November 13, 2025 • 53 mins
#319
The modern power grid faces an existential challenge: integrating massive amounts of renewable energy while keeping the lights on. Sean Kelly, CEO and Founder of Amperon, explains how artificial intelligence and machine learning are replacing outdated forecasting methods to predict energy supply and demand with precision. This conversation explores battery optimization, demand response programs, and why traditional grid models no longer work in a world of extreme weather and distributed generation.
Sean Kelly leads Amperon, a company providing AI-driven energy forecasting platforms for grid operators, battery owners, and energy traders. With nearly 20 years of experience in energy trading and risk management, Kelly has worked for major energy firms such as Tenaska, Lehman, EDF, and E.ON, where he led significant trading desk developments and transitions, including bringing nuclear power plants into new markets.
Key Discussion Points:
- Traditional forecasting models look backward. AI models look forward and learn in real time, adapting to grid changes as they happen.
- Battery storage is exploding. ERCOT expects 40% more storage in the next 5 years. Texas batteries run 1-2 hours, California runs 4-hour batteries, and companies like Form Energy are developing 100+ hour systems.
- The critical hours are 6pm to 9pm. When solar goes down at sunset and wind drops simultaneously, batteries become essential. Amperon's platform tells operators exactly when to charge and discharge for maximum profit.
- The grid needs 6 terawatt hours of energy storage but we're only one-tenth of the way there. Developers should build all projects battery-ready or include storage today.
- Demand response works like insurance. Grid operators pay large customers to reduce load during peak times. Penalties are severe for non-compliance, ensuring the system works when needed.
- Forecasting is no longer optional. Sean calls it "the operating system of the modern grid." Historical averages won't keep the lights on anymore.
- Data centers and large industrial loads play a critical role in grid flexibility, backing down from 20 megawatts to 10 megawatts when grid operators need relief.
Connect with Sean Kelly
LinkedIn: https://www.linkedin.com/in/sean-kelly-0792626/
Connect with Tim
Clean Power Hour
Clean Power Hour on YouTube
Tim on Twitter
Tim on LinkedIn
Email tim@cleanpowerhour.com
Review Clean Power Hour on Apple Podcasts
The Clean Power Hour is produced by the Clean Power Consulting Group and created by Tim Montague. Contact us by email: CleanPowerHour@gmail.com
Corporate sponsors who share our mission to speed the energy transition are invited to check out https://www.cleanpowerhour.com/support/
The Clean Power Hour is brought to you by CPS America, maker of North America’s number one 3-phase string inverter, with over 6GW shipped in the US. With a focus on commercial and utility-scale solar and energy storage, the company partners with customers to provide unparalleled performance and service. The CPS America product lineup includes 3-phase string inverters from 25kW to 275kW, exceptional data communication and controls, and energy storage solutions designed for seamless integration with CPS America systems. Learn more at www.chintpowersystems.com
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Sean Kelly (00:00):
Forecasting is no longer a nice to have. It's
(00:02):
literally the operating systemof the modern grid. The
traditional models lookbackward. Ai lets us look
forward and learn in real time,and so these historical averages
just they're not going to keepthe lights on, and we don't
always keep the lights on. Butas someone who's struggled with
this over the last four years,keeping the lights on at their
(00:22):
home. This is something that wereally, really need to pay
attention to in this new worldof extreme weather and
distributed generation.
intro (00:30):
So are you speeding the energy transition here at the
Clean Power Hour, our host TimMontague, bring you the best in
solar, batteries and cleantechnologies every week want to
go deeper into decarbonization.
We do too. We're here to helpyou understand and command the
commercial, residential andutility, solar, wind and storage
industries. So let's get to ittogether. We can speed the
(00:52):
energy transition
Tim Montague (00:57):
today on the Clean Power Hour AI forecasting for
power markets. My guest today isSean Kelly. He's the CEO of
amperon. Welcome to the show,Sean.
Sean Kelly (01:07):
Hey. Thanks so much for having me. Tim.
Tim Montague (01:09):
It's great to learn about this industry,
because I don't know very much,so I'm I'm really all ears on
this one. But why don't youstart with a little background
on yourself, and then we'll diveinto what amperon is up to in
the world.
Sean Kelly (01:24):
Yeah, absolutely. So I've been in power markets for
my entire career, since the twoweeks after I graduated from
Texas, A and M walked onto atrade floor at tenaska. At
tenaska started in ERCOT, theTexas market, happened to live,
live here most of my life, andI'm currently residing in
Houston, Texas. And went fromtenasco back to Houston, and was
(01:51):
at a great company called EagleEnergy Partners, which got
bought by Lehman Brothers. Ithink we all know how that one
worked out. And then EDFelectricity, France trading came
in and bought us. I guess whatmakes me uniquely qualified in
this is I've run more than threedozen power plants, everything
from wind and solar to fivenuclear assets, including
(02:13):
helping with the acquisitionintegration of two of them, Nine
Mile and g'day in New York. Alsomoved to Chicago in 2013 to help
set up eon, the big Germanutilities trade floor, as the
first trader in North America.
So yeah, been a really funcareer on the trading side,
which is really helps meunderstand our customer types.
(02:34):
And so I was living in New York,moved there in 2017 for a girl
who's now my wife, one of mybest trades, or my best trade
all time,
Tim Montague (02:48):
that's awesome.
Yeah.
Sean Kelly (02:50):
And in 2017 you could walk around New York and
almost trip over a datascientist. They were they were
everywhere, data engineers.
Everyone was a co founder anddisrupting and all of the
buzzwords, and that made merealize that all my friends in
Texas probably needed to be ontop of this latest and greatest
AI machine learning and and sothe joke between my co founder
(03:16):
and myself is he'll build it andI'll sell it. And so that's what
we've been doing here for almostthe last eight years at amperon.
Tim Montague (03:25):
My gut tells me there's a lot of parallels
between financial trading andenergy trading. Is that
accurate?
Sean Kelly (03:33):
There are definitely there. There are parallels, and
then there are also diversionsbetween the two. But I mean
trading is trading. It is alwaysbuy low, sell high. And a lot of
the financial trading andelectricity is very has a lot of
similarities to trading othercommodities. I mean a little bit
(03:53):
of equities, but you see thesame different options available
and things like that. So, yeah,it's a it's a really fun field.
I was the weird kid who wantedto be a trader from the time
they were about eight years old.
And here I am, guess that's justa few years later, still playing
with trader.
Tim Montague (04:11):
And what is the status of amperon? How big is
the company? And what marketsare you serving?
Sean Kelly (04:18):
Yeah, so ampron is the leading energy forecasting
company, operating at theintersection of AI and energy
data. We have raised 38 millionin venture capital. We have a
little less than 100 employees,and we are active in 23
countries with offices in bothHouston and London, but cover
(04:39):
the entire continental 48 partsof Canada, Australia and almost
20 European countries. So justactually signed our first deal
in the Middle East. And so we'vebuilt a truly global company
that can stand up forecastanywhere. Because, as you very
much know, wind and solar worksin any country. And so we our
forecast works in any. Countryas well.
Tim Montague (05:02):
All right, cool.
Well, what is something aboutthe power market that most
people completely misunderstand?
Sean Kelly (05:09):
Sean, yeah, when I started in my career, people
asked why I was working atnight. And electricity is a
physical commodity. Thesemegawatts move across wires in
real time. It can't it can kindof be stored in batteries, but
it's not like an oil or anatural gas where it can be
fully stored throughout anentire season. And so back to
(05:33):
what we talked about a minuteago. They people do think of it
like stocks or crypto. It'sfundamentally different. It's
actually more volatile. It's themost volatile commodity out
there, as we've seen in some ofthese crazy events, whether it
be polar vortex, winter storm,Yuri winter storm, Elliot and so
(05:54):
yeah, it's it's definitelymisunderstood, but it's also
it's not going anywhere. Everysingle industry is extremely
tied to electricity, as in,without it, no other industry
exists. So in my unbiasedopinion, it is the most
important thing that we shouldbe focusing on.
Tim Montague (06:12):
Yeah, it's so true, we take electricity for
granted, but literally, the goodlife completely depends on it,
and the good life goes awaypretty much instantaneously when
electricity goes away. It's adouble whammy, too, right,
right? Because then when there'sa big outage, like you're you
mentioned in Texas back in 20what? 2221 electricity prices
(06:36):
spike, because it's a very rarecommodity then, right? And so
it's also super painful forthose who have electricity. It's
not just those who don't haveit, but Well, what is, what is
the problem, I guess thatamperon was created to solve.
Sean Kelly (06:55):
Yeah, so in 2017 when I met my co founder and we
started the company in Januaryof 2018, we knew that there was
a lot of changes coming down thepipe. Smart Meters had started
getting adoption. You used tohave back in the back in the
glory days, the meter madebreaking in your backyard,
getting bit by the dog. Now youjust drive around with the truck
(07:18):
and you get on a 15 minute readfrom someone's home. We knew
that was happening. We knewrenewables were happening some
behind the meters, such assolar. You've got places like
Australia, which are all in onthat, and then Texas and
California and other westernstates that are also adopting
it. So knew that was coming.
EVs, I mean, the Tesla waspopular. Other companies were
(07:39):
talking about having EVs. Herewe are in 2025 and they're
absolutely everywhere thatchanges what your electricity
profile looks like. So we knewall this was coming. I also knew
that much of this was being donein Excel, and there were some of
the largest utilities out therethat were running these monster
models overnight and justpraying and hoping that they
(08:03):
would get an answer the nextmorning, but oftentimes the
macro would break becausethere's so many lines of just
lines on lines on lines in theseExcel spreadsheets, and that's
not going to work. And soamperon was built to deliver a
real time AI driven forecast sothat these traders and utilities
can make better decisions.
Tim Montague (08:25):
Now, back in 2018 the definition of AI was maybe
different than it is today. Imean, most knowledge workers and
prosumers are aware of llms, andmaybe, if you're knee deep in it
HRMS, which is a new model thatjust emerged. But these large
(08:46):
language models, do they haveanything to do with the
technology that you use atamperon?
Sean Kelly (08:51):
I mean, we look at them from a basically workforce
efficiency standpoint, and sothat is something we're using
just again to speed things up.
Got a very talented engineeringteam, data science team, they're
always looking at the latest andgreatest to continue to speed up
their efficiency, how many, howmuch they can crank out in the
in the hours they've committedto the team. And the way we've
(09:13):
looked at it, though, is is thereally the machine learning
standpoint? And so our firstmodel went live in November of
2018 and machine learning wasinvolved in that very first
model. And again, forecasting ishard. We're still always looking
at and tweaking the models, butour models have retrained every
(09:34):
hour since November of 2018 thatis a lot of compute cost, and so
we've continuously looked atthat. We also know that one
forecast is not always right,and so we're always running an
ensemble of four to sixdifferent forecasts. And so
looking at the simple regressionto gradient boosted trees to
(09:56):
whatever the latest and greatestis. And. That's why we're really
fortunate to have a team ofabout a dozen data scientists,
mostly PhDs, just always dealingwith this problem because it's
hard. I don't know what'shappening in your house. I need
to know when you bought an EV Ineed to know if you have solar
panels. I need to know what abattery is doing. This problem
(10:16):
is getting so much morecomplicated than back in the
day, when everyone went to workat eight o'clock, they came home
at five o'clock, every homeprofile looked very similar.
Now, home profiles are all overthe place, and so we we've got
to stay on top of that for our150 plus clients.
Tim Montague (10:36):
So let's talk about this from the perspective
of, say, an IPP, a company thatowns fleets of wind, solar and
battery projects, and let's talkabout Texas, since you you know
that market certainly very well.
ERCOT is a unique market. It isnow on a windy, sunny day, 50%
wind, solar and battery powered.
(10:58):
Who knew? Right? Ginormous. RedState is going green. Texas
eclipsed California on an on anannual basis. Now, in terms of
the install solar, right? We're,I think Texas did 11 gigawatts
last year, something crazy likethat. But if you're an IPP,
(11:18):
okay, you're in the business ofselling electricity to various
and sundry off takers. Walk usthrough that business and what
are some of the challenges andopportunities that they're faced
with on a daily basis.
Sean Kelly (11:33):
Yeah, Texas has done a great job of just really being
a pro business state and gettingan entire full text or full
generation stack. And so, Imean, we have 40 gigs of wind
and a ton of solar. And so forthese IPPs that are our
customers, independent powerproducers, for those of you who
aren't all in the acronym weeds,what they do with our forecast
(11:57):
is they give us said asset.
We'll pick a solar asset. And sowe'll give them a 15 day, 360
hour forecast for their specificsolar installation, and we'll
call it 100 megawatts. And weactually give them five minute
granularity, because, as we allknow, Sun kind of does what it
wants to. It's more reliablethan when from knowing when it's
(12:20):
going to be sunny versus not,but we've got to stay on top of
this. So a five minute forecastis the bare minimum that you
need to actually understand whatyou're doing with this, and it's
what you can actually go andschedule. And so we tell you on
this five minute granularitywhat it looks like, so that you
can tell again, what was we'repicking on, ERCOT here. ERCOT,
what you're going to be showingup within a day ahead. And so
(12:42):
every day, by about 10am youneed to tell ERCOT, hey,
tomorrow I will have this muchsolar. And so that's what
amperage forecast provides toyou. And then as we continue,
the models rerun every singlehour, and you're getting an
update there. So very similarwith wind we're telling you,
we're updating like multipletimes per hour, on five minute
(13:04):
cadence, what your wind farm isdoing so that you can best
monetize it and run itaccordingly.
Tim Montague (13:12):
And does the IPP know, though, from that model,
what kind of revenue they'regoing to be generating?
Sean Kelly (13:19):
Yeah, they can take, they can take our output that
we're telling them, and thenbasically put it with the the
clearing price, whether it's theday ahead price, if they bid it
into the Ford market, or thereal time price, if they if they
took it, if they floated it tothe intraday, and then they can
figure out what their revenuegeneration is going to be in
(13:41):
figuring out how much money ampruns helping increase on a
return on investment.
Tim Montague (13:47):
So what is let's just step back a little bit,
because I think obviously somefraction of my listeners will be
familiar with this energytrading business, but only some
fraction so in the greaterscheme of things, IPPs are
signing what kinds of contractsto they need off takers, right,
(14:11):
to develop the project out ofthe gate, frankly, right? And,
and who is, who is a goodexample of a buyer? And then
what are the what is thevariability in that contract?
Sean Kelly (14:28):
So what they're going to do is, you're
absolutely right. If I want tobuild 100 megawatt wind farm to
keep nice round numbers, I wantto finance it and not have to
foot the entire bill up front.
And so the best way to do thisis sell a certain percentage of
my of my wind going forward. Andso they'll do a PPA power
purchase agreement. And we'llsay a 10 year PPA is what
(14:52):
they're looking at. The mostnatural off taker is
municipalities. Cooperatives, sotowns. So they'll go in and say,
hey, my town would love to buythe 100 megawatt wind farm. 100
megawatt wind farm does notproduce 100 hours or 100
megawatts every hour, right?
(15:14):
It's not a base load. It'sprobably never going to produce
100 megawatts. That's just thethe nameplate capacity of it. So
I'll go in and say to Tim city,I'm going to give you 30
megawatts, like around the clockgoing 10 years. And so I would
sell you that, and you wouldsay, Great, I have 30 megawatts
coming. Well, you then get themoney from that, and you're able
(15:39):
to go build the wind farmagreement reached. Well, then
you've got to deal with theimbalance. And so if only 12
megawatts are going to show up,I own Tim City, 18 more
megawatts. And so that's where Ineed to know, with amperons
forecast, that I've only got 12megawatts tomorrow. It's just
not a windy day. I need to gobuy that like that interval of
(16:00):
18 megawatts from the market. Sothat's how, that's how these are
being built, and these are howthese are being financed. And
again, the the PPA market, it'snot, it's not an exact science.
It takes away a lot of thecommercial upside. However,
unless you have just the abilityto fork over the full amount
(16:22):
whenever you build it. Then,yeah, it makes a lot more sense
to go finance that PPA. And
Tim Montague (16:28):
then, as the asset owner, I'm getting a guaranteed
price on the wind power, but I'mbuying grid power at some
variable price, and that's alsowhere this kind of a platform
comes in handy.
Sean Kelly (16:42):
Yeah, you agree to said price, and that's where you
look at what the Ford curve istrading. So you'd say, hey,
2026, through 2025, is at thisnumber, and said like
municipality or Co Op will gobuy it. You. I guess examples of
other people buying this, orobviously, utilities will go and
buy this. Oftentimes they gobuild their own fleets, just
(17:05):
because they're normally quitewell capitalized. And so they'll
say, hey, I need to have thisfor my rate payers. And then
you'll also have now, what'sreally exciting is you've got
data centers and hyper scalerswho are coming in because
they're normally in very goodfinancial shape, and so they're
able to come in and reallyincentivize this. And many of
(17:27):
them, as we know, have madegreen commitments, and so
they're more they're more likelyto be excited about the wind and
solar if they have a greencommitment. But they're also
trying to figure out just, youknow how to, how to make sure
that said data center can canrun as needed.
Tim Montague (17:45):
And how risky is this business in the greater
scheme of things, you you knowyou mentioned, okay, if I have
100 megawatts of a wind farm, Imight be able to sell a contract
for 30 megawatts, but the windgoes up and down. Some days are
it's blowing, and some days it'sstill for a week, it could be
(18:06):
multiple weeks where there'svery little wind, and then I'm
buying power at some notnecessarily known rate. So yeah,
just how risky is this businessin general?
Sean Kelly (18:19):
I mean, power is a risky business. I mean, again,
I've been, I've been goingthrough this for 20 years. I
mean, you looked at summer 2011in ERCOT, there was not a lot of
wind. It was extremely hot.
Power prices were very high.
2014 you had the polar vortex. Ilived in Chicago then, and that
was a rough one. Being a Texanin Chicago, dealing with like
(18:41):
minus 40 feels like inFahrenheit. And so again, not
the renewable generation didn'tshow up as much then. And so
this is why having a 15 weekforecast is so important. We're
able to go out and for instance,amperon, you mentioned winter
storm area earlier, so that thatwas a little too near and dear
to me, from the standpoint of mywife, was six months pregnant at
(19:05):
the time, and the power went outat 2am on February 15 of 2021 we
had no marketing department atthe time, so I wrote a really
fun blog post on Valentine'sDay, ideas to do in the dark,
because February 14 is what dayI thought the power was Going to
go out. Went out at 2am on the15th. And so we were able to
tell our clients on February 3,our meteorologist nailed it and
(19:30):
say it's going to get reallycold. So multiple IPPs went and
bought power at $60 permegawatt. It wound up clearing
$9,000 per megawatt those days.
So because these different windfarms went and bought power,
because amperon, they did theirown research too. We don't give
full credit here, sure, butthey, but they went and bought
(19:52):
$60 power, and then it tradedall the way up to 9000 and so
that's the way that amperonReally. Really helps them
understand that, because theyknew they weren't going to be
able to supply all the PPAs thatthey had previously agreed to
that week of February 15, and loand behold, yeah,
Tim Montague (20:12):
but you literally were able to predict that the
grid was going to fail.
Sean Kelly (20:17):
That's not exactly what we do. I knew from my
experience that the grid wasgoing to fail, and literally put
out a blog post about it, or aLinkedIn post about it, that we
were going to lose power, and welost power, and then on. But we
we put that it was going to be avery, very like epically cold
event, and we started sayingthat at the very beginning of
(20:40):
February. So we gave ourclients, I mean, 12 days notice.
We'll say 10 days notice,because it really got dicey on
about the 13th on that Saturdayis when prices went well into
the 1000s, and then it justcontinued on. It was a one heck
of a Valentine's Day weekend.
Tim Montague (20:58):
And so who are the winners and who are the losers
in a stormy area scenario?
Obviously consumers are, are thebig losers, consumers and
business owners like not havingpower is painful and
uncomfortable, but yeah, justwalk us through that.
Sean Kelly (21:16):
Yeah, there were so many losers in that. And I would
say the the biggest losers, younailed it, definitely consumers,
those rate payers at the end,there are people who are on
variable contracts at their homethat were literally getting
their bank account hit, like,every like, every 15 minutes
that were on auto pay from acompany that's no longer around
(21:39):
called gritty and so, like, youliterally, like those were the
biggest losers. And the factwas, the consumers, the business
owners, obviously you lose thosedays of doing business. And then
I would also say that anyone,anyone, I mean, natural gas,
didn't have natural gas plants,also didn't all show up. So they
(22:00):
had contracts to sell that theywere they had to go buy those
back. And that wasn't fun.
Getting gas wasn't fun. Thereweren't a whole lot of winners
out of that. I mean, itdefinitely put the state in a
pretty bad situation. I do thinkit made people more resilient,
because you look in thetemperatures we had, I mean, as
you're in Illinois, they'relike, teens. Yeah, no big deal.
(22:20):
Nothing to see here. I mean,when I lived in Chicago, we had
teens for like months at a time,that 2014 and that 2015 but in
Texas, our everything wasn'twinterized properly, because we
never seen teens before. And sonow, I mean, I think ERCOT and
the PUC and the rest have done areally good job of pushing for
(22:42):
that winterization is nowwhenever we have a slight scare,
it winds up being or so far,like knock on wood, has been
pretty much a non event.
Tim Montague (22:54):
The Clean Power Hour is brought to you by CPS
America, maker of NorthAmerica's number one three phase
string inverter with over 10gigawatts shipped in the US. The
CPS product lineup includesstring inverters ranging from 25
kW to 350 kW, their flagshipinverter, the CPS 350 KW is
(23:15):
designed to work with solarplants ranging from two
megawatts to two gigawatts. CPSis the world's most bankable
inverter brand, and is America'snumber one choice for solar
plants, now offering solutionsfor commercial utility ESS and
balance of system requirementsgo to Chint power systems, com
(23:37):
or call 855-584-7168, to findout more. Yeah. All right, I
could talk about just that oneevent, probably all day. But
let's move on. So what sets whatsets April and apart, though
from other forecastingcompanies,
Sean Kelly (23:57):
yeah. I mean, what sets us apart is accuracy. I
mean, we're two to three timesbetter than grid operators.
ISOs. TSOs is what they callthem in Europe. We also, we
cover the whole value chain. Weforecast demand. But demand is
not the whole story. It's netdemand, as you and your
(24:18):
listeners know, which net demandis pulling out renewables. So
it's demand minus wind minussolar, because those show up and
and make it much easier on thegrid. And we also forecast
prices, and so you're able tosee this across the value chain.
And then this is also built bytraders and data scientists. Our
first hire was a PhD datascientist who's still hanging
(24:41):
out with the company over sevenyears later, and then my second
hire, which the investorsthought was a little weird, was
a meteorologist. And why ameteorologist? Because
forecasting, weather is so, soimportant to this. And so those
are some of the things thatsetting up. Run apart, but also
(25:01):
we get a little bit of a latemover advantage. Like I said, we
started the company in Januaryof 2018, so many of our
competitors are 20 or even 30years old. We've been cloud
native from day one. We've beenusing AI since day one. This
wasn't, Oh, we've got a goodcompany. We've got clients.
We've got to move out of, like,we've got to move out of just
(25:22):
running this off our ownservers. Oh, we've got to move
this out of Excel. Oh, we've gotto move this out of we started
with a modern tech stack fromday one, and so that's given us
a huge advantage and anadvantage that we don't take for
don't take for granted, and wakeup every single day trying to
figure out how we can improveour models and pay attention in
(25:43):
the next, latest
Tim Montague (25:45):
and greatest. And are companies building their own
systems for this? Or is it allthird party platforms?
Sean Kelly (25:54):
It's definitely both. I mean, people will look
at it as a core competency,especially utilities, but
normally they also want to havea third party come look at us.
Learn look at it. Theirforecast. The reason why is, as
I mentioned, we have 150 clientsand over 40 million meters in
(26:14):
the continental United Statesthat are in some way on our
platform, either as individualmeters or in like aggregations.
And so there's no utility thatit has 40 million meters. And so
we are able to have already seensome things. So when we walk
into utility number four, we'vealready dealt with utility one,
two and three, and so we alreadyknow what their problems are and
(26:37):
what they're seeing. And so itreally helps that kind of been
there, done that before, asopposed to having to start from
scratch. So a lot of people willstill run their own internal
models. We're going up against alot of 20 year old models that
they just don't trust because ofhow times have changed, but
they'll run those with ours. Andas I mentioned, we're a big, big
(26:58):
fan of ensembles, and soensembling is basically just
waiting different forecaststogether to make one super
forecast. In layman's terms,
Tim Montague (27:09):
I'm curious if you know the weather's changing, and
you can no longer say, well,let's just look back at the last
10 or 20 years to predict howthe weather is going to be in
February of 2026 is that afactor?
Sean Kelly (27:26):
Now, absolutely. I mean, it is. Weather's gotten
crazy. I mean, we're sittinghere talking about a one and 100
year event. I mean, I rememberearly in my career, all we were
worried about was hurricanes.
And the reason we were worriedabout hurricanes is most natural
gas came from the Gulf, and whena hurricane came in, it took out
(27:47):
gas production, and natural gaswent through the roof. That
happened in 2005 right when Istarted. And now, when a
hurricane comes, natural gasactually dips, because it takes
out load, and we don't reallydrill offshore that much, and so
that was really kind of the bigchange. Now we're sitting here,
and I can rattle off. I mean,all of the events starting, I
(28:10):
mean, ERCOT, 2011 was 62 days ofover 100 degrees in Dallas. I
mean, we've talked about 2014with the polar vortex. We've
talked about 2021, with winterstorm Yuri. I mean, blacked out
a whole bunch of people onChristmas Eve across the east
coast with winter storm Elliot.
(28:31):
I didn't know what a derecho wasuntil last May, when it came and
took out a bunch of power linesin my neighborhood and I didn't
have power for seven days. Imean, you just continue to see
these hurricanes are gettingbigger every year. It feels like
we're saying it's going to be awilder hurricane season.
Wildfires are more prevalentthan I can ever remember, just
(28:53):
in the last handful of years. SoI mean, things are getting wild.
Extreme events are here. And Imean, you just off the cuff,
reference a one in 100 yearevent. But is it really a one in
100 year event? If it happensevery two months?
Tim Montague (29:10):
Yeah, it is crazy.
So talk to us about forecasting.
How is that the role offorecasting evolving as the grid
gets smarter and more dynamic.
You know, you mentionedelectrification of
transportation, data centerscoming online, and, you know,
the economy is growing, and it'selectrification of everything.
(29:34):
It's not just EVs, it's heatpumps, HVAC, industrial
processes. So how is your How isyour business evolving?
Sean Kelly (29:47):
Yeah, forecasting started as a nice to have and
that, and now we're sittinghere, and it's literally the
foundation of daily operations.
The reason it was a nice to haveis back to what I referenced
earlier. Smart meters is youjust got a data point every
3060, 90 days, whenever theMeter Reader came around, so you
didn't really know howimbalanced you were. Now we're
(30:08):
sitting here and we're getting,I mean, some places are getting
millisecond reads, a nestthermostat or something. Of that
is picking up data all the time.
So the amount of data that wehave to take in has gone just
absolutely through the roof. Andso every single trading
strategy, dispatch decision andeven long term defense
(30:32):
investment depends on it. And soif you think of forecasting,
it's really the operating systemof the modern grid. And so what
we're doing, we knew it wasgoing to be important. But you
also, I mean, you referenced notbeing able to have look at the
last 10 years of weatherforecast. We also can't look at
the last 10 years of load data,because I didn't know when I
(30:55):
started this that covid wasgoing to be a thing. It changed
what a landscape of a citylooked like for as little as six
months to as much as threeyears, just depending on what
the like, what the lockdown,what rules were when people went
back to work, and now we'resitting here in a in a remote
economy, and we have people in,I think, 17 different countries
(31:19):
who work at amperon because Weget to hire best in class, but
they work at their house, sotheir house doesn't look like it
would have in the 1980s and sothese are all things that I
mean, all things that arechanged that when we started the
company, we couldn't haverealized we're going to hit but
continue to evolve and makethings more difficult, Which
means we have to continue to getsmarter.
Tim Montague (31:43):
And what are the what are the limitations that
you encounter with traditionalforecasting methods, and
especially with regard to AI?
How have you leveraged AI totranscend that legacy
technology?
Sean Kelly (32:00):
Yeah, and, and, and, just to be clear, we still use
the legacy, I mean, linearregression models have been
around for quite a while, and westill have that as a very core
input into our models. Butagain, being able to
automatically weight each modeland having it retrain every
single hour is not somethingthat you can go do in Excel for
(32:20):
40 million meters. It's just notcomputationally possible. And so
that's where the the AI in MLmachine learning come into play.
And so we're we're looking atthat. We're also looking at
weather. Weather has made hugestrides here, just in the last
two years. Whether it be fromlike the models that NOAA puts
(32:41):
out to you've got the like,European AI models with much
better granularity. So weatheris something we look at. We
currently use for weathervendors, and we're always
looking for like, the next,latest and greatest. So I mean,
weather is something that I'mvery, very I mean, it gets more
complicated and continues to getmore complicated, so you need
(33:02):
better models just to stay atwhere you were if things
continue to progress like this.
So that's where the AI reallydrives in. So weather has been a
huge help, and that's why weliterally, we've got a team that
goes in and tests out all thelatest and greatest AI models to
see which one's going to hit is,I mean, the the people with the
most money in the world, the theNVIDIA is, the Googles, etc, the
(33:24):
world are putting out their ownmodels, which is definitely, uh,
definitely worth diving into.
Tim Montague (33:31):
So tell us a little more about the Machine
Learning, though, to the extentyou can that you're leveraging
it's, it's built on top of someother technologies, like open AI
or Claude or Google. Yeah, I'mjust curious how that, how that
(33:51):
works?
Sean Kelly (33:52):
Yeah, so it's, it's not built on top of LLM
technology, like some of theones you mentioned, what it's
really doing is going in andwaiting our forecast and running
different forecasts and sayingthat, hey, this like grading
boosted tree needs to be withthis deep learning model. We'll
put in 15% of this one and 17%of this one. And then also
(34:14):
different clients are different.
Some, especially in the retailenergy provider world, will be
all residential. Their booklooks very different than
another client who's allcommercial and industrial. So we
again, don't have someonesitting there saying, oh, make
sure to use the commercialmodel. It just automatically
goes in and uses the commercialor the industrial model or the
(34:34):
residential model. So again, youdon't want someone sitting there
across 150 customers trying tofigure out which model to use at
which time. And so that's wherethe machine learning really
kicks in. And again, the datascience team has done a
phenomenal job of doing thissince inception. So I'm glad
this isn't something that wekind of freaked out about a year
(34:55):
ago and said, Oh my gosh, wehave to use AI in some. Thing.
One of the coolest, I guess,accolades that we've received is
Andreessen Horowitz, who we haveno money from, put out their top
50 AI American dynamismcompanies. And right when AI
became a thing about two yearsago, and there was one company
(35:16):
from Texas, and I believe therewas about, there's like, eight
to 12 energy companies, and wewere, we were on that list. And
so again, we've been using thissince very inception, and the
machine learning makes all thedifference. That's where you get
to go against a competitorrunning in Microsoft Excel. And
that's also something thatputting on an updated model, as
(35:39):
quick as weather changes, and asquick as sun and wind and all of
that changes, as opposed to somany of our competitors, we're
literally running their modelonce a day, which just does cut
it once or twice a day. And thatthat's that's not going to work
with how fast weather ischanging these days.
Tim Montague (35:59):
So when, when the next storm Yuri comes? There's,
I guess there's a couple ofthings on my mind about this.
The grid is changing. We arehardening the grid now in Texas,
we're also increasing theattachment rate of batteries,
and the percentage of wind andsolar on the grid is going up,
(36:23):
but the attachment rate ofbatteries is going way up. So
there's that. So the grid ismore resilient. And okay, maybe
our maybe our forecasting ofweather patterns is getting
better. So walk us through ascenario for another Yuri style
event. How much warning does thegrid operator have, and how
(36:47):
much, you know, flexibility dothey have? Like, one of the
things that's noteworthy aboutERCOT is it's it's famous for
being isolated from other ISOs,and so they can't necessarily
count on getting a bunch ofpower from some other pool when
the shit hits the fan.
Sean Kelly (37:07):
Yeah, I mean, and that's where, I mean, that's
where having reserves reallykicks in, and incentivizing the
reserves to act at the correcttimes. And then that's also
where demand response reallykicks in. And so a lot of the
load that we're talking about,right? You can't, you can't get
through a conference. Can't getbarely get through a podcast
without talking to demandgrowth. So might as well just go
(37:28):
for it. But a lot of the demandgrowth that we're looking at, I
mean, yes, we're seeing heatpumps, we're seeing EVs, we're
seeing demand growth at your ownhome, but the real load growth
that we're seeing is datacenters and large industrial I
mean, again, more more things onshore and just that stuff's
(37:49):
pretty flexible at the end ofthe day. And so that's something
that I think that the gridoperator is doing a good job of
paying attention to how much thegrid operator gets involved.
There's something that justpassed in Texas called SB six,
and that is basically reallypushing on data centers to have
to behave whenever times gettight. And so I think that's
(38:13):
really the next step. So ERCOT,you're absolutely right. Is more
of an island than most places,but I think they have taken the
proper responses from a demandresponse, and incentivizing
those large loads to be on goodbehavior and be good stewards of
the grid. So that helps. Andalso, if it's during the summer,
it's hot out, and that's whenthe solar is going wild, so
(38:33):
you've got a few hours at theend of the day, kind of hours in
the 1921 which is 7pm to 9pm athome, and those hours, that's
when the batteries come in. Sohaving those, although it'd be
only one to two hour batteries,ERCOT spreading those out and
kind of saying, These are thehours I need you that's really
critical to keep the gridrunning. But ERCOT had a smooth
(38:56):
summer PJM less so I think theyhad a dozen EA one alerts, and
they also had something that inJune, they blew through what
they said was their peak loadfor the entire summer, and
that's just because they havesuch a wide footprint that they
did not expect that it would gethot from Chicago all the way to
(39:18):
Virginia, and it did. So it hada heat wave that just kind of
sat over that whole Mid Atlanticregion. And so PJM saw that. But
for us, that made people realizethat you can't sleep on PJM
forecasting, either or New Yorkforecasting, because it was a
hot, hot summer. And those,they've had some pretty mild
summers for the last handful
Tim Montague (39:40):
so the grid operator sees these events
coming, right? They say, Okay,it's going to be super cold.
People are going to be runningtheir electric heaters, their
gas furnaces, whatever, right?
They have some ability topredict what the load is going
to. Be over time. And then ifthey notice that the load is
(40:01):
going to be greater than whatthey capacity of the grid is,
they will send a signal to largeoff takers and say, hey, at, you
know, four o'clock tomorrow, wewant you to throttle your data
center. Is that kind of how thatworks?
Sean Kelly (40:19):
Yeah, you, I mean, you pretty much got it there.
They go through and they say,All right, the supply stack is
this, and we think demand isgoing to be this. And this is
why having a good wind and solarforecast is so important as you
can, for the most part, tell acombined cycle when to show up.
I mean, not a lot of coal left,and not a ton of nuclear
(40:40):
unfortunately, but those arepretty reliable base loads, and
so you've got to understand whatthe difference is. And so yeah,
you tell people who are enrolledin these different demand
response programs to, hey, 4pmthat's your time. Like, go ahead
and plan ahead. Go ahead andsend your workers home. Like, I
need you to do what you can do.
So yeah, and
Tim Montague (41:03):
then if they don't do what they can do, like if
that demand response, for somereason, fails, does the grid? Is
the grid smart enough or savvyenough that the grid operator
can go, Okay, I'm going to shutdown this, this circuit to this
major off taker so that I don'thave an uncontrolled outage.
Sean Kelly (41:25):
So the way demand response works is it's like
insurance. And so I'm going topay you $10,000 to be on good
behavior when I tell you to beon good behavior. And so the
penalties are nasty if you arenot on good behavior. So if
you're sitting there and you'rerunning a data center, and you
promised that you'd go from 20megawatts an hour to 10
(41:48):
megawatts an hour, and I alreadypaid you your $10,000 it's way
more than that, but I alreadypaid you your money to back
down. Then you will back down.
And so that's where. That'swhere. Again, there's not
really, I see behaving on thegrid and very similar in, like,
especially in the East, whereyou have capacity auctions.
That's, I mean, the capacityauction was obviously a record
(42:10):
in PJM, and so they are saying,like, you are promising that you
will back down then. So, yeah,the you don't have to worry
about the good behavior. Thegood behavior has already been
paid for.
Tim Montague (42:22):
All right, so let's talk a little more about
storage. You know, storage iskind of the new kid on the
block, but there's lots of it,you know, and lots more coming.
I think I saw that there's 40%more storage coming in the next
five years in ERCOT. And ofcourse, ERCOT is just one
(42:44):
market. There's incentives inIllinois, New York,
Massachusetts, New Jersey,Connecticut. I mean, many states
now California have programsincentivizing the installation
of batteries. What should ourlisteners know about batteries
as this relates to what amperondoes.
Sean Kelly (43:05):
Yeah, I love batteries, and the reason why is
early in my career, I got tomanage part of the generation
fleet for Cobb County, soAtlanta, Georgia and we had a
bunch of hydro as part ofSouthern portfolio. And hydro
functions very similar tobatteries, in that you charge
it, ie pump storage during lowprice hours, then discharge and
(43:30):
use that generation during highhours. So I've the battery
concept, to me is made sense myentire career, and both ERCOT,
California and other places aredoing a great job, because
batteries and storage in generalis unique, because it's supply
and demand. And so the storagelets us shift energy over time,
(43:52):
not just balance itinstantaneously. And so as I
mentioned earlier, oil, gas, youcan store for a season.
Electricity. That's not thecase, but batteries are slowly
but surely helping us get there.
I'm very intrigued to see whatsome of the companies like a
form energy trying to do, like100 plus hour battery are going
to do, but, but what we haveright now in Texas is one to two
(44:15):
hours and in California, fourhour batteries. From amperon
standpoint, we're going to tellyou what hours to use those
batteries. We have well over adozen clients who are battery
operators on our platform. Andso what they're looking for is
they're looking for a time whenso sun goes down at 7pm ish,
(44:36):
it's summer right now. We'llcall it right around there. And
if wind decides to go down withit, that's when things are going
to get dicey. Because it's 7pmfamilies are still home. Kids
are up. You're using that. Youroffice also might still be on.
Your workplace might still alsobe on. And so those are the
hours that used to not get asmuch love, but now we're. Delay
(44:58):
from that like 6pm to 9pm iswhere you're paying attention.
So you're going to look onamperage platform and say, Man,
wind goes away the same timesolar goes away and loads pretty
high. This is the hour. And sothat's what I mean. Most battery
upgraded optimizers in ERCOT areusing our platform for so it's a
it's a new arbitrageopportunity. And again, I love
(45:22):
when technology can stand on itsown legs. And this is where you
want to make batteries as muchas possible, because back to
what you're referencing, we needthe more the merrier. We need
batteries to be paired withrenewable generation to make the
grids go and for operators, thisgives them more flexibility in
(45:42):
balancing these renewables andalso just smoothing out peaks so
it but it makes Sam pron moreimportant as well. So don't hate
that, because timing andforecasting become everything.
Tim Montague (45:57):
Hey, guys, are you a residential solar installer
doing light commercial butwanting to scale into large C&I
solar. I'm Tim Montague. I'vedeveloped over 150 megawatts of
commercial solar, and I'vesolved the problem that you're
having you don't know what toolsand technologies you need in
order to successfully close 100KW to megawatt scale projects.
(46:22):
I've developed a commercialsolar accelerator to help
installers exactly like you.
Just go to cleanpowerhour.comclick on strategy and book a
call today. It's totally freewith no obligation. Thanks for
being a listener. I reallyappreciate you listening to the
pod, and I'm Tim Montague, let'sgrow solar and storage. Go to
clean power hour and clickstrategy today. Thanks so much.
(46:44):
Yeah, I think I wrote in 2024that the grid needs about six
terawatt hours of energystorage, and so we are, we're
only like 1/10 of the way tothat goal, so to speak, and that
earn, save, protect, tripartitethat I drill into my listeners
(47:09):
heads. I think everybody who's aregular listener knows now earn
from grid services, save byattacking demand and capacity
charges and protect from gridoutages. But so storage is super
powerful technology, and asprices have come down, it's
(47:30):
making it much more realistic toadopt the technology. You know,
there was just an article byJohn Weaver, one of my co hosts
here at the Clean Power Hour,saying that there's good
evidence that developers shouldbuild all of their projects now
battery ready, if not includingstorage today, make them battery
(47:54):
ready so that you can affordablyAnd quickly attach storage
because the price is comingdown, and the service, the value
of storage is so great, and ofcourse, we have an extended ITC
on batteries. Well, what else inour last minute together? Sean,
should our listeners know aboutamperon? I really enjoyed this
(48:14):
and appreciate you coming on theshow. I definitely am thinking a
little deeper about energytrading. And I still don't
understand what is machinelearning when it comes to energy
trading, but we'll have to savethat for another time, perhaps.
Sean Kelly (48:32):
Yeah, I mean, I have really appreciated the
conversation. I'm definitelygoing to use the the earn, save,
protect, that's that's great,and from, from a takeaway, from
my standpoint, I mean it, thisis no longer forecasting is no
longer a nice to have. It'sliterally the operating system
of the modern grid. Thetraditional models look
(48:54):
backward. Ai lets us lookforward and learn in real time.
And so these historical averagesjust they're not going to keep
the lights on, and we don'talways keep the lights on. But
as someone who struggled withthis over the last four years,
keeping the lights on at theirhome, this is something that we
really, really need to payattention to in this new world
(49:14):
of extreme weather anddistributed generation. So
appreciate you giving me thegiving me the microphone to
share this with your audienceand learn learn some from the
conversation as well, too.
Tim Montague (49:28):
And just to be clear, earnsay Protect was a
framework created by intelligentgeneration, a software as a
service. My listeners will havecaught a recent interview with
Jay marhoefer, the founder ofintelligent generation out of
Chicago. So I just have glommedon to that and really appreciate
that framework. I'll have tolisten to that one. Yeah, do
(49:51):
check it out. It's not out yet,but I will be by the time this
interview goes live. So allright, I want to thank Sean
Kelly, CEO and founder ofamperon, for. Coming on the
show. Check out all of ourcontent at cleanpowerhour.com
Tell a friend about the show.
That is probably the best thingyou can do is just tell a
friend. There are many, manypeople in the energy industry
who do not know about the CleanPower Hour yet. So tell a friend
(50:13):
and reach out to me on LinkedIn.
I love hearing from mylisteners. Check out all of the
content at Clean Power Hour andSean, how can our listeners find
you?
Sean Kelly (50:24):
They can. I'm very active on LinkedIn, both under
my personal and as well asamperon site. And we have a new,
updated website out amperon.co
Tim Montague (50:36):
thank you so much.
I'm Tim Montague, let's growsolar and storage.
Forecasting is no longer a nice to have. It's
(00:02):
literally the operating systemof the modern grid. The
traditional models lookbackward. Ai lets us look
forward and learn in real time,and so these historical averages
just they're not going to keepthe lights on, and we don't
always keep the lights on. Butas someone who's struggled with
this over the last four years,keeping the lights on at their
(00:22):
home. This is something that wereally, really need to pay
attention to in this new worldof extreme weather and
distributed generation.
intro (00:30):
So are you speeding the energy transition here at the
Clean Power Hour, our host TimMontague, bring you the best in
solar, batteries and cleantechnologies every week want to
go deeper into decarbonization.
We do too. We're here to helpyou understand and command the
commercial, residential andutility, solar, wind and storage
industries. So let's get to ittogether. We can speed the
(00:52):
energy transition
Tim Montague (00:57):
today on the Clean Power Hour AI forecasting for
power markets. My guest today isSean Kelly. He's the CEO of
amperon. Welcome to the show,Sean.
Sean Kelly (01:07):
Hey. Thanks so much for having me. Tim.
Tim Montague (01:09):
It's great to learn about this industry,
because I don't know very much,so I'm I'm really all ears on
this one. But why don't youstart with a little background
on yourself, and then we'll diveinto what amperon is up to in
the world.
Sean Kelly (01:24):
Yeah, absolutely. So I've been in power markets for
my entire career, since the twoweeks after I graduated from
Texas, A and M walked onto atrade floor at tenaska. At
tenaska started in ERCOT, theTexas market, happened to live,
live here most of my life, andI'm currently residing in
Houston, Texas. And went fromtenasco back to Houston, and was
(01:51):
at a great company called EagleEnergy Partners, which got
bought by Lehman Brothers. Ithink we all know how that one
worked out. And then EDFelectricity, France trading came
in and bought us. I guess whatmakes me uniquely qualified in
this is I've run more than threedozen power plants, everything
from wind and solar to fivenuclear assets, including
(02:13):
helping with the acquisitionintegration of two of them, Nine
Mile and g'day in New York. Alsomoved to Chicago in 2013 to help
set up eon, the big Germanutilities trade floor, as the
first trader in North America.
So yeah, been a really funcareer on the trading side,
which is really helps meunderstand our customer types.
(02:34):
And so I was living in New York,moved there in 2017 for a girl
who's now my wife, one of mybest trades, or my best trade
all time,
Tim Montague (02:48):
that's awesome.
Yeah.
Sean Kelly (02:50):
And in 2017 you could walk around New York and
almost trip over a datascientist. They were they were
everywhere, data engineers.
Everyone was a co founder anddisrupting and all of the
buzzwords, and that made merealize that all my friends in
Texas probably needed to be ontop of this latest and greatest
AI machine learning and and sothe joke between my co founder
(03:16):
and myself is he'll build it andI'll sell it. And so that's what
we've been doing here for almostthe last eight years at amperon.
Tim Montague (03:25):
My gut tells me there's a lot of parallels
between financial trading andenergy trading. Is that
accurate?
Sean Kelly (03:33):
There are definitely there. There are parallels, and
then there are also diversionsbetween the two. But I mean
trading is trading. It is alwaysbuy low, sell high. And a lot of
the financial trading andelectricity is very has a lot of
similarities to trading othercommodities. I mean a little bit
(03:53):
of equities, but you see thesame different options available
and things like that. So, yeah,it's a it's a really fun field.
I was the weird kid who wantedto be a trader from the time
they were about eight years old.
And here I am, guess that's justa few years later, still playing
with trader.
Tim Montague (04:11):
And what is the status of amperon? How big is
the company? And what marketsare you serving?
Sean Kelly (04:18):
Yeah, so ampron is the leading energy forecasting
company, operating at theintersection of AI and energy
data. We have raised 38 millionin venture capital. We have a
little less than 100 employees,and we are active in 23
countries with offices in bothHouston and London, but cover
(04:39):
the entire continental 48 partsof Canada, Australia and almost
20 European countries. So justactually signed our first deal
in the Middle East. And so we'vebuilt a truly global company
that can stand up forecastanywhere. Because, as you very
much know, wind and solar worksin any country. And so we our
forecast works in any. Countryas well.
Tim Montague (05:02):
All right, cool.
Well, what is something aboutthe power market that most
people completely misunderstand?
Sean Kelly (05:09):
Sean, yeah, when I started in my career, people
asked why I was working atnight. And electricity is a
physical commodity. Thesemegawatts move across wires in
real time. It can't it can kindof be stored in batteries, but
it's not like an oil or anatural gas where it can be
fully stored throughout anentire season. And so back to
(05:33):
what we talked about a minuteago. They people do think of it
like stocks or crypto. It'sfundamentally different. It's
actually more volatile. It's themost volatile commodity out
there, as we've seen in some ofthese crazy events, whether it
be polar vortex, winter storm,Yuri winter storm, Elliot and so
(05:54):
yeah, it's it's definitelymisunderstood, but it's also
it's not going anywhere. Everysingle industry is extremely
tied to electricity, as in,without it, no other industry
exists. So in my unbiasedopinion, it is the most
important thing that we shouldbe focusing on.
Tim Montague (06:12):
Yeah, it's so true, we take electricity for
granted, but literally, the goodlife completely depends on it,
and the good life goes awaypretty much instantaneously when
electricity goes away. It's adouble whammy, too, right,
right? Because then when there'sa big outage, like you're you
mentioned in Texas back in 20what? 2221 electricity prices
(06:36):
spike, because it's a very rarecommodity then, right? And so
it's also super painful forthose who have electricity. It's
not just those who don't haveit, but Well, what is, what is
the problem, I guess thatamperon was created to solve.
Sean Kelly (06:55):
Yeah, so in 2017 when I met my co founder and we
started the company in Januaryof 2018, we knew that there was
a lot of changes coming down thepipe. Smart Meters had started
getting adoption. You used tohave back in the back in the
glory days, the meter madebreaking in your backyard,
getting bit by the dog. Now youjust drive around with the truck
(07:18):
and you get on a 15 minute readfrom someone's home. We knew
that was happening. We knewrenewables were happening some
behind the meters, such assolar. You've got places like
Australia, which are all in onthat, and then Texas and
California and other westernstates that are also adopting
it. So knew that was coming.
EVs, I mean, the Tesla waspopular. Other companies were
(07:39):
talking about having EVs. Herewe are in 2025 and they're
absolutely everywhere thatchanges what your electricity
profile looks like. So we knewall this was coming. I also knew
that much of this was being donein Excel, and there were some of
the largest utilities out therethat were running these monster
models overnight and justpraying and hoping that they
(08:03):
would get an answer the nextmorning, but oftentimes the
macro would break becausethere's so many lines of just
lines on lines on lines in theseExcel spreadsheets, and that's
not going to work. And soamperon was built to deliver a
real time AI driven forecast sothat these traders and utilities
can make better decisions.
Tim Montague (08:25):
Now, back in 2018 the definition of AI was maybe
different than it is today. Imean, most knowledge workers and
prosumers are aware of llms, andmaybe, if you're knee deep in it
HRMS, which is a new model thatjust emerged. But these large
(08:46):
language models, do they haveanything to do with the
technology that you use atamperon?
Sean Kelly (08:51):
I mean, we look at them from a basically workforce
efficiency standpoint, and sothat is something we're using
just again to speed things up.
Got a very talented engineeringteam, data science team, they're
always looking at the latest andgreatest to continue to speed up
their efficiency, how many, howmuch they can crank out in the
in the hours they've committedto the team. And the way we've
(09:13):
looked at it, though, is is thereally the machine learning
standpoint? And so our firstmodel went live in November of
2018 and machine learning wasinvolved in that very first
model. And again, forecasting ishard. We're still always looking
at and tweaking the models, butour models have retrained every
(09:34):
hour since November of 2018 thatis a lot of compute cost, and so
we've continuously looked atthat. We also know that one
forecast is not always right,and so we're always running an
ensemble of four to sixdifferent forecasts. And so
looking at the simple regressionto gradient boosted trees to
(09:56):
whatever the latest and greatestis. And. That's why we're really
fortunate to have a team ofabout a dozen data scientists,
mostly PhDs, just always dealingwith this problem because it's
hard. I don't know what'shappening in your house. I need
to know when you bought an EV Ineed to know if you have solar
panels. I need to know what abattery is doing. This problem
(10:16):
is getting so much morecomplicated than back in the
day, when everyone went to workat eight o'clock, they came home
at five o'clock, every homeprofile looked very similar.
Now, home profiles are all overthe place, and so we we've got
to stay on top of that for our150 plus clients.
Tim Montague (10:36):
So let's talk about this from the perspective
of, say, an IPP, a company thatowns fleets of wind, solar and
battery projects, and let's talkabout Texas, since you you know
that market certainly very well.
ERCOT is a unique market. It isnow on a windy, sunny day, 50%
wind, solar and battery powered.
(10:58):
Who knew? Right? Ginormous. RedState is going green. Texas
eclipsed California on an on anannual basis. Now, in terms of
the install solar, right? We're,I think Texas did 11 gigawatts
last year, something crazy likethat. But if you're an IPP,
(11:18):
okay, you're in the business ofselling electricity to various
and sundry off takers. Walk usthrough that business and what
are some of the challenges andopportunities that they're faced
with on a daily basis.
Sean Kelly (11:33):
Yeah, Texas has done a great job of just really being
a pro business state and gettingan entire full text or full
generation stack. And so, Imean, we have 40 gigs of wind
and a ton of solar. And so forthese IPPs that are our
customers, independent powerproducers, for those of you who
aren't all in the acronym weeds,what they do with our forecast
(11:57):
is they give us said asset.
We'll pick a solar asset. And sowe'll give them a 15 day, 360
hour forecast for their specificsolar installation, and we'll
call it 100 megawatts. And weactually give them five minute
granularity, because, as we allknow, Sun kind of does what it
wants to. It's more reliablethan when from knowing when it's
(12:20):
going to be sunny versus not,but we've got to stay on top of
this. So a five minute forecastis the bare minimum that you
need to actually understand whatyou're doing with this, and it's
what you can actually go andschedule. And so we tell you on
this five minute granularitywhat it looks like, so that you
can tell again, what was we'repicking on, ERCOT here. ERCOT,
what you're going to be showingup within a day ahead. And so
(12:42):
every day, by about 10am youneed to tell ERCOT, hey,
tomorrow I will have this muchsolar. And so that's what
amperage forecast provides toyou. And then as we continue,
the models rerun every singlehour, and you're getting an
update there. So very similarwith wind we're telling you,
we're updating like multipletimes per hour, on five minute
(13:04):
cadence, what your wind farm isdoing so that you can best
monetize it and run itaccordingly.
Tim Montague (13:12):
And does the IPP know, though, from that model,
what kind of revenue they'regoing to be generating?
Sean Kelly (13:19):
Yeah, they can take, they can take our output that
we're telling them, and thenbasically put it with the the
clearing price, whether it's theday ahead price, if they bid it
into the Ford market, or thereal time price, if they if they
took it, if they floated it tothe intraday, and then they can
figure out what their revenuegeneration is going to be in
(13:41):
figuring out how much money ampruns helping increase on a
return on investment.
Tim Montague (13:47):
So what is let's just step back a little bit,
because I think obviously somefraction of my listeners will be
familiar with this energytrading business, but only some
fraction so in the greaterscheme of things, IPPs are
signing what kinds of contractsto they need off takers, right,
(14:11):
to develop the project out ofthe gate, frankly, right? And,
and who is, who is a goodexample of a buyer? And then
what are the what is thevariability in that contract?
Sean Kelly (14:28):
So what they're going to do is, you're
absolutely right. If I want tobuild 100 megawatt wind farm to
keep nice round numbers, I wantto finance it and not have to
foot the entire bill up front.
And so the best way to do thisis sell a certain percentage of
my of my wind going forward. Andso they'll do a PPA power
purchase agreement. And we'llsay a 10 year PPA is what
(14:52):
they're looking at. The mostnatural off taker is
municipalities. Cooperatives, sotowns. So they'll go in and say,
hey, my town would love to buythe 100 megawatt wind farm. 100
megawatt wind farm does notproduce 100 hours or 100
megawatts every hour, right?
(15:14):
It's not a base load. It'sprobably never going to produce
100 megawatts. That's just thethe nameplate capacity of it. So
I'll go in and say to Tim city,I'm going to give you 30
megawatts, like around the clockgoing 10 years. And so I would
sell you that, and you wouldsay, Great, I have 30 megawatts
coming. Well, you then get themoney from that, and you're able
(15:39):
to go build the wind farmagreement reached. Well, then
you've got to deal with theimbalance. And so if only 12
megawatts are going to show up,I own Tim City, 18 more
megawatts. And so that's where Ineed to know, with amperons
forecast, that I've only got 12megawatts tomorrow. It's just
not a windy day. I need to gobuy that like that interval of
(16:00):
18 megawatts from the market. Sothat's how, that's how these are
being built, and these are howthese are being financed. And
again, the the PPA market, it'snot, it's not an exact science.
It takes away a lot of thecommercial upside. However,
unless you have just the abilityto fork over the full amount
(16:22):
whenever you build it. Then,yeah, it makes a lot more sense
to go finance that PPA. And
Tim Montague (16:28):
then, as the asset owner, I'm getting a guaranteed
price on the wind power, but I'mbuying grid power at some
variable price, and that's alsowhere this kind of a platform
comes in handy.
Sean Kelly (16:42):
Yeah, you agree to said price, and that's where you
look at what the Ford curve istrading. So you'd say, hey,
2026, through 2025, is at thisnumber, and said like
municipality or Co Op will gobuy it. You. I guess examples of
other people buying this, orobviously, utilities will go and
buy this. Oftentimes they gobuild their own fleets, just
(17:05):
because they're normally quitewell capitalized. And so they'll
say, hey, I need to have thisfor my rate payers. And then
you'll also have now, what'sreally exciting is you've got
data centers and hyper scalerswho are coming in because
they're normally in very goodfinancial shape, and so they're
able to come in and reallyincentivize this. And many of
(17:27):
them, as we know, have madegreen commitments, and so
they're more they're more likelyto be excited about the wind and
solar if they have a greencommitment. But they're also
trying to figure out just, youknow how to, how to make sure
that said data center can canrun as needed.
Tim Montague (17:45):
And how risky is this business in the greater
scheme of things, you you knowyou mentioned, okay, if I have
100 megawatts of a wind farm, Imight be able to sell a contract
for 30 megawatts, but the windgoes up and down. Some days are
it's blowing, and some days it'sstill for a week, it could be
(18:06):
multiple weeks where there'svery little wind, and then I'm
buying power at some notnecessarily known rate. So yeah,
just how risky is this businessin general?
Sean Kelly (18:19):
I mean, power is a risky business. I mean, again,
I've been, I've been goingthrough this for 20 years. I
mean, you looked at summer 2011in ERCOT, there was not a lot of
wind. It was extremely hot.
Power prices were very high.
2014 you had the polar vortex. Ilived in Chicago then, and that
was a rough one. Being a Texanin Chicago, dealing with like
(18:41):
minus 40 feels like inFahrenheit. And so again, not
the renewable generation didn'tshow up as much then. And so
this is why having a 15 weekforecast is so important. We're
able to go out and for instance,amperon, you mentioned winter
storm area earlier, so that thatwas a little too near and dear
to me, from the standpoint of mywife, was six months pregnant at
(19:05):
the time, and the power went outat 2am on February 15 of 2021 we
had no marketing department atthe time, so I wrote a really
fun blog post on Valentine'sDay, ideas to do in the dark,
because February 14 is what dayI thought the power was Going to
go out. Went out at 2am on the15th. And so we were able to
tell our clients on February 3,our meteorologist nailed it and
(19:30):
say it's going to get reallycold. So multiple IPPs went and
bought power at $60 permegawatt. It wound up clearing
$9,000 per megawatt those days.
So because these different windfarms went and bought power,
because amperon, they did theirown research too. We don't give
full credit here, sure, butthey, but they went and bought
(19:52):
$60 power, and then it tradedall the way up to 9000 and so
that's the way that amperonReally. Really helps them
understand that, because theyknew they weren't going to be
able to supply all the PPAs thatthey had previously agreed to
that week of February 15, and loand behold, yeah,
Tim Montague (20:12):
but you literally were able to predict that the
grid was going to fail.
Sean Kelly (20:17):
That's not exactly what we do. I knew from my
experience that the grid wasgoing to fail, and literally put
out a blog post about it, or aLinkedIn post about it, that we
were going to lose power, and welost power, and then on. But we
we put that it was going to be avery, very like epically cold
event, and we started sayingthat at the very beginning of
(20:40):
February. So we gave ourclients, I mean, 12 days notice.
We'll say 10 days notice,because it really got dicey on
about the 13th on that Saturdayis when prices went well into
the 1000s, and then it justcontinued on. It was a one heck
of a Valentine's Day weekend.
Tim Montague (20:58):
And so who are the winners and who are the losers
in a stormy area scenario?
Obviously consumers are, are thebig losers, consumers and
business owners like not havingpower is painful and
uncomfortable, but yeah, justwalk us through that.
Sean Kelly (21:16):
Yeah, there were so many losers in that. And I would
say the the biggest losers, younailed it, definitely consumers,
those rate payers at the end,there are people who are on
variable contracts at their homethat were literally getting
their bank account hit, like,every like, every 15 minutes
that were on auto pay from acompany that's no longer around
(21:39):
called gritty and so, like, youliterally, like those were the
biggest losers. And the factwas, the consumers, the business
owners, obviously you lose thosedays of doing business. And then
I would also say that anyone,anyone, I mean, natural gas,
didn't have natural gas plants,also didn't all show up. So they
(22:00):
had contracts to sell that theywere they had to go buy those
back. And that wasn't fun.
Getting gas wasn't fun. Thereweren't a whole lot of winners
out of that. I mean, itdefinitely put the state in a
pretty bad situation. I do thinkit made people more resilient,
because you look in thetemperatures we had, I mean, as
you're in Illinois, they'relike, teens. Yeah, no big deal.
(22:20):
Nothing to see here. I mean,when I lived in Chicago, we had
teens for like months at a time,that 2014 and that 2015 but in
Texas, our everything wasn'twinterized properly, because we
never seen teens before. And sonow, I mean, I think ERCOT and
the PUC and the rest have done areally good job of pushing for
(22:42):
that winterization is nowwhenever we have a slight scare,
it winds up being or so far,like knock on wood, has been
pretty much a non event.
Tim Montague (22:54):
The Clean Power Hour is brought to you by CPS
America, maker of NorthAmerica's number one three phase
string inverter with over 10gigawatts shipped in the US. The
CPS product lineup includesstring inverters ranging from 25
kW to 350 kW, their flagshipinverter, the CPS 350 KW is
(23:15):
designed to work with solarplants ranging from two
megawatts to two gigawatts. CPSis the world's most bankable
inverter brand, and is America'snumber one choice for solar
plants, now offering solutionsfor commercial utility ESS and
balance of system requirementsgo to Chint power systems, com
(23:37):
or call 855-584-7168, to findout more. Yeah. All right, I
could talk about just that oneevent, probably all day. But
let's move on. So what sets whatsets April and apart, though
from other forecastingcompanies,
Sean Kelly (23:57):
yeah. I mean, what sets us apart is accuracy. I
mean, we're two to three timesbetter than grid operators.
ISOs. TSOs is what they callthem in Europe. We also, we
cover the whole value chain. Weforecast demand. But demand is
not the whole story. It's netdemand, as you and your
(24:18):
listeners know, which net demandis pulling out renewables. So
it's demand minus wind minussolar, because those show up and
and make it much easier on thegrid. And we also forecast
prices, and so you're able tosee this across the value chain.
And then this is also built bytraders and data scientists. Our
first hire was a PhD datascientist who's still hanging
(24:41):
out with the company over sevenyears later, and then my second
hire, which the investorsthought was a little weird, was
a meteorologist. And why ameteorologist? Because
forecasting, weather is so, soimportant to this. And so those
are some of the things thatsetting up. Run apart, but also
(25:01):
we get a little bit of a latemover advantage. Like I said, we
started the company in Januaryof 2018, so many of our
competitors are 20 or even 30years old. We've been cloud
native from day one. We've beenusing AI since day one. This
wasn't, Oh, we've got a goodcompany. We've got clients.
We've got to move out of, like,we've got to move out of just
(25:22):
running this off our ownservers. Oh, we've got to move
this out of Excel. Oh, we've gotto move this out of we started
with a modern tech stack fromday one, and so that's given us
a huge advantage and anadvantage that we don't take for
don't take for granted, and wakeup every single day trying to
figure out how we can improveour models and pay attention in
(25:43):
the next, latest
Tim Montague (25:45):
and greatest. And are companies building their own
systems for this? Or is it allthird party platforms?
Sean Kelly (25:54):
It's definitely both. I mean, people will look
at it as a core competency,especially utilities, but
normally they also want to havea third party come look at us.
Learn look at it. Theirforecast. The reason why is, as
I mentioned, we have 150 clientsand over 40 million meters in
(26:14):
the continental United Statesthat are in some way on our
platform, either as individualmeters or in like aggregations.
And so there's no utility thatit has 40 million meters. And so
we are able to have already seensome things. So when we walk
into utility number four, we'vealready dealt with utility one,
two and three, and so we alreadyknow what their problems are and
(26:37):
what they're seeing. And so itreally helps that kind of been
there, done that before, asopposed to having to start from
scratch. So a lot of people willstill run their own internal
models. We're going up against alot of 20 year old models that
they just don't trust because ofhow times have changed, but
they'll run those with ours. Andas I mentioned, we're a big, big
(26:58):
fan of ensembles, and soensembling is basically just
waiting different forecaststogether to make one super
forecast. In layman's terms,
Tim Montague (27:09):
I'm curious if you know the weather's changing, and
you can no longer say, well,let's just look back at the last
10 or 20 years to predict howthe weather is going to be in
February of 2026 is that afactor?
Sean Kelly (27:26):
Now, absolutely. I mean, it is. Weather's gotten
crazy. I mean, we're sittinghere talking about a one and 100
year event. I mean, I rememberearly in my career, all we were
worried about was hurricanes.
And the reason we were worriedabout hurricanes is most natural
gas came from the Gulf, and whena hurricane came in, it took out
(27:47):
gas production, and natural gaswent through the roof. That
happened in 2005 right when Istarted. And now, when a
hurricane comes, natural gasactually dips, because it takes
out load, and we don't reallydrill offshore that much, and so
that was really kind of the bigchange. Now we're sitting here,
and I can rattle off. I mean,all of the events starting, I
(28:10):
mean, ERCOT, 2011 was 62 days ofover 100 degrees in Dallas. I
mean, we've talked about 2014with the polar vortex. We've
talked about 2021, with winterstorm Yuri. I mean, blacked out
a whole bunch of people onChristmas Eve across the east
coast with winter storm Elliot.
(28:31):
I didn't know what a derecho wasuntil last May, when it came and
took out a bunch of power linesin my neighborhood and I didn't
have power for seven days. Imean, you just continue to see
these hurricanes are gettingbigger every year. It feels like
we're saying it's going to be awilder hurricane season.
Wildfires are more prevalentthan I can ever remember, just
(28:53):
in the last handful of years. SoI mean, things are getting wild.
Extreme events are here. And Imean, you just off the cuff,
reference a one in 100 yearevent. But is it really a one in
100 year event? If it happensevery two months?
Tim Montague (29:10):
Yeah, it is crazy.
So talk to us about forecasting.
How is that the role offorecasting evolving as the grid
gets smarter and more dynamic.
You know, you mentionedelectrification of
transportation, data centerscoming online, and, you know,
the economy is growing, and it'selectrification of everything.
(29:34):
It's not just EVs, it's heatpumps, HVAC, industrial
processes. So how is your How isyour business evolving?
Sean Kelly (29:47):
Yeah, forecasting started as a nice to have and
that, and now we're sittinghere, and it's literally the
foundation of daily operations.
The reason it was a nice to haveis back to what I referenced
earlier. Smart meters is youjust got a data point every
3060, 90 days, whenever theMeter Reader came around, so you
didn't really know howimbalanced you were. Now we're
(30:08):
sitting here and we're getting,I mean, some places are getting
millisecond reads, a nestthermostat or something. Of that
is picking up data all the time.
So the amount of data that wehave to take in has gone just
absolutely through the roof. Andso every single trading
strategy, dispatch decision andeven long term defense
(30:32):
investment depends on it. And soif you think of forecasting,
it's really the operating systemof the modern grid. And so what
we're doing, we knew it wasgoing to be important. But you
also, I mean, you referenced notbeing able to have look at the
last 10 years of weatherforecast. We also can't look at
the last 10 years of load data,because I didn't know when I
(30:55):
started this that covid wasgoing to be a thing. It changed
what a landscape of a citylooked like for as little as six
months to as much as threeyears, just depending on what
the like, what the lockdown,what rules were when people went
back to work, and now we'resitting here in a in a remote
economy, and we have people in,I think, 17 different countries
(31:19):
who work at amperon because Weget to hire best in class, but
they work at their house, sotheir house doesn't look like it
would have in the 1980s and sothese are all things that I
mean, all things that arechanged that when we started the
company, we couldn't haverealized we're going to hit but
continue to evolve and makethings more difficult, Which
means we have to continue to getsmarter.
Tim Montague (31:43):
And what are the what are the limitations that
you encounter with traditionalforecasting methods, and
especially with regard to AI?
How have you leveraged AI totranscend that legacy
technology?
Sean Kelly (32:00):
Yeah, and, and, and, just to be clear, we still use
the legacy, I mean, linearregression models have been
around for quite a while, and westill have that as a very core
input into our models. Butagain, being able to
automatically weight each modeland having it retrain every
single hour is not somethingthat you can go do in Excel for
(32:20):
40 million meters. It's just notcomputationally possible. And so
that's where the the AI in MLmachine learning come into play.
And so we're we're looking atthat. We're also looking at
weather. Weather has made hugestrides here, just in the last
two years. Whether it be fromlike the models that NOAA puts
(32:41):
out to you've got the like,European AI models with much
better granularity. So weatheris something we look at. We
currently use for weathervendors, and we're always
looking for like, the next,latest and greatest. So I mean,
weather is something that I'mvery, very I mean, it gets more
complicated and continues to getmore complicated, so you need
(33:02):
better models just to stay atwhere you were if things
continue to progress like this.
So that's where the AI reallydrives in. So weather has been a
huge help, and that's why weliterally, we've got a team that
goes in and tests out all thelatest and greatest AI models to
see which one's going to hit is,I mean, the the people with the
most money in the world, the theNVIDIA is, the Googles, etc, the
(33:24):
world are putting out their ownmodels, which is definitely, uh,
definitely worth diving into.
Tim Montague (33:31):
So tell us a little more about the Machine
Learning, though, to the extentyou can that you're leveraging
it's, it's built on top of someother technologies, like open AI
or Claude or Google. Yeah, I'mjust curious how that, how that
(33:51):
works?
Sean Kelly (33:52):
Yeah, so it's, it's not built on top of LLM
technology, like some of theones you mentioned, what it's
really doing is going in andwaiting our forecast and running
different forecasts and sayingthat, hey, this like grading
boosted tree needs to be withthis deep learning model. We'll
put in 15% of this one and 17%of this one. And then also
(34:14):
different clients are different.
Some, especially in the retailenergy provider world, will be
all residential. Their booklooks very different than
another client who's allcommercial and industrial. So we
again, don't have someonesitting there saying, oh, make
sure to use the commercialmodel. It just automatically
goes in and uses the commercialor the industrial model or the
(34:34):
residential model. So again, youdon't want someone sitting there
across 150 customers trying tofigure out which model to use at
which time. And so that's wherethe machine learning really
kicks in. And again, the datascience team has done a
phenomenal job of doing thissince inception. So I'm glad
this isn't something that wekind of freaked out about a year
(34:55):
ago and said, Oh my gosh, wehave to use AI in some. Thing.
One of the coolest, I guess,accolades that we've received is
Andreessen Horowitz, who we haveno money from, put out their top
50 AI American dynamismcompanies. And right when AI
became a thing about two yearsago, and there was one company
(35:16):
from Texas, and I believe therewas about, there's like, eight
to 12 energy companies, and wewere, we were on that list. And
so again, we've been using thissince very inception, and the
machine learning makes all thedifference. That's where you get
to go against a competitorrunning in Microsoft Excel. And
that's also something thatputting on an updated model, as
(35:39):
quick as weather changes, and asquick as sun and wind and all of
that changes, as opposed to somany of our competitors, we're
literally running their modelonce a day, which just does cut
it once or twice a day. And thatthat's that's not going to work
with how fast weather ischanging these days.
Tim Montague (35:59):
So when, when the next storm Yuri comes? There's,
I guess there's a couple ofthings on my mind about this.
The grid is changing. We arehardening the grid now in Texas,
we're also increasing theattachment rate of batteries,
and the percentage of wind andsolar on the grid is going up,
(36:23):
but the attachment rate ofbatteries is going way up. So
there's that. So the grid ismore resilient. And okay, maybe
our maybe our forecasting ofweather patterns is getting
better. So walk us through ascenario for another Yuri style
event. How much warning does thegrid operator have, and how
(36:47):
much, you know, flexibility dothey have? Like, one of the
things that's noteworthy aboutERCOT is it's it's famous for
being isolated from other ISOs,and so they can't necessarily
count on getting a bunch ofpower from some other pool when
the shit hits the fan.
Sean Kelly (37:07):
Yeah, I mean, and that's where, I mean, that's
where having reserves reallykicks in, and incentivizing the
reserves to act at the correcttimes. And then that's also
where demand response reallykicks in. And so a lot of the
load that we're talking about,right? You can't, you can't get
through a conference. Can't getbarely get through a podcast
without talking to demandgrowth. So might as well just go
(37:28):
for it. But a lot of the demandgrowth that we're looking at, I
mean, yes, we're seeing heatpumps, we're seeing EVs, we're
seeing demand growth at your ownhome, but the real load growth
that we're seeing is datacenters and large industrial I
mean, again, more more things onshore and just that stuff's
(37:49):
pretty flexible at the end ofthe day. And so that's something
that I think that the gridoperator is doing a good job of
paying attention to how much thegrid operator gets involved.
There's something that justpassed in Texas called SB six,
and that is basically reallypushing on data centers to have
to behave whenever times gettight. And so I think that's
(38:13):
really the next step. So ERCOT,you're absolutely right. Is more
of an island than most places,but I think they have taken the
proper responses from a demandresponse, and incentivizing
those large loads to be on goodbehavior and be good stewards of
the grid. So that helps. Andalso, if it's during the summer,
it's hot out, and that's whenthe solar is going wild, so
(38:33):
you've got a few hours at theend of the day, kind of hours in
the 1921 which is 7pm to 9pm athome, and those hours, that's
when the batteries come in. Sohaving those, although it'd be
only one to two hour batteries,ERCOT spreading those out and
kind of saying, These are thehours I need you that's really
critical to keep the gridrunning. But ERCOT had a smooth
(38:56):
summer PJM less so I think theyhad a dozen EA one alerts, and
they also had something that inJune, they blew through what
they said was their peak loadfor the entire summer, and
that's just because they havesuch a wide footprint that they
did not expect that it would gethot from Chicago all the way to
(39:18):
Virginia, and it did. So it hada heat wave that just kind of
sat over that whole Mid Atlanticregion. And so PJM saw that. But
for us, that made people realizethat you can't sleep on PJM
forecasting, either or New Yorkforecasting, because it was a
hot, hot summer. And those,they've had some pretty mild
summers for the last handful
Tim Montague (39:40):
so the grid operator sees these events
coming, right? They say, Okay,it's going to be super cold.
People are going to be runningtheir electric heaters, their
gas furnaces, whatever, right?
They have some ability topredict what the load is going
to. Be over time. And then ifthey notice that the load is
(40:01):
going to be greater than whatthey capacity of the grid is,
they will send a signal to largeoff takers and say, hey, at, you
know, four o'clock tomorrow, wewant you to throttle your data
center. Is that kind of how thatworks?
Sean Kelly (40:19):
Yeah, you, I mean, you pretty much got it there.
They go through and they say,All right, the supply stack is
this, and we think demand isgoing to be this. And this is
why having a good wind and solarforecast is so important as you
can, for the most part, tell acombined cycle when to show up.
I mean, not a lot of coal left,and not a ton of nuclear
(40:40):
unfortunately, but those arepretty reliable base loads, and
so you've got to understand whatthe difference is. And so yeah,
you tell people who are enrolledin these different demand
response programs to, hey, 4pmthat's your time. Like, go ahead
and plan ahead. Go ahead andsend your workers home. Like, I
need you to do what you can do.
So yeah, and
Tim Montague (41:03):
then if they don't do what they can do, like if
that demand response, for somereason, fails, does the grid? Is
the grid smart enough or savvyenough that the grid operator
can go, Okay, I'm going to shutdown this, this circuit to this
major off taker so that I don'thave an uncontrolled outage.
Sean Kelly (41:25):
So the way demand response works is it's like
insurance. And so I'm going topay you $10,000 to be on good
behavior when I tell you to beon good behavior. And so the
penalties are nasty if you arenot on good behavior. So if
you're sitting there and you'rerunning a data center, and you
promised that you'd go from 20megawatts an hour to 10
(41:48):
megawatts an hour, and I alreadypaid you your $10,000 it's way
more than that, but I alreadypaid you your money to back
down. Then you will back down.
And so that's where. That'swhere. Again, there's not
really, I see behaving on thegrid and very similar in, like,
especially in the East, whereyou have capacity auctions.
That's, I mean, the capacityauction was obviously a record
(42:10):
in PJM, and so they are saying,like, you are promising that you
will back down then. So, yeah,the you don't have to worry
about the good behavior. Thegood behavior has already been
paid for.
Tim Montague (42:22):
All right, so let's talk a little more about
storage. You know, storage iskind of the new kid on the
block, but there's lots of it,you know, and lots more coming.
I think I saw that there's 40%more storage coming in the next
five years in ERCOT. And ofcourse, ERCOT is just one
(42:44):
market. There's incentives inIllinois, New York,
Massachusetts, New Jersey,Connecticut. I mean, many states
now California have programsincentivizing the installation
of batteries. What should ourlisteners know about batteries
as this relates to what amperondoes.
Sean Kelly (43:05):
Yeah, I love batteries, and the reason why is
early in my career, I got tomanage part of the generation
fleet for Cobb County, soAtlanta, Georgia and we had a
bunch of hydro as part ofSouthern portfolio. And hydro
functions very similar tobatteries, in that you charge
it, ie pump storage during lowprice hours, then discharge and
(43:30):
use that generation during highhours. So I've the battery
concept, to me is made sense myentire career, and both ERCOT,
California and other places aredoing a great job, because
batteries and storage in generalis unique, because it's supply
and demand. And so the storagelets us shift energy over time,
(43:52):
not just balance itinstantaneously. And so as I
mentioned earlier, oil, gas, youcan store for a season.
Electricity. That's not thecase, but batteries are slowly
but surely helping us get there.
I'm very intrigued to see whatsome of the companies like a
form energy trying to do, like100 plus hour battery are going
to do, but, but what we haveright now in Texas is one to two
(44:15):
hours and in California, fourhour batteries. From amperon
standpoint, we're going to tellyou what hours to use those
batteries. We have well over adozen clients who are battery
operators on our platform. Andso what they're looking for is
they're looking for a time whenso sun goes down at 7pm ish,
(44:36):
it's summer right now. We'llcall it right around there. And
if wind decides to go down withit, that's when things are going
to get dicey. Because it's 7pmfamilies are still home. Kids
are up. You're using that. Youroffice also might still be on.
Your workplace might still alsobe on. And so those are the
hours that used to not get asmuch love, but now we're. Delay
(44:58):
from that like 6pm to 9pm iswhere you're paying attention.
So you're going to look onamperage platform and say, Man,
wind goes away the same timesolar goes away and loads pretty
high. This is the hour. And sothat's what I mean. Most battery
upgraded optimizers in ERCOT areusing our platform for so it's a
it's a new arbitrageopportunity. And again, I love
(45:22):
when technology can stand on itsown legs. And this is where you
want to make batteries as muchas possible, because back to
what you're referencing, we needthe more the merrier. We need
batteries to be paired withrenewable generation to make the
grids go and for operators, thisgives them more flexibility in
(45:42):
balancing these renewables andalso just smoothing out peaks so
it but it makes Sam pron moreimportant as well. So don't hate
that, because timing andforecasting become everything.
Tim Montague (45:57):
Hey, guys, are you a residential solar installer
doing light commercial butwanting to scale into large C&I
solar. I'm Tim Montague. I'vedeveloped over 150 megawatts of
commercial solar, and I'vesolved the problem that you're
having you don't know what toolsand technologies you need in
order to successfully close 100KW to megawatt scale projects.
(46:22):
I've developed a commercialsolar accelerator to help
installers exactly like you.
Just go to cleanpowerhour.comclick on strategy and book a
call today. It's totally freewith no obligation. Thanks for
being a listener. I reallyappreciate you listening to the
pod, and I'm Tim Montague, let'sgrow solar and storage. Go to
clean power hour and clickstrategy today. Thanks so much.
(46:44):
Yeah, I think I wrote in 2024that the grid needs about six
terawatt hours of energystorage, and so we are, we're
only like 1/10 of the way tothat goal, so to speak, and that
earn, save, protect, tripartitethat I drill into my listeners
(47:09):
heads. I think everybody who's aregular listener knows now earn
from grid services, save byattacking demand and capacity
charges and protect from gridoutages. But so storage is super
powerful technology, and asprices have come down, it's
(47:30):
making it much more realistic toadopt the technology. You know,
there was just an article byJohn Weaver, one of my co hosts
here at the Clean Power Hour,saying that there's good
evidence that developers shouldbuild all of their projects now
battery ready, if not includingstorage today, make them battery
(47:54):
ready so that you can affordablyAnd quickly attach storage
because the price is comingdown, and the service, the value
of storage is so great, and ofcourse, we have an extended ITC
on batteries. Well, what else inour last minute together? Sean,
should our listeners know aboutamperon? I really enjoyed this
(48:14):
and appreciate you coming on theshow. I definitely am thinking a
little deeper about energytrading. And I still don't
understand what is machinelearning when it comes to energy
trading, but we'll have to savethat for another time, perhaps.
Sean Kelly (48:32):
Yeah, I mean, I have really appreciated the
conversation. I'm definitelygoing to use the the earn, save,
protect, that's that's great,and from, from a takeaway, from
my standpoint, I mean it, thisis no longer forecasting is no
longer a nice to have. It'sliterally the operating system
of the modern grid. Thetraditional models look
(48:54):
backward. Ai lets us lookforward and learn in real time.
And so these historical averagesjust they're not going to keep
the lights on, and we don'talways keep the lights on. But
as someone who struggled withthis over the last four years,
keeping the lights on at theirhome, this is something that we
really, really need to payattention to in this new world
(49:14):
of extreme weather anddistributed generation. So
appreciate you giving me thegiving me the microphone to
share this with your audienceand learn learn some from the
conversation as well, too.
Tim Montague (49:28):
And just to be clear, earnsay Protect was a
framework created by intelligentgeneration, a software as a
service. My listeners will havecaught a recent interview with
Jay marhoefer, the founder ofintelligent generation out of
Chicago. So I just have glommedon to that and really appreciate
that framework. I'll have tolisten to that one. Yeah, do
(49:51):
check it out. It's not out yet,but I will be by the time this
interview goes live. So allright, I want to thank Sean
Kelly, CEO and founder ofamperon, for. Coming on the
show. Check out all of ourcontent at cleanpowerhour.com
Tell a friend about the show.
That is probably the best thingyou can do is just tell a
friend. There are many, manypeople in the energy industry
who do not know about the CleanPower Hour yet. So tell a friend
(50:13):
and reach out to me on LinkedIn.
I love hearing from mylisteners. Check out all of the
content at Clean Power Hour andSean, how can our listeners find
you?
Sean Kelly (50:24):
They can. I'm very active on LinkedIn, both under
my personal and as well asamperon site. And we have a new,
updated website out amperon.co
Tim Montague (50:36):
thank you so much.
I'm Tim Montague, let's growsolar and storage.
Popular Podcasts
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
Ruthie's Table 4
For more than 30 years The River Cafe in London, has been the home-from-home of artists, architects, designers, actors, collectors, writers, activists, and politicians. Michael Caine, Glenn Close, JJ Abrams, Steve McQueen, Victoria and David Beckham, and Lily Allen, are just some of the people who love to call The River Cafe home. On River Cafe Table 4, Rogers sits down with her customers—who have become friends—to talk about food memories. Table 4 explores how food impacts every aspect of our lives. “Foods is politics, food is cultural, food is how you express love, food is about your heritage, it defines who you and who you want to be,” says Rogers. Each week, Rogers invites her guest to reminisce about family suppers and first dates, what they cook, how they eat when performing, the restaurants they choose, and what food they seek when they need comfort. And to punctuate each episode of Table 4, guests such as Ralph Fiennes, Emily Blunt, and Alfonso Cuarón, read their favourite recipe from one of the best-selling River Cafe cookbooks. Table 4 itself, is situated near The River Cafe’s open kitchen, close to the bright pink wood-fired oven and next to the glossy yellow pass, where Ruthie oversees the restaurant. You are invited to take a seat at this intimate table and join the conversation. For more information, recipes, and ingredients, go to https://shoptherivercafe.co.uk/ Web: https://rivercafe.co.uk/ Instagram: www.instagram.com/therivercafelondon/ Facebook: https://en-gb.facebook.com/therivercafelondon/ For more podcasts from iHeartRadio, visit the iheartradio app, apple podcasts, or wherever you listen to your favorite shows. Learn more about your ad-choices at https://www.iheartpodcastnetwork.com
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.