Hearing damage before it becomes visible: Drone-based inspection of wind turbines

Initial situation: Maintenance at great heights

Wind turbines are complex systems whose reliable operation largely depends on the functionality of individual components such as gearboxes and bearings. To prevent failures, regular inspection is essential.

The challenge lies not only in the technology itself, but also in accessibility. Many of the relevant components are located at great heights and can only be reached with considerable effort. Traditional inspection methods are therefore often time-consuming, costly, and associated with increased safety requirements. At the same time, many types of damage develop gradually and initially become noticeable not visually, but acoustically.

Technical challenge: Reliable detection of sound

In industrial practice, acoustic measurement methods are already successfully used to detect changes in gearboxes or bearings at an early stage. Unusual sound patterns can not only indicate damage, but also help pinpoint its exact location within the system.

However, transferring this approach to wind turbines introduces additional challenges. At great heights, environmental noise, wind, and movement affect the measurement. At the same time, the measurement platform itself—in this case a drone—generates noise that can interfere with data acquisition. The goal was therefore to develop a system that delivers precise and reliable acoustic data even under these conditions.

The solution: Drone as a mobile measurement platform

To capture acoustic changes directly at the turbine, a drone was developed as a flexible inspection platform. Equipped with a highly sensitive directional microphone, it is capable of recording sounds from specific components. The drone was also designed with particularly quiet propellers to minimize self-generated noise and improve data quality.

A key component of the solution is signal processing. Since the characteristic sounds of the drone are known, they can be specifically filtered out. By using acoustic filters, the natural frequencies of motors and propellers can be isolated, allowing the remaining signal to be clearly analyzed. This makes it possible to detect even small acoustic deviations and accurately assign them to the corresponding component.

Result: Precise, flexible, and safe

The use of the drone enables flexible, location-independent inspection without requiring complex infrastructure or direct intervention on the turbine. The combination of targeted sensor technology and intelligent signal processing ensures that acoustic changes remain clearly identifiable despite complex environmental conditions. At the same time, safety is increased, as no work at great heights is required.

Outlook: Acoustics as a key to condition monitoring

The project demonstrates the potential of combining mobile robotics with acoustic analysis. Especially in areas where visual inspections reach their limits or are only partially feasible, new possibilities for condition monitoring emerge.