Provided is a method of controlling at least one wind turbine in case of a rotational overspeed situation, the method including: determining a current state related to the wind turbine; providing data related to the current state as input to a turbine model; predicting a load of at least one wind turbine component and power output of the wind turbine using the turbine model provided with the input for plural control strategies; comparing the predicted load and power output for the plural control strategies; and selecting that control strategy among the plural control strategies that satisfies a target criterion including the load and the power output.

The present invention provides a method of estimating an orientation of a wind turbine. The method comprises determining, using a polarising light compass of the wind turbine, a sun polarisation value, and determining a yaw angle of the wind turbine associated with the sun polarisation value. A sun direction vector is determined based on the sun polarisation value and the associated yaw angle; and an orientation of the wind turbine is estimated relative to a fixed direction using the sun direction vector.

Systems and methods to monitor a wind turbine azimuth drivetrain. Azimuth variation characteristics data are accumulated from wind turbines over a period of time. Clusters of values within the azimuth variation characteristics data are identified and a respective condition of the main drivetrain is associated with different clusters of values. After the associating, a measured set of azimuth variation characteristics data is received. A cluster corresponds to values in the measured set of azimuth variation characteristics data is determined and a condition associated with that cluster is determined to be a condition associated with the subject main drivetrain. That condition is then reported.

Provided is a method for monitoring one or more wind turbines in a wind farm, each wind turbine having a rotor with rotor blades which are rotatable around a rotor axis, wherein one or several times during the operation of the wind farm a process is performed that includes i) obtaining a digital image of the respective rotor blade, the image being a current image taken by a camera looking at the respective rotor blade; ii) determining one or more operation characteristics of the respective rotor blade by processing the image by a trained data driven model, where the image is fed as a digital input to the trained data driven model and the trained data driven model provides the one or more operation characteristics of the respective rotor blade as a digital output.

A method for computer-implemented maximization of annual energy production of several wind turbines of a wind park is provided. The method considers the impact of individual turbine manufacturing tolerances on the turbine performance, thereby avoiding under-utilization of those wind turbines. The method includes: receiving, by an interface, one or more actual manufacturing tolerances of characteristic values for each of the number of wind turbines; determining, by a processing unit, for each of the number of wind turbines a power versus wind speed map which is calculated from a given turbine model with the one or more actual manufacturing tolerances of the respective wind turbines as input parameters; determine, based on the power versus wind speed map of each of the number of wind turbines a respective performance measure; and assign a selected siting position for each wind turbine in the wind park according to its determined performance measure.

The invention relates to a method of determining the vertical profile of the wind speed upstream from a wind turbine (1), wherein wind speed measurements are performed by a LiDAR sensor (2), then the exponent α of the power law is determined by an unscented Kalman filter and measurements, and the exponent α is applied to the power law in order to determine the vertical wind speed profile.

The invention relates to a method for acquisition and modelling of an incident wind field by a LiDAR sensor. Acquisition and modelling include a step of estimating the wind amplitudes and directions for a set of discretized points, and a step of incident wind field reconstruction in three dimensions and in real time. The invention also relates to a method of controlling and/or monitoring a wind turbine equipped with such a LiDAR sensor from the incident wind field reconstructed in three dimensions and in real time.

This method allows determining the operating point of a hydraulic machine in a considered operating range, such as turbine mode, and comprises steps that consist in a) determining two coordinates (N′11, T′11) of a first series of potential operating points of the hydraulic machine for the orientation affected to guide vanes of the machine, b) measuring the rotation speed of the machine, and c) determining the torque exerted by water flow on the machine. The method further includes steps consisting in d) calculating two coordinates (N11, T11) of a second series of potential operating points of the machine in function of the rotation speed (N) measured at step b) and the torque determined at step c), and e) deducing the two coordinates (N11_real, T11_real) of operating point that belongs both to the first and the second series in the said considered operating range of the machine.

Provided is a method for computer-implemented monitoring of a component of a wind turbine, having access to a trained machine learning model which has been trained for one or more components of the same type of wind turbines. The trained machine learning model is configured to provide an output referring to a predetermined fault occurring at a component of a wind turbine by processing vibration signals in a predetermined domain which are measured in the vicinity of the component during the operation of the wind turbine. Vibration signals are mapped to corresponding vibration signals valid for the component based on one or more given kinematic parameters of the component and one or more given kinematic parameters of another component. The machine learning model is applied to the vibration signals valid for the component, resulting in an output referring to the predetermined fault occurring at the another component.

Disclosed is a method for analyzing wind turbine blade coating fatigue due to rain erosion. According to the method, a stochastic rain field model is established, and the coating fatigue life of the wind turbine blades is calculated based on a crack propagation theory. The present patent innovatively develops a stochastic rain field model considering the shape, size, impact angle, and impact speed of raindrops to simulate the raindrop impact process, analyzes the impact stress of raindrops on the blade coating by using a smooth particle hydrodynamics method and a finite element analysis method, calculates the impact stress of all raindrops in the random rainfall process by using a stress interpolation method, and carries out fatigue analysis for the blade coating based on the impact stress.