Skip to main content

Understanding and predicting the wind behind floating wind turbines

By REALTEK-redaksjonen

Havvind
Photo: Shutterstock

What happens to the wind behind a floating wind turbine when it moves with the waves? Haris Hameed Mian at NMBU has investigated these changes and developed faster ways to predict them.

When wind passes through a turbine, some of its energy is converted into electricity. This leaves a trail of slower and disturbed air behind the turbine, much like a boat that leaves a trail in water. This disturbed airflow is called a wake. It matters because other turbines operating in this disturbed air may produce less electricity and experience changing forces.

In his doctoral research at NMBU, Haris used high-fidelity computer simulations, validated against wind-tunnel experiments, to investigate how pitching and rolling motions of the floating turbine affect its wake. He found that different movements produce different patterns. Pitching motion causes wake pulsation, while rolling motion induces wake meandering.

Understanding these changes is important, but detailed computer simulations can take several days to get an answer, even on high-performance computers. The doctoral research therefore developed a neural-network framework called MoC-PiRNN, which combines the motion information with physical principles to predict how airflow changes over time.

“My aim was to use computational fluid dynamics simulations to understand how the motion of a floating wind turbine changes its wake and then use that to develop faster ways to predict these changes without repeating expensive simulations,” explains Haris.

Building on this framework, Haris developed another approach that learns to predict only the changes caused by turbine movement, rather than the entire airflow and called it “Perturbation Learning”. In selected test cases, this reduced the prediction errors by approximately half.

To make the research more accessible, Haris has also developed an interactive online demonstrator showing how neural networks can be used to predict wind turbine wakes.

Explore the demonstrator.

The research provides new insights into how floating wind turbines affect the air behind them and could contribute to faster tools for analyzing future offshore wind farms.

Portrett av mann med kort, mørkt hår og blå glidelåsgenser. Bakgrunnen er grå.

FACTS:

Haris Hameed Mian

  • PhD: NMBU's Faculty of Science and Technology. Employed by Aker Solutions.
  • Previous education: Master in mechanical engineering.
  • Title of thesis:
  • Norwegian: Høypresisjonsmodellering og fysikkintegrerte nevrale surrogatmodeller for flytende havvindturbiner
  • English: High-Fidelity Modeling and Physics-Integrated Neural Surrogates for Floating Offshore Wind Turbines

Haris Hameed Mian will defend his PhD thesis on 3 December 2026.

Published - Updated