Provided is a method for controlling a wind turbine, in particular an electric generator of said wind turbine. The method includes an optimization during which a suitable operating parameter for controlling said wind turbine or generator thereof is determined, in particular in an iterative manner. The optimization includes providing a multidimensional space comprising a plurality of parameters; providing an objective function for said multidimensional space, e.g., a simplex has a shape of a triangle or a tetrahedron; and determining one parameter of said multidimensional space as a suitable operating parameter by applying said objective function to said multidimensional space, in particular in an iterative manner. The method includes selecting a suitable operating parameter as an operating parameter for said wind turbine or generator thereof; and operating said wind turbine or generator based on said operating parameter, in particular by controlling a converter connected to said generator.

Embodiments herein describe a hydraulic pitch system where a velocity (e.g., the velocity of a hydraulic cylinder or the piston rod in the cylinder) is fed forward and combined with a setting outputted by a pitch controller. The velocity of the hydraulic cylinder is derived from the reference pitch angle or a continuous pitch signal (e.g., a cyclic pitch or ramp rate) in the control system. In either case, the velocity can be determined by monitoring the change in the reference pitch angle or the continuous pitch signal. Using a gain control, the velocity is converted into a position setting of the hydraulic pitch system (e.g., a spool setting in a valve) which is combined with another position setting generated by the pitch controller.

A system and method are provided for controlling a wind turbine of a wind farm. Accordingly, a controller implements a first model to determine a modeled performance parameter for the first wind turbine. The modeled performance parameter is based, at least in part, on an operation of a designated grouping of wind turbines of the plurality of wind turbines, which is exclusive of the first wind turbine. The controller then determines a performance parameter differential for the first wind turbine at multiple sampling intervals. The performance parameter differential is indicative of a difference between the modeled performance parameter and a monitored performance parameter for the first wind turbine. A second model is implemented to determine a predicted performance parameter of the first wind turbine at each of a plurality of setpoint combinations based, at least in part, on the performance parameter differential the first wind turbine. A setpoint combination is then selected based on the predicted performance parameter and an operating state of the first wind turbine is changed based on the setpoint combination.

A neural network method, comprising: modeling an environment; implementing a policy based on the modeled environment, to perform an action by an agent within the environment, having at least one estimated dynamic parameter; receiving an observation and a temporally-associated cost or reward based on operation of the agent in the environment controlled according to the policy; and updating the policy, dependent on the received observation and the temporally-associated cost or reward, to improve the policy to optimize an expected future cumulative cost or reward. The policy may represent a set of parameters defining an artificial neural network having a plurality of hierarchical layers and having at least one layer which receives inputs representing aspects of the received observation indirectly from other neurons, and produce outputs to other neurons which indirectly implement the policy, the plurality of hierarchical layers being trained according to respectfully distinct training criteria.

A method for operating and maintaining a wind farm comprising a plurality of wind turbines includes determining an odometer for one or more components of at least one of the pluralities of wind turbines in the wind farm, the odometer representing operational usage of the component(s). The method also includes tracking the operational usage for the component(s) using the odometer and a usage threshold. Further, the method includes predicting an expected time frame for one or more preventative maintenance actions based on a comparison of the tracked operational usage and the usage threshold. Moreover, the method includes triggering scheduling of the one or more preventative maintenance actions when the prediction indicates that the tracked operational usage will exceed the usage threshold. In addition, the method includes shutting down the wind turbine or idling the wind turbine once the one or more preventative maintenance actions are scheduled.

A computing system is configured to detect irregular yawing at wind turbines. To this end, the computing system (i) for each respective turbine of an identified cluster of wind turbines: (a) obtains yaw-activity data indicative of the respective turbine's yaw activity during a window of time, and (b) based on obtained yaw-activity data, derives a yaw-activity-measure dataset having measures of the respective turbine's yaw activity during time intervals within the window of time, (ii) based on the respective yaw-activity-measure datasets for the turbines in the cluster, derives a cluster-level yaw-activity-measure dataset, (iii) evaluates the respective yaw-activity-measure dataset for one or more turbines in the cluster as compared to the cluster-level yaw-activity-measure dataset, (iv) based on the evaluation, identifies at least one turbine of the cluster that exhibited irregular yaw activity, and (v) transmits, to an output device, a notification of the irregular yaw activity at the at least one turbine.

Provided is an apparatus and method for cooperative controlling wind turbines of a wind farm, wherein the wind farm includes at least one pair of turbines aligned along a common axis approximately parallel to a current wind direction and having an upstream turbine and a downstream turbine. The method includes the steps of: a) providing a data driven model trained with a machine learning method and stored in a database, b) determining a decision parameter for controlling at least one of the upstream turbine and the downstream turbine by feeding the data driven model with the current power production of the upstream turbine which returns a prediction value indicating whether the downstream turbine will be affected by wake, and/or the temporal evolvement of the current power production of the upstream turbine; c) based on the decision parameter, determining control parameters for the upstream turbine and/or the downstream turbine.

A wind turbine control apparatus, method and non-transitory computer-readable medium are disclosed. The wind turbine control apparatus comprises a generator connected to a wind turbine with a drive train. The drive train comprises a rotor, a low speed shaft, a gear box, a high speed shaft, and a controller module. The controller module is configured to obtain a maximum power within a large range of varying wind velocities by operating the rotor at a neural network determined optimal angular speed for the current wind velocity.

A system and method are provided for operating a wind farm. Accordingly, a wind direction affecting the wind farm is determined. Based on the wind direction, a controller identifies a turbine cluster, which is a subset of a plurality of wind turbines of the wind farm. The subset includes at least an upwind turbine and a downwind turbine that is affected by a wake emanating from the upwind turbine. With the turbine cluster identified for the given wind direction, the controller then determines a difference between a freestream maximal cluster power output and a wake-affected cluster power output for the turbine cluster. The controller then determines a mitigation setpoint combination for the subset of wind turbines. The mitigation setpoint combination is configured to establish a mitigated cluster power output. Mitigated cluster power output has a difference from the freestream maximal cluster power output that is less than the difference between the freestream maximal cluster power output in the wake-affected cluster power output for the turbine cluster. Based on the mitigation setpoint combination, an operating state of at least one wind turbine of the turbine cluster is changed.

The present disclosure provides a method and an apparatus for self-adaption of a cut-out strategy. The method may include: predicting, using a wind speed prediction model, a wind resource parameter of a wind turbine at each machine location; predicting, using a load prediction model, a fatigue load and a limit load of the wind turbine based on the predicted wind resource parameter and an air density; comparing the predicted fatigue load and limit load with a reference load; and determining the cut-out strategy based on a result of the comparison, wherein determining the cut-out strategy includes determining a cut-out wind speed and an output power.