Efficient electricity generation through wind power, even in icy temperatures
An interdisciplinary team of experts from VERBUND, the AIT Austrian Institute of Technology, the University of Vienna and Meteotest is researching the use of intelligent rotor blade heaters in wind turbines in the joint research project SOWINDIC. These heaters are intended to ensure efficient electricity production in winter temperatures.
Alongside hydropower, wind energy is Austria’s most important alternative source of energy. In the winter months, the alpine climate causes not only windscreens and car doors to freeze up, but also the rotor blades on wind turbines. When this happens, wind turbines have to be brought to a stop and defrosted, either with a rotor blade heater or by natural means. This shutdown causes unpredictable production losses. To nevertheless ensure a stable electricity supply, the generating companies have to purchase expensive electricity or resort to conventional power plants such as, for example, gas power plants.
The SOWINDIC - Smart Operation of Wind Turbines Under Icing Conditions research project launched in April 2021 investigates the intelligent operation of rotor blade heaters for wind turbines, as a means to use these with maximum efficiency. Financed through the Climate and Energy Fund and under the consortium leadership of VERBUND, Austria’s leading energy company, experts of the AIT Austrian Institute of Technical, Austria’s largest non-university research facility, the research association “Data Science @ Uni Vienna” and Meteotest AG, a specialist for weather, climate and environmental data, are working on the development of an optimised rotor blade heater control system.
The project aims at delivering two initially independent approaches for the optimised operation of the rotor blade heater, which will later be merged within the scope of a so-called hybrid model: on the one hand, the working group “Applied mathematics with a focus on optimisation” of the University of Vienna will examine and further develop existing Machine Learning strategies for their suitability; on the other hand, Meteotest will research an empirically based approach that builds on physical models. Both strategies pursue the goal of being able to predict the ice-caused production failures of wind turbines as precisely as possible using control, sensor, weather and market data. A network component specially adapted by AIT will capture the data streams as close to the system as possible, implement both algorithms on the wind turbine with real-time capability, and use them for automated, optimised control of the rotor blade heater. As an operator of wind turbines, VERBUND will mainly focus on validating the developed models and bring its many years of operating knowledge to bear in the project.
Austria will be further strengthened as a knowledge, business and climate location through the cooperation of leading, Austrian companies and research institutions with strong international partners such as Meteotest.