# Navigating the Future of Solar with Autonomous Drones & Data Analytics

- Event: NestGen ’24
- Published: March 28, 2024
- Length: 32:00
- Watch: https://nestgen.org/library/watch/I2K0Tse8J9E/

> 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.

## About this session

The session delves into the pivotal role of autonomous drones in the solar industry's quest for efficiency and scalability. With a focus on utility-scale solar, the speakers highlight how drones equipped with AI revolutionize construction, maintenance, and inspection processes. By leveraging resident drones and comprehensive data analytics, companies can optimize resources, enhance productivity, and maximize asset performance. The future lies in harnessing technology to mitigate labor shortages and address the exponential growth of solar installations. This innovative approach promises to redefine industry standards and propel sustainable energy initiatives forward.

To know more, visit:

=====
Subscribe to FlytBase YouTube -

Follow Us On -

LinkedIn -
Facebook -
Twitter -
Instagram -

## Transcript

**[0:20]** hello everyone and welcome to another engaging session at N24 I'm RHA product designer at FlytBase and I'm delighted to welcome you to this session I hope you're enjoying Nan so far please let us know in the chat and do share your feedback at the Nan wall of love in this session we're going to dive deep into how drones are popularly being used in the Solar industry the global push to expand and scale renewable energy capacity is on a rise and solar power is undoubtedly leading the way and as the industry exper expands an increasing number of solar companies are adopting drones to enhance their operations and boost overall efficiency so to shed some more light on this let's welcome feda velli strategy and Business Development lead at above above is the solar industry's leading aerial solar inspection and data analytics company having backgrounds in both Investments and entrepreneur Venture Venture building fed drings a fresh perspective and strategic

**[1:25]** Vision to the company she makes sure above becomes the sun's personal rzi ensuring everything is shining bright and running efficiently just kidding but before I let feda take the stage I'd like to share a few housekeeping rules please type in your questions in the Q&A tab of the session and it'll be answered at the end of the presentation the session is getting recorded and will be published soon there's an ongoing poll during the session so please do participate and last we have an above booth in the Expo session so please make sure you visit it during the n networking and both visit slots and finally if you find the session insightful then please share it on LinkedIn by using the hashtag NJ so before we move on I just want to thank DJ our T title sponsor and sunflower Labs our goal sponsor thank you for supporting us in this Mission with that over to you feda thank you so much Natasha for the kind

**[2:27]** introduction let me quickly share my screen all right so hi everyone today we're going to be talking about how to navigate the future of utility scale solar by using autonomous drones and data analytics I'm feder I work at above and I work on developing new and Innovative products for our clients so before I dive into our work with autonomous drones I wanted to give you a little bit of a background on Who We Are so we are a software Force company that delivers solar plan data and analytics and we do this by using drones to collect this data and we apply AI to transform this data into actionable insights we operate globally we're headquartered in the UK but we really have a global reach with a really capillary Network and we span from the US to Australia to South America and

**[3:28]** Europe of course we work with different types of solar players so from independent power producers utilities EPC operations of maintainance players and asset owners but what is our purpose well at the heart really of above's mission is innovation and pushing the boundaries of the status quo we are involved in numerous projects that are really at The Cutting Edge of our field and because we have this deep domain expertise on solar it really allows us to be Pioneers in finding innovative solutions through Tech to solve current challenges of the industries I'd like to show you a quick video of one of our clients that uses our software and services during the construction phase of an asset it really explains how we leverage drones with the whole aim of driving the digitalization of the solar

**[4:29]** assets using the digital construction management that um that actually above provides what is what are the benefits that you are seeing from using this service first and and foremost benefit I mean it's it's again about the the digital attendance that I mentioned also previously and then it's it's a a very good way of of independent uh uh and very close supervision of the construction companies on site uh you can U measure relative easily both progress of construction but also quality of construction without having to walk along uh the the the newly erected mounting structures every day and and it's it's uh we we we start uh taking data and recording data uh if if you want you can call it the digital life

**[5:35]** cycle record of of the plants from the very beginning from the day the first pile is is red into the ground and and the first modules are mounted the first cables are are connected and uh you can easily also assign tickets to to individual uh observations to to individual people so that the electrician checks if if it seems that that some cables are not uh uh mounted as they should or or if uh some some mechanical structures are not aligned uh then it's maybe another subcontractor that needs to look after that so so it's it's just a very accurate way of of monitoring the progress in construction I'd like to start with a quick poll now what do you think is the ratio of megawatts to human needed for a solar plant

**[6:37]** to operate efficiently or if you want how many megawatts can a single person or engineer be responsible for guys I put the poll in the poll section so you can actually go and vote curious to hear your thoughts and this is a really important question that we need to answer and understand why is important oh all right the maximum votes go to B 27 27 the result the real answer actually is only three megawatts and why is this question so important because exactly this point is going to be one of the biggest challenges and bottlenecks that the industry will face as it scales and the reason why is that the issue with solar is that it's an exponentially

**[7:38]** growing Market which is however fundamentally under supplied in fact it has strong labor shortages which are only forecasted to get worse as demand increases installed solar globally is expected to grow from the current 1.6 terawatts to 8.25 ter WS by two 2030 and this is a fivefold increase in the next six years to put it very simply we basically need to deploy three times more per year compared to what we did last year and this is a huge growth and what this implies going back to people is that if solar now employs around 4.9 million people in six years it will have to employ and find a 20 million people and this is a really big amount in an industry that is already under supplied it really risks to be the biggest constraint and bottleneck for the industry this means that we need to learn how to improve that

**[8:40]** ratio of megawatt per engineer if we want to succeed in this growth we really need to learn how to do more with less and once more to aggravate the situation is that solar companies are expanding globally meaning standardization and quality assurance is really really harder to deliver these plans are getting bigger and it means that it will be very hard for the engineers on site to be everywhere at all times technology is also constantly evolving meaning the complexity of the project has increased along with higher costs and thinner margins the good thing is however that recent Tech advances especially in the Drone space have brought us to an inflection point of what this Tech can do autonomous drones in utility scale solar means that we can now have a resident drone a resident drone that is always on site in its own charging box what we do at above then is

**[9:45]** together with Cutting Edge imaging technology and their proprietary AI we analyze all this data and by doing this we can really revolutionize the construction maintenance and the inspection of solar Farms and this is because we're basically unlocking a new way of doing operations currently whenever we run an inspection we have to send out a very expensive pilot to site however a resident drone breaks this business model because every marginal mission that we run the cost of this extra mission is infinitely small compared to having to send a classic pilot there meaning it will really Le a new way of doing inspection of the type of inspections we do and the type of data that we collect that before we couldn't even consider because it was not financially interesting to do so what this means is that for example we can start flying at really high frequency intervals and we don't have the cost of that pilot we can fly

**[10:49]** on a specific and recurring schedule times we can also FY on triggered based uh or certain activities when certain activities and events are happening in the moment essentially we will be able to do new type of inspections and data collections that were not possible before because of the cost and what this does is it allows us to uncover all those unknown unknowns on a solar plant and have access to really novel data fundamentally if we go back to our question on dilemma which is how do we do more with less and how can we improve that 3 megawatt per one engineer ratio well we really think that autonomous drones are a key element in doing that so firstly a resident drone will support Engineers when on site to increase their productivity and reach just think about it for example an engineer is walking and

**[11:52]** looking at the back sheet of the modules whilst the drone at the same time is looking at the front or it could also have help the engineers do the homework beforehand so collecting all the information needed and identifying the issues before they go to site so that once they do go to site they can be prepared with the equipment they need ready at hand to fix it and with the correct knowledge of what to do and where to go what the resident drone allows is effectively as we said cheaper frequent inspections and this is because every marginal extra flight taken has a residual cost compared to having to dispatch a pilot this means that we can really increase the frequency of monitoring of the asset and this in the long term can really lead to a reduction of the energy you'll loss and the prevention of that happening because we can have high and quick response times to all the issues happening and finally because we can access now all this wealth of data we can

**[12:57]** really start to tap into that big data and Predictive Analytics to really maximize the performance of an asset and increase the lifespan so by putting all these pieces together we're developing an autonomous drone service that allows our players to win in an undersupplied market and this service is fundamentally made up of four steps first off is the creation of a digital twin of the plant here all components and their electrical hierarchies are defined and put into our software second thanks to the knowledge of this digital twin which is a digitalization of the asset we can really design all these missions to perform very specific tasks around certain components and areas thirdly there is the deployment of the resident drone to collect all the data that

**[13:58]** we need unfor L the most important step is all the data that we collect is reported back and pinpointed in the digital twin of the plant under software and we analyze all the data that comes in without the analysis our clients really risk of being overwhelmed by the amount of data that they can get what we really want to do is distill all the noise that comes in and really provide them with only actionable insights about what is actually going on as above we can operate throughout all the life cycle of the plant so from the development phases to construction to the commissioning and the operations of an asset of course the earlier you get in as Robin was saying in the video and start collecting the data in the digital twin the more complete data set you will have to manage the plant more eff more efficiently you have more data you can probably set

**[14:59]** set it better you can handle insurance claims and so on so what are some of the examples and activities that we think that the resident drone could do what data can we analyze and what insights are we getting out of it one application is definitely construction management and this is about doing frequent inspections to monitor the progression of the works and monitor really the quality of what is being built in construction management specifically we track the progress of mechanical work electrical and infrastructure items we can also help for example spot early on any area that might cause significant delays and major deviations from the initial design or cat drawing of the plant that might interfere with the yield of the asset that you're building here for example you see a solar plant during its construction phase

**[16:02]** on the left there are the tree lines the cable trenches and the module tables which are shown in a cad layer which is the colored lines and these are superimposed over a visual orth the Mosaic of the side on the right instead you have the same image taken a month later and you will see if you look closely that all the modules tables have been constructed and are shown in the updated visual mosaic visual orth the Mosaic another product that we have is aerial thermographic inspections to really detect anomalies on solar PV modules examples of what we might see are on the bottom right and we process that data with a machine based learning algorithm that allows us to understand what issues there are and what really the root causes of the those issues are with an an autonomous drone what we can do is run thermography a lot more frequently

**[17:07]** we can run missions on specific areas instead of doing it on the whole side we can also do them based on scada systems which are basically uh which are monitoring and control mechanisms that monitor the heartbeat if you want of the solar plant and we can use those signals to trigger the Drone and check out what is happening the output of this thermographic inspection is an identification of the issues on the digital twin in this example for example uh we have identified a hot spot caused by arboreal shading other possible autonomous missions and products that we're exploring are centered around allowing these really highly qualified and expensive Engineers to focus their limited time on more high value added activities whilst letting the

**[18:08]** drones do the time consuming and low valued ones a good example of this is conducting a security fence sweep every morning you really don't need an engineer to say whether a fence is still there working or not our objective is the one to report the inspection with the ratings on the conditions of the items that we're looking at similarly we could use the res the resident drone for specific visual missions in the case here you see that we run a mission to check on a beehives condition and this is just an example of various Vision inspections that the engineers need to do or personnels needs to do which don't actually require their highly value their their their skills right their time is very limited and so stretched we're really designing New Missions that allows them to focus and prioritize on what actually matters

**[19:12]** most the whole idea is that the resident drone unlocks a new array of visual inspections like checking the fence or the Beehive or helping with the green keeping or after modules or before the modules have been cleaned before the arrival of the autonomous drones all these missions were not possible because of the high cost of having to send there a pilot what we aim to do is report all this information back in our digital twin in our platform to really pinpoint as you see the exact location and also time intervals and how this changes throughout time you with annotations of what actually is happening what we're really creating is is a life Dynamic passport of the asset itself to conclude as we have seen the issue with solar is that it's an

**[20:14]** exponentially growing Market which is however fundamentally under supplied with strong labor shortages the aim of our solution is really the one of optimizing Resources by reducing labor cost and improving that megawatt to human ratio we want to increase the energy yield of a plant by providing real time data driven insights and we want to bring together all the data into the digital twin to really provide a comprehensible and consistent data set that then allows us to tap into that Predictive Analytics to really maximize the performance of an asset and increase its longevity and we can do this now because the autonomous drone allows us to collect a lot more data on the con on the conditions of a solar plant that were not financially accessible before we can really uncover those unknown

**[21:17]** unknowns thanks a lot for listening please feel free to reach out if you want to chat some more in this topic and I'd be really happy to take some questions now yeah thanks petria for sharing these insights I'm sure it it'll be much beneficial to our audience who are in the solar industry uh so yeah right uh we would now like to open the house for a round of Q&A so everyone you can go to the Q&A Tab and you can ask away right so we do have a few questions right the first one says What proportion of the high frequency faults at the solar Farms can be detected through above's approach as drones can only collect data from the sky is that a significant drawback of that approach so how do you solve this for inspecting electronic components like laid out wires

**[22:17]** Etc yes thank you for the for the for the question um so our solution as I was mentioning in the construction management what we really aim to do is that we the digital twin that we have really shows all the components and also the electrical hierarchies down to string level monitoring so we can show throughout the cad designs what the deviations have been and we really focus on measuring deviations that can really let's say impact on the quality of what is being built so what is really important for us is to track the progress uh um of the work that has been done uh on a side and as I was saying uh what we do now is that we send Pilots to site right but the inspections only take place at certain intervals but with a resonant drone what this can now allow is that you can do it a lot more frequently so you don't risk in missing out important steps in between

**[23:21]** because you can have much more real- time information coming in and you don't risk missing out any important steps uh during the construction um during the construction uh project right thanks so we have another one how are you going to understand what new type of activities or missions should be done on a solar site yes so uh this is a interesting question we what we have done is we've just launched a an autonomous drone lab uh as part of our Innovative projects and what this is is that we've brought together let's say some of our key most important clients that really want to push the boundaries and that are facing these these issues that we've just talked about about Labor shortages and so on and what we're doing is we're running multiple workshops with them uh both virtually and we're also

**[24:22]** going to be deploying them uh on site and conducting experiments and what we're doing is really understanding what are your key challenges when building a plant when operating a plant and how can we solve those issues through using autonomous drones and big data analytics because as I was saying before a lot of this is all new what has been happening in solar so far has been very limited because of the cost Associated to doing all these new activities but the the autonomous drone there is so much that can be done so many so much data that we can really get out so what we do is we we run workshops we really dig down on root causes and issues that our clients have we running an asset and we then what we do is that we distill them down on new type of products that can come out of this and once we have identified some Key Products we're going to run beta trials on all of our clients s

**[25:26]** um and this is how we're approaching let's say building MVPs as quickly as possible because as we said we really want to be the pioneers and the first ones to really push the boundaries of what is happening great so we have one we're seeing a surge in residential solar rooftop installations in the country will the offering also help in catering to their needs additionally do the drones really clean the panels yeah this is I mean this is a good point definitely residential solar is a is a very big uh is a very big big Market um right now what we're doing is focusing on utility skill solar and this is because the the technology that is out there and the hardware the Drone in the box is quite big and it is an expensive piece of of Hardware so you really need to look at that pigb period the hardware is placed on a side and because it stays there maybe from

**[26:26]** the start of the side throughout its whole operational life you can really see that the payback is going to happen on residential the hard thing is that you might have to move it uh you don't know where to place it there are certain like security uh considerations to take into place so we're not for now applying this to uh residential uh Solar Solutions and we're only focusing on utility scale because it's it's one of the the biggest constraints that these these plants have they're growing so quickly and being so big that the engineers themselves cannot physically be everywhere and everywhere at the same time uh do drones really clean the panels uh so our drones don't um I've seen Solutions uh out there um there are some Innovative startups that are trying to do that with water and cleaning the panels uh personally we haven't looked uh into to that as I said we are

**[27:30]** a software first company so what we really want to do is allow our clients to have as much varied data but insightful data that tells them information about what to do what to prioritize and when and then they can act upon it so we leave the operations to to other players in the in the market okay and uh there's another one what are the type of missions that you think are going to emerge and be important yes so the missions that are going to emerge I think are going to be really centered around what is an engineer doing that he shouldn't be doing because it doesn't let's say fit in his highly qualified skills that they're Highly Educated people and they're also expensive resources for the company themselves so for example doing a fan sweep which has to be done frequently is really not adding any value and it's

**[28:32]** wasting this person's time so I really think that a lot of the ideas that we've seeing out there is doing those repetitive tasks at high frequency intervals that really do not add any value uh and also it's a lot about preparing beforehand the engineers so missions that will allow the engineer once they go on site Maybe reduce the times they go to site because they can bundle in all together a lot of let's say activities they need to do in one day because they can go there prepared with all the information that they have and all the knowledge on how to fix it and and and where especially because these assets are huge but a digital twin can really help them in being a lot more productive when when outside good okay let's take one last question um what do you think is going to be the most valuable part of your new autonomous

**[29:36]** solution so definitely the most important part is number four so the analysis of the collected data this is because as I was saying we can do a lot more inspections with at higher frequency right but this means that our clients will be bombarded by images bombarded by data and they're already scarc in their time and how they manage these asset that they will really not have time to go through all this so we've had our clients being like we really want this but we don't know what to do with it it's too much to handle so what we're going to do is analyze this data apply AI to all this imagery to tell them exactly what is it saying for example if we do a fence sweep there's no use showing them 100 pictures every day of this fence we're just going to produce a report using AI that says there's no new changes the fence is still the same exactly as the day before so we really want to delete all this noise around all this huge amounts of data that

**[30:42]** can come in to make their life easier this is essentially our job and the the highest value out of our solution is really getting all these images because if not they're just flying images of a of a plant and telling them what they are saying in a very Sy quick and immediate way that they can react to okay thanks feda right so I think we're done with the Q&A uh feda do you have would you like to share any closing thoughts uh no it was uh it was great to be here thanks a lot for inviting us it it's a great event and yeah please feel free to reach out if you want to continue chatting I'd be happy I'd be happy to and other than that hope you have a lovely weekend right so um moving on I just wanted to tell everybody that we have launched the nesin community an exclusive invite only Community for all the Drone autonomy enthusiasts so please find the form in the chat window and you can join the wait

**[31:45]** list so that's it thanks everyone for joining us today hope you're having a great time at nesg that being said we have multiple upcoming sessions so please go to the sessions Tab and stay tuned see you there bye-bye