Bright prospects for the future of wind energy

Wind turbines provide more than one third of renewable energy throughout Europe. However, so far, their maintenance has been expensive and their lifespan rather short. Eleni Chatzi, Professor of Structural Mechanics and Monitoring at ETH Zurich, developed a series of methodologies in a research project funded by the ERC, allowing for intelligent, low-priced wind turbine monitoring during operation.

The clouds in the winter sky have slightly cleared. Eleni Chatzi is standing below her wind turbine on the Taggenberg outside Winterthur, Switzerland, and has her photograph taken together with two of the WINDMIL team members: a souvenir picture. Because the WINDMIL Project, its full title being «Smart Monitoring, Inspection and Life-Cycle Assessment of Wind Turbines», will be completed in April after five years of collaboration. «How often did I come here? Not often!» Eleni Chatzi says, laughing. Thanks to telemetry, the remote data transmission, she can record and monitor data comfortably from her office.

The wind turbine in Winterthur was bought by the group two years ago and serves as a type of production laboratory. The plant was manufactured by Aventa and stands on a concrete tower that is 18 metres high; its rotor diameter is 12.9 metres. What is more, the researchers have a wind turbine model standing at ETH Zurich, which is much smaller but allows for controlled simulations. Its tower diameter is only two centimetres and its height is 1.6 metres. The rotor blades have a length of 80 centimetres. On this model, one can for example make the tower base swing and modify individual parts to explore performance and test damage.

Monitoring, inspecting, enhancing

However, the principal supplier of research data for the project are the wind parks in all of Europe. The objective of the project is to develop a method for the surveillance of the plants, the inspections of the individual parts and the assessment regarding a wind turbine’s life cycle. WINDMIL should deliver a more efficient and cheaper method for the maintenance of wind turbines, including at the same time the building structure itself and thereby maybe even expanding the turbine’s service life.

As yet, most wind turbines are still equipped with so-called Supervisory Control and Data Acquisition (SCADA) systems. They provide information about rotor operation, performance and wind. The building structure itself and the underlying soil condition are not part of this monitoring system and the wind turbines must be shut down for inspection, which makes maintenance costly.

«The operators of wind energy
plants maintain an open attitude
towards innovation.»

The timing for this project was well chosen: a whole set of wind turbines across Europe has currently neared or arrived at the end of its nominal life. Both operators of wind energy plants and political decision-makers are therefore highly interested in reliable methodologies for life-cycle assessment – and a monitoring method able to keep up with the constantly improving wind turbine technology. The production of wind energy is an industry of the future; as the availability of other energy resources is gradually depleted the green and sustainable energy generation will become more important than ever.

Difficult search for partners

Partners of the project are Aventa and Siemens Gamesa as well as wind parks in Germany, OSMOS in Greece, Vattenfall in Sweden and Ørsted in Denmark. How do you find so many partners? «It is always time-consuming to find suitable structures, in the field, which we can work with,» Eleni Chatzi states, having previously conducted research exclusively on more traditional building structures, such as bridges and other buildings.

For the wind turbine project, the search for partners was particularly difficult, according to Chatzi, as many different people are involved. There are the proprietors themselves, but also the engineers operating the wind parks, as well as the authorities that have to approve the plant construction. Chatzi found her partner companies via European projects as well as existing collaborators, for example in Greece. «If it would have been easier, we would have included some further wind turbine locations in Switzerland as well,» she says and adds, «Admittedly, there is not as much interest in wind power in Switzerland, as it is primarily a hydropower country.»

During the first two years, the group was almost exclusively devoted to finding the right partners. After two and a half years, by half-time of the project, they were ready. The measuring instruments were in place and started to deliver data. Before, the entire research was based on simulations.

The diagnosis and prediction module of WINDMIL was validated by means of various records. This included datasets generated by large numerical simulations, laboratory data from the small wind turbine model, SCADA data generated by wind parks in Europe as well as structural monitoring data from sensors installed on the wheels of the wind turbine.

Achieving the objective using new methodologies

The project ran on two tracks, a purely data-driven and a model-based track. The first relies on the processing of the information gathered by the sensors for tracking the evolution of the wind turbines’ dynamics and searches for signatures of the system’s performance and condition by purely looking into signals. The latter track relies on use of computational models that are able to account for the uncertainties relating to diverse loadings caused by wind and waves, etc., as well as interactions between soil, structure and foundation.

Additionally, during the WINDMIL Project a new procedure was developed building on the combination of both types of simulations, that is on the merger of data with models. This kind of procedures, linking physics to data, is what Chatzi and other researchers in the field of system identification call a hybrid approach. Thus, PhD student Konstantinos Tatsis combined wind turbines simulation models with data received from sensors attached to the wind turbine and tried to develop predictive indicators of the condition of the wind turbine in terms of machine fatigue and other damages.

The project is located at the interface of various disciplines: the structural analysis requires the knowledge of constructional engineering; motion recording demands mechanical engineering expertise and electrical engineering is needed for signal processing. Correspondingly, the research group of the young Greek scientist is quite heterogeneous.

«Our approach has the potential
to revolutionise
wind energy systems.»

Hurdles were mainly found in the administrative area. Wind energy plants observe strict regulations. In order to introduce a new technology, many norms and standards had to be revised. This led to delays. «The operators of wind energy plants, however, maintain an open attitude towards innovation,» Chazi emphasises.

Now, the project is almost completed and the result is available: a «protective suit», as Chatzi calls it. It consists of hardware in the form of sensors and a modular software package called WINDMIL-RTDT that is easy to implement. The assembly kit shall completely redefine the status quo of current monitoring and data recording systems.

Eleni Chatzi is visibly proud of her achievement. «I know it sounds somewhat technical», she says. «But with our technique we offer a system capable of supervising and extending the lifespan of wind turbines, as structures. So far, weight was put on power production and the mechanical components, such as the gearbox, but not the structure itself – and this structure is basically a bottleneck when we ask about the duration of a wind turbine’s operational capability. Thus, our approach has the potential to revolutionise wind energy systems.»

A new research field is created

During the last two months of the project duration, it is mostly about final reports. And then there is the legacy of the project: WINDMIL yielded an instrument with the capacity of being adopted in wind parks across the globe. This is why Eleni Chatzi and her colleagues are currently in the process of establishing a spin-off company with which they can enter the industrial market.

At the moment, Imad Abdallah, the group’s senior scientist, is preparing the proposal for ETH Zurich. The PhD student Charilaos Mylonas, who calculated the lifespan of wind turbine material based on simulations during the project, also envisions his future in this industry. Konstantinos Tatsis, however, intends to conduct further research in this field. And the future looks bright: the technology of wind turbines keeps getting better so that the existing turbines can be replaced by larger and stronger versions that produce more energy; so much more energy, in fact, that the change from the old to the latest models is also financially worthwhile.

What is more, the project generated several follow-up and cooperation projects, for example Aerosense, a project funded by the Swiss National Science Foundation: Eleni Chatzi’s group and other partner groups are currently collaborating with the Eastern Switzerland University of Applied Sciences (OST) to produce a first microelectromechanical intelligent measuring system based on surface pressure and acoustics for the monitoring of wind turbine blades.

Other joint projects have been started with industrial partners from the field of wind energy. «The WINDMIL Project may come to an end soon, but is has created a prospering research field with numerous opportunities to transfer our results into current practice,» Eleni Chatzi is pleased to state.

Interview with Eleni Chatzi
Eleni Chatzi

Eleni Chatzi is Associate Professor of Structural Mechanics and Monitoring at the Institute of Structural Engineering of the Department of Civil, Environmental and Geomatic Engineering at ETH Zurich. She was born in Athens, Greece, in 1981 and studied at the National Technical University of Athens, where she obtained her Bachelor Diploma magna cum laude in 2004 and concluded her studies with Honours with the Master of Science in Civil Engineering in 2006. She continued her studies in the Department of Civil Engineering & Engineering Mechanics at Columbia University (New York), where she obtained a Master of Philosophy in 2018 and her PhD Degree with Distinction in 2010. Later that year, she was appointed Assistant Professor at ETH Zurich.

Horizon 2020 Project

WINDMIL: Smart Monitoring, Inspection and Life-Cycle Assessment of Wind Turbines

  • Programme: ERC Starting Grant
  • Duration: 1 May 2016 – 30 April 2021 (60 months)
  • Contribution for ETH Zurich: 1’486’224 €