Listening for damage to wind turbines

Maintaining wind turbines and detecting potential vulnerabilities is expensive and time-consuming, especially when they are located offshore. As a result, rotor blades are often simply replaced, a costly process, when damage is only suspected. The Fraunhofer-Institut für Integrierte Schaltungen IIS, in collaboration with the Fraunhofer-Institut für Windenergiesysteme IWES, has developed a solution that allows remote detection of cracks and fractures in wind turbine blades at an early stage.

Cracks, fractures and erosion in rotor blades are common causes of failures in wind turbines. Offshore wind farms are at particular risk because they are exposed to exceptionally strong winds, rain and other severe weather conditions at sea. Because wind turbines are difficult to access, inspections are costly and time-consuming. Therefore, rotor blades are sometimes simply replaced when damage is only suspected. The cost can amount to over €200,000 per blade per incident.

Structural noise sensors in wind turbines

A division of Fraunhofer IIS is working with Fraunhofer IWES on a solution. "Our aim is to use acoustic emission sensors to reliably monitor damage to rotor blades remotely, so that wind turbines remain available and do not fail," says Björn Zeugmann, group manager at Fraunhofer IIS in the area of analogue integrated circuit design, about the project's objectives.

Special chip

To this end, he and his colleagues have jointly developed a special chip to be used in the sensors. The sensors, which are attached to the inside of individual rotor blades, absorb sound waves travelling through the structure of the blades. One challenge is that, unlike, say, a steel beam, the material is not homogeneous. In fact, rotor blades consist of several layers. The newly developed chip captures signals known as surface waves, which occur when damage such as a crack occurs. It then transmits these signals, for example via mobile communication.

Abnormal information

What makes this technique so special is that unlike conventional measurement systems such as radar systems or drones that collect and transmit all raw data, the Dresden researchers' new chip only transmits anomalous information. "We use an acoustic system that detects damage to the wind turbines based on the sounds it makes, so that it can tell the difference between a crack forming in the rotor blade and a fracture, for example," Zeugmann explains.

Characteristic features

Fraunhofer IWES developed this acoustic solution in an earlier project. By extracting characteristic features, the data volume can be significantly reduced so that it can be transmitted over a mobile network at all. "Our chip is always listening, which means that ideally it can classify and transmit information about possible damage from the rotor blade itself."

Monitoring wind turbines

In future, this should make it possible, firstly, to determine whether damage has actually occurred and whether the wind turbine should be shut down in the worst case. Secondly, it will also help reduce unnecessary maintenance missions to hard-to-reach offshore wind farms and optimise maintenance, as damage can be monitored over a longer period of time. If the damage gets worse and produces noise, the noise will be detected so technicians can carry out a targeted inspection and repair the problem if necessary.

Small and energy-efficient solution

Compared to existing measurement methods, the new solution is smaller, more energy-efficient and consumes significantly less data, as smaller data packets are transmitted. This also means that no broadband internet connection is needed to transmit relevant information from the wind farms to the mainland.

Extending the system

In two previous projects, Fraunhofer IIS and Fraunhofer IWES have already collaborated to develop an initial prototype chip that can be used to detect damage. The follow-up project, now under way, started on 1 June. In the project, the researchers plan to expand the overall system so that it can also detect lightning strikes - and their possible consequences - in the future. Until now, this has not been possible, but especially for offshore wind farms, this additional information is crucial.

Creating value

Zeugmann is pleased with how far the new technology has come: "I find it fascinating to work in a future-oriented field like the energy transition and thus create value for society."

Photo: Fraunhofer IWES/Gerrit Wolken-Möhlmann

Read also: AI system developed to predict wind turbine defects early