 Time to have such a full house. I'm happy to see and I'm happy for to see our speaker I don't want to talk a bit around the bush beckles are not a nice thing Everyone possibly saw one or experienced one. You don't want to have them and How to actually cause one and how to prevent one? Whatever will be told by our next speaker, but he is Daalheimer Yeah, day four You're planning your trip home. So thank you for coming in so many numbers So this shows that this might be a significant topic for some of you for most of you. So usually a talk about blackout starts with About a book by Mark Ellsberg called blackout. I like this book way better. This is the basis of this novel from the office for Technical analysis It's a scientific analysis of what are mechanisms of blackouts Large large scale power blackouts so After about five days, they say we would have like Situation that is pretty war like here in Germany. So there's a risk there So was it was it woohoo or I don't know but the risk is pretty small, but the impact would be very large All right how how Predictable is our power supply network. So there's a organizations in Europe that Calculate the system average interruption duration index. It's a pretty hard to Pronounce but this is just a number that says within a year. How many minutes? Was there no power available for a consumer And if we have a look at this It's around about 13 minutes. So it's for Germany 2008 to 2013 plotted for this duration duration Renewable energies Where we're like improving a lot. So about 30 percent of our energy comes from renewable energy sources So you can see at these numbers you can't prove You can't you can't prove like renewable energy energy sources Endanger the network But let's stay at this Saidi number. How do we calculate that? So there's paragraph 52 For the producers whenever there's like more than three minutes a blackout of more than three minutes Then they have to tell someone to tell to the Bundesnet's again tour. They have a XMI web service that Receives that number and then saves it to a SQL database and then create a report and Calculate the Saidi number So So I requested this database it took about nine months using the freedom of information act So It was kind of a painful experience for me because the Bundesnet's agency and me we were like of different opinions like who was able to access that content So I thought I thought I thought it should be Freely available, but I only got a subset for now of these data So if we have a look at this the first blackout is from 2008 the last one is from the end of 2013 So this is the basis for my data analysis. So the reasons for blackouts. I just added up here So one thing is they have a problem with utf of utf-8 encoding obviously So there's the curious curious reason for a blackout called please choose It's the last one on the list Let me let me phrase it this way the input validation maybe might have been Improvable in this case So but let's get serious the data is kind of Interesting This is plotted One point is a blackout The y-axis is the power that has been missing times the duration So how large was the blackout x-axis is the time and we can see Like most most stuff is happening on low Network load layers Voltage layer So this is of course, this is this is the others have more redundancies and more failsafe so this is like to be expected So this this looks possible, but I don't really trust this data because the reason for this is the following This is kind of complex on the x-axis. We have the time and the y-axis. We have the cumulated number of blackouts So what I would expect is a line like the red one. So just like continuously Outages like go up go up go up over time and the red one is actually the other line for all the producers but the black lines I Just selected ten rendered randomly ten producers and added them up and Just look at this this value 641 That's a curve. That's looks pretty weird 2008 this producer had like four outages 2009 it was 29 2010 it was 211 57 and so on and what made me pause was this jump in 2010 That's actually a line exactly that that actually changes changes inclination Directly when the years are changing. So this is not something that would be explainable physically more like organizationally So the bundles nets are going to I don't we don't know how this how they score these data But I have some questions for this. So if someone from the agency is Able to explain this to me. I would be very grateful So the official statistics and the official monitoring I think it's kind of curious I Don't want to say it's bullshit, but I have some questions So how does it work? How does the planning work the the reliability or the dependability in our generation network or in the distribution network, there are some criteria and the The most important criteria is the n-1 security is it says if I have n N big devices and one fails the rest has to work and that's necessary because I need to do maintenance to fix a high-volage line Something like that and with these in this Here we have three phases that are transmitted and two Two wires transmit one phase and the other wire can fail and the other one will still work if if there are reserves So So let's take a step back How does it work the power generation system and this is from Wikipedia? And I really like to use that because it is quite complex even if it's a Very a very abstract level. What is there in a in a electricity network? There is the network itself of the grid in different voltage levels high voltage 220 and 380 kilo volts then the high voltage and medium voltage and lower voltage and It branches out into the Households and what we get is low voltage and the highest voltage One tries to reduce the losses on the the line losses by the higher voltage and the whole operation is is very strictly regulated and by the federal agency for a network for electricity gas telecommunications, etc And you have different generators on different levels of the network With the big nuclear power stations and coal fired power stations and they are on the highest level and they are medium Medium level power plants on the medium or high voltage level. They are wind wind generation parks and big photovoltaic or solar power stations and what is new and where the Renewables really make problems in the grid is if that is that you have Generation units on the lowest level on the low voltage level, but this is just a side remark I'm not going to get into details about that and then you have the consumers there are industrial consumers even on higher network levels and but mostly on the municipal networks on the lowest level and and there is also generation on that level So if you want a slide for the workings of the power generation at work and Then it must be this slide the generated power must equal the consumed power there is practically no way to store electricity in this in this on This scale if when I turn on my TV is somewhere more power must be generated. Otherwise, this wouldn't work and The network frequency of 50 Hertz They are the indicator for imbalance for power imbalance and you can Think of it as like this scale if generation and consumption Then it's 50 Hertz and everything is fine. And if you have too much load Over the generation then my network frequency goes down on the other hand If you have if I have too much generation and too little load then the frequency goes up and that is the indicator Which can be used for regulation How much power is required and you measure the network frequency and you know how many do I need to fire more coal and when I started Started getting into this into the into this topic. There were no freely available data Many companies measure that and many service providers measure that for Electricity for the electricity market, but freely available and high resolution There were no data available and I said, okay, I'm going to build this myself It's just a micro microcontroller. It's a my Raspberry Pi and can still be optimized, but for more than a year and a half. I have Data records of different devices and I had to talk about that about this project how that worked in the background and So but for this talk I only use the data that I have recorded from that project so that was that about Network grid the power grid and let's get back to the blackout The scenario that we always Here in the press the hacker attacks a nuclear power plant Is that something special? Is that a problem or what what does the network do in that case? What you can observe that here's a here's a nuclear power plant It's a grundrammingen and on the 25th of 25th of March there was a Scram emergency shutdown. There are three reactor blocks. The first is It's down and the second was not not operated the third was operating and their maintenance caused so cost an outage of the pressurized air supply and there was a Was an emergency shutdown a trip of this nuclear power power plant and how much Was the How much power was lost at that point and how much power did it put out and on a nuclear power plant? There you look at the available cooling power and can I get rid of my excess heat? that's one of the important factors in a nuclear power plant and I looked around with weather data and asked and Okay, they said it was at maximum performance maximum power. It was 1.3 gigawatts at the time of the outage and this is the Frequency and you can see on the x-axis is the time and the y-axis is the frequency and you can see in the beginning It was very close to 50 Hertz as it should be as it is normally and then doing in a very short time This is 19 seconds. There's a very quick drop down to the frequency. It's about 50 millihertz And after that the network regulates that again and if the outage of a single power plant is is more or less noise and there are Events that are much bigger and this is not a big problem And we can look what forces are at work here and I used this frequency measurement and smoothed it a bit and on the smooth data the blue line is the smooth data and made the first differential on of that and third derivative and The gradients the power gradients were quite a big it was five gigawatts per minute peak Peak change and these are huge mechanical stress on the on the power plant and the The operator of this power power plant wasn't happy when when an intern or someone played around with a pressurized air And how does this stabilization work? How could the grid do that? And there are three effects and I Roughly sketch those three effects and in the following I always talk about the UTC Reference case there they start with a relatively low network load of 150 gigawatts and assume the outage of a two of a double block Three gigawatts, maybe a nuclear power station or a bit more completely goes down and there are three effects which were cumulatively and You always have to Imagine those three curves being added First is the blue that is the use of rotational energy and the red one is the Rotational energy plus network self regulation regulatory effect and the black is the primary Regulation that is an actively regulated component and the other two are just physical effects to rotational energy This is the turbine or an X turbine of block to Felix Burke This has a mass of about 190 tons and rotates at 25 rotations per second there's a huge generator axis on that and Generator shaft and this is one device in this power plant in this rotational Motion there's a lot of energy stored That and this is the momentary reserve and what it does is the blue line And this says how fast the frequency Degrades and this turbine rotates with a multiple of the frequency of the network grid all Synchronized across Europe and if some something goes down then all masses Braked and decelerated more or less Unison and this momentary reserve is the inertia of all generators But also on the consumer side are some Inertial masses there are some masses such as pumps or compressors They also contribute to that where I have a rotational machine. They all contribute to that now we look And the addition of all three effects How does that combine if you use this if you change this rotational energy and it is characterized by Nets unloved site and if That is small small I have a steep decline of the initial frequency and If it is higher then I get the more smooth Black one and these are empirical estimates, but it's about in the right range in Europe So the red curve is the network self regulatory effect And I have 1.5 to 2 percent Reduction of percent for every percent of reduction in frequency and If you imagine about an asynchronous motor and if I if I And it uses less energy if I reduce the rotational speed except it is really There's a frequency transformer, then that's not no longer true, but everything that is really synchronous Contributes to this effect There's the same plot the same graph this is the network Coefficient the red line is the initial Then there's the initial dip higher and if it's The coefficient is higher than the initial dip is small And now we get to the first active Compound so far it was only physical effects that are just part of a European network and now we are in the regulatory system in a huge power plant for example a nuclear power plant and There is a mechanism called primary regulation and Because if there is a frequency dip and we are trying to stabilize that and There's a circuit From a circuit diagram from a paper that I can recommend from from Elgard And I recommend that if you want to know more about it This was my source and what I did at one there's a frequency sensor and I look what is the actual network frequency and transmit that to to there is a regulatory system and there's the Speed of the generator and compare the two the frequency and see how that works together And what is my set point and what is my current value and in the end? We open or close a valve and more or less steam is admitted to the generator and that way you can Can you correct that or by the inertia of the rotational turbines and There is a certain time it takes Around 30 seconds delay you have in this system until the primary regulation really works From the regulatory strategy. It's a simple proportional Regulator of feedback control loop. This is from the European Commission How this should be structured and in principle If you have a positive frequency deviation, then the power is reduced and if you have a negative Frequency deviation, then you increase the power. It's quite simple and the the slope of this This Graph is Is determined for each power plant or each energy company? Okay, all right, so back to gun drumming and I Wrote a model that I wanted to that I showed to you right now that creates those plots About the regulatory behavior of these things There's no there's one Important thing that we need to actually run this and this the network load So how much energy was produced in the European Union or was actually consumed in the European Union? So on the x-axis we have the time of the day. So you see those are UTC hours The others is the network load in megawatts and this curve just moves around in For different like in the summer we need less light than in the winter and so on and also it depends on the time of the day Now if you roughly look at the so 25th of March 735 when this was about 405 gigawatts and of consumed energy Yeah, okay So this is my measured data laid over the model the green line is the actually measured network frequency And the other lines you already know If you look at the black line at the top The hockey form stick there. That's pretty Well balanced on one Level like almost before the outage So if I compare my measurements with like I can actually predict my measurements with this with this model what I can't predict is the actual shape of the Measurements because there's other things happen in Europe other than this event So it's impossible to actually predict this curve exactly but like I'm roughly speaking. This is pretty well fitting The model isn't really calibrated Well, but I I would need I would need more more like atomic power plants to quickly shut down to trip there to trip their reactors to actually calibrate this but The reason for this work is to have to calculate measurement data what happens when atomic power plants have outages and Then how can I improve this model in the future? So just so we have a measurement like how how big is how how big is the delay to? How big is the delay to re-establish a stability in the network? How large are these how large a delay? Can we actually survive so what happens if consumers have disturbances? producers sorry So physical effects stabilize the network in the first ten seconds So without physical effects, this would not work because we don't have a regulation that happens that fast then After that the primary regulation takes effect. So if I wanted to cause a blackout, I want I would have to Get a great a huge jump in the network frequency So the primary regulation like would take in like take take effect to last the same happens for Consumer outages. So for example, if I go in a like industries or private homes, whatever Whatever consumes power whenever I just disconnect that from power then the frequency goes up, but basically it's the same effects Good Next up next point here is what about the power lines Or the transnist transmission network so the transmission network kind of have outages as well for example November 4th 2006 about 10 past 10 p.m. The Norwegian pearl this very pretty ship was actually Finished by the worst in Was finished in partner work. They don't have a direct access to the North Sea So they have to transport those ships to the North Sea somehow This is this is made more complicated by the fact that there's a high voltage line power line This is a picture there there was there was these red and white These red and white pylons weren't there back then because it was lower down. So the ship was actually too high so the Shipyard actually called the power energy company and sort of Coco be like actually disable those power lines Yes, Ian did that you can see the red line there. This is roundabout where the line goes There's the Ems goes into the North Sea and they talked about it and they on just Switched off this line And there was a small fuck up there Aon had on his side They plan to to distribute the electrical power over other high voltage lines And on Eons side of one of those lines there was a they assumed that the trip value for Circuit breaker was 3000 Ampers and on the other side on Rve was the same Circuit breaker with 2100 amperes and Aon made a calculation and said well, okay We are at 2500 amperes and it's maybe it's just an imagined value to five or something But it's below our trip value. So it'll be safe and of course this line tripped and Because the situation in the grid just Transferred a lot of power. There was a chain reaction And an outage cascade and every number in this slide is the trip of one high voltage line This was across the Republic Across the country and the consequence was that that the European grid Separated into three areas with different frequencies and here you can say areas such as Spain France had an under-frequency Hamburg had over frequency and Indian in an easterly direction and in the south-easterly direction was under frequency again and Ungood really so and there were three network frequencies in the graph in the plot And That wasn't quite as good and one they tried to fix it the blackout was around 10 past 10 in the evening and around 2234 so 25 minutes later they tried to Reconnect two areas and Didn't really work and it really took nine attempts to physically reconnect the network networks and They discovered quite a lot of effects of Frequency oscillations the frequency went up and the wind turbines in the north Had a different frequency and Turned off and the frequency went down again and the and the wind turbines found the Frequency to be fine and connect it again and well, it's not it's a non-trivial system and it's a highly complex system and there's a lot of lot of laws and regulations and we are talking about Europe and in Italy, they're quite different threshold values and To make another another abstraction here and make it better understandable. I thought about another scenario This is a little Python script which Implements simulates the IEEE 24 reliability test system. So there are Power flows in the American High-voltage network. No idea what it's actually called there, but you have to imagine behind Every one of these little green circuits. There's a network segment. There's a lot of there's a lot that's behind that and in a normal operation For example between six and ten the line Has over capacity and is operated over capacity that happens sometimes and it's just a small over capacity it could also be more and So I shut down the line between 16 and 14 16 14 is still Still supplied and this is the n-1 security and it continues working and if I now Shut on the line between 11 and 14 14 is isolated But there are further effects the line between 11 and 13 doesn't Transmit any energy at all anymore and why or how and where which power plant is and how it was supplied before the You can't see that in this in this example, but if there are non-intuitive connections and Now I trip the line between three and nine and now a lot of stuff happens Now I have between eight and seven An outage there's a good it's sort of in a completely different place in our place and now of between 16 and 19 There's a line which slowly heats up And between one and three there's a line which is operated at 250 percent of nominal capacity And so we assume that will soon go down as well And now we have the case. Well, it's just just an example to make it more understandable and now we have the case that One has two working lines, but cannot be cannot be supplied because of physical effects of the underlying physical effects and and the conclusion for the grid is electricity network is extremely complex and every Every event every action for every switching action will be checked and for the consequences on the net and Usually that works everything goes fine and Such a cruise ship is not delivered every day. So but inherently these nets Are prone to cascade outages or cascade effects and if something goes wrong more can go wrong and the Behaviour of this network is not intuitive. So you can Look at the network plan and make an X and you can understand everything that goes on But but you really have to make a complex simulation All right, this is exacerbated by The increasing transport of energy There's a liberated European energy markets energy is actually being brokerage at the At the So a French people could actually buy Austrian Energy and this let them send the energy across the networks, but the networks actually have to transport this energy. So platform This is from the network transparency platform of the European power producers This is just a random screenshot the data is online. You can have you can have a look at them and this is a situation where The energy is being distributed through the through the true transnet DE from France to this the Switzerland and Austria. So just business as usual this just So this is completely Irrelevant what else is happening in the network of transnet. So there's a brokerage of energy So the possibilities also to to move around these this energy and this and an additional access abation of the networks So when we look at the energy brokerage So this is an event That's just random from my data set This is not something very important or something very Like this brings it to either this is a situation from September 2014 shortly after midnight the frequency dips rapidly It's from like the very top to the very bottom. We're about a hundred sixty or a hundred seventy millihertz So about three atomic power plants worth that just break away. This is something that's very normal It's a brokerage artifact that doesn't have a physical reason behind it Nothing broke down, but it's a consequence of energy brokerage. So I found lots of these events and And just Calculated a median network Frequency across the times of the day. So I took my data and just like calculate a median Frequency and plot of that and this is the graph that comes out there So what I was expected is that it would just result in about 50 Hertz But as you can see the middle the network frequency has structures very visible structures And so if we have a look a closer look at this So this is zoomed in then we'll find every hour. There's an event So in the evening I have to go back, sorry They go down in the evening and in the morning they go up In the midday there's less events But like in the morning and afternoon and the morning and evening this they events are larger So this is a the reason behind this is this energy is brokerage in hours or quarter hours and you see Single products in this market Being traded and the next the next power plant like taking over the power production for this So this is my recipe for a blackout You need a Jump in energy in produce and this has to be quicker than the frequency regulation so the primary existent can't take effect and I have to find a constellation where maybe I can Find a outage cascade in the power network that like may exacerbate this even more So what I wanted sort of this I want to disturb the equilibrium between creation and consumption So concretely what if I wanted to do something I would find a day with lots of wind because Wind power plants are not part of the primary regulation. So this means that if I have a wind Power plant I have less rotating mass and the other power plants and this is Bad, then I want to find a position with the trading Data where lots of energy is exported in different networks for example from France to Great Britain which which are part of the power network and I would try to find a change change between hours like a The time of the day was the loss of irregularity anyway, and then I just need to quick jump in Network energy, so this is my plan of attack So shortly before Christmas Hacker Infiltrated parts of the US power grid Okay, obviously if it was hackers or governments or some Non-transparent organizations whoever that is it's a software system. So it is being attacked somehow So additionally the software systems in the power grid Production it's a monoculture. So off of 800 Produces 300 of them use the ideas high light system, which is a play on words in German so if I am in the highlights product and I find a a A Attack point I can I can also attack the other 299 system. So that's a monoculture smart meter. It's also a With a German smart meters you cannot just Turn off the entire house, but you can control individual devices in the household, but it's maybe easier in the other In the other countries they have Smart meters which are protected even less and you just don't attack the smart meters But the control center of the smart meters and everything that is green There is a smart meter infrastructure which allow Allow the household to be cut off from the network remotely so potentially without Without having looked at the systems in detail Not not in any detail, but this is a computer system and there's a possibility to switch to switch rates and tariffs and then maybe you can Pre-define the times at which those change and maybe you can find some scripts and And influence that and maybe this that's a good point in time and maybe you shut down half of France But it's it's even simpler And I Want to talk about a short term Jump in power in the network and I can just manipulate part of the network I have power lines which transport this kind of power and they are somewhere in the forest And I'm just looking for a nice place and destroy mechanically the infrastructure and if I Do it good and if I just destroy that part of the infrastructure that is the most critical And that is just the homework that I'm not going to do But there is there are military weapons for that they're called graphite bombs they have graphite powder or carbon fibers and they are They are fired upon High-volage lines or power stations and create a short circuit that we used in the second Gulf War they were used in Kosovo and It's really well. It's a mature technology that is available and Yeah, what is my critique? Okay, what is my criticism here in the beginning I said there's this n-1 criterion and one part Goes down and the rest has to continue working, but what happens around that? What does the energy grid do and where are the weak points and this systemic weakness are not I'm not taking into account for that. We have the power brokering we have High load on the network and this creates power jumps and I know that beforehand and the power jumps They cascade effects are not really locked at in in in In the current research and these cascade effects have not been explicitly locked at and there are further effects But I don't have measure data of measured data for that, but there are frequency oscillations and the frequency Frequency oscillates between Portugal and Turkey. Maybe there's a there's an oscillation of the network frequency And if that is really stable, I don't know There are some There are some records in the grid operators, but I haven't found I don't have any Any measuring points outside of Germany, but if you want to do that Just contact me and I'd be very grateful. But what can we do? Dare to allow more decentralization There are studies a study at the from the 40 a And how do we integrate renewable energies into our power distribution system not only electricity, but also heat and but The suggestion is to have autonomous decentralized centers which sells which Exchange power, but our autonomous So we don't have one huge system across Europe, but small-scale structures. What else do we need? we need stabilization on the low voltage level, but the System hasn't been built like that, but we create the energy at the high level and Distributed to the lower level, but right now we also generated the low level and send it upwards and So there's no primary regulation in a photovoltaic rectifier, so And we have to do things there, but there are efforts To get active there and to create new regulation And I'd really like to have autonomous cells in the power grid cities such as Kaiserslautern with the surroundings could Supply its own power and maybe it can exchange power with the neighboring cell But if there's fine weather and everything works, then then that thing can really work autonomously Okay, that's really the end of it and I'd really like to thank Those people who send emails to me before this talk and I heard that my thoughts aren't that wrong in principle, but there are lots of people who Who talk about this and think about this topic and if you want to look at the data if you want to know more about the project, this is called net-zinos and Thank you very much. So the URL is net net-zine dot us and ET Z S I N dot us. Thank you very much You do not have to ask Matthias now questions who Matthias does not want to ask questions, but want to go out Please do that quietly and with as little as possible, turn around a little bottle So once again a very hearty thank you Eighth bottle just kept so questions, please are you ready? Okay, let's go. Thank you very much for the for the presentation Have a question about autonomous small networks if you want to buy Electricity you need to Synchronize the frequency so you know big rectifiers All right, so with your new renewable energy you have this anyway because this happens with these devices anyway But I don't know where you feel where your question goes to So how how can that be done? Is that possible today? So feasible possible we are Recreating our energy system anyway right now It's technically possible But we're doing a lot to integrate photovoltaic power plants anyway right now So the additional effort isn't that much so the biggest Obstacle is probably that the people who have a finished power network They don't want this to change and there's very large lobbies working to To push this in the in the into the opposite direction. Yeah, thank you Okay, would be great great sign of respect Towards those who are interested to leave the room quietly from the internet, please That's okay. The internet would like to know if with distributed and synchronized measurements of the network frequency to find out Where or which power plant has gone down? Well speed of light isn't infinite cool question especially because in the Transformator like collections of transformators. They actually also influence the timing in delays So it's a nice thought but possibly we would have to have a very details map Where the lines are where the transformators are how the lines are being used and so on the data is freely available though So if someone knows how then to do it go check send me an email What would it be like if we if we Attack the metal melting Plants of the industry so that suddenly there's a lot less consumption This is basically the same just like the energy goes down so the if I have like a loss of energy being produced or a loss of energy being confused the curves go curve goes Down so this is basically the same the the only difference is that These 50.2 Hertz problem where wind power plants just shut down this changes though I'm not sure if it's like already being done or if it's like in the future, but this Very fixed point where they shut off. They want to actually remove that from the infrastructure. So just turning off If consumption goes down is is no longer the thing so with wind power plants They will just reduce their power output and with photovoltaic Plants it's the same already. So it's not a it's not a fixed point, but it's a gradient The gradient reduction Okay over there Roughly How much deviation in percent do you need to have a complete blackout of the grid and How long would it take then to? Start the network again to get it going again. All right. Give me a second, please All right This is just a frequencies where something happens in the energy grid. It's a lot of text, but more or less if I'm below 49.5 or above 50.5 0 Hertz then things happen, so Segments are thrown off or and so so a blackout is being produced to stabilize the frequency or Power plants are actually being shut down So this actually is in place already But this means the system would protect itself from a total European blackout So and it would just kick out Hamburg from the from the network. I'm I'm back would just be noise, but But for example, you go to steel melting facilities and just steal their energy There's like special contracts that allow this they like to just like close off their energy for a short time so this means That we we would have to trip this very quickly to actually like so this so sorry If you have a law big load shedding then what do you do against that if against this over frequency? Because it takes a while when you shut it down up until the inertia is gone. Is there a huge? Huge consumer. Do you turn on a huge light bulb or something? So I know actually energy companies that have water cookers that are as much as one megawatts to take part in the energy brokerage so I'm actually actually like they think like I'm guessing actually I'm actually gaining money if I use up energy So I'm actually activating the water cooker But what they think is they would usually reduce the output of power plant This is actually happening in inside the power plants by the primary regulation already. So the frequency actually regulates this Okay for the 50.2 Hertz problem Problems as far as I know this has already been done and but all the all the inverters had to be changed But by wind turbines It will is also implemented and there are these nice linear curves and the regulation Regulatory power with and wind turbines. I just read at MP on is also being implemented So everything like that is coming now and this wasn't a question. That's just an Summon more information. Is it known if it's possible to to get to the exchange the phases To permutate the phases at the stations would be funny. What do you mean permutate? I've thought about if that was possible to with the existing switches to to Switch the phases there in the wrong way Well, I don't want this to exclude this like whether if you just like throwing a spanner into the system That would actually like forge itself then into the system But this would be very small and very local effect. So not in this size From the internet, please. Okay, the internet would like to know if you have data about the so-called Earth hours where people say between Today and between seven and eight everyone turns the light off. I have I have data But it's on the website, but I haven't actually looked at it I just looked if the event detector that I am that I'm actually like running if that if there's something happens during the earth hour and There was no impact that I could measure or discerned, but take the data. Have a look yourself. Don't trust me Okay, in Austria There's diesel and and petrol They create much more power than over the electricity grid and if you try in the next few years, maybe 20% of the Cars to convert them from combustion to electric then I wonder what happens to the electric grid Do you have models or thoughts about that or what would happen? For example, if if a controlled car is suddenly switches from charging to Discharging or vice versa So there's a lot of people who are thinking about exactly this topic and I'm not that deep into it But the nice thing about a battery is that it can actually in a very short duration It can like shed a lot of load or receive a lot of load. So if I am able to like In a large scale manipulate in an energy like a mobile Cars I can actually create such an energy fluctuation as well So there's no difference for me if I attacked smart meter or electric cars over here, please I have another and a question about the Informative freedom of information Request to the federal government. Can can we find the data? All right, so I haven't uploaded the data, but it will be on my github thing, which is linked on by the website and Like the letters exchange isn't finished finalized yet It's on frag den start dot the either documented and not right now But give me a two to three days and then this will be the case Okay, you said five days until civil war and how many candles and ravioli and marty did you stockpile? Nothing at all Candles ravioli and so on doesn't really there's not enough because my heating doesn't work if I don't have any gas I can't can't continue getting I can't can't go to the ATM at the regular bank They wouldn't give me anything so Mike Catastrophe prevention is pretty bad But it's also bad for others here as well. So the internet, please as long as it's still as it's still there Okay, the internet would like to know how well this Network Frick network voltage Changes can be used as a fingerprint or for example Find it out from the hum in the audio signal Alright, so there's two things in my head right now. Alright, so can I Take take the hum of audio and then like recreate like the idea like When was this and by matching like this to the background time of the network? And I think it's pretty as feasible But you read well high resolution measurement data and the other thing that's going through the media every time is like If I distinguish like the light and dark Pattern in like TV shows Then when was this actually to determine when was this actually being watched? I try to have a look at this I can't really understand it I would like to know How the synchronization of the generators works between the different power plants? Sorry So you gotta you gotta just like stop me Yeah, but it's interesting. All right, so on the left side you see the just an image of Like the the power network and the round things that generators and you can just imagine this The L I let L1 line. They are just like coupled through the L1 line And so my network cyanide is just like oscillates over the network And so the generators just synchronize over the network cyanide and if they wouldn't do that Mechanically, there would be lots of stress. I have because I have the electrical force. That's like being Good generated by the generator from the from physical force So if one generates more than the other then it actually drags the other so this is actually like You can imagine like two locomotives that are being coupled with feathers That's how I like how you can imagine this. There's no It's just an effort frequency that does this my second question of before soon we have a Network that is not operating. How long does it take to get into operation or Europe? No one knows this good question though The black state scenarios There's a special sequence how to like restart the network This probably works over special kinds of power plants with the pump storage that works Thank you And that's like how you like you try to like recreate the the landscape of producers If this actually works and how it works and how electric cars behave. No idea We have to accelerate this for a few more questions analogous to network topology on the other side We have this hierarchical networks and on the other side we have mesh networks and you prefer decentralization for the future and looking at so if we have meteorological extremes between Greenland and Iceland and So this argument of decentralization. Do they really take it seriously? So in my experience, it's that the Electricity companies are very very slow to react to input from the outside regardless of where it comes from or why It's all of the reason it comes So I can't say that they that they're like more slow to respond for this exactly exact topic But I think they just want to save their There's the status quo and without like thinking about this problem exactly Okay, so this story about the market the brokering and that's what the power companies do and you could Regulate that in advance and you know in advance that something's coming other systems for that and would that be possible financially Okay, so lots of stuff was happening with those brokerage effects They were way larger a few years ago There's actually a group of people around Professor Fonda and Stuttgart. So and they to do something But there's a big misunderstanding The the power companies are not the ones who actually do the park worker brokerage But people who actually broke brokerage and trades is actually like power plans with your energy supply company, so people actually like Like on the network actually can't take influence on this. This is actually being regulated by the EU This distinction between the rules So there's no exchange here. Okay, two more quickly and Maybe you can talk to him afterwards Okay short question if someone in this project wants to participate in this project How can I find the information and what does it cost? I'd love to go to the websites There's a Tab the project and there's a list of things where I need help right now the thing it it costs because it's a Raspberry Pi it's a kind of expensive, but I think we could like reduce the cost It's just right now. It just works and it just like click it together I could also use people have to have like idea of data analysis and Who know a little about like electrical energy then about measurements stations people People who create hard and software for this. This is all on the website. So it's all all on get up as well Okay last question as a follow-up if I have a dense sensor network you could Measure the the length of the line I have no clue if this is electrically possible in detail, but I love the thought though because it's very nice Actually, it came to me as well. So let's try this All right