So, we have stopped somewhere between regional and global perspective.
Let's go global.
The energy transition implies that thousands power systems around the world should be transformed
with a pace which has never been seen before.
And while we know what should look like the picture on the global level, that is still
a question how should it be translated into regional levels.
And what is special about this global scale energy planning problem is that we should
plan decades ahead under deep uncertainties.
And basically, we have quite an experience of energy policy failures.
There have been quite a few cases when energy policy measures looked quite reasonable in
advance but have resulted in failures, didn't lead to results which have been expected and
these programs should be stopped.
And that is why actually we need large scale energy modeling.
We can replace this painful experience of rail war cultures by playing with the modeling,
with energy models.
And obvious advantages of open source, of open modeling and open data for energy planning
has lead to a rapid increase in interest towards open energy modeling.
And currently we have dozens of open energy models.
We have a lot of open data sets relevant for energy modeling.
But the picture is very incomplete and very patchy and there are regions in the world
where we do not have even a net zero plan not to say open net zero plan.
And that is exactly the gap which we are addressing as independent research initiative.
PIPESIMEDS Earth aims to provide every part of the world with the open and reproducible
and accessible energy systems model.
What we are doing can be divided by three blocks.
First of all we are doing open coding indeed.
We are working with open data and we support open energy modeling community.
So just a reminder about energy systems model.
There are I would say power engineering models that is tools which we mainly have discussed
today and there are also academic integrated assessment models.
Academic integrated assessment models relate to the whole world and model global scale
large scale interconnections between economics, environment and energy.
And energy systems model that is kind of tool which translates these results of global assessment
into plan of actions on the regional scale and obviously energy systems model should
contain should reproduce in realistic way behavior of power systems.
So that is what our workflow what our architecture look like.
We have data block, we have modeling block and we have optimization block.
Processing is orchestrated by snake mate and well probably the most trying part of the
whole picture is work with data.
There are different groups of data which effort operation of power system and there is also
quite trivial but very impactful moment which relates directly to open data licensing.
Basically we have starters data kit which we provide with the model to facilitate starting
with modeling and I think the most frequent how to start request is about loading this
starter kids data and many troubles by created by the fact that some open light some licenses
of open data set do not allow redistribution or hosting of the data.
So for some data we can collect data set and transform it in the form which is needed for
energy system model to run while for others we do not have right to redistribute and have
provide data to sources and connect them with the scripts to clean data and to prepare them
to format which can be used by energy modeling and that is exactly chain of the whole link
which breaks most frequently.
So just open data in action.
Environmental and climate that is part of the data workflow where we are truly grateful
to open science community and to geophysical community.
Basically that is the most unproblematic part of the whole workflow and we have package
which translates geophysics to energy related parameters and basically that's it.
Mainly it just works but as for electricity demand here the biggest problem is data availability.
Indeed well what we need are hourly demand profile for every country of the world at least
at national at aggregated national level.
Indeed they data exist but they are not openly available and so we have a model we have machine
learning model which has reproduced synthetic lot profiles but we would be very happy to
improve flexibility and geographic coverage of this approach and access to the data to
original lot profiles that is bottleneck currently for this group.
Another important part, another part which is crucial if you are interested to model
a power system of some arbitrary country is data on power infrastructure especially on
grid and here we have used open street map database and developed a dedicated package
which extracts power futures and allows to prepare model of grid topology.
A part of that we have packages from pipes ecosystem which provides data on power plants
on installed generation and a data set which collects and curates data on technology costs
including forecasts for technology of technologies development.
So and that is what modeling workflow look like.
We take preprocessed data for power infrastructure and simplified topology preserving electrical
properties of original power grid then cluster it to make the problem tractable and the next
point is the most challengeable from the perspective of open source because open solvers are still
overplayed by commercial solutions and here there is some room for improvement and we
are collaborating with developers of open solver to improve the situation and now once the workflow
has been established we had to be to ensure that it is actually possible to apply our model for
every country in the world in the most literal sense.
It has it took about almost a year work to introduce all the necessary fixes which account for
different special futures and now it is done that is linked to it's another report which
contains schemes for power systems of every country of the world of 193 United States country
and we also have the code the source code which we have used to produce these schemes as images
and if you are interested in model any country of the world please feel free to do that.
Now let's look what actually can we obtain if we apply this approach that is net zero study for
Nigeria which we have used in course of development the model which we have used as kind of proof of
concept and the lessons which we have learned the most interesting output of this study has been
that well net zero power system for Nigeria can be actually a little cheaper as compared to
state of as compared with status quo indeed we haven't included we haven't accounted properly for
uncertainties which exist for energy demand for Nigeria and this work should be certainly
continued and applied to every country of the African continent but that is what does it look
like that is which well that may be helpful to shift a paradigm and that is actually what is it all
about and that is a study which has been done by in collaboration of pipe summits earth and open
energy transition and a German think tank agor energy vendor they have considered Kazakhstan power
system and the question is if it is feasible to implement solar in and wind faster as compared to
with the current Kazakhstan current national development plans and the results are quite
encouraging and being currently discussed on policy level and that is output of a master study for
Saudi Arabia and that is a country where 99% of energy mix relays on fossil fuels a study which
an author has done using pipes earth has shown that wind and solar actually can have quite well
quite a place in power system of Saudi Arabia and it isn't so expensive as it could be expected
that is the case when data accessibility data availability is a big issue so this this results
are quite preliminary because more advanced optimization methods are needed to account for
this uncertainty and also account for all pathway all transformation pathway but what is important
what is an effect what is an impact of this study is translating conversations translating
discussion about possible futures for fossil fuel relate countries from purely
hypothec hypothec level to a level of numbers and that is a case for Bolivia case when South
America when networks of South America are considered and that is region where data of
open street map data have not so good quality so it has been needed to introduce quite some
tricks to restore topology and the resulted model has been successfully validated for energy
for dispatch on the national level so it works even if you don't have data of such excellent
quality in open street map and that is a case for Malaysia we have considered decrobanization of
industry and in Malaysia the local the local feature is well renewable sources renewable
potential is not so excellent so we have shown that it is basically possible to decrobonize one
bright branch of energy sector but if we would speak about the whole national economics it looks
like it makes absolute sense to include into modeling into discussion not only traditional
on wind off wind and photovoltaic also something more exotic like floating solar or probably to
consider cross-country interconnections so and last but not least community is essential part of
the whole story we have different channel of communications and we are very interested that
is essential for now for us to build global community as we have seen there are some countries of
the world where there are still a lot of modeling evidence available where efforts of researchers
and developers are still focused but energy transition is a global thing and if we wanted to
work we need to provide tool we need to involve people around the whole world and we can unfortunately
confirm that there is definitely a gap geographic geographic gap in free and open source software
community Tobias has talked about during free about that during previous first day and now I think we
have some understanding about reasons which which are behind this gap and that is basically quite
quite simple people around people in different regions just have different patterns of communication
and that should be accounted for if you want to build inclusive community and another part of the
story is that many things which we take for granted like education or even stable internet
connection cannot be taken for granted in too many parts of the world so but the good news is that
actually those problem which cannot be solved alone can be perfectly can be solved if we join efforts
and well we are doing it we are solving them we still have a lot to do there are research tasks
there are validation tasks because we can build power system model for every country of the world
but it would be nice to understand how close are we to reality what are errors what are modeling
errors for each of the components for power grids model for installed capacity how far we are from
reality in demand profiles and that validation task it is huge if you're interested to join please
feel absolutely free we would be happy to accommodate you and another big task is to increase
usability in particular condo environment and version conflicts inside all our pytonic soul that
is still a big questions and we have we would be very happy to improve it somehow and another part
relates to capacity building relates to improvement of documentation and to spreading the world spreading
knowledge so again we are very happy to accommodate any suggestions and we are inviting contributions if
you are interested please do not hesitate to ping us using any of our channel our communication
channel so just a reminder that energy transition is a global thing and can be tackled effectively
only together thank you very much and I am very happy to take your questions
what's the role of Earth observation for this models do you use satellite data to track transit
lines or look you look for wind turbines or solar cells or is in this data set you use is just you
use official data sets for your modeling thank you we do not use satellite observations directly
but we are using for power grid we are using open-street map data only while it would be great to
supplement them by satellite images we had some stream which have been focused on addition and on
adding actually satellite data to open-street map but the this team currently is not very active so
that is perfectly that would be perfect that would work definitely but we just don't have capacity to
do that right now also we would be happy to revive it and as for installed capacities we are using
fusion we are using merging of a number open data sets on power plants I am not sure if satellite
observations have been used of in any of this data sets but at least we don't make yet satellite
processing ourselves and we do not use them directly also I agree with you that it would be
very interesting idea and it would be also perfect academic topic
there are some countries which really don't want to collaborate I don't know quite like North Korea
or other other countries where we don't get any data well to answer that directly we have data for
Northern Korea but we would be very careful about using them because when you are applying if you
are modeling especially well specific countries I would be very much concerned about safety people
who are affiliated with those countries and that also goes for China for example because for China
there are some local regulations which basically forbid going into too much details of power system
for people who are not approved by national government so I would be very careful about
delicate areas of the world but technically yes it is possible so my feeling is that correct
approach would be to try build collaboration in more or less safe way providing tools to people
who are safe using these tools for example if there is there is some group in China who is
approved as approved by national authorities as experts in power systems as people whom whom they
trust then we may provide tool and support them in using it in the right way also I agree that it
is complex question and it may go a little bit complicated first let me remark agor energy
vendors a very good name in Germany so Congress and getting them using that and my question now
are you also doing storage like water reservoirs or millions of batteries distributed well I
agree that storage is one of the key question when we are speaking about energy transition and we
include a number of different storage technologies and currently that is one of the key points of
energy transition we are able to capture them and if you're interested please feel free to
investigate the details we would be very happy very happy to obtain your feedback and suggestions
and contributions if you see that something can be improved actually we have a huge poor request
which should provide interface to a big list of different storage technologies and it would be
perfect if you could revive it I was just interested I've got a friend who I'm doing who's a
research who's doing geothermal in Nigeria and you have to yeah well he's doing it on me do you have
geothermal resources in there as well okay all right good thank you yes we have geothermal and we
have quite recent request from from Kenya where people are interested to include geothermal in a
more sophisticated way thank you