Why we write about this topic:
This year, the Dutch cabinet decided to increase the amount of installed offshore wind power capacity even further to 21 GW by 2030. If everything goes according to plan, there will be approximately 1,700 wind turbines operating on the Dutch North Sea. Right now, the blades of wind turbines are all subject to visual inspections – using drones and maintenance engineers. When thousands of blades are in operation at sea in eight years’ time, that will no longer be feasible in terms of costs and time. Moreover, inspections are often periodic and repairs are carried out reactively. A wind turbine breaks down, is shut down and will only then get repaired.
Photonic sensors and AI
Start-up Tarucca wants to do things differently. Photonic sensors combined with AI should be able to detect any defects ahead of time, making inspection of wind turbines less time-consuming and ensuring that the blades operate as optimally as possible.
Wind turbines are complex, says co-founder Hans van Beek. “We have been able to monitor electromechanical installations effectively for much longer. You can very easily measure vibrations at certain points in a machine. Wind turbine blades are made of composites, such as polyester. This makes it much more difficult to carry out measurements for condition monitoring, because it is not nearly as straightforward.”
Measuring vibration patterns
Vibrations provide an indication of the status of a blade. If a blade is damaged, it will vibrate in a different way. In order to inspect blades properly, small shifts in vibration patterns need to be detected. This requires very accurate measuring equipment and high-level AI software and algorithms. Van Beek: “What’s more, nothing else attracts as many lightning strikes at sea as a wind turbine. So, you want to have as little electronics in those wind turbine blades as possible.”
The sensors that Tarucca uses tackle both of these problems. Photonic sensors work on the basis of light, and light, unlike electronics, is not encumbered by lightning. Moreover, these sensors can measure much more precisely and accurately than traditional optical sensors.
Efficient repairs
Photonic sensors are installed inside the blades and reflect light signals back through fiber optics. This is how they collect the data, which is sent to the cloud via 4G or 5G. There it is processed by the AI algorithm. “If something is broken or in danger of breaking down, the wind farm operator receives a notification. Sometimes the repair needs to be done as soon as possible, while other times it can wait a few weeks. This is how these repairs can be scheduled as efficiently as possible. During a period when the wind is at a lull, for example,” Van Beek explains.
‘The flywheel needs to start turning’
The high-tech start-up is only in its infancy and needs to convince potential customers, little by little, that it all really does work. Van Beek compares it to a flywheel that they have to get to start turning. “Installing our system costs money, but this is the only way we will be able to generate data. Because, the more data we generate, the more conclusions we can draw. But installing it costs money. Breaking through that circle is our biggest challenge.”
From time to time, Van Beek finds it frustrating how long the start-up phase takes. “Raising venture capital in the Netherlands, even across Europe for that matter, is a lengthy process. That’s also the big difference between a start-up that builds a website, or start-ups in deep- and high-tech.”
Sensors in all Dutch wind turbines
The first photonic sensors will be installed early next year in the wind turbines that their partners provide. The start-up will start with one wind turbine and work up to a research pilot with five wind turbines. “So that we can really see step by step what goes well and what needs to be improved. In five years time, we hope that our sensors will be in all of the wind turbines in the Netherlands and that we will also be sitting around the table with manufacturers so as to integrate the sensors at an early stage during their construction.”
“In five years time, we hope that our sensors will be in all of the wind turbines in the Netherlands.”
Hans van Beek
Brainport region as conducive environment
After twenty years of living and working abroad, Van Beek has now been back in Brabant for two years. He experiences the Brainport region as a conducive environment where no one does anything on their own, but where everything is interconnected. “Take, for example, the Eindhoven AI Systems Institute (EAISI) at TU/e. A lot of knowledge and research on AI is brought together there. The mindset of cooperation is very powerful there.”
At the cutting edge where Tarucca is positioned – that of photonics and artificial intelligence – a lot is happening in this region. Therefore, there is a good chance that the photonic sensors and equipment hailing from Denmark that the start-up makes use of will also eventually come from the Brainport region in the future. “The sensors that we import are still very new, and that technology is not readily available everywhere. We are going to work with PhotonHub to see if there are other options compared to the sensors we use now.”
Technical facilities and a better connection to science
Van Beek is in a building at the Eindhoven University of technology (TU/e) during our interview. Workplaces are plentiful in the region, yet the entrepreneur had been missing a technical workshop. “As a start-up, you don’t necessarily have money to rent something right away, but you actually want to have a technical space where you can set up and test your setup. There should be more facilities like this in Brainport.”
Van Beek also thinks the connection between start-ups and the university could be even better. “At TU/e, there is a huge treasure trove of scientists that we would love to work with as start-ups. The university already does a lot for start-ups, but if you do not come from the world of academia, it is not so easy to always make the right connections, it often still comes down to who you know. It would be nice to have more cooperation and opportunities to establish links between the university and start-ups from outside TU/e.”
Collaboration
Related Posts
