I did actually learn a little something so it wasn't a total waste, just mostly a waste of time.

 

[jghrist]

This seems to go back to that old problem that you have to let someone without experience prove to themselves that their wonderful new cheap money savings ideas don't work.

It's amazing how many of these things will work with someone else's money!]

 

[slavag]

NO, thanks.
WiFi for control of power plants and substations.
EE have a good stuf for the telecontrol, telemonitoring and good protocols like to DNP, IEC-104 etc.
Thanks IT guys, we don't need your help in this issue.
SMART GRID!!! :-(.

 

[tlrols]

So pwrtran would you mind taking a moment to briefly discuss what problems you had using unlicensed radio? In my experiences we always got the spectrum we needed--not that we didn't have issues anyway with the "usual suspects" in regards to using radios...

I agree that WiFi is really a dubious solution for the control of anything that is important. However, IT managers are quite adept at infiltrating organizations and laying claim to areas that, perhaps, are outside of their area of expertise. Sometimes people and projects have to fail before learning occurs--this is unfortunate, but pretty much a fact of life.

An observation: For the early part of the 21st century it looks like the term "SmartGrid" will be a new entry to the corporate game of "BS Bingo"...sort of like the horribly abused word "paradigm" was in the 90's.

 

[amptramp]

What about the issue of signal jamming? Wifi jammers can be had on the internet for $400. Considering the US electrical grid could (should) be considered as a national security priority, is it wise to install systems that can easily be affected or shut down by third parties?

Didn't the IT guy see the movie "Live Free or Die Hard".

 

[tlrols]

Maybe we should include a ban on Pringle's cans as part of the implementation!? I will have to get a "must watch" list of movies on the desks of the implementors...

 

[amptramp]

Only Pringles cans that come with a set of instructions (found in the Anarchist's Cookbook)...

But seriously, shouldn't RF Denial of Service be considered as a viable security risk.

I found the info at the link below interesting.

http://www.cisco.com/en/US/prod/col...3/ps9394/prod_white_paper0900aecd807395a9.pdf

 

[jayp99]

Ruggedcom has some WIMAX stuff out working in the 3.65 GHz range that is more secure than WIFI 802.11.

I am looking at & comparing similar systems as we speak. As my office has received Smartgrid grant.

It's a confusing mess that's for sure. IT guys think everything should be WIFI til you hack their nannycams!

 

[jayp99]

PS - interference in the WIFI 802.11 is a big concern for me

 

[pwrtran]

The antenna guy (also the radio consultant) did not know the root cause yet, however, interference appears to be a very generic explanation. We found in some days, when our unlicensed radios feel in a good mood, they talk to each other smoothly. The signal strength level is about -70dbm, which is very good. However, in most of the month, they are in a very low spirit and the signal strength can go down as low as -88bdm which is unacceptable. When it goes down to -80dbm, we start having big problems to control of devices. The output power of our unlicensed radio is only 1W, I think it is regulated, the maximum output power for the unlicensed radio could be 3W I am not sure. Anybody can do anything on that frequency range. If someone is stronger than you then they blast you down.
So, we need to send guys to a unmanned station if have to during the "bad weather".

 

[amptramp]

In a manner similar to Richard Feynman's famous rubber band experiment, maybe a wifi setup being blitzed by a bluetooth enabled cellphone might have some impact on the IT guy and all the non-technical bean counters involved in the decision making process.