Mad Science: Smart Grid

Uploaded by IBMSocialMedia on 16.12.2009

>>Tim: We've got a problem. Cities are growing too fast and they're using too much electricity
Cities consume 75% of all the world's energy
but 400 billion kilowatt hours are wasted
each year because we don't use information effectively.
That's enough energy to power almost 40 million households.
And that's a problem. But we got a solution.
We're gunna tell you what it is and then
crazy John's gunna make lightning with a fork.
[Music playing]
Hey we're here again with John Cohn
IBM fellow, mad scientist, superhero, and swimsuit model.
John we're talking a lot about smarter planet
what IBM's doing that cities can learn from.
Explain this idea of a smart grid.
IBM is like a city. We buy our power
straight off the New England grid just like
power utility. Now, an electric utility has to generate
both for its average load, and for its peak load.
>>Tim: When you say peak load, what are we talking about?
>>John: Oh, like on a hot summer's day, people are all
turning their air conditioners on
>>Tim: OK
>>John: Takes extra power. They have to plan for that capacity.
What they tend to do is they generate power
at their average load as cleanly as they can
and then they bring on auxilliary things like, uh
oil fired, or gas fired generators
to make up for that peak. Now that's
more expensive, and it adds more carbon.
Now what'll happen is on a hot muggy day
the electric grid will send out an alert, saying
that hey, we got a high demand situation, and they ask
us to load shed. Because we've instrumented all
our circuits over there, we know all exactly down to the watt
where all the power is going. We can determine
what pieces of non-critical equipment we can shut down
to save power. And in turn they'll write us a big check
for saving that power, cause they didn't have to spend that money.
and if we can do it, cities can do it.
>>Tim: Look, this intelligence stuff, it intrigues me, I mean can you show me more?
>>John: Absolutely.
>>Tim: And then can you make lightning?
>>John: For you Tim, yes.
So, Tim, these are the hydro-electric generators I was telling you about.
These just happen to be down the street from us at IBM.
These beautiful machines have been in place since 1917
generating power 24 hours a day just from the Winooski River.
These are new generators over here, and these are deisel run
and these are just used on those peak load conditions.
Now, what we're able to do by load shedding
is we're able to rely more on these and less on those
This is completely green,and is free.
This puts carbon in the air and costs money to run because of deisel fuel.
>>Tim: Yo, my Mom's from Cleveland. Why do you ask?
>>John: So Tim, this is one of those giant air handling units
I was telling you about. Like a really fancy air conditioner
It supplies ultra pure air to the process lines below
Takes quite a lot of power.
Now, if you're in one of those load emergencies
We can turn this thing off for a couple of hours
as long as we turn it back on later
>>Tim: So you're just moving the times when you run things
to be more efficient. It's like my washing machine, right?
I mean, I still have to run that, but instead of running it
during the day when my AC is on, I wait and I run it at night.
>>John: Yeah, exactly
>>Tim: Or, I could not run it at all and just keep wearing the same pants.
>>John: So Tim, this is where we make the whole process smarter.
Remember I told you that every piece of equipment here is instrumented?
>>Tim: Yeah
>>John: So we know down to the watt, how much energy
the entire plant is using.
Not only that, we can predict it. That's that blue line.
And for example, you can tell, this was Monday
That was kind of a warm and muggy day.
We were using almost 60 mega-watts. That's a lot of power.
Where today it's cooler outside and we're using
somewhere around 48 mega-watts.
Now, if it had been an emergency
like a hot muggy day, and the New England Grid had said
Hey, we need load shedding
we would know exactly whic pieces of equipment
We would know that we could turn that stuff off
and that would prevent the electric grid from having
to start some of those backing generators
and put carbon in the air.
>>Tim: You know that's fascinating and all, but you
know what this trend line looks like to me?
>>John: What?
>>Tim: Lightning.
Alright doc, I get this idea about the smart grid
I'm just trying to wrap this all up, and would love to crystalize
it with some sort of visual, you know, some sort of model,
some sort of demo of, oh I don't know...lightning?
Would that be representative?
>>John: No, man, but it's cool!
>>Tim: Let's hit it.
>>John: Ok.
[Lightning buzzing and mad scientist laughing]
>>[John Crazy laugh]
[Music plays]