S2E15

Dynamic Grids for a Dynamic Load: E-Mobility, DOEs & AI | B. Herndler & B.V. Rao, AIT | S2E15

with Barbara Herndler, Bharath Varsh Rao· AIT Austrian Institute of Technology· 65m

TL;DR

Up to 70% of renewable connection requests could go unconnected if European grids are not reinforced over the next one to two decades — and the planning methods still in use make that worse by sizing for a worst-case day. Rao and Herndler make the case for automated planning that treats batteries and flexibility as alternatives to copper, claiming up to 30% off total infrastructure cost. On AI they are precise about the boundary: agents orchestrate the workflow, physics simulators do every calculation, and no European DSO will let a language model near a SCADA system.

Synopsis

Barbara Herndler and Bharath Varsh Rao of AIT join The Energy Bridge to argue that the energy transition is won or lost in the distribution grid. They lay out the connection-queue crisis — up to 70% of renewable requests unconnectable without reinforcement — and why traditional worst-case planning, built on a flat load-growth assumption, leaves efficiency on the table. Rao describes the AIT spin-off he is founding, Rapson, which automates network development planning and weighs non-wired alternatives such as battery storage against digging up a street in Vienna. Both are careful about AI's limits: language models orchestrate simulations but never compute numbers, and DSOs will accept an advisory layer beside the control room but nothing inside it. The conversation also covers vehicle-to-grid versus vehicle-to-home, optimal sensor placement, climate-driven outage risk in Turkey, and the tacit knowledge retiring out of grid operators.

Key metrics

by the numbers · 6

70%
share of renewable generation connection requests that could remain unconnected if grids are not reinforced
3%
flat annual load growth traditionally assumed in worst-case grid planning
30%
total infrastructure cost Rao expects automated planning with non-wired alternatives to save
10%
rule-of-thumb share of grid nodes needing sensors to achieve full observability
99.9998%
grid reliability in Austria and Europe, which Rao credits for DSO conservatism
400 billion
euros Europe is already investing over ten years in electrification, grid expansion and retrofit

Topics

8 tags

grid planningdistribution gridse-mobilityartificial intelligenceflexibilityvehicle-to-griddigital twinsgrid connection queues

Stats

Duration: 1h 05m
Words: 12.1k
Questions: 41

Timeline

12 chapters

The transition happens underneath Vienna
Rao frames the episode: Europe has bet its competitiveness and sovereignty on the energy transition, and the power grid is the bottleneck.
Rapson, the spin-off
Rao is leaving AIT to commercialise eight years of research as an automated grid-planning company, named after the mathematician behind the Newton-Raphson method.
The connection queue
Up to 70% of requests to connect wind and solar could stay unconnected without reinforcement — the status quo, not a worst case.
What worst-case planning costs
Assuming a flat load growth and the worst summer and winter day leaves inefficiency on the table and puts copper where demand is not.
Agentic planning, with physics doing the maths
Agents orchestrate load flows and reinforcement algorithms and can draft the report, while every number comes from a physics simulator and every step is logged for audit.
Non-wired alternatives
Why a small battery supporting a local grid can beat digging up a Vienna street, and how planning tools must weigh both options.
Vehicle-to-grid versus vehicle-to-home
Herndler places V2G firmly in research — open questions from manufacturers down to operating mode — with vehicle-to-home and fleet depots the nearer opportunity.
Climate risk on exposed networks
A Turkish pilot on the Karaburun Peninsula, where overhead distribution meets extreme wind, storms and wildfire, and tower sensors enable predictive maintenance.
Where to put the sensors
State estimation means observability does not require a meter at every node; roughly 10% of nodes, placed algorithmically, is enough.
Why no LLM goes near the control room
SCADA is critical infrastructure and AI is a black box, so the realistic role is an advisory system beside it with an expert operator deciding.
Tacit knowledge walking out
Experienced grid staff are retiring with knowledge that was never written down, and new technicians cannot absorb a system this complex quickly.
Quick-fire round
Blockchain as the most overhyped, a faster load-flow solver as the most underrated, and why Rao thinks a billion euros is the wrong order of magnitude.

Key insights

5 takeaways

01
The connection queue, not generation cost, is the binding constraint
Rao cites estimates from European institutions, research organisations and DSOs that without reinforcement over the next 10 to 20 years up to 70% of renewable connection requests will remain unconnected. The projects exist and the capital exists; the grid is what stops them, which is why he calls the grid the place the transition is actually decided.
02
Worst-case planning is why reinforcement money lands in the wrong places
Traditional practice assumes a flat load growth figure and plans against the worst summer and winter day. That over-reinforces areas without demand while under-serving hot spots, and Herndler's team instead does regionalised techno-economic analysis to tell DSOs where adoption of EVs or heat pumps will actually concentrate.
03
The credible AI architecture keeps the model away from the arithmetic
Rao is explicit that language models never compute: agents orchestrate load flows, optimal power flows and reinforcement algorithms, and the numbers come from physics simulators that DSOs have trusted for decades. Every step is recorded for audit so the resulting network development plan can be reviewed — reproducibility being the condition for using any of this on critical infrastructure.
04
DSO conservatism is earned, not cultural
Rao says he used to call DSOs conservative and no longer does. At 99.9998% reliability, the mindset protecting that number is rational, and it explains why operators want pilots and proof before adopting new tools — and why no European DSO is interested in AI anywhere near a SCADA system.
05
The staffing gap is a knowledge-transfer problem, not just a headcount one
Both guests return to the shortage of qualified network planners, but the sharper risk they name is tacit knowledge: experienced technicians are retiring with expertise that was never documented, and the replacements cannot quickly comprehend a system this complex. Automation is pitched partly as a response — freeing planners to work as grid architects rather than running simulations by hand.

Pull quotes

6 quotes

Why 8 years? You can ask yourselves because that's what it takes.
Bharath Varsh Rao
So we will not use LLMs to compute anything.
Bharath Varsh Rao
We cannot plan the grid with the tools of 1990s to the plan the grid of the 2040s.
Bharath Varsh Rao
99.9998% reliability is unbelievable.
Bharath Varsh Rao
Blockchain is definitely over.
Bharath Varsh Rao
Everybody can be a superhero in the energy transition
Barbara Herndler

Transcript1827 cuesClick a timestamp to jump

Welcome to a new episode here in the
energy bridge. Uh today we have the
pleasure to have Barat and Barbara from
Austria Institute of Technology AIT. In
this episode we have many surprises and
the first one is that Barat is going to
make the context of this episode what is
important what we are going to talk
about. So pleased about that.
>> Really good to be uh back
with uh with you guys the energy bridge
here in Vienna actually. Beautiful sunny
afternoon.
>> Yeah, finally it's wonderful.
>> Spring starting finally.
>> Yeah. I am Paratra. uh with Austrian
Institute of Technology in the group of
power system digitalization uh as well
as uh soon to be a bigger group
combining two different groups which is
power system planning or operation which
will be merged into power systemization.
I think the new name is power system
planning operation and digitization.
>> Yeah, we're still discussing the name.
>> All right. Okay. [laughter]
Yeah. Uh I would love to provide the uh
the context uh for today. So the energy
transition that uh that that's happening
right now uh Europe is going all in. So
Europe is pretty much fully invested in
energy transition. So we are betting our
our future our economy uh for energy
transition our competitiveness as well
as our sovereigntity. So we need to make
sure that uh we can deliver this and uh
right now uh in the parliaments in
boardrooms in in in the editorial boards
of newspapers everybody is discussing
about exactly this topic. How can we
deliver energy transition without
bankrupting ourselves? Right now I want
to make a a humble statement that uh
energy transition
is going to happen in the power grid.
>> So it happens right underneath us here
in Vienna kilometer by kilometer. So
this is where the energy transition is
happening. This is the bottleneck. So
this is the context that I'd love to to
talk more about today with with the
three of you. So just on that note um
very welcome guys and to you as well
Barbara you are very summerlike today
and yeah completely how the weather is
in in Vienna but uh thank you very much
guys and and also for making the context
>> let's dive in deeper but uh before we
dive in deeper into the the the big
context that you made we would love to
know about you please introduce yourself
about your role and uh how you have come
across doing what you do currently. We
we start with Barbara.
>> So, good afternoon everybody and thank
you for having me here today. I'm very
excited to be here. This is my first
podcast. So, let's see how this
develops. But yeah, so um as I already
mentioned, my name is Barbara. I was
born and raised in South Africa, so a
long way from home, but I've found
myself here in Austria um with a lot of
new opportunities and a lot of new
challenges. I'm very happy to be here.
So my current context um in the role of
the power system and planning and
operation at AIT I'm a research
engineer.
>> Um I also focus a lot on the e-mobility
topics looking at various interactions
between the DSOS the TSOS also um
focusing on how we can integrate new
technologies with into the power grid
and yeah also trying to serve as a good
um project manager together with my
teams.
>> And you you did your study at in South
Africa.
>> Yes that's correct. So I was born and
raised Johannesburg but uh studied in
Cape Town. So
>> and what did you study your initial
study?
>> So electrical engineering. So I'm very
much a power systems engineer. Yeah.
Hardcore engineering. [laughter]
>> Yeah.
>> We can go deep.
>> Yeah. And and you Barrett?
>> Yeah. I already uh mentioned uh about my
role uh at uh within the group of power
systemization. Uh I'm a scientist and
business developer there. So um I do a
bit of project management research
projects as well as try to commercialize
the uh the solutions that uh we produce
across many research projects um in our
entire team. Uh the bigger team or
bigger group is called power and
renewable gas systems. So it's uh it's
both electrical energy system as well as
uh renewable gas system. So it's it's
both. So it's quite a large group uh
that we have at at AIT. And my
background is in power system
engineering. Um so I have a bachelor's,
masters and PhD in in power system
engineering. 10 years of experience a
lot of uh different uh Austrian European
international level projects. So right
now I'm in the process of uh actually
quitting AIT.
So creating my own spin-off and uh
hopefully next year spin out of AIT and
create a a company of my own.
>> Yeah, that's the other that was other of
the surprises. Today we are breaching
academia, industry, policy makers and
innovators within energy sector and you
are pretty much both of you reaching
from the research um bringing solution
innovative solutions to the industry and
we have something here on the table
which is the Hushka prize that um
Barbara won with your team last year and
recently yesterday was right
>> the day before yesterday.
>> The day before yesterday you won it
again. So we have some applauses
[laughter]
but uh yeah if if you can tell us uh a
bit what is the prize about and you can
tell us about the spin-off of your own
startup out of 80.
>> Yeah we were very humbled uh to receive
the the award for 2026. Uh so for the
project uh called Wterara. So Wterara is
a research innovative
that was within AIT. So it was part of
many research projects across uh eight
years approximately uh with many funded
projects um uh national Austrian as well
as European projects and also one
international project. So uh we were
able to fully develop the solution from
let's say TRL one like the fundamental
idea that we came up with uh to right
now TRL 7 or 8 very close to the to the
market. So we're able to fully develop
this. Why 8 years? You can ask
yourselves because that's what it takes.
Yeah. Because this is a very difficult
field. We need to be able to do it
correctly. Develop the solutions that is
trustworthy to our clients who are
distribution system operators
>> who manage a critical infrastructure.
>> So we we have to make sure that
everything works perfectly and it needs
to be meticulous and it is a slow
process unfortunately. That's how it is.
Yeah. To to come to a point where you
can sell it and then actually put it in
the real world and we need to make sure
that everything is is is correct and
that's the reason it takes a lot of
time. Yeah. And regarding the spin-off
um so the spin-off is going to be called
Rapson. So named after Joseph Rapson
who's a mathematician.
>> He's known for Newton Rapson method
which is
>> Yeah, I remember it. Yeah. which is the
the most important uh mathematical tool
for computing uh load flows or power
flows in in power system and Newton is
quite famous Rapson actually not not so
much not very well known so we decided
that we would give him give a bit of a
homage
>> to just rapson and uh that's why it's
called Rapson and um I would say a lot
of tools that were part of Voltera will
be taken into Rapson and and we'll
create a commercial solution out of it
and hopefully it'll help a lot of our
clients who are distribution system
operators of Austria, Europe and also of
course worldwide.
>> Amazing. And and for you Barbara, what
was the project about and and and you
and [laughter]
which among your team was working on the
project?
>> Um so basically it was actually last
year um that but with the project is
still ongoing. So it's actually called
the Medusa project which focuses on the
development of a multimegawatt charging
system and how to integrate that within
the power system as well. So we also see
that the directions of uh e-mobility for
electric vehicles not only for the
vehicles themselves but also towards
trucks and buses.
>> So this is also becoming quite a big
challenge and the industry is moving
quite fast in this regard. Um but also
what does it actually mean for the power
system and in this project we're
developing the solutions and
investigating how to integrated in a
better way. At the beginning as Barat
mentioned it's uh we are navigating in
this energy transition and to deploy
more copper is much slower than utilize
the grid more smartly. So you are using
AI to do that and also Barbara you
mentioned that you are um working in the
grid planning. So how do you take into
account these new solutions to how do
you plan a grid?
>> Yeah. So this is quite a challenging
question and it's not a simple answer.
>> So I mean I think every like system
operator has their own approaches
because the as we already know the
networks are very heterogenous. So each
let's say system operator has their own
unique challenges. Um but I mean at the
same time the energy transition is
transitioning into a way where we have
to accept and adopt new technologies
such as photovalttaic electric vehicles
and of course heat pumps just to mention
a few
>> and I mean I think all DSLRs are seeing
this challenge and I think connection
times are also becoming a big problem.
So waiting times, how to connect in in a
fast way. Um we see that waiting times
are accelerating because to keep up with
this technology, I mean it's not it's
not very linear. It's like accelerated
in an exponential level. And I think
it's good that people are accepting the
new technologies and adopting them, but
at the same time it's very challenging
for our DSOs.
>> Yeah. And no a following question that
you mentioned permitting because uh I I
also obtain permits in Chile for for
projects. So I know the pain of the
whole process and
>> and now Barrett you mentioned in the
last forum um that the grid operators
use the worst case scenario to give
permits
>> but what would it be if to this approach
would it be
>> optimized with AI forecast or whatever
how do you see that
>> yeah I mean that's that's a valid
question um so right now just again uh
to to take a step back a bit uh so right
now uh so we can estimate or eur
European and institutions as well as a
lot of research organizations and also
DSOs are estimating that if we don't
upgrade the grid in the next 10 15 20
years up to 70% of all the requests for
renewable energy generations for example
wind parks solar parks will be uh
remained unconnected so you cannot
connect them because the grids are not
ready right now
>> so we have% below the
>> so 70% of of connections could just stay
there and then the projects cannot be
realized
because the grid is not ready. This is
the the status right now we're in at the
moment up to 70% if you don't reinforce
the grid.
>> Yeah. Yeah, and that's where all this
you know challenges about uh uh the the
permission and then the connection
requests comes into picture because you
want to be able to very quickly perform
the simulations and and you know provide
accurate information to DSOs where to re
reinvest the grid or reinforce the grid
and when to reinforce the grid and how
how much to reinforce the grid because
there's only a certain amount of uh
funding that goes every year to grid
reinforcement. you want to make make
sure that it goes in in the right
places,
>> right? So that's super critical and you
don't want to re reinforce the grid
where there is not a lot of demand and
that's where the linearity comes into
the picture
>> because right now uh traditionally grid
reinforcement is done with just by
considering uh you know let's say 3%
load growth in a certain region and this
is you know it's called worst case
planning. You just assume that uh there
is the worst day in summer and worst day
in winter and then you do you know
planning based on that. Mhm.
>> Yeah. This is a lot of inefficiency just
sitting on the table. You're
reinvesting. You're spending a lot of
you know again copper and uh you can do
it better and more intelligently. I
think this is where you know one of the
projects that Barbara was part of and
and her team in power system planning
and operation uh they did it the right
way.
>> So we're also looking at ideas of more
like in a way of regionalization
approaches.
example identifying where the hot spots
would be for to certain technologies and
prioritizing where the reinforcement
requirements actually are. So I mean
identifying certain areas for example
which we might foresee a higher um
adoption of electric vehicles for
example in the future what the demands
there would be in terms of network
reinforcement and how much it would
actually cost. So we can do this so to
say a technoeconomic analysis of how um
these approaches should be done and help
the DSOS prioritize what their needs are
and where. But it's not only the fact
that it's just the where the funding
comes from. It's also about the
manpower. So I mean also it also needs
some effort from personnel. Um not
everybody can be everywhere at once but
this again helps to identify where the
priority areas are.
>> You both don't work in the same team
right or you it's like a same umbrella
team that you work in.
>> So when we combine uh these two groups
she's going to be my boss. [laughter]
>> But we work very closely together. So I
mean within power system planning and
operation we cannot achieve it without
the digitalization aspect. So I think
the the two are very much overlapping.
We have many overlapping projects
together publications and so on. So I
think it's it's a good combination and
it would be a nice way move forward to
combine the two teams.
>> That's what I wanted to understand as
well like how exactly a function and how
exactly you guys complement each other
in your function and in your projects.
>> Yeah. Uh it's a very uh uh very close uh
collaboration between power system
planning or operation and power system
digitalization groups here at IIT
because digitalization is key in pretty
much every field right now.
>> Mhm.
>> And especially in in power system uh
planning and operation. So that's where
a lot of projects Walter is the best
example right. So we have a lot of AI
going AI solutions in Walter
>> and and we work closely with the
colleagues from the Barbara's team.
Yeah. to to realize the uh because they
bring the power system expertise and
then we bring the power expertise of
course with the AI and visualization
expertise.
>> Barrett, what could you tell us about uh
the interaction or the intersection
between AI and network planning for
example?
>> Yeah, I mean that's uh that's something
that we are uh looking into at the
moment especially with the all this uh
agentic approaches where uh the the name
of the game is automation. And this also
ties into something that I mentioned
earlier uh also the industry issues with
uh with not enough you know people who
are uh experienced and expert in this
topics uh or mainly network planners.
Yeah. Uh and with uh with this agentic
approach we can uh you know calculate
multiple a multitude of you know tools.
So for example, if you want to do
network planning, you need to be able to
first run a load flow, right? You need
to run like a few more load flows to
figure out what the uh what the issue is
in the grid and then you need to run an
optimal reinforcement algorithm for
example. So all of these are very
mathematical and uh it needs a lot of
you know uh um input configuration and
all of this can be fully automated with
AI. So this is what we're trying to do
with Rapson as well. Mhm.
>> So where we have like a fully end toend
automated uh tool where a genetic
systems orchestrates different
simulations that that need to
[clears throat] run and ultimately can
also write a report for you. So that's
that's that's the whole automation uh
that we are trying to bring to uh grid
planning and then AIS can really help
with this. Yeah. And of course we need
to be careful. Yeah. Every step needs to
be uh uh recorded for audits and
reviews. Yeah. to make sure that the the
network development plan that comes out
of this uh uh tool is actually valid.
Yeah. And that's the reason LLMs never
do any kind of computation. So all the
numbers everything that are computed are
backed by physics simulators. So physics
inform simulation.
>> Yeah. They are not very good in maths
>> of course. [laughter] So we will not use
LLMs to compute anything. So it's it's
traditional power system tools
>> like optimal power flow load flow behind
which actually does the calculations.
>> So the agents only call those
simulations and then orchestrate the
whole flow
>> and that's the reason we can ensure that
the uh the answers are are actually
accurate
>> and and you would say that the AI based
stimulator the the math based stimulator
would be better than the AI based
stimulator in a way
>> all the time of course. Yeah because
also the trust is is important.
>> Yes. because the grid operators trust
these tools that have been around for 50
60 years.
>> Mhm.
>> Yeah. And then we uh cannot assure that
the uh the results coming out of an AI
LLM is accurate which is obvious and
this this we know it elucidates a lot.
Yeah. And uh this is critical
infrastructure. So we need to be able to
prove and reproducible results is is the
key.
>> And and at AIT uh do you have a
specialized AI team or that's like
everyone is working?
Yeah. I mean um AIT is part of um uh I
>> mean it's AIT [laughter]
factory.
>> AI factory. Exactly. Exactly. So uh
there's a group called uh digital safety
and security. So they are responsible
for the uh AIAT which is the AI factory
of Austria.
>> Uh it's still under construction at the
moment. um and they are responsible for
pretty much uh boosting uh the entire AI
industry of Austria. So any startup any
industry uh uh uh you know players want
to develop and and use compute
infrastructure so this is possible to
get from the air factory
>> and AIT is the uh founding member of a
>> yeah it's a leading leading member uh
for the infrastructure as well as a lot
of you know services coming out of the
uh the air factory but every single
research institution university are
stakeholders of of the air factory
>> but you you have been able to uh start a
company within AIT. Uh would you walk us
through how does uh the whole
entrepreneurship system or a catalyst? I
heard there's something called startup
catalyst within AIT. How does it
function and uh how does the spin off
from AIT? Are they do they actively uh
uh help spinning off uh from AIT and how
are the ecosystem working there?
>> Right. Yeah. Yeah, I mean it's it's
actually really wonderful because our uh
managing director uh one of one of them
we have three of them. Yeah. Uh shared
responsibility for AIT, one of them is
Alexander Spakovski. is um uh extremely
supportive for uh startups and and
spin-offs coming out of AIT because he
strongly believes that uh research uh
that we are developing amazing research
projects anyways uh should go to the
market and this is the this is part of
the the DNA of the uh of the managing
directors and that's the reason there is
a very strong entrepreneurship uh
support system built into a AIT and Hans
Peter Blhovski is the head of
entrepreneurship at IIT
and he is uh uh currently um deeply
involved in something called as founders
journey. So the founders's journey
begins actually at the beginning of a
PhD at AIT. So PhD students when when
when they start their PhD that's when
the journey begins and AIT supports them
throughout the whole journey until they
become an entrepreneur and then they
spin out of AIT. So this is the whole
whole process. So there's at every step
of the way there is support uh from the
entrepreneurship team.
>> Yeah. And I was asking the question
because a lot of people just they know
AIT is like something some institution a
lot of smart people sitting over there
like where it takes a bit longer to get
there but [laughter]
>> yeah it's a bit far away from
>> the effort to come and visit
>> but but uh but how how exactly you you
kind of collaborate with industries and
and the project that you work on how how
how it exactly gets implemented by
industries because you work together
with a lot of industries right? Do you
want to take a crack at that?
>> Yes. I mean, of course, we work together
a lot with our our industry partners as
well and I think it's very good
collaboration. So, we're learning a lot
from them and they're also we're helping
them find good solutions and I think the
collaboration up until now has been very
good for the most part and we're very
excited to actually work closely with
industry because that means we can solve
real world problems. I mean, we're also
quite advanced in terms of our
technologies and in terms of research,
but to actually see things being
implemented and actually utilized, it
makes makes all the difference. And I
think that's what makes our jobs very
exciting because we're very close to the
real world as well. And of course, also
with the new technologies keeping up
>> from the perspective of research, so
we're very much innovative, but at the
same time, we're able to realize some of
the solutions and at least knowing that
we have a good impact.
>> Yeah, absolutely. because AIT is uh a
research organization, right? So we are
not a university or educational
institution. So we uh if you if you look
at the technology readiness level,
right? So from from let's say TRL1 to
TRL 9 which is market.
>> So we operate between four five six
seven approximately. So we're closer to
the industry than uh the universities
are. And that's the reason the the first
question that we always ask ourselves uh
what does the industry need
>> that's our starting point and then we
build research projects together with
the industry partners. So in our field
we work very closely with a lot of
distribution system operators of Austria
and across Europe and uh they tell us
what they need and then we perform
research based on their needs. Of course
we go way beyond because that's what
being innovative is. Yeah. because they
of course are already overloaded with
the issues that they have.
>> They're putting out fires every day. So
we are trying to solve those immediate
issues and of course look beyond that
into the future as well and anticipate
where the market is going and where the
technology should follow. I just wanted
to
uh let or leave an idea that you
mentioned you both mentioned before that
now we are passing through uh from a
system which was a black box. So now we
are making it transparent box kind of or
it has to be uh transparent and smart
with full of lights and sensors and
everything like for everyone. So it's uh
super exciting and just to know your
your feeling that how is today the the
grid planning and how do you think that
it could it could be or it should be. So
to uh connect some of the the the the
crumbs that I was you know throwing
around. So uh Voltera a lot of it is on
control algorithms realtime operation of
the of the grid and right now we're
talking about grid planning. So what's
happening right now is there's a bit of
a unification at the moment. So the uh
the planning departments and planners
want to move away from all this
traditional planning approaches where
looking at just the worst case scenario
for example or reinforcing uh the
traditional infrastructure like the
cables uh underground cables overhead
cables or transformers all this you know
components right so we want to move away
from from that and also to your point
where you're just you know putting
copper is not always a solution you're
losing a lot of efficiency and leaving
it on the table so we want to
nonwired alternatives
which are flexibilities.
So that's that's what I talk about
everything coming together. Yeah. So we
want to use for example battery storage
systems to support the grid instead of
reinforcing the grid.
>> Mhm.
>> For example, in Vienna it's more
expensive to dig the cable and get it
out of the of the of the streets and put
in new cable versus installing a small
battery system that somehow supports the
the local distribution grid. So this is
what uh is the the evolution from the
traditional planning approaches to this
new paradigm where non-wired
alternatives are becoming more
predominant. Yeah. And this is exactly
the platform that Rapson is going to be.
So it's going to be a fully automated
grid planning tool which considers both
traditional planning approaches
>> and non-word alternatives. So we can
save up to 30%
in total uh uh infrastructure costs that
the uh the DSOs right now use per year.
>> Wow.
>> 30%. So we can reduce the capital costs
by by a certain amount and then at the
same time the OPEX is going to increase
right because DSOs have some operational
aspects now they need to be able to
operate the battery and
>> so we can combine both of them and play
around with the simulation and the and
the numbers. So we can save up to 30%
totex
>> across the board. So you so just to
paraphrase let me know if it's right but
we are transitioning from a predictable
load with a black box that it was
planned just um for long years because
you could know what were going to happen
and now we are with unpredictable load
distributed generation and consumption.
So your approach is that that the only
way to address it is with AI or or some
>> it's it's a combination
>> automated so solutions
>> the automation part is coming from the
AI. This touches like multiple points
because it's datadriven.
>> So you can see patterns that uh you know
typically you can't see in like
traditional equations. So we can you
know recognize patterns with a lot of
data. We have a lot of smart meter data
which we can use right now with the new
regulation uh from from the DSOS from D
control uh and this we can use to to
leverage a lot of uh you know hidden uh
background information that you
typically can't see and at the same time
AI can also automate a lot of processes
where uh the the DSO planning department
and the planning teams do not have to do
the tedious work that they currently do
by doing a lot of you know manual work a
lot of manual simulations. So right now
they can transition into becoming grid
architects rather than actually doing
you know very tedious simulations by
hand. So this is what we want to
transition transition to and this is
what Rapson uh can can do that
>> and could it be used also for for um
molecules like gas pipelines or other
kind of consumption or you are focused
now only in electrical and electrons.
Yeah, right now it's uh electricity
because uh electricity is where the the
major battleground is
>> for energy transition. Yeah. And we are
a startup so we need to pick [laughter]
>> the uh the minimum viable product
>> but in the future you are thinking into
>> I mean we will go wherever our clients
want us to go. So it's it's really
market driven. So we really very closely
listen to what our clients want.
>> Yeah. You definitely would also a
business developer at AAT.
>> [laughter]
>> Thank you.
>> Hey uh and also Barbara like just g
getting back to because you you also
focus on e-mobility uh at AIT. So just
wanted to understand that uh the load
growth from housing and industry I think
uh Daniel was touching upon before it's
changing and we we had a couple of uh uh
other startups like Podero which is also
from Austria they they are looking into
the flexibility assets uh
>> uh from um and from the e-mobility also
trying to get V2G
>> like what when there was a similar
projects which is between TUV and AIT
and also a company that we know KW
solution. what what what is the status
of V2G currently and are they are they
functional or are they already being
able to integrate or uh I'm I'm sorry
for going out of line but just wanted to
understand
>> yes I mean vehicle to grid is very much
a research topic for everybody at the
moment so I think it there's many open
questions so for example even ranging
from the technologies perspective from
so for example even from the
manufacturers of the vehicles themselves
even down to how do we ensure what
happens happens when the vehicles um are
operating in vehicle to grid mode for
example. But I think the biggest
opportunity here at the moment is to see
the aspects from vehicle to home where
people can actually utilize their their
own vehicles as a battery storage system
that I mean that can be operated. But I
mean there's many use cases that we're
currently investigating especially when
it comes to larger for example electric
vehicles like the trucks and buses. What
is the potential there? Because of
course they're bigger batteries and have
more a lot of potential there. And what
does it mean for them? And I think this
is also an interesting topic for us.
Yeah.
>> So is it one or another? You cannot have
vehicle to grid and vehicle to home.
>> Well, I mean the use case would mostly
be from a vehicle to home perspective
because then you can just basically use
your own vehicle to supply your own
household needs to some degree at least.
So I mean we also have research
questions coming into the perspective of
what it actually means for using uh
electric vehicles as a let's say a grid
to load perspective and we have a few
projects that are investigating it and
it's also a very interesting topic but
there's still many unknowns as well in
this in this area. So I think it's
something that we still have to keep our
eyes on. I mean electric vehicles for
example have a lot of potential there
and I mean especially when it also comes
to supporting the grid and also from a
black start perspective because I mean
identally they are basically batteries
on on wheels basically. So [laughter]
>> so yeah
>> right but how does it u kind of alter
the strategic model that you're working
on with within smart charging or V2G
like within the grid planning that you
were telling as a context. So I mean
we've also looked at it from a
perspective of using optimized um let's
say charging perspective and then moving
towards what happens actually when we're
using a vehicle to grid perspective. So
it's more on the idea of how we can
support the grid.
>> Um but there's also yeah many qu open
questions of what it means when we have
optimized charging and who's actually
benefiting is it the customer is it the
DSO and so on and so on. there's a lot
of questions open and how we actually
going to manage it in the future um is
still an open question especially here
in Austria and in in in other parts of
the world
>> are some projects ongoing that we could
ask you about in the future
>> there also from the electric vehicle
perspective um but yeah we haven't
implemented anything in the field at the
moment but it's still a lot in
simulations and investigations yeah
>> but currently uh V2G is not u
operational as such it's just pretty
much also in the project stages right
like it's not getting implemented
anywhere at the at a at a DSO level or
at a local level.
>> So the uh we had some projects
investigating it and we had some
demonstrations. Um there were a lot of
ch a lot of challenges in that regard
but we managed to get it. I think the
project came at an earlier stage. Um it
was called car to flex. It was it was a
very successful project at the end also
won many awards. So we were quite proud
of it and that was led um by Tuv at the
time. Um but yeah I mean it was yeah
something that we investigated in a
demonstration mode with different use
cases. So again, it was more from an
individual household perspective, car
ownership as well as other use cases
such as car sharing because that's also
becoming a good topic at the moment when
it comes to um how we want to use
e-mobility and or mobility in general
amongst the cities and then also from a
fleet perspective and again from a fleet
perspective here. We also see a lot of
potential here where for example
businesses want to electrify their
fleets and then they can use these
potentials like let's say for example
overnight where you can have a vehicle
to grid supporting system when the cars
are not being used basically.
>> Yeah. Just to understand the concept but
uh car sharing you mean literally share
the car and not own it.
>> Yeah. So
>> okay because I heard also like cars
changing charging cards that's another
word then car sharing in this context
car sharing. the sharing cost basically
many you have basically a fleet of
vehicles that people can use to share
amongst a community for example.
>> Yeah. And I heard I've heard only that
in only in Germany it's allowed now
vehicle to grid and they are making
tests but it's not spread uh yet and
also you you both have been in this uh
scalability and replicability analysis
SR
a sorry
>> uh so why would you say that these
analysis are so critical from a
perspective that you are coming from a
theorical
and research approach into a real world
problem if you could tell us more about
it.
>> I mean this is entirely barbarous topic.
>> [laughter]
>> Well, I mean it's both topic but it's an
interesting topic and I think
scalability and replicability analysis
is a very important topic as you already
mentioned and I mean looking just for
example at the scalability so for
example we might be able to develop
certain solutions but what does that
mean when we actually start implementing
them in the real world at a larger scale
so I mean I think adopting certain uh
technologies and approaches also needs
to be tested in such a way so we like to
do scalability and replicability
analysis um and what does that really
mean and also looking at because the
networks for example are very
heterogenous and there's no
one-izefits-all. So when we look at
replica replicability analysis, what
does it mean if we want to implement
certain solutions in different types of
grid for example in different regulatory
um aspects or in different countries
under different scenarios um and testing
if this actually really works and how it
works and how it how it differs for
example. Now just uh thinking about we
went to AIT one month ago and we were
shown this uh EV charger which is for a
project for India actually
>> and
>> low voltage
>> low voltage I think yeah but it's in DC
>> that was the the megawatt charger yeah
which is one of the the winners for last
year Cisco prize was it in the green
>> so so you had to study the the other
countries reality to to know how to
implement it. Yeah. So I mean like I
said every country has their own
challenges and they might have their own
regulations for example. They might have
their own grid codes. Um also just from
a a technical perspective they might
have different challenges because the
grids are different. They different
voltages for example different ages more
rural uh networks more urban networks
depending on where you want to place
these um charges for example in this
case. Um but it's just in general with
solutions because there's no
oneizefits-all. I mean we try to make
solutions that are applicable and
modular for most cases but of course we
also have to understand what the
differences are if they are and how they
actually impact our solutions and if
they do or if they don't.
>> Super cool. [laughter]
>> Super cool.
>> Just to understand the grids better as
well. Um my contention is that as you're
saying that the grid from different
countries will be different as well. So
within Europe there there might be some
some sort of homogeneous
uh genity within the the grid system.
But when you have to compare that to
let's say somewhere from Asia where the
it's it's a bit younger grid let's call
it like that. So
>> I mean also different challenges exist
in different countries. I mean just
talking for example from a South African
perspective there we also have see
different challenges there. I mean there
we have yeah different voltage levels
for example the infrastructure might be
different aged in a different way. We
have different technologies. Um so
there's challenges that differ from
country to country and I mean that's one
aspect where we see for example the
ISCAN initiative which is the
international smart grid action network
and here we also collaborate with many
countries across the world and
identifying what the challenges are
learning about their lessons learned for
different technologies we have different
working groups there in particular I'm
working in the working group six which
is also on transmission and distribution
grids and there again we allow knowledge
sharing activities to to happen and here
we can exchange and yeah share sharing
ideas
>> and you're the representative from AIT
or from Austria.
>> Um so I'm just one of the members of the
working group six and I'm running an
activity there on grid forming units in
particular but I mean there are many
different activities ongoing within
ISCAN itself. Yeah.
>> And and the the the the institute itself
the Ikan is for all countries.
>> Yeah. It's depends on the member states.
So every country can can apply and yeah
it's a little bit yeah complicated huh
>> that's exactly the uh the reason why
SRRA is so important and actually
Barbara also did the SR for this project
it's called gritcloud
>> mhm
>> yeah it's uh Aeronet project uh funded
by three funding organizations the
Austrian German as well as uh the
Turkish uh funding organization and we
had the uh I think the second biggest uh
distribution system operator in Turkey
uh GDC electric as part of the the
project 50 million customers. So it's
>> wow
>> quite large compared to the uh the
Austrian uh grid uh and distribution
companies.
>> Um so there we performed or Barbara and
a few other colleagues performed the
scalability and replicability analysis
to ensure that the solutions that were
developed in in Austria can actually be
extended to a country that is uh close
to Europe but is still beyond Europe. uh
to make sure that um these solutions can
actually be implemented and deployed uh
in Turkey.
>> Look, just putting more ingredients in
the soup, but I remember you mentioned
something that uh
>> you also take into account climate
changes into the electrical system to
address risks. For example,
um uh stronger winds or more
thunderstorms or more like
>> Absolutely. How how how do you address
all of this?
>> Yeah, it's it's a challenging task
because right now uh we are seeing that
uh some regions across the world uh and
especially right now in in in Turkey
where uh one of the pilots of the of the
project was it's u it's called the
Karabun Peninsula which is an hour away
from Ismir on the on the Mediterranean
Sea. That region is especially hit by
adverse uh climatic events like extreme
winds, thunderstorms,
um significant rain, uh and their their
entire distribution infrastructure is
above ground. So if it was underground,
then of course it would be affected
less, but they're above ground. Uh so
the cables are exposed and their the
distribution towers are exposed to all
this
>> the heat
>> adverse heat and then everything. It's
it's rain and then wind. Wind is
actually the major issue. wildfires.
>> And wildfires. Yeah, absolutely. Yeah.
And uh that's the reason we wanted to
install sensors on top of these
transmission or distribution towers and
to be able to do predictive maintenance.
So the goal is to be able to predict
>> and to figure out when a fault could
occur because of this climatic event. So
we track a lot of different data points
so that the the teams who are
responsible for going and fixing the uh
the grid if there is a breakdown. So
they're already prepared. So it's just
the point is to to minimize this this
downtime
>> as much as possible.
>> So that's the the whole point of the of
the project.
>> Just one following question because we
we read that you are
pretty much expert in to place or to um
estimate where to place the sensors. If
you have to place uh five sensors in a
load line for example
>> in a feeder.
>> Yeah.
>> And which sensor what do you sense
where? Yeah,
>> there's no obvious answer for this
question. It really depends on on the
specific uh feeder that you're looking
at. Yeah, because uh what we found out
on on many research projects a few years
ago and also continuing the work right
now. Also part of the Husco by the way
with the Volter solution is uh some of
our colleagues who are experts on on
data analytics and and AI. So they were
able to develop a a state estimation
method
>> which uh which proved that we actually
don't need uh measurement devices or
sensors at every single node in the
grid.
>> So we don't need need this. So we only
need it at some critical points in the
in the grid. So with with the help of
these minimal number of sensors you know
five six whatever it is
>> we can uh estimate the uh the voltages
the currents the active power the
reactive power and any other uh grid
parameter with the help of this this
small subset of of measurement devices.
>> So
>> and we're able to like prove this and
and that's where the the ideal sensor
placement is important.
>> Mhm.
>> Yeah. So we can algorithmically
calculate
um based on some methods that we have
where is this you know this ideal
location using which we can predict uh
the other uh points in the grid.
>> So it's kind of have a minimum cost
digital twin.
>> Exactly. So you're you're you don't need
to invest uh millions uh in in full
observability of the grid. This is not
needed. So you can do let's say 10% is
is the rule of thumb uh is enough to be
able to get full observability of the of
the grid.
>> Dynamic operating envelopes are a hot
topic. Barbara, your team is developing
method for them. How do they actually
change the conversation between DSO and
the proumer?
>> Yeah. So this is a very interesting
topic for us and I think it's also
coming up a lot in international um
perspective. So for example in Australia
they're very far advanced when it comes
to the operational envelope topics but
not to mention within our team itself we
also have many projects that are
investigating this topic and I think
it's very interesting because we also
want to see how we can adapt from this
so-called static approach to a more
dynamic approach to identifying when
users are utilizing the grid and at what
time of the day and how this can be used
by the DSOs in a flexible way. So for
example having a fixed connection point
versus having a more dynamic connection
point and how this changes is something
that needs to be investigated and
further developed within various
countries.
>> Yeah. Yeah. Energy management solutions
for home and and there are there whole
bunch of them within Austria within
within Europe right now. Do you think
that they will be uh going forward
within the utilities themselves them
having a software and then it's it's
like a more market driven question but
like they will die out or you think they
everyone will have a place because of
the decentralized system.
>> So let me let me understand this
question again. So you're asking me uh
uh so you have a lot of different uh
software developers and and you know
vendors supplying energy mining
solutions. Yeah, like there there many
many so flexibility there there's a
flexibility for the bigger assets there
flexibility for the home assets and
there there are many software with the
optimizing what what Barbara was
touching upon as well but going forward
this
>> at the grid level or at the house level
>> at the house level
>> you got to look at yeah behind the meter
>> behind the meter or
>> the meter one is a system level and one
is more like based on behind the meter
>> so so for example optimizing your
battery your you know rooftop solar
panel and your electric car.
>> Yeah, I think yes.
>> Yeah. And yes, there are a lot of them a
lot of you know companies and startups
>> and initiatives which have solutions for
this uh market.
>> Yeah. I mean uh because sometime I just
feel that uh they there's so many of uh
that optimization going on but uh on on
what level and
>> yeah I mean it's um it's it's it's an
interesting uh market to be in because
uh the uh barrier of entry is I would
say lower than the companies who do
system level optimization because that
needs a lot of expertise in part system
engineering in uh and in software
development and you know these two needs
to like these two domains need to like
coincide really well
>> and that's the reason solutions like
Wterara and Rapsson and all of that can
only come out of uh research
organizations like AIT or utilities
themselves or DSOs themselves. So all
the all the market participants and or
the market itself or the distribution
industry tends to be a bit more
challenging to enter and that's the
reason there are not many players in the
distribution energy management scale
>> and and you will also talk to the DSOS
directly you because
>> we always talk to the DSOs
>> and and it it will limit to them that if
they're going to extend the benefit of
the 30% and uh efficiency that you're
telling
>> that's the planning.
>> Okay.
>> Yeah. So the market of the grid planning
is different from the operation uh
software market
>> right. So so wter uh with the host price
that we won it's it's a control solution
for the operations
>> Mhm.
>> itself. So operation of different assets
in the grid and how we manage it and how
the voltages and then the loading of the
grid uh in is you know behaving in terms
of uh uh uh you know different
flexibilities that that are available
and how we can it's like a control
day-to-day operation of the of the grid
>> and Rapson is is a planning tool so it
helps DSO teams plan better for the
future.
>> Okay. And
>> two different market segments here.
>> Yeah.
>> Yeah. And in terms of in in the market
of grid operators in your opinion what
is the actually or or how actually helps
these LLMs in the control room?
>> Yeah.
>> For grid operators but where I mean how
actually help but and what is the
the border where the operators will say
okay LLM doesn't work here. This is
>> yeah I mean this is this is an
interesting question right so we're also
trying to answer this question um at IT
in our in our teams right now the uh
control room of uh of the DSOS's uh
let's say a scattera system or an
advanced distribution manage system um
and some of the biggest you know
companies in the energy industry uh they
have their own products in this regard
and these are extremely closed systems
because they're part of a critical
infrastructure so right Now no DSO in
Europe at least based on you know all
the DSOs that we have spoken to is
willing to or even interested in putting
any kind of AI or large language model
>> anywhere close to the scholar system
>> this is not happening
>> also any securities around like
>> this is a part of a critical
infrastructure so any decision that you
make it's it's super critical that that
it doesn't lead to some kind of adverse
effect on the grid yeah and AI is
blackbox so they don't know how the
decisions of AI and and what kind of
decisions it's making and how that would
affect the the grid. So this is not
happening in anytime future. Uh but
there there could be um some kind of
supervisory an advisory system which is
parallel to this kind of system which
doesn't have any kind of real control
over any of the assets
>> but just advising the the controllers
the operators of the of the grid in the
control room to take one action or the
other. So this is possible. there
already tools out there which provide
this kind of let's say supervisory and
advisory support
>> mhm
>> for the uh for the operators and then
they still have to have the expertise
which they do to say yes okay this is
the right decision uh which is coming
out of an AI LLM or any other AI tool uh
and if they want to execute this
decision in the real SCA system which
has a direct impact on the on the grid
or not
>> so we are changing one black box from
the electrical grid to a black books in
AI.
>> Yeah. [laughter]
>> Okay.
>> It was very interesting. Uh we asked
this question to also uh Philip Han from
Cyber Grid
>> and and uh and it was interesting. He
said he would always like a human to be
>> at the end of the levels.
>> Exactly.
>> He would not at least at this level he
would not entirely leave it to the
system.
>> Yeah.
>> Fully agree. [laughter]
>> Yeah. Fully agree. Barrett, you when you
pitch to offer digital services to TSOs
and DSOs, what are the honest interest
from these companies to acquire digital
services?
>> Mhm.
>> And what interest they show you
>> with this? Yeah, I mean right now a lot
of uh DSOs as well as TSOs are under a
lot of pressure to uh to adapt to this
changing uh world that we are uh
increasingly living in uh energy crisis
uh crisis in terms of grid complexity.
So right now the figures are up I think
five million devices in terms of
electric vehicles, photoable tech
panels, heat pumps, uh wind parks that
are uh waiting to be connected every
year and the grids are not prepared for
for this kind of uh uh uh uh transition.
Yeah. And um they are openly uh uh and
increasingly getting more interested in
this new approaches because uh we cannot
uh plan the grid with the tools of 1990s
to the plan the grid of the 2040s.
>> We agreed that it was built in the 70s
>> with the grid that was built in
[laughter] the 70s. Exactly. Yeah.
Exactly. So uh and that's the reason uh
increasingly seeing a lot of DSOs who
are extremely interested in those
digital tools this innovative tools that
are coming up which is amazing for us
right because we've been working on all
the solutions for the last 10 years and
and now the the pressure is really
mounting and then they want to see these
tools in reality they want a lot of lot
more pilot uh interest to actually see
it in real life because uh um of course
they want to make sure uh that the grid
is still stable to when they adopt this
the solutions So rightfully yeah. So
because a lot of uh people I myself in
the past have have uh repeatedly
mentioned that the DSOs are conservative
but now I'm after talking to them like a
lot in the last you know year or so.
>> I I I totally understand why this
conservatism comes from. Yeah. It's
>> because the grid is is in in Austria and
in Europe are one of the best in the
world. Yeah. The the the numbers speak
for themselves. You know 99.9998%
reliability is unbelievable. That kind
of mindset is still here because we
still want to maintain this reliability,
this level of reliability. Yeah. And I
totally get it now. Yeah.
>> And have you seen some kind of
difference like generationally like the
old experts from the old grid and now
with new professionals with new tools
what is the resistance and the
acceptability? It's every every
generation of of uh grid planner, grid
operator is understands the the issue
that we're facing right now and they
want to change and they want to adapt
new solutions. But the the main issue
that the DSOs are also facing is that
all the experience uh uh experienced
older generation professionals are
leaving and retiring.
>> Yeah.
>> So then what happens? Yeah. I mean
that's that's kind of a bit of a risk,
isn't it? Because if you're a grid
operator,
>> you have such experience the technicians
and professionals who are taking all
this knowledge that's basically not
written down anywhere. Yeah.
>> And then uh the new technicians are
coming in they they they simply cannot
you know comprehend because it's a very
complex system. So they need to like
impart this knowledge to the new
generation of technicians and then
retire. So this is a huge risk right now
in in Europe and uh it's you know
commonly known as tacet knowledge
>> in in AI space. So we weren't able to
like capture that somehow and then
deliver or pass it on to the next
generation of power system engineers.
>> Yeah. Well, recently we published in in
our social media uh research about
energy skills and in in the last two
years in Germany and and Austria and
it's very aligned with uh international
energy agency report that it's a lack of
professionals and skill in energy
sector. So we are addressing this in the
bridge. We are building um community
intelligence and we want to capture and
keep involved the old professional that
that are in this retiring period and to
accelerate the learning curve for the
new ones. Y
>> so yeah keep in touch with the because
we are going to bring some news about it
that we are building this uh energy
brain
>> and Barbara you're a task leader in IEA
ISCAN working group six that you earlier
mentioned so what role does ISCAN play
in bridging the gap between the national
grid initiative and the global knowledge
sharing and how much of the sharing
really happens because a lot of the time
like uh what is going on is that the
even who DASOS do to not share so much
of knowledge among themselves. So we we
are talking about knowledge sharing
between countries. So how much of the
knowledge
>> I mean of course it's not specific data
sharing um but it's more about
approaches and lessons learned from
implementing certain technologies. So I
think that's what the biggest benefit
again like I already mentioned as taking
aspects from use cases regulatory
sandbox aspects bringing together what
is been identified it's I mean it's a
wide range of topics. is not only the
from the technical perspective but also
from the adoption of actually people
from a a customer perspective what the
challenges are there and how to address
them and I think those lessons learned
are valuable especially to identify what
countries are doing what and how how
their solutions can be implemented so
it's more not it's I mean some in some
cases it can be very technical but also
on a policy level what policy measures
should be changed how to accelerate
certain adoption of certain technologies
>> yeah and Barbara you're also a national
expert
uh in cigarette C6 what what are the
emerging topics right now within the
community and what do you think is the
community underestimating
>> yes this is also a very big topic um I
think flexibility is still a hot topic
but also I see a lot of interest moving
towards this grid forming topic so also
identifying what it means for the DSOs
and also on the TSO level and I think
this is something that's very much
emerging it's underestimated because
there's still so many unknowns and we
not really sure of what's going to
happen in the future but it's something
that I see emerging and coming up as a
hot topic and it's something that should
be addressed.
>> What do you think about the current
increasing interest into having more
solutions in DC? I mean we were talking
about the electrical system is super
conservative and it has been worked with
AC for many many years and now we are
coming back with Edison using DC.
>> So Tesla for the win Edison is coming
back. Yeah.
>> Yeah. Exactly. So what do you think
about it?
>> Yeah. So definitely we also see that DC
is becoming a hot topic now and there's
also a lot of many aspects that are
still unknown and challenges. So for
example things like uh rectifier options
things like protection is still a big
question. Um but these technologies are
definitely coming up as a solution and
DSOS are interested in it seeing what is
what's going to happen. So for example
we have many use cases. So for example
again like the megawatt charging system
what it actually means for the grid
operators where they can also look at it
as an alternative to network
reinforcement because of course then we
can have a higher capacity transfer
within the grid and this topic for
example even ships aeroplanes uh DC hubs
and so on.
>> Um we're looking at data centers coming
up in the future and so on. Yeah.
>> Yeah. Actually we are having a session
in the forum that we are having in June
and one session will be only about data
centers and there is a company which is
>> calculating where to place them also
optimally. So it's yeah super exciting.
>> Yeah that's the uh grid planning teams
uh nightmare to to place data centers.
>> Yeah sure
>> and also of course DSOs to to approve
the the project and upgrade the grid.
Yeah.
>> Yeah. And that's a paradox right because
data centers are paying for the DSO but
they are bringing AI which is the
solution. So [laughter]
>> yes yeah for sure yeah the timing is the
the issue right so it's yeah we need the
data centers right now but the grid is
not ready and uh
>> plan better to have more data centers of
course yeah AI is is great and we need
more data centers for sure. Now we go to
the last part which is the quagfire
round and the third question is what
would you do with 1 billion euros to
accelerate energy transition.
>> Okay. So for me personally I think
electrification is still a big topic. So
this idea of energy poverty we see many
in many countries where many people
still don't have access to electricity
and I think putting the investment there
is also a big topic that we should be
considering. So for example like in
emerging countries where some people
don't even have the basic essential of
turning on a light bulb for example and
that's something we should address in
applying and allowing people to have
this idea of energy security.
>> Now Europe is electrified like in 22 23%
approach do you know or how much do you
think that Europe could be electrified?
No
>> that's a good yeah big question. I think
it also depends on industries and
updating the industries there for
example. But I think I mean a for
example and Europe in general is quite
established when it comes to
electrification especially on the
household level. But I think moving
forward to applying for different
industries as industry is growing. Um
also finding ways to make them more
efficient. I think that's a bigger topic
at the moment.
>> Yeah. But could it reach 30% any soon?
Because China, we went to China recently
last year and they are near to 40
>> electrified they say.
>> I mean I think I think these are
challenges that the DSOs are addressing
here within Europe. So I think it's very
much on their radar as well alongside
all the other challenges that they're
addressing. Yeah.
>> So electrification that's pretty much
that's my investment for the future.
Yeah.
>> You did you say 1 billion or 1 billion?
>> Wow. 1 billion is a very small amount.
You're not you're not
if you say if you say one trillion then
then we can talk.
>> Okay.
>> Yeah. Because because Europe Europe is
uh already investing 400 billion in the
next 10 years for electrification and uh
grid expansion and grid retrofit.
>> Oh wow. Yeah.
>> So the scale is 1 trillion. Yeah. Then
then I would put all of that into
carefully uh uh retrofitting and uh
reinforcing uh the grid so that we can
already uh integrate a lot more
renewable energy and and speed up the uh
energy transition
>> and unlock all the projects waiting in
the
>> exactly there's a lot of projects
waiting uh in in the in the queue. So to
already unlock this as quickly as
possible because this ties into the
whole energy sovereigntity, energy
poverty pretty much uh all the u the big
bets that have placed on energy
transition. So that would unlock
everything. That's where I would put the
one trillion
>> one trillion.
>> Yeah.
>> Cool.
>> If you had to pick one technology or
method that will reshape distribution
planning by 2030, what would it be? And
how much of the TSOS and DSL are into
network planning? Are they planning for
2030 or they're planning further?
>> Yep. Yeah, that's a wonderful question.
Um, a super relevant question as well.
So, uh, uh, I think I spoke to 17 DSOs
in the last, let's say, three or four
months for, of course, Rapson, my
startup.
>> And, uh, uh, every single one of them
uh, mentioned to me that they're right
now so stressed with so many connection
requests that are getting at the moment.
So, they're really putting out fires. uh
every day to to be able to accommodate
all these new requests. Uh according to
the European regulation as well as a lot
of national uh regulations, they are
required to produce
>> mhm
>> uh a national development plan NDP every
two years which also includes uh future
planning as well. So how the grid will
evolve in the in the future and they do
this based on a lot of different
methodologies. You it really depends on
the on the DSO themselves. uh but they
uh or at least what I found out was they
absolutely uh would love to have a tool
that can automate a lot of these things
because they're of course you know also
hit by the same challenges that uh the
whole industry is is experiencing not a
lot of engineers who can uh help with uh
with you know doing all this very
tedious calculations
>> and Rapson can somehow automate it and
and by the way in Rapson uh we will have
uh just one button if you just click on
this button instantaneously you'll get a
fully developed network development
plan. Wow, that's nice.
>> And what would you say is the most
overhyped technology currently?
>> Blockchain is definitely over. So that I
can certainly uh say that for sure.
>> I mean it's overhyped for anything like
not [laughter]
>> Yeah. But it was especially high for
energy right and and also uh from the
grid uh point of view and that is
definitely not happening. Yeah. And in
the energy sector maybe the most
underrated concept that is now what
would you say that is the hidden gym?
>> So we have one of our uh PhDs within our
group uh uh especially specifically
focusing on how to uh speed up uh a load
flow method.
>> Okay.
>> Yeah. It's it's such a fundamental uh
tool that everybody uses in in power
system engineering. every single DSO TSO
every single day someone is you know
calculating a load flow
>> right and then there are uh standard
state-of-the-art methods
>> and uh our colleague Alper
uh he's working on speeding this up by
let's say a margin of 10
>> mh
>> and the implications are actually quite
significant with this just a 10 10 time
uh increase in in the conversion speed
of of a loadflow is extremely
significant because we are if you want
to use AI solutions and if you want to
uh calculate a lot of you know load
flows in loop this really adds up at the
European scale. So I would say uh a
humble loadflow tool to to make it
faster I think this is like super
underrated right now in my opinion needs
to get a bit more attention.
>> Yeah. If uh young electrical engineer
ask you today whether to go into grid
planning physical grid operation or AI
for energy what would you tell them? I
mean in general what advice you would
give to the young engineers getting into
this field? Yeah, I mean I think we've
identified already that there is a bit
of a shortage gap coming up in the
future and trying to encourage younger
generation to follow through and
becoming more engineering uh topics such
as the ones that you've already
mentioned and I think this is the big
challenge um trying to make power
systems look sexy is a bit yeah
challenging for us but I think going
forward this is the future and I think
everybody has a major role to play and I
think the sooner we address this the
better and everybody can be a superhero
in the energy transition and I think
that's what makes the And I think making
sure that the people are and the younger
generation is more aware of these topics
and how they can actually contribute to
creating this change is what makes it
more challenging and is the way forward
basically.
>> And you
um I would say education talking more
about uh u energy transition uh which is
happening already. I think uh Gen Z uh
is extremely climate conscious which is
very hopeful. Yeah, I'm very happy about
that.
>> Yeah. and uh they are uh extremely uh
more active compared to other
generations going out and doing
something about it which is also very
hopeful. Yeah. And uh I also see that
there is I would say an uptick in uh
interest in the engineering uh streams.
>> Okay. So in that note, thank you very
much for coming today. We are very
exciting also to having you both in the
forum in June. Barrett, you're going to
maybe moderate the biggest session. It's
It looks very good. Uh so yeah, thank
you very much. I hope you enjoyed it as
much as we did. Um looking forward for
seeing you in the forum and congrats
again for the Galardon. [clears throat]
>> Thank you so much. Yeah, it's always
wonderful to to talk to you guys and
also be back at the Energy Bridge
headquarters here in Vienna. Uh yeah. Uh
I'm I'm really looking forward to the uh
to the sessions that uh that you were
I'm still in a bit of a dark darkness
here, so I don't know what to expect,
but I'm sure it's going to be amazing.
>> Yes,
>> you will see.
>> Thank you very again.
>> And just from my side, thank you very
much for the invitation. It was very
exciting to be here and talk about my
favorite topics. [laughter] So I'm
looking forward to seeing what comes out
of this energy bridge um podcast in the
future. And yeah, I wish you guys all
the best.
>> Thank you very much, Barbara.
>> Thank you. And thanks for the comm to
the community for watching the episode
until the end. And keep tuned for next
interviews and for the [music] forum in
June. So see you. Chowo chiao.

Dynamic Grids for a Dynamic Load: E-Mobility, DOEs & AI | B. Herndler & B.V. Rao, AIT | S2E15

The connection queue, not generation cost, is the binding constraint

Worst-case planning is why reinforcement money lands in the wrong places

The credible AI architecture keeps the model away from the arithmetic

DSO conservatism is earned, not cultural

The staffing gap is a knowledge-transfer problem, not just a headcount one