# Monitoring Hydropower Plants in Swiss Mountains with Dji Dock 2

- Event: NestGen ’25
- Published: March 7, 2025
- Length: 32:30
- Watch: https://nestgen.org/library/watch/U13oihqBK4g/

> The transcript of a recorded NestGen session, for reading or for use with an AI tool. It comes from the video's
> captions; automatic captions can mishear names and terms.

## Transcript

**[0:20]** hello everyone and welcome to NES gen 2025 my name is Neil and I'm from the solutions engineering team at FlytBase and it's great to have you all here today so let me start by saying this imagine that you're Soaring Over the stunning Swiss mountains and high up above Cutting Edge drones are hard at work inspecting reservoirs checking embankments and ensuring security and smooth operations for power plants and substations well that's the kind of success story and case study we're diving into today before we get started I can see a lot of people have joined the session so please let us know in the chat section section where all are you coming from what excites you the most about this session so as you can see today's session is titled monitoring Hydro power plans in Swiss mountains with DJ dog 2 and today we are going to explore how Expo group is using DJI dog2 as a Cornerstone in their Innovative Hydro power monitoring strategy

**[1:25]** over the next few minutes we'll discuss the challenges in Today's Energy sector the impact of you know docking stations for autonomous drone operations and how Expo is Paving the way for a sustainable future in renewable energy but before we jump into the heart of today's session I'd like to take a moment and set some you know housekeeping rules so if you're digging what you're hearing on this session go ahead and drop a LinkedIn post tag FlytBase and use the hashtag nesg your engagement fuels our community and also helps us keep the momentum going and there's also some inside scoop so top three posts will score exclusive n gen ampers so make sure you get posting because trust me like the hampers are really good in addition to this uh we also have a Q&A corner so on the top right you'll be able to ask all the questions that you want to ask to our speakers and to make the session more interactive we also going to have some

**[2:28]** interactive polls last but not the least a quick word of thanks to our sponsors who make events like nesg especially the fourth edition um you know a success so we are grateful for their support and having said that let's get right into it uh I am excited to learn more about ao's Innovative approach to Hydro power monitoring and with that uh it's time for us to introduce our speakers for today so first up is klous uh Lighty so klous is a project manager at axo AO group and he has over 25 years of experience uh with G power alstom power and clous joined Expo in 2021 he's been a driving force in merging technology and traditional swis engineering to transform hydrop power operations so hi clous good morning really glad that you were able to join us here today

**[3:39]** yeah CL it looks like you're not audible can you just check your audio once while clous ising yeah yes okay very good not good so yeah welcome to n CL and we hello here audible loud and clear all right so joining clous is [Music] also you would probably need

**[4:57]** to yeah all right and I'll get started with the introductions so along with clous we also have Anna so Anna is a traine in digital Innovation at Expo group and she recently completed her Masters in mechanical engineering with a focus on at ETA focus on energy at eth Zurich and during her studies she was a part of a team that developed a four passenger electric airplane called eastling how cool is that and since October an I has been leading the proof of concept for deploying DJI dog 2 at a Hydro power plant in the Swiss mountains her fresh perspective and Technical expertise at tremendous value at Expos digital initiatives so hi CLA and Anna first of all thank you for taking the time to join us at nesen and show the world what's cooking at Expo and we're all super excited to have you here and can't bear for you to get started and share your insights on how you're getting cutting Ed Solutions

**[6:00]** and Technologies are shaping the Hydro power monitoring you know uh industry and to our audience please buckle up for some really interesting insights uh with that Anon clous I hand it over to you so I'll stop sharing my screen and you can get started uh clous Hannah I believe clous is not audible sorry guys we facing some technical difficulties in the meantime let's let's keep the session engaged let us [Music]

**[7:03]** know can can you hear me now can you hear me okay I'll use my audio no problem good so yeah we're sitting next to each other so we try now the other microphone then so so yeah thank you for the kind introduction and good morning afternoon or evening to everyone again um it's a pleasure for us to here to show at next gen how we are redefining hydrop power plant operation with drones so we just have been introduced so my colleague Anna and I will now share how axo is leveraging drones in the swi Alps and let's dive right into it first I want to give you a brief overview of axo axo is Switzerland largest sustainable energy producer we have over A Century of experience in operating power plants so we actually had the first hydrop Power Plant within Switzerland and uh but even we are

**[8:07]** that old we are not too old to focus on Innovation and efficiency the Swiss mountains and what I show you next digitalization so in 2019 we had the first digital Hydro power plant in Switzerland it was called under a project we called Hydro 4.0 and there our goal was to create the Hydro power plant of the future currently we're rolling this out to all our Fleet and this digitalization is based on on four main pillars one is infrastructure so we ensure that we have data reception all over the power plant if it's in the very far out in the mountains or deep down in a tunnel or a dam so we have communication infrastructure all over the place then we're concentrating on Workforce so that

**[9:13]** people go digital they don't have paper to do their checklist we have they need to do any work on it it can be reached by QR codes on the equipment analytics it's a big important part where we get real-time data for dat we going to do today a bit deeper deeper is Robotics and here we're going to show you how we use drones in our power plants traditionally our inspections are um sending people buy food into hard environment so it's as you can see here snowy ter and and quite tough environment where you have to go this was time consuming it's risky and also inefficient if you call now what we're doing we're doing it with drones so people can conduct the

**[10:17]** inspections much faster safer and also with real-time data so we have data where we can rely on also in the future as you all know grone flying is very easy but there's a caveat to it if you want to integrate this into about health and safety of your of your Pilots you need to ensure that the pilots are train there are a lot of regulations you have to follow there are permits you require you have no fly zones or if you want to fly BV loss it can be quite difficult also which drones are the right ones where is the limitation of the Drone how can you follow up with Technical Innovations and then also no network you can imagine on our side we have quite often only a few pilots on a power plant maybe two or three pilots and they're also part-time Pilots so they don't have really a network where they can change and share the experience so we as a company decided we

**[11:21]** want to centralize a big portion of this backup background support and here comes our service what we call hydon into play so we take care our regulations we get the permits we [Music] see on procuring the drones which drones is the right one which technology should be used for which purpose we have a big training portfolio where we really train based on the needs of the pilots on the sides Tech technology Anna will give you deeper the inside what we do training with on on flight routes or if drones are crashed we're helping we're helping the people on site and then also the community where we have a big Community regular Community meetings where we update what's going on in the community so our

**[12:27]** mission is quite simple so you fly and and we keep you safe so we provide the tools knowledge and support needed for safe and reliable drone operations so we have more than 120 Pilots which Trust on us so these are not only Hydro power plant um Pilots but also on nuclear and a grid division so axo has a bigger Division and also we support external not axo power plants but we also don't only support Pilots we also support organizations on proof of Concepts and as you now will see hydrop Power drone to automated drone flights with a drone in the Box solution and Anna will take over now and show you how we approach this change thank you CL it's I will now present our proof of concept drawn in a box the

**[13:28]** goal of this proof of concept was to gain hands-on experience with drone in the Box Solutions at Hydro power plants so we decided to directly install a dock at a Hydro power plant throughout this winter the dock was placed um next to a water reservoir um in grunon in Switzerland and this water reservoir is about half a car ride away from the main power plant Central in the map here from there the dock and the Drone can monitor this Valley which is shaded in red and specifically it can inspect different water catchments which are marked in blue and these water catchments they need regular inspection already Additionally the the Drone can also inspect other items such as snow Stakes which is relevant in Winter then during this winter now the hydrop power power plant already fully operated this do right on a regular

**[14:32]** basis Technologic technological wise we installed a doc 2 and a mat 3dt from DGI and for this proof of concept we are using Flight Hub 2 but we are also um evaluating other softwares um in the near future in this video you can get a good impression of what the ter is just off and is off to inspect one of the surrounding water catchments you might have seen that while we were flying over the lake um it was very fast but this is because the video is in an increased speed of up to 10 times the normal speed here we've reached the water catchment and we are now taking different pictures from different angles and once that is completed the Drone flies back to its home to its dock

**[15:32]** in general we want to evaluate four different criteria for one it's the usability then second it's the techn technological capabilities and limitations third it's the regulatory aspects and fourth it's the economical feasibility so far we've been able to um fully evaluate two out of these so the usability and the regulatory aspect as we don't know yet which software we would like to proceed for example um but now I will quickly go into the usability and the regulatory aspects but then really focus on the technological challenges that we are still facing forther usability the biggest use comes from Saving Time intensive and tedious travel distance this is mostly this half

**[16:33]** War this then goes hand inand with the second point which is to save important time in case of an event this event can be an emergency for example an avalanche where if the dock isn't covered by the Avalanche we can get almost instantly first insights of the situation using the dock this point also goes hand in hand with the third one which is to increase work safety so in case of such an emergency a worker no longer has to reach the water reservoir or additionally also generally in harsh weather climates the worker does not need to go there and the fourth point is we can have an improved datab basis because we get regular images and video material from the dog for the regulatory aspects we follow here in Switzerland we follow European laws of Aviation and we have created our own specific operations risk assessment and then applied and obtained a location specific permit from the federal office of civil

**[17:35]** aviation for our and we had to guarantee that we can monitor 50% of the participants so this was a technological challenge as well so that's where I'll continue now what are the technological challenges from our use case for one it's how can we guarantee that we monitor the the third point is we have some harsh climate um in winter so is the dock really winter prooof and then fourth do all functionalities work properly as for example the return to home functionality for the air traffic control we found a really simple plug andplay solution where we mounted a flarm sender onto our drone and then a flarm antenna at our ground station as you can see here on the picture on the left and this data then

**[18:35]** automatically flows to Safe Sky which you can see on the right safe Sky it's a platform designed to enhance air traffic awareness and safety for pilots and it is commonly used for um by helicopters and paragliders these are our biggest um air traffic participants in our Valley so with this solution here we can ensure that we can detect other flight participants as we constantly monitor safe Sky while on a mission and we can also guarantee that other participants that use safe Sky they can detect us another oh um another question was um whether we have reception and as the furthest point that we have to reach is over 4 almost almost in this mountainous area so we rely on mobile data reception luckily there's a

**[19:37]** swisscom mobile antenna which is located right at the top of the lake and this guarantees us a good St mobile data reception and we were able to test that the reception it holds until the furthest point that is away but we also realized that at this point um we are getting some gaps so it's really this is almost the limit then our one of our biggest challenges it comes from the weather as you might have already seen in the video we have snow in the winter and also here this can be seen in these two pictures up here which were taken from the camera of the dock on the left hand side you can see how snow accumulates on the dock cover and on the right hand side you can't really see anything because there is so much snow but so far we were able to evaluate that if you place the dock on a p pedestral which has the same dimensions the dock is O is able to fully open as it's no no

**[20:38]** longer surrounded by snow however as you can see on the bottom right we have still some challenges that the snow would fall into the dock as there's lock remotely it is not ideal that a worker would have to come and remove the snow currently we're still looking for for Solutions we are evaluating heating mats which could um remove the snow or include some metal sheet barriers on top of the dock which could prohibit the snow to fall into the dock but here we would really like to reach out to the community if you have heard of any solutions of how to make the dock snow prooof we'd be very happy if you would reach out to us I think this is a very um you know our last technological challenge it comes from the return to home functionality during our missions the Drone it can increase in altitude with over 120 M with respect to the dock and if at such a high point an

**[21:44]** return to home is initiated for example due to a signal loss the Drone would horizontally fly back um to where the dock is this poses uh regulatory difficulties as it is illegal to fly higher than 120 m above ground in Europe currently once the signal is back the pilot can take manual control and then descent the Drone in order to avoid reaching such high high altitudes however we would we would um find it beneficial if there was a functionality where the Drone can is a safe and pre-programmed route to summarize um we have now we've solved the um issue of air traffic control and found a good technical solution there we verified that there is re reception throughout the entire Valley for the weather we are still looking for solutions to make the dock fully

**[22:49]** snowproof and we would uh like to find a new return to home functionality implemented in in a software Zoom for an Outlook so first we started our proof of concept at a Hydro power plant back in October of 2024 currently we are finalizing our evaluation so that we can make a final decision on whether to proceed with these drone in a box Solutions at hydrop power plants this would be in March of 2025 and if we do proceed uh then we would like to look into um alternatives for the doc software here focusing mostly on the user friendly less and the data security we would also investigate into other proof of Concepts such like a drone in a box solution at a nuclear power plant which comes with new um criterias and also a proof of concept in the field of grd here at Expo but we are open to any kind of proof of Concepts um throughout

**[23:51]** Switzerland thank you very much I will now hand something when when you see on your on your end that you want to have a solid proof of concept if drone or drone in the box is useful this is for sure something we have now quite a big expertise in and um so we are now open to to start the questions and and back to Neil great thanks Anna and CLA for you know such an in insightful session you know giving us a complete R down on what all you have implemented the challenges that you've solved and I have a lot of questions that I want to ask but give the audience so let's pick one question from the chat um we still muted on our loud speakers so yeah you're Audible and yeah you're

**[24:54]** both Audible so we'll go with the first question which is uh is there a cost saving with the B method oh could you please repeat the question the um sound just came back right now sure so the this question is from Simon and Simon is asking is there a cost saving with BB loss method of inspection compared to the traditional methods so yeah this is a clear Yes actually compared to the traditional methods the distance are so far so we are really within yeah I think we don't have any proper use case from the traditional inspections where we can't uh where we can fly Voss so we actually need need um yeah we have BV loss flight all over the place then of course if you have like a house roof you can still see the Drone but in the environment

**[25:54]** we're working for we're very after a couple of hundred meters distance um what we want to save boss is a requirement it's a it's a must otherwise there's no use case for us within the Drone all right understood so we'll go to the next question which is from Mario so Mario is asking you know on one of the challenges that you have been facing when it comes to the weather so speaking of harsh weather conditions what are your constraints when having bad weather conditions from a regular and a technical standpoint here for one we follow the technical requirements also provided by by DJI which we also placed in our um Zora which we handed in to the um to the federal office of civil aviation and these are yeah provided by DJI we have a wind speed um Mark we have for icing conditions we have conditions

**[26:57]** and temperature yeah and these we follow all right so I'm assuming you have your own internal Sops that you follow uh obviously from you know it's a combination of the uh Sops that the doc manufacturer or the software provider has given and then you have a you know your own custom Sops as well so it's kind of a hybrid model we really got difficulties to to get we only got the last piece of the question sorry yeah so I I was just summing it up no worries uh I'll go ahead and ask a few other questions so uh one question is like when it comes to the data processing workflows for data worklow that you implemented during the POC the images that you're capturing uh

**[27:58]** let's say uh in the water catchments you're looking for something right so are you using a tool to analyze or is that something that you've been doing manually yeah currently we are um on the proof of concept it's more visual inspection and then the the the pictures are taken and saved but we have for our drone operation itself so for the handheld drone operations we have a tool where we automatically transfer inspection insides what is called system so where we transfer the relevant pictures into the cloud and then either on the grid part we have an AI which then detects which which component we are seen and then um so it's already mapped on on the hydro part it's more because it's so different areas what we see you can then have on the same tool we have kind of marking tool where you can take what you are seeing if

**[29:00]** it's critical if you just want to do there there not so many things where we say it's um you can use AI to detect what you're seeing so it could be lens lights it could be avalanches and so on and most of the information is okay how much material do I read to remove away and and all this kind of things so this is very difficult with with um with AI it's more for preparing the actual work that we have all this information but we have a all right that's you know thanks again for I think it's really important for the audience to also know that the data that you're capturing you know how are you actually analyzing it and uh one last question

**[30:01]** before you know I think we're almost up on time so any best practices that you can share from your own Po's let's say if somebody is out there who wants to start you know their PES so what are the best practices that you would recommend from you know start to end uh that let's say an end user or a drone solution provider should follow I think the main best practice is really go inside and test it where the people are and how the people are using it I think it's it's very difficult if you have a technical organization or people who are not really in the field and know how people are using it and what they it don't go there and tell them how they should use it see observe how they use it and then you can start okay is it possible to use it differently but that's the focus Point really go on onite or go to the customer themselves and and let them

**[31:05]** decide what's important for them and don't tell them what's what you feel it's important for them totally agree CL you know I also believe that the more you are connected to the problem statement and you know the more validation from a technical standpoint you do you'll be able to have a good POC with some good results and finally take things into production so yeah so I think we're almost up on time we we are up on time clous and and the audience but uh again thank you so much for sharing all of your insights I'm sure the audience you know found it really valuable I did for sure and uh Mario Simon if you have any more questions uh you can see you know Anna like the email ID of Expo so you can reach out to them directly and take things forward so um I think it's time to wrap it up thanks Anna CLA for your time uh we really appreciate it and thanks for you know showing the world about your POC and what the

**[32:11]** future holds for the renewable sector energy perfect thank you thanks as well thank you bye everyone goodbye