Slashdot Mirror
Wireless Sensors To Monitor Power Grids
Roland Piquepaille writes "Major power outages like the ones which affected the New York state last month or Western Europe ten days ago are becoming more frequent — even if their causes were different. In some cases, the utility companies have to dispatch electricians all over the place to discover the cause of the power failure or simply to restore power. Engineers at the University of Buffalo think they have a better solution: deploy wireless 'nanotech' sensors to monitor the networks and to find the exact location of a failure. They also say that even if the technology is almost available, several years of research are necessary before such a solution can be used by electrical companies. Read more for additional details about this attractive solution."
... But what happens when the batteries die?
Online Starcraft RPG? At
Dietary fiber is like asynchronous IO-- Non-blocking!
Why nanotech? Power infrastructure isn't exactly tiny. Why not just normal wireless sensors? Buzzword much...?
Cemil.
Hope they run on battery!
One of the many implications for the developing nanotech sensors is their ability to pinpoint the exact location of a power outage
I'll give you a hint. It's the area where nobody seems to be able to use any electrical equipment.
The theory of relativity doesn't work right in Arkansas.
and call its by its name
Read more for additional details
or perhaps Roland is a script ? from the google search results it seems it just copies large chunks of other peoples articles and presents them on an advertising laden website and intersperses them with 20 word linking statements
should take about 5min for a perl programmer to replicate this Roland script
Just more proof that big brother is spying on you and your electricies! Damn that George Bush! What ever happened to our right to power outages since 9/11?? Might as well burn the Constitution.
For all of the technical details given in either article, they might as well propose monitoring the lines with an army of fairies that communicate using magic pixie dust, deployed via unicorns.
Help save the critically endangered Blue Iguana
to his blog? Isn't this another one of those self-promoting submissions? :P couldn't resist.
what about wired sensors? The infrastructure is already in place. When the sensors stop sending data, the power must be down at that location. It would be easy to map out which sensors arent responding.
Utility companies have been deploying Digital Fault Recorder systems in transmission substations and generating plants for 20 years. The DFR units are usually powered by a bank of DC batteries. The Digital Fault Recorders are wired into the transmission circuits and constantly collect waveform data from them. A satellite-based clock signal is used for a time reference. The DFR will typically monitor line voltages and current levels, line frequency, and harmonic distortion, sampling and testing the values 6000 times a second. If an error or "fault" condition is detected (blackout/brownout/line down/generator failing), the DFR will have a detailed snapshot of the event for review. From this digital snapshot, the location and cause of the fault can be determined mathematically.
When the batteries die, the "picotech" sensors on the nano-power-grids will send out warnings to the electric company...
Until their batteries die, that is...
Inductive charging works fine, put the unit on the line... Check EOS manufacturing: http://www.eosmfg.com/news/index.html
Sure, you would not want to power small devices straight from 400kV or whatever power lines, but it is very easy to get smaller voltages off a power line using corona charging currents etc.
Engineering is the art of compromise.
The article keeps referring to these sensors as nanotechnology.. since when were 2-3 inch devices considered "nanotech?"
New webcomic updated on Sundays: HERE
www.msfirefox.com
So I was reading this magazine called "Optimize". It's one of those freebie trade rags where you simply have to sign up as a "CTO / CIO" to get, along with about 5 lbs of junk mail every day.
Anyway, I was reading about so-called "autonomic computing" with "dynamic resource allocation" and "self-healing capabilities". It was this fluffy, buzzword-laden stuff that just didn't quite dig with me.
Just when I thought that there might actually be something here for me to look into, I noticed an example and jumped on it.
The example was of an "enterprise" backup that had to be done nightly, and some tech weenie had to remote in at 1:00 AM every night to check disk space and kick off the backup. How did they do it the autonomic way? Well, they set up a background scheduler that would automatically check for disk space and start the process!
Yep, that's right. A cron job that did about 5 lines of shell scripting. WTF?
This sounds to be just as buzzword laden and content poor. I've come to conclude that the number of buzzwords used to describe a particular application are inversely proportional to the substance of said application.
I have no problem with your religion until you decide it's reason to deprive others of the truth.
Power utilities are already(and have been for some time) deploying automated meter reading systems like DCSI TWACS and Tantalus(even Wikipedia knows about automated meter reading), and... wait for it... they are already using them for prediction and analysis of blackouts by pinging them. And if they lose power, and so they don't answer? That's Algorithms 101 for the non-thinking impaired to figure out what that means.
While, all these technologies can be improved, there is no need for a separate system, between automated meter reading(which has cost benefits other than outages... like not driving around all day writing numbers down) and substation monitoring, outages can be pinpointed very closely(I have seen it personally, and it gets 95+ % accuracy in identifying the location of the outage).
As an alleged researcher in this buzzword rich field, might I say what a bunch of hot air this is. A cynical person might say that this entire field exists only to extract funding from gullible funding bodies... and after a few years in the field I must say that my cynicism is growing fast.
i am not an electrical engineer. my question may be simple but implementation may be difficult.
the blackouts that happen are mostly cascaded effects with one circuit overloading and all others tripping and everybody has no power. is there any way of limiting the tripping to just a particular area or circuit instead of tripping the entire system? or there are systems in place to avoid that (which i believe is likely) but it doesn't work as planned? it always confuses me why a small problem causes huge disruptions (been in that situation a lot of times.) the behavior appears that there is just one circuit for all.
that said, will it be cheaper for that system instead of putting sensors all around? and as a side note, it is good that cities would experience power failures from time to time in order to provide people with awareness about disaster preparedness especially critical infrastructure such as medical and communications.
Live your life each day as if it was your last.
.. welcome our nanotech deploying overlords..?
As someone who works in the electrical power industry I can say, yeah, we have wireless sensors already. Hell, we have wired sensors too, because there's all these big frickin' wires already running all over the place anyway, don'tcha know!
...One of the many implications for the developing nanotech sensors is their ability to pinpoint the exact location of a power outage...
>
Uh.. we also have such technology already, and in fact it's quite old. The same signal reflectors that are in a LAN cable tester and tell you the length of a cable, are used on an industrial scale to tell you where the end of a power line is. Program the monitor with "this line is 9,374 feet long" and it sees "8,124 feet long" then it can, in fact, tell you exactly where the break is, right down to the foot! Now, these industrial grade units are highly expensive (partly by their shear power and range, because I'm grossly underestimating; line lengths can reach over 20 continuous miles), so it MIGHT be news if these little "nano" buggers are cheap and plentiful but can still do the job.
Virtually every piece of equipment we have on the line has remote monitoring capability. Now, whether the power companies are USING it is another matter, because of cost and infrastructure and such. My own company has substations we have no remote monitoring on, just because they were deemed low priority enough to not spend money on enabling it. So needing to send crew door-to-door to find a downed line or a damaged power box is just not necessary (though barring major disasters, it can be cheaper than installing all that remote monitoring equipment).
The one thing I do see in the "additional details" article is the idea of using these things, because they're so small, to monitor every single home and business on the grid. That's something we don't currently do, mostly for cost reasons. We can see a neighborhood is down, but not a house. THAT would be news worthy I suppose. Otherwise, I see nothing in this article that is new, just "we've made it smaller!" and they therefore tacked the "nano" buzzword onto it and acted like it was the first time anyone ever created such a device.
Important Stuff
* Please try to keep posts on topic.
* Try to reply to other people's comments instead of starting new threads.
* Read other people's messages before posting your own to avoid simply duplicating what has already been said.
* Use a clear subject that describes what your message is about.
* Offtopic, Inflammatory, Inappropriate, Illegal, or Offensive comments might be moderated. (You can read everything, even moderated posts, by adjusting your threshold on the User Preferences Page)
Problems regarding accounts or comment posting should be sent to CowboyNeal.
Is battery nano too? There's no valuable information in the articles at all. As far I know, wireless sensors aren't that small, and battery makes them very bulky. If they are using some new technology, where's proof?
Who's gonna change battery in nano device?
A system in British Columbia already exists..
Sensors with GPS units (used for their accurate timers) are placed on various different towers and connected via radio. They detect power spikes resulting from a line breaking / lightning hitting. Based on the times recorded when the spike passes over the two closest towers, the exact position of the fault can be deduced. If it's a break, a crew is sent out to repair the damage. In the case of a lightning strike, line switches can be signaled to disengage on either end of the fault before the effects of the fault pass over them. Once all is ok, the line switches can be reengaged.
They had all this working over 10 years ago when I did a little tour of their R&D lab.
Many areas of California have sensors deployed that run their health status over the powerlines themselves. making them "wirefree" (although not wireless as in radio). But these are not "nano", nor are they used to detect dangerous storms. They just sit on the pole and tell the power company if they need to send someone out to repair an issue. they do detect more than just power outage problems, enabling the power company to keep a high quality of service in remote regions. surprisingly this was done as a cost cutting measure, and has been successful. I wonder if a hall effect transistor or something on each "node" would let it detect storms without having to use some cutting edge technology as proposed in the article.
“Common sense is not so common.” — Voltaire
I recently heard about a guy defending his PhD thesis about technology to pinpoint the source of the failure by "listening to what the power failure sounds like" at the "powerline endpoint". I don't have any references available but that sure sounds much more attractive than having some nanobots around!
Roland's blogging on ZDNet's payroll now.
I wonder if he used "250+ submissions accepted on Slashdot" as a bullet point on his resume...
And for those who don't know the backstory, Roland Piquepaille and Slashdot will fill you in.
Thanks to the War on Drugs, it's easier to buy meth than it is to buy cold medicine!
the blackouts that happen are mostly cascaded effects with one circuit overloading and all others tripping and everybody has no power. is there any way of limiting the tripping to just a particular area or circuit instead of tripping the entire system? or there are systems in place to avoid that (which i believe is likely) but it doesn't work as planned? it always confuses me why a small problem causes huge disruptions (been in that situation a lot of times.) the behavior appears that there is just one circuit for all.
There's something better than just limiting blackouts, generate at least some power locally using solar or wind genies. Those who had such systems during the blackout in the Northeast and California a few years back had a easier tyme of it and had at least some power.
FalconShould there be a Law?
Link
TXU is supposed to deploy broadband over its power lines to 70 percent of its system. Why not just figure out a way to use this to see where the power went out. Put some devices at specific points that can be pinged, or make it a routed network and use traceroute to see how many routers it goes through. When the stuff that is supposed to be working, isn't, that is where the problem is.
The plus side is, they can fire a bunch of meter readers, have the above troubleshooting added...and they can sell the service and make the customers pay for it all!
Transporter_ii
Doctors destroy health, lawyers destroy justice, universities destroy knowledge, religion destroys spirituality
And they already do this via microwave and radio. More specifically, they monitor a lot of their substations and large important objects via a telecom link and have RF or microwave transmitters as a backup. I worked for the department that monitored the transmission and distribution of power and I got to see exactly how their entire monitoring system for their entire grid works. Every point in their grid that can cause havoc is monitored in real-time (4 second interval). I don't exactly see what this article is getting at. In fact, this kind of monitoring is a FERC requirement. While not on a scale of what this article is saying, they can already pinpoint outages to the closest substation or transformer.
Perhaps the article is the output from an "infinite number of monkeys" simulation?
Step 2: Use the word nano an awful lot
Step 3: Come up with an outrageous bill and say you need political support
"The necessary research must be completed, four to five years, at (a cost of) five to six million dollars per annum here at UB," Sarjeant said.
Check out the USA's flagship neutron scattering facility, which will churn out hundreds of scientific papers per year and put the USA at the front of materials science, on real science at the atomic scale (not inches, Sarjeant) and consider that 30 million dollars would kit them out with a whole suite of instruments for the entire scientific community to use (based on the quality of the proposals, of course). Sarjeant, you're a complete... no, I can't even be bothered to say it...
This message was scanned by European governments and contains no terrorism.
As for residential, google Scientific-Atlanta's LOCATE system. 10+ years ago they were putting small wallplug boxes in homes that used the telephone line to call in to a server if there was an outage. This could be connected to a GIS, to pinpoint the feeder that was down. This was used for things like fused lines, that you typically wouldn't have SCADA on. The wallplug box was charged by the electrical line, but held a 1 farad 5V capacitor inside. That was ample power for several tries to call in to the server.
...but not using nanotech. Instead organizations who buy and sell electrical power pay people, like farmers, who have power lines running through their land to locate a sensor near the line to detect whether power is running or not. The obvious advantage to this is that if a known line that is connected to a known power generator is not running then they can bid up prices at another power generator.
A hand up and a foot on every chest...
With most power companies rolling out ( or planning to roll out ) smart meters that can be read remotely in the next few years, all they need to do is ping their customers periodically. They can tell from who responds and who doesn't where the break is.
Regardless of the cause it is still not a case for extensive monitoring of the network. My point was that the Americans have problems identifying where their problems begin - in Germany and Europe that does not seem to be the case.
P.S. Thanks to Deutsche Welle TV I know a little about the German windpower issues. However what is the difference between storing energy production from wind turbines and solar power? Would you not need to "store" it either way? After all solar energy is hardly a 24 hour pursuit?