Yaw Error Detection and Why Turbit Deprioritizes It

At Turbit, we’ve always been committed to providing scientifically validated, high-impact solutions for wind turbine optimization. Over the years, we’ve invested significant expertise into understanding yaw errors and their impact on turbine performance. However, after extensive research and real-world testing, we’ve deprioritzed the further development of yaw error detection modules. Here is why:

Deep Expertise in Yaw Errors—With Key Limitations

Our team, especially our founder Michael, has extensive experience with yaw errors. He has worked on LiDAR-based systems for yaw misalignment detection and even developed a method to detect yaw errors using tower vibration measurements. This expertise gave us an early lead in exploring solutions for identifying and correcting yaw misalignment.

The Complexity of Proving Performance Gains

Yaw error corrections promise power output improvements, but in reality, detecting these improvements is extremely challenging. At Turbit, we specialize in predicting turbine power output under various conditions. We are able to predict power output with AI with the highest precision world wide. However, we’ve found that:

The natural variability in power induced by wind speed, turbulence intensity, temperature dependency and other factors is far greater than the systematic improvements that might be achieved by correcting yaw errors.

Even with advanced AI models, the signal of an improvement is difficult to isolate from normal fluctuations, especially when only 10 Min average data is available.

Theoretical models suggest that even a yaw error greater than 15° may yield improvements of less than 3%—a change so small that it becomes practically undetectable with any method.

SCADA-Based Yaw Error Detection—A Partial Success

We also developed a method to detect yaw misalignment using only SCADA data, successfully identifying errors that were later confirmed with LiDAR measurements. However, while we could detect yaw errors, proving an actual power curve improvement remained elusive. Despite our best efforts, no scientific validation could demonstrate a measurable power gain from correcting yaw misalignment in real-world conditions.

Staying True to Scientifically Proven Solutions

At Turbit, we have a strict principle: we don’t sell anything that isn’t scientifically proven. While we acknowledge that other companies may develop alternative methods to detect yaw errors, we choose to focus on solutions with a clear, measurable impact. We dedicate our efforts to areas where we can confidently drive real performance improvements—such as predictive maintenance, optimizing throttling strategies, and reducing downtime risks through continuous monitoring of turbine power, main components, blades, and more components in the future, to reduce operation risks. We believe that the bigger chunks of performance improvements lie in these areas.

Conclusion

We believe that wind operators deserve solutions that deliver tangible, scientifically validated benefits. By focusing our efforts on high-impact areas, we ensure that our customers get the best possible return on their investments in predictive maintenance and performance optimization.