WEBVTT 00:00.000 --> 00:11.800 Good, everyone, my name is Andreas, and in today's talk, I'm also going to talk about 00:11.800 --> 00:16.480 why society needs free and open power grid data. 00:16.480 --> 00:21.640 So now we just had lunch, so some of you might be a bit sleepy, so I'll ask you a question. 00:21.640 --> 00:27.400 What do you think is the total worldwide length of transmission lines, so how big is the 00:27.400 --> 00:28.400 grid in the world? 00:28.400 --> 00:38.080 Just give you a few seconds, and the correct answer is, we don't know at all, but a seemingly 00:38.080 --> 00:43.120 basic number, and yet there isn't a common definition used for transmission, there isn't 00:43.120 --> 00:48.360 a common way to calculate it, or using the sag or not, there isn't a common open standard 00:48.360 --> 00:51.960 framework for these types of statistics and this data. 00:52.360 --> 00:57.480 What I can tell you though is that one of these numbers, which is one that we've made 00:57.480 --> 01:04.760 and is in a data set, is at least open free and more importantly, very viable. 01:04.760 --> 01:09.320 And so very vitality and openness are going to be key to this talk, but also key to the 01:09.320 --> 01:15.320 development and maintenance of power grids and energy systems, notably because we are starting 01:15.320 --> 01:22.120 to see a democratization of our energy system, so from a more centralised to a more 01:22.120 --> 01:28.120 decentralised system with more people, more actors, different actors, whether public or 01:28.120 --> 01:30.560 private. 01:30.560 --> 01:34.960 And this democratisation can also be seen with the data. 01:34.960 --> 01:41.440 And without this open grid data, many, a lot of research, many innovator digital products, 01:41.480 --> 01:46.520 market applications are just simply infeasible because there's nothing else. 01:46.520 --> 01:51.680 And unfortunately, the global energy industry hasn't really come up with a standard global 01:51.680 --> 01:56.800 framework for this type of data, which is why, fortunately, the open data community has 01:56.800 --> 02:01.600 stepped in specifically open treatment. 02:01.600 --> 02:02.600 Thank you. 02:02.600 --> 02:10.120 And we also got a pen strip map, which is a special case in this landscape, because it's 02:10.200 --> 02:15.800 the most applicable database we can use to get data from different countries and most of 02:15.800 --> 02:18.680 all worldwide scale. 02:18.680 --> 02:23.080 I am myself contribute of open treatment, and I've been contributing for 15 years now, mainly 02:23.080 --> 02:31.000 on infrastructure, on utility and mainly on progress, defining not only the ontology 02:31.000 --> 02:35.840 but also contributing in France, Europe, and thanks to map your grid in many countries. 02:35.880 --> 02:44.160 It shows that even if the mapping started years ago, we assume it's 28, it's almost 02:44.160 --> 02:49.160 complete and very detailed thanks to the ontology thanks to a common language, we can 02:49.160 --> 02:55.200 use to make those grid comparable, which is not achievable in the national datasets we 02:55.200 --> 03:01.760 used to work with in many opportunities. 03:01.760 --> 03:08.040 We miss something, it may miss something, we can seek to assess the quality of this digital 03:08.040 --> 03:15.680 command, but from now, open strip map became a very applicable solution to get those data 03:15.680 --> 03:18.600 and that's what we are going to show. 03:18.600 --> 03:26.080 Last year, map your grid initiative started, kicked out with a very basic set of features, 03:26.160 --> 03:34.000 trying to assess quality, trying to encourage contribution in mostly in low-one medium-income 03:34.000 --> 03:40.560 countries, we've mapped so much things we get some details here, but it's already the figures 03:40.560 --> 03:43.840 after one year of contribution are outstanding. 03:43.840 --> 03:50.880 So this is a really interesting solution, we can get into for the next year's to achieve 03:50.880 --> 03:56.160 these things about interoperability. 03:56.160 --> 04:00.120 And so where is and who is using this open grid data? 04:00.120 --> 04:05.880 So from research and academia for end-system modeling to forecasting, network operators, 04:05.880 --> 04:11.160 grid operators, for example, ministries, local authorities for certain studies, electrification 04:11.160 --> 04:17.040 for example in certain developing countries specifically, but also energy companies and all 04:17.040 --> 04:19.880 of this for large-scale end-system planning. 04:19.880 --> 04:26.880 Thanks to Russ Garrett, who's someone in the room, we know that a lot of it is more used 04:26.880 --> 04:32.960 by generation, more than network operators, as well as consultancies and engineering companies 04:32.960 --> 04:38.640 and organizations, but it's also used for small-scale end-system planning and site planning. 04:38.640 --> 04:44.200 So we've actually explained to quite a few private actors who have taught us how reliant 04:44.200 --> 04:48.960 they are and how useful this data is for them for specific site planning, for example, 04:48.960 --> 04:57.920 so the site planning or best budget system and without it it wouldn't really be feasible. 04:57.920 --> 05:04.400 So just to say that the point is that there is already an ecosystem and hopefully it can grow 05:04.400 --> 05:07.640 in industry and academia both. 05:07.640 --> 05:13.320 So a few more concrete examples, for example Microsoft or Google using it for training AI, 05:13.320 --> 05:17.280 then you have also other cases, disaster response, so measured response for like, mid-salsa 05:17.280 --> 05:25.720 content and so whole ecosystem and all of these for example are users, which is fine, 05:25.720 --> 05:31.080 but the goal is what to have users and contributors contributing back and that is possible 05:31.080 --> 05:34.880 and it does exist with energy for example. 05:34.880 --> 05:39.560 Yeah, a few words about this in the in a discoloration in France. 05:39.560 --> 05:45.320 Maybe six years ago we started a discussion about how to get the internal data, an open-street 05:45.360 --> 05:51.960 map data on the same plan to achieve completeness of some crucial sets of the overhead network 05:51.960 --> 05:55.880 from the distribution group because in addition to some of you may have heard before 05:55.880 --> 06:02.240 and this is the largest DSO in France and they run tremendous amount of overhead lines 06:02.240 --> 06:08.160 in major voltage and the Benzreet map is helping them to gather knowledge of the ports, for instance. 06:08.240 --> 06:17.440 We currently run a community with more than 2,200 people that seeking ports in the country 06:17.440 --> 06:23.880 to improve the quality of both of a Benzreet map and the internal data. 06:23.880 --> 06:29.240 So it can be useful for disaster response, for grounding as well. 06:29.240 --> 06:36.560 France is doing great grounding so it's important to know the state of the overhead network. 06:36.560 --> 06:40.440 And that's where Benzreet map is incredibly useful. 06:40.440 --> 06:45.520 That's something we could achieve with a very motivated community. 06:45.520 --> 06:52.520 26 and five years a day and there is a year and 24 hours a day. 06:52.520 --> 06:59.280 It's no only days, no weekends with people contributing from their surroundings or their 06:59.280 --> 07:06.000 trips or whatever from public data. 07:06.680 --> 07:10.320 And so a very key component, as I said, is a very far ability. 07:10.320 --> 07:16.320 So all of this data is basically verified so OSM, Open3M map has strict rules. 07:16.320 --> 07:20.880 So on the left you can see you can use hints so that's a public data open data reports and 07:20.880 --> 07:26.880 use but that is only used as a hint to fall in the mapping which is basically needs to be verified 07:26.880 --> 07:31.000 via open satellite imagery, open street view imagery or ground surveys. 07:31.000 --> 07:32.600 And this is then validated. 07:32.600 --> 07:37.240 So if you're example map you're greeted and the general open street map power community 07:37.240 --> 07:42.600 that's called it, do this manually and so at the end there's this validation and 07:42.600 --> 07:47.880 there's a very high ability, reading increases the quality of what's actually being done. 07:47.880 --> 07:51.760 So it's not a copy paste of like a generated data. 07:51.760 --> 07:57.440 And then importantly as well, this open data is open and it's under an ADBL license which 07:57.520 --> 08:07.760 is also very important and yes, okay, regarding this particular thing about very 08:07.760 --> 08:16.720 flability we can talk about grounded features that can be seen from the usual people. 08:16.720 --> 08:25.360 Open street map actually have a great amount of underground features but we don't want 08:25.360 --> 08:32.720 to map secrets so those features come from a pen data or works of a moments at which the 08:32.720 --> 08:34.640 underground features are visible. 08:34.640 --> 08:39.920 We are not using or even markers, you have some markers but we on the other side we don't want 08:39.920 --> 08:45.040 to map things that could be secrets also and we if we don't have the suitable documentation 08:46.160 --> 08:52.480 or the appropriate information we have in license in particular we won't map it and it's 08:52.560 --> 09:00.000 regarding grounded features it's obviously a local decision from the community that act in the 09:00.000 --> 09:06.240 country that could be difference between countries we have some data in France we don't have 09:06.240 --> 09:13.120 another country is also on so that's something we care in a local context so that's why we you 09:13.120 --> 09:18.640 won't you are not supposed to find any secrets in open street map if you find it in open street map 09:18.720 --> 09:22.480 it should be properly sourced and documented appropriately. 09:25.280 --> 09:32.240 Oh yeah exactly let's try to assess why we are in this situation with 09:32.880 --> 09:39.520 ID difficulties regarding different data sets for countries or so on and why open street map is different. 09:40.400 --> 09:49.200 The first point is for historical reasons the asset management regarding monopolistic centralized 09:49.200 --> 09:54.560 energy system is not an easy thing you have to maintain that a basis you have to 09:56.160 --> 10:01.360 recolse it some works that has been done etc and the operators may have difficulties to do so. 10:02.080 --> 10:07.760 So first of all it's history. Second one as Andrea has explained the energy system 10:08.400 --> 10:17.440 is becoming way more competitive with many more actors that are committed to deploy 10:17.440 --> 10:25.440 decentralized energy system and very different way of producing energy and the second point 10:26.400 --> 10:30.560 make the things even more difficult because as you are many more people you are not 10:30.560 --> 10:37.600 yet able to recolse it and make things on a common plan so that's why open street map 10:37.600 --> 10:45.600 is relevant as well and three third point it's secretly by obscurity it exists actually in 10:45.600 --> 10:52.960 some places but it could disrupt valuable processes at stake here because you will 10:53.040 --> 11:00.560 prevent that's a good people from collaborating and producing really even data because you will 11:01.200 --> 11:06.000 let's say build walls and that's also a process to consider when you are trying to assess 11:06.000 --> 11:13.760 the reasons of the situation we are dealing with. So a little bit more on the security side 11:14.800 --> 11:21.600 so this is the infrastructure is too big to high festival the internet provides a massive amount 11:21.600 --> 11:25.680 of news of reports of details, intricate details of substations, power plants, 11:26.320 --> 11:32.480 transmission lines etc also mapping services so this is more for the location of substations 11:32.480 --> 11:37.200 for example also provide these locations because there are physical places that people need to go to 11:38.480 --> 11:44.880 and so for example on Google Maps as well you even have images of inside of the infrastructure 11:46.560 --> 11:51.360 and also in real life you can see the stuff like it's not hidden or anything 11:51.840 --> 11:55.920 and also more and more on LLMs you can actually find all of this detailed information 11:57.040 --> 12:00.800 not to mention that many countries not necessarily Western actually but 12:01.360 --> 12:07.200 really quite accurate power maps of their grid which also includes other data that certain 12:07.200 --> 12:13.840 countries might find sensitive and others don't and but what about the bad actors 12:15.040 --> 12:20.080 so all of this information has been online for decades and professional bad actors have more 12:20.080 --> 12:25.600 access to data reports about all this infrastructure so what public data actually really does 12:25.600 --> 12:30.800 is reveal certain vulnerabilities that can then would otherwise remain hidden and so 12:32.000 --> 12:37.520 it really helps basically for for this infrastructure and and resilience in terms of security 12:38.880 --> 12:45.360 and it also increased the pleasure to close these vulnerabilities and I'm professional bad actors 12:45.360 --> 12:52.480 as well if a professional bad actor can can harm the infrastructure then then a professional 12:52.480 --> 12:58.720 one can do it in a large scale so it just a little way to illustrate this as well is just for 12:58.720 --> 13:03.680 example the homeland security of the US who currently isn't very popular but they conducted 13:03.680 --> 13:07.840 some assist assessments of risks and benefits of this data see if publishing it 13:09.040 --> 13:13.520 would basically if the benefits would outweigh the risks and from 2022 13:13.600 --> 13:17.600 they basically have been publishing open data of the entire transmission lines 13:17.600 --> 13:22.560 and the grid of their country so seemingly coming to the conclusion that the benefits are 13:22.560 --> 13:30.240 either risks and another actually side note is that 30% more or less of the data from the 13:30.240 --> 13:39.760 homeland security open data comes from open street map so it's quite useful yeah 13:41.760 --> 13:46.480 we could really simply ask the question does everything needs to be open and the answer is 13:46.480 --> 13:52.480 obviously no we have to focus on what deserve to be open as to ensure the right level of 13:52.480 --> 13:59.040 interoperability between actors but these actors can perfectly and they should carry it 13:59.040 --> 14:05.280 keep the secrecy on what deserve to be secret um in open street map we thought about this 14:06.080 --> 14:10.880 point particularly about the features that are difficult to survey I already talked about the 14:10.880 --> 14:17.760 grounded features and these are obviously a case for us so without about solution 14:18.800 --> 14:24.240 an effective solution that is already used by the community is to build logical 14:24.320 --> 14:32.320 relation between assets that can be easily linked by a physical link I will try to be more precise 14:32.320 --> 14:38.080 in some countries you have signage about relation between substation you have colors and 14:38.080 --> 14:43.120 towers you have markers on ground you have some hints that a link exists but you are not able to 14:43.120 --> 14:48.000 create the physical link because you don't know what's the power cable route actually he's 14:48.960 --> 14:53.520 so in open street map we are able to build a relation without involving the exact route 14:53.520 --> 15:00.880 so that's the solution that's a practical solution we could and force to find the right compromise 15:00.880 --> 15:09.920 and find let's say a right and a middle solution to prevent secrets to be revealed just for 15:09.920 --> 15:18.240 sake of building an open map and create the relevant data to be used in coordination with any 15:18.240 --> 15:25.200 other piece of software or private data as well we should note that in open street map we better 15:25.200 --> 15:32.400 at putting identifiers on things instead of describing them all we are not forced to put every 15:32.400 --> 15:37.840 attributes we could find on features with better putting identifiers to link between data sets 15:37.840 --> 15:45.200 especially with the private ones that's our actors and why not most of you could use the 15:45.200 --> 15:50.960 open street map data by using the identifiers we could data in a t-fayers for instance or business 15:50.960 --> 15:56.320 in a t-fayers we have many countries where we add the public reference of assets as to 15:56.320 --> 16:01.840 a low the bridge to be to be built between the private data on open street map data as well without 16:01.840 --> 16:12.160 revealing anything and another key point as well is that this so a lot of the data can be 16:12.160 --> 16:16.960 private a lot of the data of certain t-fayers minestries might not want to share certain stuff 16:16.960 --> 16:21.440 but it doesn't necessarily mean that this is incompatible with open data so there's the whole 16:21.440 --> 16:31.040 idea of interoperability where basically these two can so close data sets etc or from t-fayers for example 16:31.040 --> 16:37.040 can technically benefit from this openness and so for example we we've got a bridge type of a 16:37.040 --> 16:42.080 matcher so basically we would match open street map data with this for example close data set 16:42.800 --> 16:47.600 which might have certain attributes that they don't want to to be open such as capacity 16:48.800 --> 16:56.240 and then to come up with a higher quality closed ideally open but closed a data for that 16:57.600 --> 17:03.760 institution for example so so it doesn't necessarily mean that for example if there's uncertainty or 17:03.760 --> 17:11.920 or scared about let's say using open data don't be because there is there is a technical solution 17:11.920 --> 17:21.040 to this but for this to have a bigger impact and to create a long-art long-lasting ecosystem 17:21.040 --> 17:28.160 really the grid data needs you it needs you in terms of training and services for for mapping 17:28.160 --> 17:33.840 for data integration and implementation which for example matcher grid sort of goes into that 17:34.880 --> 17:39.520 it needs you in terms of funding so I think a lot of lots of us know how difficult it can be 17:39.520 --> 17:45.200 fun get funding for open source projects but funding is obviously needed in terms of yeah training 17:45.200 --> 17:50.880 digital infrastructure as well and it also needs you in terms of empowering communities to do this 17:50.880 --> 17:57.440 because at the end of the day the goal isn't to have a very good 2026 grid it's to have an ecosystem 17:57.920 --> 18:05.680 where the grid is constantly validated constantly mapped and it takes people and time and so that's 18:05.680 --> 18:14.160 also very key component and it needs you in terms of what needs the data users as well to not just 18:14.160 --> 18:19.360 benefit from from from the open data but it also contribute back because if anything they're in a 18:19.360 --> 18:24.480 really good position to do that as they're quite familiar with the data so that can be in terms of 18:24.480 --> 18:34.800 of creating back data code tooling for example and I think yes so yeah inclusion join us join 18:34.800 --> 18:41.200 that for grid join the ecosystem we at matcher grid provides specific services in terms of all of 18:41.200 --> 18:50.640 the data mapping tooling training etc and yep so also come talk to us we've got a few people 18:50.640 --> 18:59.680 from the team here Tobias Francois I think you talked to Max as well Ross and yeah thank you very much for 18:59.680 --> 19:03.680 this time 19:06.560 --> 19:11.840 questions but it's five minutes nice so there is a problem because we in relation you have the 19:11.840 --> 19:20.080 people on this okay okay you will have to choose to home together to make this 19:20.400 --> 19:31.120 we have extra what 19:37.840 --> 19:38.800 there 19:50.080 --> 19:57.080 What's the biggest broker to get people to contribute instead of just using it? 19:57.080 --> 20:04.080 That's a really good question and map your grid actually, I'd great time to think about this question. 20:04.080 --> 20:11.080 Upends grid map as experience in such a process, so we are deploying appropriate documentation. 20:11.080 --> 20:13.080 We are delivering up so tools. 20:13.080 --> 20:20.080 That when people are downloading the data, we try to provide some ends about the quality and how they can improve. 20:20.080 --> 20:25.080 We add this bridge between upends grid map and pipes are software, for instance. 20:25.080 --> 20:30.080 And we plan to inform the consumer when they don't load the data for given country. 20:30.080 --> 20:36.080 Okay, you are able to use this country, but know that we are almost 60% complete and you can contribute. 20:36.080 --> 20:42.080 So that kind of message can be a clear encouragement to get people into the right pipeline. 20:42.080 --> 20:49.080 But I think also the biggest obstacle is first finding who the users are. 20:49.080 --> 20:58.080 Because this is open data and at the end of the day, we've had to go through a lot of research to just find so all this list was done like a year. 20:58.080 --> 21:05.080 Just because first of all, many people don't respect licenses and don't give any attribution. 21:05.080 --> 21:11.080 So you really need to dig in like I've gone through like specific data sets where it was suspiciously matching OSM. 21:11.080 --> 21:14.080 And it clearly was. 21:14.080 --> 21:19.080 And then suddenly on their website it says, oh, OSM attribution. 21:19.080 --> 21:27.080 So I think finding it is the first thing or ideally the data users would see this and then come towards us. 21:27.080 --> 21:32.080 But I think that's another big obstacle is just the whole attribution behind it. 21:32.080 --> 21:40.080 You mentioned there are many people on the ground. 21:40.080 --> 21:44.080 Actually, they're fighting against fire. 21:44.080 --> 21:45.080 It's great. 21:45.080 --> 21:49.080 There are a lot of people on the ground. 21:49.080 --> 21:51.080 How many people are there? 21:51.080 --> 21:53.080 Where are they? 21:53.080 --> 21:55.080 You know, how many people are there? 21:55.080 --> 22:05.080 The question was how many people. 22:05.080 --> 22:18.080 The question was how we know how many people are contributing and how are they actually on the ground. 22:18.080 --> 22:22.080 On the ground, it's difficult for the on the ground. 22:22.080 --> 22:30.080 But map your grid spend grand time to improve the ability to monitor open straight map contribution. 22:30.080 --> 22:38.080 In other talk today, I present the PODOMAS software, which is a generic solution we use to monitor where the data is headed by whom. 22:38.080 --> 22:40.080 So we know how many members there. 22:40.080 --> 22:42.080 That's how I know that within it is. 22:42.080 --> 22:45.080 I have 2,000 people that are seeking for polls. 22:45.080 --> 22:49.080 So we are first of all monitoring the open straight map contribution. 22:49.080 --> 22:55.080 And secondly, for the underground, we can check for the source that is provided with change sets. 22:55.080 --> 23:01.080 So people are directed to states from which source they get the knowledge. 23:01.080 --> 23:05.080 So we can know if it's come from survey, from a imagery, from a pen data, whatever. 23:05.080 --> 23:09.080 So open straight map already provide the framework to get this information. 23:09.080 --> 23:12.080 But we still have to be improved. 23:12.080 --> 23:15.080 But you have the dashboard online, we can give you the link. 23:15.080 --> 23:19.080 But we have the dashboard online that has been updated every day. 23:19.080 --> 23:23.080 And for each country, you get the figures of how many people where they are doing, etc. 23:23.080 --> 23:25.080 It's public. 23:25.080 --> 23:28.080 So I'm part of the members. 23:28.080 --> 23:30.080 I really like the perspective of the stream. 23:30.080 --> 23:33.080 I'm a little scared that I'd like to do that. 23:33.080 --> 23:38.080 I get so in front of some of the media's work. 23:38.080 --> 23:43.080 And I'd like to see the panel and the numbers and we have certain occasions. 23:43.080 --> 23:46.080 This is, I find this street complete. 23:46.080 --> 23:51.080 Yeah, in the car and I get a pop up like, as a panel, is it good? 23:51.080 --> 23:52.080 Is it metal? 23:52.080 --> 23:57.080 Are you working with things like street complete to make sure that the data quality you see? 23:57.080 --> 24:01.080 It's actually been requested by entrepreneurs. 24:01.080 --> 24:04.080 Yeah, so the answer is yes. 24:04.080 --> 24:09.080 And the question was, do we work with software like street complete? 24:09.080 --> 24:14.080 Because as you must know, open street records, very different edit hours. 24:14.080 --> 24:16.080 You can separate it all. 24:16.080 --> 24:19.080 And street complete is one of them when you answer just answer question. 24:19.080 --> 24:20.080 Okay, what's the surface of the road? 24:20.080 --> 24:22.080 What's the material of the poll and so on? 24:22.080 --> 24:33.080 So yes, we are working with them to improve the quests and ask more and more questions in ground context for people who are directly seeing the figures. 24:33.080 --> 24:37.080 Yeah, if you want to level up, you can use other editing so much more powerful. 24:37.080 --> 24:41.080 And we've got some YouTube videos on that if you want to start a kit. 24:41.080 --> 24:42.080 Time to go, guys. 24:42.080 --> 24:43.080 Thank you.