Researchers have utilized a genetic learning algorithm to discover the best pitch profiles for the blades of vertical-axis wind turbines, which, despite their high energy potential, have previously been susceptible to strong gusts of wind.
When you think of an industrial wind turbine, you probably imagine the windmill design, technically called a horizontal-axis wind turbine (HAWT). However, the very first wind turbines, developed in the Middle East around the 8th century for grinding grain, were vertical-axis wind turbines (VAWT), meaning they spun perpendicular to the wind, rather than parallel.
VAWTs are quieter and more efficient than HAWTs due to their slower rotation speed, and they require less space for the same amount of output, whether on- or off-shore. Additionally, the blades are designed to be more wildlife-friendly by rotating laterally, making it easier for birds to avoid them compared to HAWTs which slice down from above.
Despite these advantages, VAWTs are not widely used in today’s wind energy market. Sébastien Le Fouest, a researcher in the School of Engineering Unsteady Flow Diagnostics Lab, attributes this to an engineering problem related to air flow control. However, he is confident that this can be solved.A recent article in Nature Communications discussed the use of sensor technology and machine learning to develop two optimal pitch profiles for VAWT blades. These profiles resulted in a 200% increase in turbine efficiency and a 77% reduction in structure-threatening vibrations. According to Le Fouest and UNFOLD head Karen Mulleners, this study is the first experimental application of a genetic learning algorithm to determine the best pitch for a VAWT blade. Le Fouest also explains that Europe’s installed wind technology was turned from an Achilles’ heel into an advantage.The growth of energy capacity is currently at 19 gigawatts per year, but it needs to increase to 30 GW in order to meet the UN’s 2050 goals for reducing carbon emissions. The main obstacles to achieving this are not financial, but social and legislative. There is a lack of public acceptance for wind turbines due to their size and noise, according to experts. Despite their benefits, VAWTs (Vertical Axis Wind Turbines) have a major drawback: they require moderate and consistent airflow to function effectively. The vertical axis of rotation causes the blades to constantly change position in relation to the wind, making them less efficient in strong gusts.The blades experience dynamic stall when the airflow angle changes, creating vortices that cause structural loads the blades can’t handle. To address this issue, the researchers attached sensors to a blade shaft to measure air forces. They then adjusted the blade’s angle, speed, and amplitude to create different pitch profiles. Using a computer and a genetic algorithm, they conducted over 3500 experiments to find the most efficient and robust solution.The researchers took the unique approach of analyzing the characteristics of various pitch profiles and combining them to create new and improved “offspring.”
This method not only helped the researchers identify two pitch profile series that significantly enhance turbine efficiency and robustness, but also allowed them to turn the main weakness of VAWTs into a strength.
According to Le Fouest, “Dynamic stall, which is the same phenomenon that destroys wind turbines, can actually propel the blade forward on a smaller scale. We use dynamic stall to our advantage by redirecting the blade pitch forward to generate power. Most wind turbines currently angle the force generated by the blades upwards.”, which does not contribute to the rotation. Altering the angle not only creates a smaller vortex – it also pushes it away at the exact right moment, resulting in a secondary area of power generation downwind.”
The Nature Communications article showcases Le Fouest’s doctoral research in the UNFOLD lab. He has since been awarded a Swiss National Science Foundation (SNSF) BRIDGE grant to develop a prototype VAWT. The objective is to install it outdoors for real-time testing under real-world conditions.
“We anticipate that this method of controlling airflow can advance VAWT technology to make it more efficient and reliable.”Le Fouest states that the technology can soon be available for commercial use.
Journal Reference:
- Sébastien Le Fouest, Karen Mulleners. Optimal blade pitch control for enhanced vertical-axis wind turbine performance. Nature Communications, 2024; 15 (1) DOI: 10.1038/s41467-024-46988-0