Wind Farm Turbine Control Using Predicted Downwind Wind Loads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wind turbines downwind in a wind farm face increased loads due to uneven wind conditions, and existing methods for predicting wind changes are either unreliable or costly, as they rely on anemometer measurements or complex LIDAR sensors.
Innovation Solution
A control system and method that derive current wind estimates from operating parameters of multiple wind turbines, using a prediction model to generate control signals for turbines downwind, reducing loads by adjusting pitch angles and power consumption based on predicted wind conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If anemometer measurements are used to predict wind conditions, then wind speed can be measured, but the measurements are distorted by rotor movement and are not sufficiently reliable
Solution Approach 1:
The patent uses operating parameters of wind turbines (rotor speed, pitch angle, generator torque) as intermediary variables to infer wind conditions indirectly, rather than relying on direct anemometer measurements that are contaminated by rotor effects. This mediator approach allows accurate wind estimation without the distortion problems of direct measurement.
Solution Approach 2:
The patent replaces the mechanical anemometer measurement system with a computational approach that uses turbine operating parameters and prediction models to estimate wind conditions. This substitution eliminates the physical measurement distortion caused by rotor movement while maintaining the ability to predict wind speed and direction.
2Reliability
If LIDAR sensors are used to measure wind conditions upwind, then reliable wind data can be obtained, but the system becomes complex and expensive
Solution Approach 1:
The patent makes the wind turbines themselves serve as the measurement system by using their own operating parameters (rotor speed, pitch angle, torque) to infer wind conditions. This self-service approach eliminates the need for separate LIDAR sensors or anemometers, reducing system complexity and cost while maintaining measurement reliability.
Solution Approach 2:
The patent enables wind turbines to perform multiple functions: generating electricity and simultaneously measuring wind conditions for prediction purposes. The operating parameters collected for power control are reused for wind estimation, eliminating the need for dedicated measurement equipment and reducing overall system complexity.
3Reliability
If wind conditions are predicted for future time, then downwind turbines can be prepared to reduce loads, but the prediction must be accurate to be effective
Solution Approach 1:
The patent uses feedback from multiple wind turbine operating parameters to continuously refine wind condition estimates. By collecting data from both upwind and downwind turbines and comparing predicted versus actual conditions, the system improves prediction accuracy over time, ensuring reliable load reduction when future wind conditions are forecasted.
Solution Approach 2:
The patent performs preliminary wind condition predictions using current operating parameters from multiple turbines to forecast future wind states. This preliminary action allows downwind turbines to adjust their operating parameters (pitch angle, rotor speed) in advance of anticipated wind changes, optimizing load reduction before the actual wind conditions occur.
4Reliability
If multiple wind turbine operating parameters are used to derive wind estimates, then a broader information base is obtained, but data processing complexity increases
Solution Approach 1:
The patent merges operating parameter data from multiple wind turbines (both upwind and downwind) to create a comprehensive information base for wind prediction. By combining rotor speeds, pitch angles, and torque data from several turbines, the system obtains a broader spatial view of wind conditions, improving prediction reliability for the entire wind farm.
Solution Approach 2:
The patent transforms multiple operating parameters (rotor speed, pitch angle, generator torque) into a unified wind speed and direction estimate through prediction models. This parameter transformation consolidates complex multi-dimensional data into actionable wind condition predictions, reducing processing complexity while maintaining the benefits of multiple data sources.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for operating a plurality of wind turbines (14, 15, 16) in which a first current wind estimate is derived from the operating parameters of a first wind turbine (14) and a second current wind estimate is derived from the operating parameters of a second wind turbine (15). A prediction model (28) is applied to derive a wind prediction for a third wind turbine (16) valid for a future time (25) from the first and second wind estimates. The wind prediction is processed in a controller (24) to generate a control signal for the third wind turbine (16) that takes effect before the future time (25). The invention also relates to an associated control system. By predicting the wind conditions for a future time (25), the load on certain wind turbines can be reduced.