“Wind turbines provide investors with the most profit when they are run at optimum efficiency, maintaining the right balance between revenue and costs,” explains Dr Boris Fischer, in press release from the Fraunhofer Institute for Energy Economics and Energy System Technology (IEE), providing background to the KORVA research project.

Fischer’s team spent three years developing software based on machine-learning that enables operators to strike the optimum balance between energy output and wear and tear. It does this by extending the service life of the turbines, achieving higher energy yields, reducing inefficiencies and increasing the overall profitability of the wind farm.

The researchers used meteorological data from four wind farm regions within Germany to map different location characteristics. Fluctuations in wind strength per site, operating costs (particularly wear and tear) and the volatile nature of the electricity market had to be included in calculations. Because of the high level of complexity, artificial neural networks were deployed to estimate expected loads and yields per turbine.

In real terms, the IEE’s optimisation tool resulted in turbine service life being extended from 20.7 to 30.4 years (in one case study). Data from that wind farm showed an energy yield increase of 122 percent across service life and a resulting increase in rate of return of 312 percent. The data produced several insights: for example, it certain conditions such as high turbulence, it would make economic sense to temporarily take the turbines offline.

The team hope KORVA could be applied to other systems with load-dependent aging effects including batteries or electrolysers. The project was funded by the German Federal Ministry for Economic Affairs and Climate Action and involved multiple and international partners including Nordex (manufacturer), ABO Wind (operator), Steag and Statkraft (resellers) and the TÜV Süd (certification body).