30results about "Neural network algorithms" patented technology

A method, system, and device for wind speed prediction and layout optimization in wind power generation are provided. The method includes: obtaining a basic wind resource dataset of a target region; constructing a physics-informed neural network model based on the basic wind resource dataset; obtaining wind speeds data at a specific location in a velocity field based on the physics-informed neural networks and constructing a training dataset; training the physics-informed neural network model based on the training dataset; reconstructing a wind speed distribution within the velocity field and predicting wind speeds for a next time period with a wind farm using the trained physics-informed neural network model; and optimizing a layout of a wind turbine cluster based on a reconstructed wind speed distribution within the velocity field. The present application reconstructs a two-dimensional velocity field of the wind farm by training the PINN and enables accurate ultra-short-term wind speed prediction.

The invention belongs to the technical field of wind power control, and discloses a wind driven generator fixed time adaptive neural network variable pitch control method, which comprises the following steps: step 1, establishing a kinetic model of a wind driven generator, and initializing a system state and control parameters; 2, aiming at the wind driven generator system with external interference, designing a fixed time sliding mode surface based on a sliding mode thought; and step 3, designing a fixed-time adaptive neural network variable pitch controller. The method has the advantages that under the conditions that external interference exists in the wind driven generator and parameters are uncertain, rapid convergence of the rotor rotating speed tracking error within fixed time is achieved, and the convergence time does not depend on the initial state.

A method for determining operating states of a wind turbine (2) comprises providing an AI model (10), wherein the AI ​​model 10 is trainable to determine at least one current and / or future operating state of the wind turbine (2) on the basis of input data containing current and / or future environmental parameters (11a) and / or control data (12a), and determining the at least one current and / or future operating state (13) of the wind turbine (2) using the AI ​​model (10) on the basis of the input environmental parameters (11a) and / or control data (12a).

The application relates to the technical field of wind turbine data analysis, and discloses a system for analyzing and predicting offshore wind turbine operation data. The system comprises a marine environment data integration module, which collects data to generate a multi-source spatiotemporal synchronous data set; a multi-modal feature fusion module, which extracts cross-modal correlation features to generate a high-dimensional fusion feature tensor; a dynamic fault prediction module, which constructs a bidirectional gated recurrent network model to predict the degradation probability and residual life of key components of equipment; and a self-adaptive optimization control module, which constructs a multi-objective dynamic programming model to optimize the operation strategy of the wind turbine. In addition, the system is also provided with a feedback correction module to correct the parameters of the prediction model, and a virtual sensor module based on a physical information neural network to monitor the tower stress and diagnose sensor faults. The system realizes comprehensive monitoring, accurate fault prediction and optimized control of offshore wind turbines, effectively improves the operation efficiency, reliability and safety of the wind turbines, and reduces the operation and maintenance cost.

A method for monitoring a rotor assembly of a wind turbine includes receiving, via an imaging analytics module of a controller, thermal imaging data of the rotor assembly. The thermal imaging data includes a plurality of image frames. The method also includes automatically identifying, via a first machine learning model of the imaging analytics module, a plurality of sections of a rotor blade of the rotor assembly within the plurality of image frames until all sections of the rotor blade are identified. Further, the method includes selecting, via a function of the imaging analytics module, a subset of image frames from the plurality of image frames, the subset of image frames comprising a minimum number of the plurality of image frames required to represent all sections of the rotor blade. Moreover, the method includes generating, via a visualization module of the controller, an image of the rotor assembly using the subset of image frames.

The invention provides a wind turbine method that includes receiving alarm state data indicating that the wind turbine has entered an alarm state in which operation of the wind turbine has stopped, and receiving sensor data from a plurality of sensors of the wind turbine indicative of operating conditions associated with the wind turbine. When the alarm state data is received, the method includes executing a trained machine learning model based on the received sensor data and the alarm state to obtain an output, where the machine learning model is trained based on historical data associated with a plurality of wind turbines, the historical data being indicative of the plurality of wind turbines previously being in the alarm state. The method includes providing, based on the obtained output, an action to be performed to allow the wind turbine to resume operation.

It is described a method of predicting a frequency value of a utility grid (3) to which a wind park (1) is connected, the method comprising: obtaining plural utility grid measurement values (7) pertaining to a predetermined time range before and until a present point in time; obtaining plural wind park measurement values (12) pertaining to the time range; feeding the plural utility grid measurement values (7) and the plural wind park measurement values (12) into a recurrent neural network trained to output the frequency value at at least one next point in time, the next point in time being in particular between 0.5 s and 2 s after the present point in time.

The invention discloses a digital intelligent simulation method for a hydroelectric generating set speed regulating system based on hydroelectric-mechanical coupling, and aims to solve the problems of low hydroelectric-mechanical coupling degree, poor controllability of a simulation tool and insufficient consideration of a water hammer effect in traditional simulation. The method comprises the following steps: establishing a hydraulic electromechanical coupling model containing hydraulic power (simulating hydraulic transient by a characteristic line method), machinery (a T-S fuzzy model self-defined speed regulator) and electricity (PSS / E tidal current + generator equation); constructing a load flow calculation interface layer for secondary development of a PSS / E Fortran API, and integrating intelligent algorithm modules of GSA, DEPSO and BP neural networks; carrying out real-time interactive simulation and optimizing PID parameters of the speed regulator; and verifying dynamic characteristics of the model based on a water hammer effect analytic solution. The method realizes hydroelectric-mechanical-electrical deep coupling, improves simulation controllability and precision, adapts to unit debugging and power grid dispatching verification, and ensures safe and stable operation of the hydroelectric generating set and the power grid.

The application provides a neural network-based yaw control method and device for a wind turbine, and relates to the technical field of wind turbine yaw control. The method comprises the following steps: obtaining state parameters of a target wind turbine; taking the state parameters of the target wind turbine as input, outputting a first predicted yaw angle of the wind turbine through a first yaw angle prediction model, and outputting a second predicted yaw angle of the wind turbine through a second yaw angle prediction model, wherein the first yaw angle prediction model and the second yaw angle prediction model are obtained by training different deep learning algorithms through historical state parameters and corresponding yaw angles of different wind turbines; determining a predicted yaw angle difference between the first predicted yaw angle and the second predicted yaw angle, and determining a target predicted yaw angle of the target wind turbine based on the first predicted yaw angle and the second predicted yaw angle according to the predicted yaw angle difference. The application improves the accuracy of yaw control for the wind turbine, and improves the efficiency and reliability of wind power generation.

Individual pitch control method and associated wind turbine controller The individual pitch control method (100) comprises an online phase and an offline phase. The online phase comprises acquiring (130) a current value of a parameter corresponding to the wind speed, selecting (140) a current control law and generating (150) a control signal based on the current control law and a current azimuth angle of the blade. The offline phase (110) comprises obtaining a trained reinforcement learning model. Selecting a control law in the online phase consisting in running in inference the trained reinforcement learning model to estimate the current control law from the current value of at least one parameter corresponding to the wind speed.

A wake evaluation apparatus includes: a wake region detection unit 13 that analyzes an SAR image observed using a satellite's synthetic aperture radar using a machine learning model and detects one or more wake regions included in the SAR image; a wake position specifying unit 14 that specifies the position information of the wake region detected by the wake region detection unit 13 based on geographic information of the observation area measured by the satellite; and a wake region association unit 15 that associates the wake region with a wind power plant at a corresponding location based on the specified position information of the wake region and position information of the wind power plant specified in advance, and is configured to be able to evaluate wakes generated around a wind power plant without providing a transect by using a novel analysis method that is executed by applying the SAR image to the machine learning model.

Correcting measured wind characteristic of a wind turbine It is described a method of correcting a measurement value (101) of least one wind characteristic, in particular wind speed and / or wind direction, related to a wind turbine (1) having a rotor (3) with plural rotor blades (5) at least one having an adaptable flow regulating device (7) installed, the method comprising: measuring a value (101) of the wind characteristic; obtaining state information (107) of the adaptable flow regulating device (7); and determining a corrected value (111) of the wind characteristic based on the measured value (101) of the wind characteristic and the state information (107) of the adaptable flow regulating device (7).

A method for determining operating states of a wind turbine (2) comprises providing an AI model (10), wherein the AI ​​model 10 is trainable to determine at least one current and / or future operating state of the wind turbine (2) on the basis of input data containing current and / or future environmental parameters (11a) and / or control data (12a), and determining the at least one current and / or future operating state (13) of the wind turbine (2) using the AI ​​model (10) on the basis of the input environmental parameters (11a) and / or control data (12a).

The invention relates to a method for determining a spatial position of a portion of an infrastructure element (10), especially a wind energy plant (12). The method comprises obtaining first data (D1) indicative of an operational parameter of the infrastructure element (10) and / or indicative of an environmental parameter of the environment in which the infrastructure element (10) is located. The method further comprises inferring second data (D2) indicative of a position of the portion of the infrastructure element (10) based on the obtained first data (D1) and based on a trained data model (66), wherein the data model (66) is trained to provide a spatial position of the portion of an infrastructure element (10) based on the first data (D1). Moreover, the invention is directed to a method for monitoring the infrastructure element (10), to a method for training the data model (66) as well as to a data processing apparatus (30), a computer program and a computer readable storage medium for executing said methods. Furthermore, a system (28) for monitoring an infrastructure element (10) and an infrastructure element (10) are described.

The invention relates to a method for determining a spatial position of a portion of an infrastructure element (10), especially a wind energy plant (12). The method comprises obtaining first data (D1) indicative of an operational parameter of the infrastructure element (10) and / or indicative of an environmental parameter of the environment in which the infrastructure element (10) is located. The method further comprises inferring second data (D2) indicative of a position of the portion of the infrastructure element (10) based on the obtained first data (D1) and based on a trained data model (66), wherein the data model (66) is trained to provide a spatial position of the portion of an infrastructure element (10) based on the first data (D1). Moreover, the invention is directed to a method for monitoring the infrastructure element (10), to a method for training the data model (66) as well as to a data processing apparatus (30), a computer program and a computer readable storage medium for executing said methods. Furthermore, a system (28) for monitoring an infrastructure element (10) and an infrastructure element (10) are described.

A method for controlling wind turbines. Incident signal data is obtained from wind turbines and fed to an artificial intelligence (AI) model in order to identify patterns in the incident signals generated by the wind turbines. One or more actions are associated to the identified patterns, based on identified actions performed by the wind turbines in response to the generated incident signals. During operation of the wind turbines, one or more incident signals from one or more wind turbines are detected and compared to patterns identified by the AI model. In the case that the detected incident signal(s) match(es) at least one of the identified patterns, the wind turbine(s) are controlled by performing the action(s) associated with the matching pattern(s).

The invention provides a wind turbine generator shutdown protection method and system and electronic equipment. Collecting multi-dimensional working condition data of the wind turbine generator, processing the multi-dimensional working condition data, and constructing a multi-dimensional state vector for the wind turbine generator; based on the multi-dimensional state vector, using a time sequence prediction model to analyze and generate a prediction state trajectory; calculating and generating a dynamic safety boundary matched with the current state of the wind turbine generator according to the real-time health index of the key component in the wind turbine generator and the future environment prediction information of the position of the wind turbine generator; and calculating and generating a comprehensive safety margin index of the wind turbine generator based on the predicted state trajectory and the dynamic safety boundary, and triggering the wind turbine generator to execute different levels of shutdown protection response mechanisms based on the comprehensive safety margin index. According to the method, future risks can be predicted, the safety boundary can be dynamically adjusted, cascading failures caused by coupling of multiple factors such as part health deterioration and severe environment are effectively prevented, and the viability and operation safety of the wind turbine generator are improved.

The application relates to the technical field of water turbine control, in particular to a water turbine speed regulation optimization control method and system based on a neural network, which comprises the following steps: collecting each kind of monitoring data of each group of data under each kind of working condition of a water turbine, obtaining interference significant coefficients of each modal component of each kind of monitoring data in each group of data under the same working condition, performing frequency domain conversion on each modal component of each kind of monitoring data, obtaining interference influence coefficients of each modal component of each kind of monitoring data in each group of data under the same working condition, then obtaining adjustment coefficients corresponding to each modal component, performing noise reduction processing on each modal component to obtain adjustment thresholds of each modal component of each kind of monitoring data in each group of data under the same working condition, performing regional division after linearization processing of each kind of monitoring data, constructing a multi-objective optimization function of a linear model of each region, and then performing speed regulation optimization control on the water turbine through the neural network. The application improves the precision of water turbine speed regulation optimization control.

In a first aspect, a wind turbine blade inspection system for detecting defects in a wind turbine blade is provided. The wind turbine blade inspection system comprises a directional light source, a diffuse light source, an image-capturing device and a controller. The controller is configured to analyze images from the image-capturing devices to detect a defect. In a further aspect, a computer-implemented method for detecting defects in a wind turbine blade is provided. In a yet further aspect, a computing system comprising a processor configured to perform a method according to any of the examples herein is provided. In yet a further aspect, a computing program comprising instructions, which, when the program is executed by a processor, cause the processor to carry out a method according to any of the examples herein is provided.

This invention relates to the field of defect detection technology and discloses a method and system for detecting welds and bolts in wind turbine towers. The method is executed by a data processing device within a wind turbine tower weld and bolt detection system, which also includes a wall-climbing adsorption robot communicatively connected to the data processing device. The method includes: acquiring ultrasonic images of the wind turbine tower surface; obtaining the ultrasonic images by scanning the wind turbine tower surface using the wall-climbing adsorption robot; performing anisotropic diffusion denoising processing on the ultrasonic images to obtain a smoothed ultrasonic image; and processing the smoothed ultrasonic image using a defect detection model to obtain the defect detection result. The defect detection model is a neural network model pre-trained using sample images of the wind turbine tower surface. This solution offers high accuracy and efficiency in detecting welds and bolts in wind turbine towers.

A wind turbine pitch control method and system. The method comprises: acquiring operation parameters of a wind turbine, determining whether current operation of the wind turbine is normal, and if the wind turbine is currently faulty, stopping the operation of the wind turbine; and acquiring first pitch parameters of a first wind turbine that receives wind, and on the basis of horizontal coordinate parameters, sending second pitch parameters to a wind turbine adjacent to the first wind turbine that receives wind. By this method, the first wind turbine that receives wind can be used to collect parameters, and the parameters are processed by using positional relationships between wind turbines and then sent to all wind turbines in the entire section, so that the wind turbines can make pre-pitch adjustments before the wind arrives, thereby saving pitch adjustment time, enabling the wind turbines to be in an optimal pitch state, and further improving power generation capacity.

A method for monitoring a rotor assembly of a wind turbine includes receiving, via an imaging analytics module of a controller, thermal imaging data of the rotor assembly that includes a plurality of image frames. The method includes automatically identifying, via a first machine learning model of the imaging analytics module, a plurality of sections of a rotor blade of the rotor assembly within the plurality of image frames until all sections of the rotor blade are identified. Further, the method includes selecting, via a function of the imaging analytics module, a subset of image frames from the plurality of image frames, the subset of image frames comprising a minimum number of the plurality of image frames required to represent all sections of the rotor blade. Moreover, the method includes generating, via a visualization module of the controller, an image of the rotor assembly using the subset of image frames.

The present invention refers to a method for at least partially decentralized calculation of the state of health (SH1, SH2) of at least one wind turbine (1) based on corresponding wind turbine individual data (D) to enable a diagnosis as well as prescriptive information to enable avoidance of occurrence of pre-known specific failure modes.

A method for predicting power production of a wind turbine (10) includes obtaining local weather data at an area of the wind turbine over a first time period, obtaining ice formation data of ice formed on blades (30) of the wind turbine (10) over a second time period, obtaining power production data of power output of the wind turbine over a third time period, and predicting power production of the wind turbine at a future time point based on the obtained power production data, the obtained ice formation data, the obtained local weather data as well as current weather forecast data for the future time point and current ice formation data.

The invention relates to the technical field of wind power generation, in particular to a control method and device of a high-altitude wind power generation system. The method comprises the following steps: acquiring historical wind regime data of a preset duration; wherein the historical wind condition data comprises historical wind speed data and historical wind direction data; inputting the historical wind regime data into a preset joint model to obtain current wind regime data; wherein the current wind condition data comprises current wind speed data and current wind direction data; inputting the current wind condition data into a preset cable speed prediction model to obtain a first cable speed of the high-altitude wind power generation system; and controlling the cable of the high-altitude wind power generation system based on the first cable speed, so that the power generation efficiency of the high-altitude wind power generation system can be improved.

A method for controlling wind turbines (2) of a wind park (1), the wind park (1) comprising a plurality of wind turbines (2), is disclosed. Incident signal data is obtained from a plurality of data providing wind turbines (5), the incident signal data including incident signals generated by the data providing wind turbines (5), and the incident signal data is fed to an artificial intelligence (AI) model (9) which is by means of the incident signal data in order to identify patterns in the incident signals generated by the data providing wind turbines (5). One or more actions are associated to the identified patterns, based on identified actions performed by the data providing wind turbines (5) in response to the generated incident signals. During operation of the wind turbines (2) of the wind park (1), one or more incident signals from one or more wind turbines (2) of the wind park (1) are detected and compared to patterns identified by the AI model (9). In the case that the detected incident signal(s) match(es) at least one of the identified patterns, the wind turbine(s) (2) of the wind park (1) are controlled by performing the action(s) associated with the matching pattern(s).

A method for computer-implemented controlling of wind turbines in a wind farm is provided. The wind farm includes an upstream first and a downstream second wind turbines, wherein the following steps are performed: i) obtaining environmental data and stress data of the first wind turbine, the environmental data and the stress data being taken; ii) determining a status information indicating whether or not a predetermined event is present at the time of taking the data, wherein the predetermined event requires immediate controlling of the first wind turbine; iii) broadcasting a message which contains environmental data and a timestamp as event information; iv) evaluating the event information whether or not the predetermined event at the first wind turbine will hit the second wind turbine; v) generating a control command for controlling the second wind turbine in case the evaluation holds that the predetermined event will hit the second wind turbine.

The invention provides a method for controlling noise of wind turbines in a windfarm. Measured weather data measured noise data are obtained (O_WND) from a plurality of sensor nodes arranged at different positions in relation to the windfarm. Further, weather conditions (O_W_WT) are obtained at least at one wind turbine, e.g. from sensors / predictors in the wind turbine. The noise and weather data from the sensor nodes and wind turbine(s) along with distances (D) between at least some of the wind turbines and at least some of the sensor nodes are used to train a learning algorithm. Further, the learning algorithm is used to predict noise (P_N) at a position of one of the sensor nodes based on measured or predicted weather conditions at the windfarm. Next, extrapolating (EX_N) the predicted noise at said position of one of the sensor nodes to a predicted noise at a point of interest in surroundings of the windfarm. Further, changing mode of operation (C_M_O) of one or more wind turbines of the windfarm based on the predicted noise at the point of interest, especially if the predicted noise exceeds a noise limit for the point of interest P_I. The method allows compliance with noise limits e.g. by slightly re-orienting one or more wind turbines to reduce noise, if it is predicted that a noise limit is exceeded, and this only costs a slightly lower than optimal energy generation in a limited time period.

The individual pitch control method (100) comprises an online phase and an offline phase. The online phase comprises acquiring (130) a current value of a parameter corresponding to the wind speed, selecting (140) a current control law and generating (150) a control signal based on the current control law and a current azimuth angle of the blade. The offline phase (110) comprises obtaining a trained reinforcement learning model. Selecting a control law in the online phase consisting in running in inference the trained reinforcement learning model to estimate the current control law from the current value of at least one parameter corresponding to the wind speed.

The invention discloses a wind power plant unit hybrid control method and device suitable for a wake effect, which comprehensively considers the influence of the wake effect in a wind power plant, constructs a data-model hybrid control method, combines the advantages of model driving and data driving, and performs optimal control on the pitch angle of a wind power unit group and the torque of a generator. The unit group fatigue load is reduced under the wake flow effect of the wind power plant, and technical support is provided for optimized operation of the wind power plant.