Episode Transcript
[00:00:10] Speaker A: Welcome to the EU Energy Projects podcast, a podcast series from Enlit and friends focusing on the clean energy transition for the European Union and the EU Commission funded energy projects that will help us achieve it. My name is Areti Daradimo. I am the editor of the EU Energy Projects podcast and your host.
The crucial role of solar power in the energy transition requires no introduction, but the ever increasing scale of solar pv operations has thrown up a range of challenges. Today, we're diving into the world of large scale solar energy with the Talos project. Talos is developing cutting edge robotics and artificial intelligence to automate operations and maintenance tasks in photovoltaic plants. These sprawling installations, often located in remote areas, face challenges like finding skilled labor and ensuring worker safety. In greek mythology, Talos was a giant bronze automaton made to protect Crete from pirates and invaders. He circled the island's shores three times daily.
It seems like an appropriate name for the robots being developed to protect the pv panels.
My co host, Ross Hasty, spoke to senior project manager Daniel Albuquerque in Lisbon. To find out more, we'll hear about the unique challenges of different environments, including dealing with bird droppings on floating solar panels.
Stay tuned for a conversation about the future of solar power and the robots that will help us harness it.
[00:01:59] Speaker B: Daniel, welcome to the EU Energy Projects podcast. We're chatting about the Talos project.
Why don't you give us a quick overview of the project's objectives?
[00:02:11] Speaker C: Hi Ross, and thanks for having me. It's a pleasure to be here with you today and talk a little bit more about talos. So I think before the objectives, I usually like to start by giving a little bit of the motivation that is behind the project. So basically, our main drivers are everything there is to do with the size of these pv plants. So we're talking about photovoltaic plants that are growing quite a lot in the last decade, especially in the last two years, we believe that the size of these pv plants grow by four times. So it's quite, quite a lot of area that these pv plants now cover. And at the same time, the fact that these parks are located in remote locations. So due to some land costs and even some land competition, that exists also, we have been feeling this for quite a while now, a shortage of skilled workforce. This is something that is already happening, and the trend is only to increase. So this is something that also worries us and then the inner dangers that exist of working in such a place. There is a power plant. So giving this motivation, we have this objective, which is to develop cutting edge robotics and also some digital tools that can perform operation and maintenance tasks in pv plants in a completely autonomous way and without human intervention. In Talus, this, in fact, is a reality that spans across three different pv scenarios. So we'll have the land based, which is the conventional pv that we see across fields in the countries, and then we have the floating pv. These are pv panels that are installed above the water bodies of lakes, or even water reservoirs of dams. And then we'll have another scenario, which is Agri PV or agrivoltaics, which is this duality between production of energy and food at the same time.
[00:04:22] Speaker B: Very interesting. So tell us a little bit about the differences between the on site operations between land and floating and agri. I imagine the technical challenges that you have to overcome, and each of those is slightly different.
[00:04:41] Speaker C: It is in fact, so for the conventional pv, I think that the biggest challenge now is the scale of the projects. The utility scale projects are very, very big. We are talking now about, for instance, China is planning on having multi gigawatts of pv plants installed in the remote areas, deserted areas of the country. And when we think about maintaining this kind of assets, this is a huge challenge. So that's why for this specific case, we think that robotics can also be applied, especially for the cleaning part, and also, for instance, vegetation management. Then for the floating part, we are facing quite a big challenge that was not foreseen at all, which is about the bird droppings. One of our assets, I mean, I'm talking now as an EDP employee, right? And one of our assets in Alkyv is facing this challenge for quite a while now. And again, we have this objective of having every operation and maintenance tasks being more or less dealt with by robots. And finally, the aggravoltx is quite of a new field. And we are also trying to explore how this duality can work together, because it's difficult to have some panels that are more or less mixed within crops, because they have the shedding effect on the crops, etcetera. So these panels can affect or not, let's see, the crop healed. And this is something that we want to assess. So, on this agrivoltaic part, we are more on the, on the pilot phase. We are just testing some small configurations and see if they will be scalable or not in the future.
[00:06:31] Speaker A: Did you know that Enlit has another podcast? The Energy Transitions podcast is a broad ranging bi weekly podcast about the people accelerating the energy transition in Europe and beyond. You can find it on Spotify, Apple, or wherever you enjoy your podcast.
[00:06:52] Speaker B: NiT became aware of this project, and we got involved in covering the project when Enit went to our edPs, a floating voltaic project in al Qaeva. Tell us a little bit more about EDP's involvement in the project.
[00:07:06] Speaker C: Overall, the whole objective of the project is very important to any energy utility that focuses mainly on renewable energy sources. And EDP is one of those energy utilities. But in fact, the concept of Talos, the initial idea and the scope, they were crafted more or less in an ideation process that involved several companies within ADP, together with several partners that in fact ended up being our partners in this consortium in Talus. But this ideation process mainly was a process that was managed by three companies. So EDP new, that's the company I work for. We are basically the R and D center of EDP, and then we ideate this together with renewables and EDP generation. So to have conventional pilots onboard of the project, the same with the floating from the EDP generation. We are owners of three of the four pilots within this project. So we have a big presence on the project. And we also want to work quite a lot for the success of this project, because as end users of this project, all the solutions that will be developed and tested during the project, we hope that we can use them also in the future.
[00:08:30] Speaker B: And who are some of the other partners?
[00:08:32] Speaker C: So we have different kinds of partners. We have universities, other end users too, in the project. So we have a big operation and maintenance service provider, which is now.
Then we have universities and R and D centers. We have the University of Wagningen in the Netherlands. We have R and D centers, such as Certh from Greece and Dinesh tech from Portugal. Then we have different industries and SME's and startups, more on the robotics part and the AI tools development.
[00:09:11] Speaker B: So let's get into the tech itself a little bit. So Talos is about robotics and maintaining and cleaning. You mentioned specifically cleaning of photovoltaic panels. Maybe we can go back to the problems and specifically what the robotics aim to solve.
[00:09:31] Speaker C: Okay, so the robots cannot do it just by themselves. They need kind of an intelligence that is behind the robots. So there are two main developments within Talus. We have the autonomous robots from one hand, and from the other, we have the digital platform.
And we can think of this digital platform as the brain of talos that will operate both the analytics part and all the AI procedures. Let's call it like this. And also some robot fleet management from the robot side. I've mentioned the cleaning part, but in fact, we are here trying to automatize different levels of ONM tasks. So for one part, we have the monitoring, then we'll have the inspection of the assets.
We'll have the cleaning, as you said. And then we also have this vegetation management. Of course, this vegetation management doesn't apply to old pilots. For instance, floating. It doesn't make any sense. So we have these tasks that are being considered within talus. And for them, we'll be developing different types of robots. In fact, we are kind of co designing these robots during the development phase of the robots. We are at the same time gathering a lot of information from the future users of the robots. Namely on the specific user needs that these users might have in the future expectations about the robots. And also some features that they wanted to have in the robots. So we are gathering this information and we are adapting our developments to take this into consideration. So this is a co design thing. This was also something that was mandatory by the call for this specific project.
So, regarding the robots, we'll have different kinds of robots for the different scenarios. For the conventional pv, we'll have drones that can operate autonomously. Again, for doing some aerial inspections over the pv sites. Then we'll have a cleaner. We'll be developing a cleaner and a mower to do this vegetation management for this conventional pv. Since we are talking about huge power plants, these robots are also very big. So the idea is for the cleaner to have the capability to clean all these areas in a quite fast place. Way faster than with humans, to be honest. And also the mower will have the capability to both cut and collect all the vegetation that has cut for the floating. The challenges are a little bit different. So we want also to develop a drone that can do these around inspections. But this drone will be assisted by an autonomous boat that can basically deliver the drone into the platform. So that the battery of the drone can be whole, used to inspect the plant. And not for doing those trips to the platform and then coming back to a charging station. So we'll have these additional boats that will transport this drone into the platforms and then back to the docking station. In this specific scenario, the cleaner will be quite smaller. So the cleaner will be on top of the panels. So it will be a lot smaller and also lighter. So that the cleaner can operate between rows in automatic way without making any damage to the panels. Our last scenario, the agrivoltaics one. Here, the challenges are also different. We'll have, again, the drone to do their own inspections over the pv panels. But then we'll have an autonomous ground vehicle that will do some monitoring and will be used for recommendation purposes. So these robots will do some detections, both of the plant and the crop that is being used there. We'll try to, will have obstacle avoidance and also human detection to also avoid any contact with humans. So it is more or less this. So we'll monitor the crop health and the fruit size, and we'll provide some recommendations, for instance, for spraying and some yield forecasts too.
So this is more on the robotic side. I will now go a little bit deeper into the digital platform, the brain of talos, that will operate all of this and will orchestrate all these robots in one hand. We'll have these AI algorithms that will detect defects on pv panels based on the imagery that is gathered by the drones.
And this will help predictive maintenance of our pv parks.
And then we'll have also a tool that will support the decision making with regards to scheduling of every operation and maintenance tasks, and also the robot fleet management. So to make sure that the robots doesn't interfere with each other, this interface tool will be available for the end users. And now I'm talking about the operation and maintenance teams or even farmers if we think about the arrival tags.
[00:14:52] Speaker B: Okay, what about the future of PV? Obviously, the industry is growing very quickly, and as you mentioned, the size and scale of these installations are getting bigger and bigger.
What does a project like this mean for the future of PV? Is it obviously it's going to help accelerate the efficiency of these projects?
[00:15:14] Speaker C: Yeah, I think it will do that. So I think the deployment of larger pv plants will be a lot easier with this kind of solutions. I think the, especially with the way that we are dealing with this paradigm of having robots and human collaborating within the same space, I think we'll have a better understanding of how this interaction will happen in the future. So I think we'll provide good insights also for the industry. If we think of this interaction between these two components, of course, that we are improving a lot, I would say the general efficiency and the general productivity of these kind of energy assets, because most of the times these PV plants, and especially the o and m tasks, are reduced to a bare minimum amount per year, and the PV plants are usually running well below its optimal point. And we believe that having these robots with a more frequent operation of maintenance tasks, we are producing quite a lot of more energy and running these plants in a very close to the optimal point. So with that, we also increased the profitability of these projects. And we can also try to create these new opportunities within PV and different kinds of projects. For instance, when we are studying these robots in the floating scenario in the area voltaics, we believe that the, the general confidence of the sector in developing pv plants in these locations will also grow. And especially for these two scenarios, we are also giving a hint on how to have multiple uses on the same piece of land. This is something that the European Union is also trying to promote.
In fact, this is something that we are taking very seriously, seriously in this project because there will be a lot of opportunities in these sectors, definitely.
[00:17:32] Speaker B: Let's chat a little bit about the workforce required. But the results of the research and your project will obviously be very interesting because it will imply a fundamental change in the workforce required at these. So, you know, at these pv installations, because you're going from an unskilled workforce that's, you know, cleaning and removing vegetation, to a skilled workforce which is managing a fleet of robots.
[00:17:59] Speaker C: Yeah, that's totally true, but we at the same time believe that the skilled workforce is something that goes a little bit natural. So we don't believe that there will be quite.
I mean, we are not talking about the developers of the robots, we are talking about users of the robots. And the skills that are needed to operate a robot is very different from the ones that are creating the same robot, the robot developers. And we believe that anyone can start, can start working in these fields with robots. To be honest, I think some training will be definitely necessary, but not a lot of it.
We in fact predict that a similar learning curve and acceptability from society will happen in this specific scenario. As it happened with drones.
Initially it was very difficult and it was a niche to have drones and to operate drones, but nowadays they are quite widespread and everyone uses either for recreational purposes or even professional ones. So I would say that it will be more or less similar to that part. But I mean, from one hand, we have also the safety concerns in mind when we talk about the introduction of robots in our assets, because again, the safety is quite a concern because power plants are dangerous. And one of the reasons why we want to develop robots is also to decrease the exposure of humans to this kind of harsh environments that we may face, especially in the future, and also to avoid, to have people working in monotonous and very repetitive tasks.
So in the long term, we want to have a safer operation and also more pleasant working conditions for our people. I give you the example of the floating pv because I think that's a very good one. I mean, the platform is there and is more or less stable, but is not as stable as on the ground. Right.
So we have also this problem on having people above a platform that is a little bit unstable. It's not, as I said, it's not walking on the ground. And we are doing some, for instance, cleaning actions. So we are playing with water and electricity production at the same time in the middle of a water body. So the scenario is not perfect in terms of safety. And we want to make sure that we don't have any kind of accidents in our assets.
[00:20:46] Speaker B: Exactly. Tell me a little bit about the Telos community. I'm on your website and seeing a mention of Telus community. Is this for people to get involved in the project?
[00:20:57] Speaker C: Yeah, it's for people to get involved and to get more information about the status of the project and how the development developments are ongoing. And we want to create this community also to give some notoriety to the project. And also because of our open call, we'll have an open call this fall, between September and December of 2024. And this open call will have around €2 million to fund european SME's and startups, so they can develop some solutions for Talos too. So this will be based on some challenges that we are in fact now developing and trying to define very concretely, which will be the challenges, and we'll pose these challenges to the market and see if there are european startups and SME's that can propose some goods and very innovative solutions for Talos. So Talos had these initial scope that I just talked about it, but we want to broaden it a little bit. We want to make talos as like an ecosystem where we want to develop different kinds of solutions for different scenarios. For instance, we believe that one of the challenges can be related with the floating pv. So this floating pv, these platforms have anchoring and moving system below it to position the platform in this hydrazam. And one of the challenges that we believe will give to the market is to develop some ROV's, so some underwater robots that can perform autonomous inspections on these marine lines and see if there are some algae production or too much of algae, etcetera.
[00:22:47] Speaker B: Thanks, Daniel. Can you give us some timelines around your ambitions for completing the project and then, you know, rolling the technology out?
[00:22:56] Speaker C: Yeah. So the project has a total duration of three years. It started in October last year, 2023, and it will last until September 2026. So we believe that until then we'll develop all these robots and then we'll have a full year of monitoring where these robots will be fully operated within our pilots, and after that year of monitoring and pilot phase, we'll treat all the data and we believe that by the end of the project we'll have all these solutions in the very mature level so that they can also be brought up to the market.
[00:23:43] Speaker B: Great. Daniel, thank you very much for your participation in the EU Energy Projects podcast.
[00:23:49] Speaker C: Thank you so much.
[00:23:52] Speaker A: You've been listening to the EU Energy Projects podcast, a podcast brought to you by enlit and friends. You can find us on Spotify, Apple, and the enlit world website. Just hit subscribe and you can access our other episodes too. I'm Areti Darademo. Thank you for joining us.
