The erosion of wind turbine blade material is a costly problem. “The blade material erodes due to the effect of rain, hailstones, and sand dust, which significantly reduces the service life of wind turbines. Accelerated replacement of turbines becomes expensive: up to 2–4% of the value of all wind-generated power is lost as a result of this problem,” says Principal Scientist Anssi Laukkanen from VTT, a Finnish research and development company.
“It is a question of a classic problem within this particular industry that costs billions of euros and brings additional costs to all wind energy. As wind turbine sizes increase and wind farms are placed out on the sea in increasingly demanding conditions, the significance of the problem becomes emphasised”, continues Laukkanen.
In its antiAGE project, VTT modelled the material problem and set out to solve it virtually using artificial intelligence (AI) for the first time for developing a material solution in this scale. In principle, it is possible to find an unlimited number of different variations for the material used in wind turbine blades composed of the same material components but differing slightly from one another. Of all these alternatives, one would need to find the one that is best suited for this particular purpose and meets its operational requirements.
“Human perceptive skills are insufficient to visualize all the dimensions related to the optimization of material solutions. AI, on the other hand, is capable of unraveling very complicated cause-and-effect relationships, simulating solutions and going through an infinite number of alternatives to find the one that works best in relation to the requirements set”, Laukkanen points out.
Using AI, it is possible to find a tailored material solution optimized for a specific purpose for every component of any single product. However, manufacturing of such highly tailored materials is difficult if traditional manufacturing techniques are used. Here, three-dimensional printing, or additive manufacturing, offers an opportunity to produce the desired material in any shape without unreasonable costs.
The results of the antiAGE project exceeded expectations, according to the company. Through a design process using virtual testing and machine learning, VTT was able to develop an optimized solution to a very difficult material problem in less than a year: a highly durable material that hardens when exposed to mechanical stress.
“When we published the news about our solution, wind turbine manufacturers became immediately interested in it. We are now negotiating details with commercial operators,” Laukkanen reports. VTT is also applying for additional funding for the project, since there are plenty of targets for optimized material solutions in other sectors of industry as well. The more complex the product, the more expensive it is to develop materials suited for a specific purpose, and the more difficult it is in general to find material solutions that perform well. AI allows such problems to be addressed with efficiency.
