Proactive support for wind farm condition management

Similar to bridges, wind farms are demanding, large-scale items of infrastructure. Today, the platforms alone are nearly 100 metres long and they normally weigh over 20 tonnes. In Finland, wind farms are exposed to factors such as large temperature differences, snow and ice, which can cause mechanical and corrosion damage and change the stress and fatigue resistance of structures.

We have studied the condition of Finnish wind farms in the GenerIoT project funded by Business Finland. In addition to Granlund, the partners of the national consortium include Tampere University, Loopshore, Unikie and Tietoevry. The modelling of wind farm structures has produced information on the kind of dynamic loads that the structures are subjected to in different wind conditions, for example. We have also assessed the condition of wind farms on the site from the perspective of building physics and MEP.

Based on the results of the project, the condition of wind farms should be assessed more frequently than periodically when it comes to structures as well as building and electrical technologies, which must withstand the extremely difficult northern conditions. As the nacelle with its hubs and blades is located at a height of 150 metres, all of the above-mentioned technical areas have to withstand harsh conditions during their service life.

The developed software and structural analyses make it possible to improve the operations related to the maintenance and repair of wind farms.

That is why we are developing proactive condition management software that collects information on wind farms’ IoT actuators, condition surveys and corrective measures. In addition, we have created a material-independent and numerically based FEM (element method) model, which is well-suited to assessing the condition and mechanical properties of both steel and hybrid towers. The model provides information on factors such as the effect of dynamic stress on the long-term durability of the tower.

We use neural networks to process data and conduct structural analyses on the basis of the data. They enable proactive condition management of wind farms and earlier detection of damage to structures. It reduces unexpected power outages and prevents safety risks.

The developed software and structural analyses make it possible to significantly improve the operations related to the maintenance and repair of wind farms. Simultaneously, risks are reduced and expected profits are increased. This is a benefit to municipalities and wind power companies both.

Want to know more?

Esko Sistonen

Tutkimuspäällikkö

Granlund Oy

040 481 2468

etunimi.sukunimi@granlund.fi