Effective wind speed estimation: Comparison between Kalman Filter and Takagi-Sugeno observer techniques

Area of Science:

• Control Engineering
• Renewable Energy Systems
• Nonlinear Observers

Background:

• Accurate wind speed estimation is crucial for advanced model-based control of wind turbines.
• Existing methods like Kalman Filters have limitations in handling nonlinearities and uncertainties.

Purpose of the Study:

• To benchmark a nonlinear Takagi-Sugeno observer against enhanced Kalman Filter techniques for wind speed estimation.
• To assess the performance and robustness of these observers against model-structure uncertainties.
• To evaluate the impact of different modeling assumptions on estimation quality.

Main Methods:

• Implementation of a nonlinear Takagi-Sugeno observer.
• Application of Linear, Extended, and Unscented Kalman Filters.
• Benchmarking using reduced-order models of a reference wind turbine.
• Numerical evaluation of wind speed reconstruction accuracy.

Main Results:

• The Takagi-Sugeno observer showed improved performance and robustness compared to Kalman Filter variants.
• Estimation quality varied based on modeling details and design assumptions.
• Wind speed estimation significantly benefits wind turbine control, as shown in a feedforward loop example.

Conclusions:

• The nonlinear Takagi-Sugeno observer is a promising technique for accurate and robust wind speed estimation in wind turbines.
• Careful consideration of modeling assumptions is necessary for optimal observer design.
• Accurate wind speed estimation enhances the effectiveness of wind turbine control strategies.