CN117967502A

CN117967502A - Yaw angle estimation method and system of wind generating set

Info

Publication number: CN117967502A
Application number: CN202211289416.0A
Authority: CN
Inventors: 陈志文; 岳红轩; 陈卓; 王真涛; 杜洋; 王爽; 韩义
Original assignee: Beijing Huaneng Xinrui Control Technology Co Ltd
Current assignee: Beijing Huaneng Xinrui Control Technology Co Ltd
Priority date: 2022-10-20
Filing date: 2022-10-20
Publication date: 2024-05-03

Abstract

The application discloses a yaw angle estimation method, a yaw angle estimation system, electronic equipment and a storage medium of a wind generating set. Wherein the method comprises the following steps: acquiring a plurality of distance values between a target wind generating set and other wind generating sets in a wind power plant; acquiring a plurality of yaw angles of a plurality of wind generating sets at a plurality of preset time points; acquiring a first distance feature vector based on a plurality of distance values; acquiring a plurality of first yaw angle feature vectors based on the plurality of yaw angles; taking the first characteristic value of each position in the first distance characteristic vector as a weight, and weighting a plurality of first yaw angle characteristic vectors to obtain a plurality of second yaw angle characteristic vectors; and acquiring yaw angle estimated values of the target wind generating set at the current time point based on the plurality of second yaw angle feature vectors. By the technical scheme, the yaw angle of the target wind generating set can be estimated, and the normal operation of the target wind generating set is ensured.

Description

Yaw angle estimation method and system of wind generating set

Technical Field

The application relates to the technical field of wind power generation, in particular to a yaw angle estimation method and a yaw angle estimation system of a wind generating set.

Background

When the wind generating set operates, the yaw angle of the wind generating set can also change along with the change of the external wind direction so as to maximize the generating efficiency. In the related art, wind direction is generally measured by a wind vane on a wind turbine, so that a yaw angle of the wind turbine is adjusted according to the measured wind direction. However, when the wind direction of the wind generating set cannot be measured due to damage of the wind vane or other reasons, the wind generating set may not normally operate, and therefore the generated energy is lost.

Disclosure of Invention

The application provides a yaw angle estimation method, a yaw angle estimation system, electronic equipment and a storage medium of a wind generating set. The yaw angles of the target wind power generator sets can be estimated based on the yaw angles of other wind power generator sets except the target wind power generator set in the same wind power plant and the distances between other wind power generator sets and the target wind power generator sets, so that the normal operation of the target wind power generator sets is ensured.

In a first aspect, an embodiment of the present application provides a yaw angle estimation method of a wind turbine generator set, including: acquiring a plurality of distance values between a target wind generating set and other wind generating sets in a wind power plant; acquiring a plurality of yaw angles of the wind generating sets at a plurality of preset time points, and acquiring a first distance feature vector based on the plurality of distance values; acquiring a plurality of first yaw angle feature vectors based on the plurality of yaw angles; taking the first characteristic value of each position in the first distance characteristic vector as a weight, and weighting the plurality of first yaw angle characteristic vectors to obtain a plurality of second yaw angle characteristic vectors; and acquiring yaw angle estimated values of the target wind generating set at the current time point based on the second yaw angle feature vectors.

According to the technical scheme, the yaw angle of the target wind generating set can be estimated according to time sequence dynamic association characteristics among a plurality of yaw angles of a plurality of wind generating sets at a plurality of preset time points, and a first distance characteristic vector is obtained based on a plurality of distance values between the target wind generating set and other wind generating sets, so that weighting processing is carried out based on the first distance characteristic vector, accuracy of an estimation result is improved, and normal operation of the target wind generating set is guaranteed.

In one implementation, the acquiring the first distance feature vector based on the plurality of distance values includes: acquiring a distance input vector based on the plurality of distance values; and inputting the distance input vector to a sequence encoder to obtain the first distance characteristic vector.

In an alternative implementation, the sequence encoder includes a full-concatenated layer, a one-dimensional convolution layer, and a normalized exponential Softmax activation function, wherein the inputting the distance input vector to the sequence encoder, obtaining the first distance feature vector, includes: inputting the distance input vector into the full connection layer to obtain high-dimensional implicit features of second feature values of all positions in the distance input vector; inputting the distance input vector into the one-dimensional convolution layer to obtain high-dimensional implicit correlation features among the second feature values; acquiring a second distance feature vector based on the high-dimensional implicit feature and the high-dimensional implicit associated feature; and inputting the second distance feature vector into the activation function to obtain the first distance feature vector.

Optionally, the formula for acquiring the high-dimensional implicit feature is:

Wherein Y is a high-dimensional implicit feature, X is the distance input vector, W is a preset weight matrix, B is a bias vector, Representing a matrix multiplication; the acquisition formula of the high-dimensional implicit association features is as follows:

Wherein a is the width of the convolution kernel of the one-dimensional convolution layer in the X direction, F (a) is a convolution kernel parameter vector, G (X-a) is a local vector matrix operated by the convolution kernel function, w is the size of the convolution kernel, and X is the distance input vector.

In one implementation, the obtaining a plurality of first yaw angle feature vectors based on the plurality of yaw angles includes: arranging the yaw angles to obtain a plurality of yaw angle input vectors; inputting the plurality of yaw angle input vectors to a multi-scale neighborhood feature extractor to obtain the plurality of first yaw angle feature vectors.

In an alternative implementation, the multi-scale neighborhood feature extractor includes a first convolution layer and a second convolution layer, wherein the inputting the plurality of yaw angle input vectors into the multi-scale neighborhood feature extractor obtains the plurality of first yaw angle feature vectors includes: inputting the yaw angle input vectors into the first convolution layer to obtain a plurality of yaw angle feature vectors of a first neighborhood scale, wherein the first convolution layer has a first one-dimensional convolution kernel of a first length; inputting the yaw angle input vectors into the second convolution layer to obtain a plurality of yaw angle feature vectors with a second neighborhood scale, wherein the second convolution layer has a second one-dimensional convolution kernel with a second length; and cascading the yaw angle feature vectors of the first neighborhood scale and the yaw angle feature vectors of the second neighborhood scale to obtain the yaw angle feature vectors of the first neighborhood scale.

Optionally, the acquiring formula of the yaw angle feature vector of the first neighborhood scale is:

Wherein a is the width of the first one-dimensional convolution kernel in the X direction, F (a) is a parameter vector of the first one-dimensional convolution kernel, G (X-a) is a first local vector matrix operated with a convolution kernel function, w is the size of the first one-dimensional convolution kernel, and X is the yaw angle input vector; the obtaining formula of the yaw angle characteristic vector of the second neighborhood scale is as follows:

Wherein b is the width of the second one-dimensional convolution kernel in the X direction, F (b) is a parameter vector of the second one-dimensional convolution kernel, G (X-b) is a second local vector matrix operated with a convolution kernel function, m is the size of the second one-dimensional convolution kernel, and X is the yaw angle input vector.

In one implementation, the obtaining, based on the plurality of second yaw angle feature vectors, a yaw angle estimated value of the target wind turbine generator set at a current point in time includes: obtaining a plurality of local metric factors based on context statistics for the plurality of second yaw angle feature vectors; weighting the second yaw angle feature vectors by taking the local measurement factors as weights to obtain third yaw angle feature vectors; two-dimensionally arranging the plurality of third yaw angle feature vectors to obtain a feature matrix; inputting the feature matrix into a convolution feature network to obtain a decoding feature matrix; wherein the convolutional feature network has learned the ability to generate the decoding feature matrix based on the feature matrix; inputting the decoding characteristic matrix into a decoder to obtain the yaw angle estimated value.

In an alternative implementation, the local metric factor is calculated by the formula:

wherein v _i is a plurality of third eigenvalues at each position of the second yaw angle eigenvector, μ is a mean value of a set of the plurality of third eigenvalues, σ is a variance of the set of the plurality of third eigenvalues, L is a length of the second yaw angle eigenvector, exp () represents an exponential operation.

In an alternative implementation, the yaw angle estimation value is obtained by the following formula:

Wherein Y is the yaw angle estimated value, W is a preset weight matrix, X is the decoding feature matrix, Representing a matrix multiplication.

In a second aspect, an embodiment of the present application provides a yaw angle estimation system of a wind turbine generator system, including: the first acquisition module is used for acquiring a plurality of distance values between a target wind generating set and other wind generating sets in the wind power plant; the second acquisition module is used for acquiring a plurality of yaw angle first processing modules of the wind generating sets at a plurality of preset time points and acquiring a first distance characteristic vector based on the plurality of distance values; the second processing module is used for acquiring a plurality of first yaw angle feature vectors based on the yaw angles; the weighting module is used for weighting the first yaw angle feature vectors by taking the first feature values of all the positions in the first distance feature vector as weights to obtain a plurality of second yaw angle feature vectors; and the third processing module is used for acquiring yaw angle estimated values of the target wind generating set at the current time point based on the second yaw angle feature vectors.

In one implementation, the first processing module is specifically configured to: acquiring a distance input vector based on the plurality of distance values; and inputting the distance input vector to a sequence encoder to obtain the first distance characteristic vector.

In an alternative implementation, the sequence encoder includes a full-concatenated layer, a one-dimensional convolution layer, and a normalized index Softmax activation function, and the first processing module is specifically configured to: inputting the distance input vector into the full connection layer to obtain high-dimensional implicit features of second feature values of all positions in the distance input vector; inputting the distance input vector into the one-dimensional convolution layer to obtain high-dimensional implicit correlation features among the second feature values; acquiring a second distance feature vector based on the high-dimensional implicit feature and the high-dimensional implicit associated feature; and inputting the second distance feature vector into the Softmax activation function to obtain the first distance feature vector.

Optionally, the formula for acquiring the high-dimensional implicit feature is:

In one implementation, the second processing module is specifically configured to: arranging the yaw angles to obtain a plurality of yaw angle input vectors; inputting the plurality of yaw angle input vectors to a multi-scale neighborhood feature extractor to obtain the plurality of first yaw angle feature vectors.

In an alternative implementation, the multi-scale neighborhood feature extractor includes a first convolution layer and a second convolution layer, and the second processing module is specifically configured to: inputting the yaw angle input vectors into the first convolution layer to obtain a plurality of yaw angle feature vectors of a first neighborhood scale, wherein the first convolution layer has a first one-dimensional convolution kernel of a first length; inputting the yaw angle input vectors into the second convolution layer to obtain a plurality of yaw angle feature vectors with a second neighborhood scale, wherein the second convolution layer has a second one-dimensional convolution kernel with a second length; and cascading the yaw angle feature vectors of the first neighborhood scale and the yaw angle feature vectors of the second neighborhood scale to obtain the yaw angle feature vectors of the first neighborhood scale.

In one implementation, the third processing module is specifically configured to: obtaining a plurality of local metric factors based on context statistics for the plurality of second yaw angle feature vectors; weighting the second yaw angle feature vectors by taking the local measurement factors as weights to obtain third yaw angle feature vectors; two-dimensionally arranging the plurality of third yaw angle feature vectors to obtain a feature matrix; inputting the feature matrix into a convolution feature network to obtain a decoding feature matrix; wherein the convolutional feature network has learned the ability to generate the decoding feature matrix based on the feature matrix; inputting the decoding characteristic matrix into a decoder to obtain the yaw angle estimated value.

In a third aspect, an embodiment of the present application provides an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of estimating yaw angle of a wind park according to the first aspect.

In a fourth aspect, an embodiment of the present application provides a computer readable storage medium storing instructions that, when executed, cause a method as described in the first aspect to be implemented.

In a fifth aspect, an embodiment of the application provides a computer program product comprising a computer program which, when being executed by a processor, implements the steps of the yaw angle estimation method of a wind park according to the first aspect.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the application or to delineate the scope of the application. Other features of the present application will become apparent from the description that follows.

Drawings

The drawings are included to provide a better understanding of the present application and are not to be construed as limiting the application. Wherein:

FIG. 1 is a schematic diagram of a yaw angle estimation method for a wind turbine generator system according to an embodiment of the present application;

FIG. 2 is a schematic diagram of another yaw angle estimation method for a wind turbine generator system according to an embodiment of the present application;

FIG. 3 is a schematic diagram of a yaw angle estimation method for a wind turbine generator system according to an embodiment of the present application;

FIG. 4 is a schematic diagram of a yaw angle estimation method for a wind turbine generator system according to an embodiment of the present application;

FIG. 5 is an application scenario diagram of a yaw angle estimation system of a wind turbine generator system provided by an embodiment of the present application;

FIG. 6 is a system architecture diagram of a yaw angle estimation system for a wind turbine generator set according to an embodiment of the present application;

FIG. 7 is a schematic diagram of a yaw angle estimation system of a wind turbine generator system according to an embodiment of the present application;

FIG. 8 is a schematic block diagram of an example electronic device that may be used to implement embodiments of the present application.

Detailed Description

Exemplary embodiments of the present application will now be described with reference to the accompanying drawings, in which various details of the embodiments of the present application are included to facilitate understanding, and are to be considered merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the application. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

Wherein, in the description of the present application, "/" means or is meant unless otherwise indicated, for example, a/B may represent a or B; "and/or" herein is merely an association relationship describing an association object, and means that three relationships may exist, for example, a and/or B may mean: a exists alone, A and B exist together, and B exists alone. The first, second, etc. numbers referred to in the present application are merely for convenience of description and are not intended to limit the scope of the embodiments of the present application, nor represent the sequence.

Referring to fig. 1, fig. 1 is a schematic diagram of a yaw angle estimation method of a wind turbine generator set according to an embodiment of the present application. As shown in fig. 1, the method may include, but is not limited to, the steps of:

Step S101: a plurality of distance values between a target wind turbine in the wind power plant and a plurality of other wind turbine sets are obtained.

For example, a distance value corresponding to each of the target wind turbine generator set and the other wind turbine generator sets in the wind power plant may be obtained by manual measurement, so as to obtain a plurality of distance values between the target wind turbine generator set and the other wind turbine generator sets.

In an embodiment of the present application, the target wind turbine generator may be a wind turbine generator that cannot acquire a wind direction bearing itself.

Step S102: a plurality of yaw angles of a plurality of wind turbine generator sets at a plurality of predetermined points in time are acquired.

Wherein, in the embodiment of the present application, the predetermined time point is a time point (for example, a plurality of time points spaced by 10 minutes) within a predetermined period of time (for example, 1 hour) before the current time point.

For example, the yaw angles of each wind turbine generator set at a plurality of predetermined time points may be obtained by angle sensors preset on the wind turbine generator sets, so as to obtain the yaw angles of the wind turbine generator sets at the plurality of predetermined time points.

Step S103: a first distance feature vector is obtained based on the plurality of distance values.

For example, feature extraction is performed on the distance values between the target wind generating set and each other wind generating set to obtain feature values corresponding to each other wind generating set, and the feature values are arranged to be a first distance feature vector.

Step S104: a plurality of first yaw angle feature vectors are acquired based on the plurality of yaw angles.

For example, feature extraction is performed on a plurality of yaw angles corresponding to each wind generating set, so as to obtain a first yaw angle feature vector corresponding to each wind generating set, thereby obtaining a plurality of first yaw angle feature vectors.

Step S105: and taking the first characteristic value of each position in the first distance characteristic vector as a weight, and weighting the plurality of first yaw angle characteristic vectors to obtain a plurality of second yaw angle characteristic vectors.

For example, the feature value corresponding to each wind generating set is used as a weight, and the first yaw angle feature vector corresponding to the same wind generating set is weighted to obtain a second yaw angle feature vector, so as to obtain a plurality of second yaw angle feature vectors.

Step S106: and acquiring yaw angle estimated values of the target wind generating set at the current time point based on the plurality of second yaw angle feature vectors.

For example, a plurality of second yaw angle feature vectors may be input to a pre-trained neural network model to obtain yaw angle estimates for the target wind turbine generator set at a current point in time.

By implementing the embodiment of the application, the yaw angles of the target wind generating set can be estimated according to the time sequence dynamic correlation characteristics among the yaw angles of the wind generating sets at a plurality of preset time points, and the first distance characteristic vector is obtained based on a plurality of distance values between the target wind generating set and other wind generating sets, so that the weighting processing is carried out based on the first distance characteristic vector, the accuracy of the estimation result is improved, and the normal operation of the target wind generating set is ensured.

In one implementation of the embodiment of the present application, a distance input vector may be obtained based on a plurality of distance values, and the distance input vector is input to a sequence encoder, so as to obtain a first distance feature vector for performing subsequent steps. As an example, please refer to fig. 2, fig. 2 is a schematic diagram of another yaw angle estimation method of a wind turbine generator system according to an embodiment of the present application. As shown in fig. 2, the method may include, but is not limited to, the steps of:

Step S201: a plurality of distance values between a target wind turbine in the wind power plant and a plurality of other wind turbine sets are obtained.

In the embodiment of the present application, step S201 may be implemented in any manner of each embodiment of the present application, which is not limited to this embodiment, and is not described in detail.

Step S202: a plurality of yaw angles of a plurality of wind turbine generator sets at a plurality of predetermined points in time are acquired.

In the embodiment of the present application, step S202 may be implemented in any manner of each embodiment of the present application, which is not limited to this embodiment, and is not described in detail.

Step S203: a distance input vector is obtained based on the plurality of distance values.

For example, a plurality of distance values are arranged to obtain a distance input vector.

It can be understood that due to the influence of factors such as wake effect and topography effect, the real-time wind directions born by different wind power generating sets in the same wind power plant are different, and when the yaw angle of the target wind power generating set is estimated, the yaw angles of other wind power generating sets which are closer to the target wind power generating set are closer to the yaw angle of the target wind power generating set. The more distant the yaw angle of the other wind turbine generator sets from the target wind turbine generator set is, the larger the yaw angle difference from the target wind turbine generator set is. Therefore, in the technical scheme of the application, the effect of prediction is optimized by utilizing the characteristics by introducing information characteristics related to the distance.

Step S204: the distance input vector is input to a sequence encoder to obtain a first distance feature vector.

In an alternative implementation manner, the sequence encoder includes a full-connection layer, a one-dimensional convolution layer and a normalized index Softmax activation function, where the distance input vector is input to the sequence encoder to obtain a first distance feature vector, and the method may include the following steps: inputting the distance input vector into the full connection layer to obtain high-dimensional implicit features of second feature values of all positions in the distance input vector; inputting the distance input vector into a one-dimensional convolution layer to obtain high-dimensional implicit correlation features among second feature values; acquiring a second distance feature vector based on the high-dimensional implicit correlation feature and the high-dimensional implicit correlation feature; the second distance feature vector is input into a Softmax activation function to obtain a first distance feature vector.

Wherein, in an embodiment of the present application, the Softmax activation function is the last layer activation function of the sequence encoder.

It can be appreciated that the embodiment of the present application activates a function by Softmax to map the feature values of each position in the second distance feature vector into the probability space of [0,1], so as to facilitate the subsequent weighting process.

Optionally, the formula for obtaining the high-dimensional implicit feature is:

Wherein Y is a high-dimensional implicit feature, X is a distance input vector, W is a preset weight matrix, B is a bias vector, Representing a matrix multiplication; the acquisition formula of the high-dimensional implicit association features is as follows:

wherein a is the width of the convolution kernel of the one-dimensional convolution layer in the X direction, F (a) is a convolution kernel parameter vector, G (X-a) is a local vector matrix calculated by a convolution kernel function, w is the size of the convolution kernel, and X is a distance input vector.

Step S205: a plurality of first yaw angle feature vectors are acquired based on the plurality of yaw angles.

In the embodiment of the present application, step S205 may be implemented in any manner of each embodiment of the present application, which is not limited to this embodiment, and is not described in detail.

Step S206: and taking the first characteristic value of each position in the first distance characteristic vector as a weight, and weighting the plurality of first yaw angle characteristic vectors to obtain a plurality of second yaw angle characteristic vectors.

In the embodiment of the present application, step S206 may be implemented in any manner of each embodiment of the present application, which is not limited to this embodiment, and is not described in detail.

Step S207: and acquiring yaw angle estimated values of the target wind generating set at the current time point based on the plurality of second yaw angle feature vectors.

In the embodiment of the present application, step S207 may be implemented in any manner of each embodiment of the present application, which is not limited to this embodiment, and is not described in detail.

By implementing the embodiment of the application, a multi-scale neighborhood feature extractor can be used for extracting a plurality of first yaw angle feature vectors from a plurality of yaw angles, so that time sequence dynamic association features of a plurality of wind generating sets among the plurality of yaw angles at a plurality of preset time points are obtained based on the first yaw angle feature vectors, so as to estimate the yaw angle of a target wind generating set, and a first distance feature vector is obtained based on a plurality of distance values between the target wind generating set and other wind generating sets, so that weighting processing is carried out based on the first distance feature vectors, the accuracy of the yaw angle estimation result of the target wind generating set is improved, and the normal operation of the target wind generating set is ensured.

In one implementation of the embodiment of the present application, a plurality of yaw angle input vectors may be obtained based on a plurality of yaw angles, and a plurality of first yaw angle feature vectors may be extracted from the plurality of yaw angle input vectors using a multi-scale neighborhood feature extractor. As an example, please refer to fig. 3, fig. 3 is a schematic diagram of a yaw angle estimation method of a wind turbine generator system according to another embodiment of the present application. As shown in fig. 3, the method may include, but is not limited to, the steps of:

step S301: a plurality of distance values between a target wind turbine in the wind power plant and a plurality of other wind turbine sets are obtained.

In the embodiment of the present application, step S301 may be implemented in any manner of each embodiment of the present application, which is not limited to this embodiment, and is not described in detail.

Step S302: a plurality of yaw angles of a plurality of wind turbine generator sets at a plurality of predetermined points in time are acquired.

In the embodiment of the present application, step S302 may be implemented in any manner of each embodiment of the present application, which is not limited to this embodiment, and is not described in detail.

Step S303: a first distance feature vector is obtained based on the plurality of distance values.

In the embodiment of the present application, step S303 may be implemented in any manner of each embodiment of the present application, which is not limited to this embodiment, and is not described in detail.

Step S304: and arranging the yaw angles to obtain a plurality of yaw angle input vectors.

For example, a plurality of yaw angles of a plurality of wind turbine generator sets at each preset time point are arranged to obtain a yaw angle input vector, so as to obtain a plurality of yaw angle input vectors.

Step S305: a plurality of yaw angle input vectors are input to a multi-scale neighborhood feature extractor to obtain a plurality of first yaw angle feature vectors.

For example, a plurality of yaw angle input vectors are input to a multi-scale neighborhood feature extractor to extract associated features of a yaw angle multi-scale neighborhood of the plurality of yaw angle input vectors over different time spans using the multi-scale neighborhood feature extractor as a feature extractor, thereby obtaining a plurality of first yaw angle feature vectors.

In an alternative implementation, the multi-scale neighborhood feature extractor includes a first convolution layer and a second convolution layer, where the inputting the plurality of yaw angle input vectors into the multi-scale neighborhood feature extractor obtains a plurality of first yaw angle feature vectors may include the following steps: inputting a plurality of yaw angle input vectors into a first convolution layer to obtain a plurality of yaw angle feature vectors of a first neighborhood scale, wherein the first convolution layer has a first one-dimensional convolution kernel of a first length; inputting a plurality of yaw angle input vectors into a second convolution layer to obtain a plurality of yaw angle feature vectors of a second neighborhood scale, wherein the second convolution layer is provided with a second one-dimensional convolution kernel of a second length; and cascading the plurality of first neighborhood scale yaw angle feature vectors and the plurality of second neighborhood scale yaw angle feature vectors to obtain a plurality of first yaw angle feature vectors.

For example, each yaw angle input vector is input into a first convolution layer to obtain a corresponding yaw angle feature vector of a first neighborhood scale; respectively inputting each yaw angle input vector into a second convolution layer to obtain a corresponding yaw angle feature vector of a second neighborhood scale; and cascading the first neighborhood scale yaw angle feature vector and the second neighborhood scale yaw angle feature vector corresponding to the same yaw angle input vector to obtain a corresponding first yaw angle feature vector, thereby obtaining a plurality of first yaw angle feature vectors.

wherein a is the width of the first one-dimensional convolution kernel in the X direction, F (a) is a parameter vector of the first one-dimensional convolution kernel, G (X-a) is a first local vector matrix calculated by a convolution kernel function, w is the size of the first one-dimensional convolution kernel, and X is a yaw angle input vector; the acquisition formula of the yaw angle feature vector of the second neighborhood scale is as follows:

Wherein b is the width of the second one-dimensional convolution kernel in the X direction, F (b) is a parameter vector of the second one-dimensional convolution kernel, G (X-b) is a second local vector matrix calculated by a convolution kernel function, m is the size of the second one-dimensional convolution kernel, and X is a yaw angle input vector.

Step S306: and taking the first characteristic value of each position in the first distance characteristic vector as a weight, and weighting the plurality of first yaw angle characteristic vectors to obtain a plurality of second yaw angle characteristic vectors.

In the embodiment of the present application, step S306 may be implemented in any manner of each embodiment of the present application, which is not limited to this embodiment, and is not described in detail.

Step S307: and acquiring yaw angle estimated values of the target wind generating set at the current time point based on the plurality of second yaw angle feature vectors.

In the embodiment of the present application, step S307 may be implemented in any manner of each embodiment of the present application, which is not limited to this embodiment, and is not described in detail.

By implementing the embodiment of the application, the yaw angles of the target wind generating set can be estimated according to the time sequence dynamic association characteristics among the yaw angles of the wind generating sets at a plurality of preset time points, and the first distance characteristic vector is obtained from the distance input vectors obtained from a plurality of distance values based on the sequence encoder, so that the weighting processing is carried out based on the first distance characteristic vector, a plurality of second yaw angle characteristic vectors are obtained, the accuracy of the estimation result is improved, and the normal operation of the target wind generating set is ensured.

In one implementation manner of the embodiment of the present application, a feature matrix may be obtained based on a plurality of second yaw angle feature vectors, and the capability of decoding the feature matrix to obtain a decoded feature matrix by using a convolution feature network, and then the decoded feature matrix is input to a decoder to obtain a yaw angle estimation value. As an example, please refer to fig. 4, fig. 4 is a schematic diagram of a yaw angle estimation method of a wind turbine generator system according to an embodiment of the present application. As shown in fig. 4, the method may include, but is not limited to, the steps of:

Step S401: a plurality of distance values between a target wind turbine in the wind power plant and a plurality of other wind turbine sets are obtained.

In the embodiment of the present application, step S401 may be implemented in any manner of each embodiment of the present application, which is not limited to this embodiment, and is not described in detail.

Step S402: a plurality of yaw angles of a plurality of wind turbine generator sets at a plurality of predetermined points in time are acquired.

In the embodiment of the present application, step S402 may be implemented in any manner of each embodiment of the present application, which is not limited to this embodiment, and is not described in detail.

Step S403: a first distance feature vector is obtained based on the plurality of distance values.

In the embodiment of the present application, step S403 may be implemented in any manner of each embodiment of the present application, which is not limited to this embodiment, and is not described in detail.

Step S404: a plurality of first yaw angle feature vectors are acquired based on the plurality of yaw angles.

In the embodiment of the present application, step S404 may be implemented in any manner of each embodiment of the present application, which is not limited to this embodiment, and is not described in detail.

Step S405: and taking the first characteristic value of each position in the first distance characteristic vector as a weight, and weighting the plurality of first yaw angle characteristic vectors to obtain a plurality of second yaw angle characteristic vectors.

In the embodiment of the present application, step S405 may be implemented in any manner of each embodiment of the present application, which is not limited to this embodiment, and is not described in detail.

Step S406: a plurality of local metric factors based on context statistics for a plurality of second yaw angle feature vectors are obtained.

For example, a local metric factor of the context-based statistics for each second yaw angle feature vector is calculated using a calculation formula for the local metric factor, thereby obtaining a plurality of local metric factors of the context-based statistics for a plurality of second yaw angle feature vectors.

In an alternative implementation, the local metric factor is calculated as:

Wherein v _i is a plurality of third eigenvalues at each position of the second yaw angle eigenvector, σ is a mean value of a set of the plurality of third eigenvalues, σ is a variance of the set of the plurality of third eigenvalues, L is a length of the second yaw angle eigenvector, exp () represents an exponential operation.

Step S407: and weighting the second yaw angle feature vectors by taking the local measurement factors as weights to obtain third yaw angle feature vectors.

For example, the local measurement factors corresponding to each second yaw angle feature vector are respectively used as corresponding weights, and the second yaw angle feature vectors are weighted to obtain a plurality of third yaw angle feature vectors.

Step S408: and two-dimensionally arranging the plurality of third yaw angle feature vectors to obtain a feature matrix.

Step S409: and inputting the feature matrix into a convolution feature network to obtain a decoding feature matrix.

In the embodiment of the application, the convolution feature network has learned the capability of generating a decoding feature matrix based on the feature matrix.

For example, inputting the feature matrix into a convolution feature network to perform convolution processing on the feature matrix to obtain a convolution feature map; pooling the convolution feature map along the channel dimension to obtain a pooled feature map; ; non-linear activation is carried out on the pooled feature map so as to obtain an activated feature map; so that a decoding feature matrix can be derived based on the activation feature map.

It may be appreciated that, in the embodiment of the present application, based on the cross-sample arrangement of the second yaw angle feature vectors, each second yaw angle feature vector is regarded as a separate feature descriptor of a cross-sample dimension as a pressed representation of a cross-sample dimension scene of the second yaw angle feature vector, and the local scene metric factor of the context statistics is calculated for each second yaw angle feature vector, and the local scene metric factor is used to weight the plurality of second yaw angle feature vectors, so that the global context correlation between the weighted third yaw angle feature vectors is improved, thereby improving the decoding accuracy of the decoding feature matrix.

Step S410: the decoding feature matrix is input into a decoder to obtain a yaw angle estimation value.

It can be understood that, in the embodiment of the present application, the decoded feature matrix is input into the convolution feature network serving as the feature extractor to perform feature extraction, so as to extract the implicit correlation feature of the distances between the target wind turbine generator and each wind turbine generator set in the other wind turbine generator sets, thereby obtaining the decoded feature matrix, and the decoded feature matrix is decoded to obtain the decoded value for representing the yaw angle estimated value of the target wind turbine generator set at the current time point. And expressing the correlation characteristic between the target wind generating set and each other wind generating set by utilizing the distance characteristic between the target wind generating set and the other wind generating sets, so as to optimize the yaw angle estimation accuracy of the target wind generating set.

In an alternative implementation, the yaw angle estimate is obtained by the formula:

Wherein Y is a yaw angle estimated value, W is a preset weight matrix, X is a decoding feature matrix, Representing a matrix multiplication.

By implementing the embodiment of the application, the yaw angles of the target wind generating set can be estimated according to the time sequence dynamic association characteristics among the yaw angles of the wind generating sets at a plurality of preset time points, the second yaw angle characteristic vectors are weighted to obtain the third yaw angle characteristic vectors, and the global context association among the third yaw angle characteristic vectors is improved, so that the accuracy of the yaw angle estimation result of the target wind generating set is improved, and the normal operation of the target wind generating set is ensured.

Referring to the drawings, fig. 5 is an application scenario diagram of a yaw angle estimation system of a wind turbine generator system according to an embodiment of the present application. As shown in fig. 5, in this application scenario, the actual yaw angles of other wind power generation sets (e.g., H2-Hn as illustrated in fig. 5) located in the same power generation field as the target wind power generation set (e.g., H1 as illustrated in fig. 5) at a plurality of predetermined time points within a predetermined period are acquired by a plurality of angle sensors (e.g., A1-An as illustrated in fig. 5), and the distances between the target wind power generation set and each of the other wind power generation sets are acquired. The above data is then input into a server (e.g. S as in fig. 5) deployed with a yaw angle estimation algorithm for the wind turbine, wherein the server is capable of processing the input data in the provided yaw angle estimation method of the wind turbine according to any of the embodiments of the present application to generate a decoded value representing the yaw angle estimate of the target wind turbine at the current point in time.

Referring to fig. 6, fig. 6 is a system architecture diagram of a yaw angle estimation system of a wind turbine generator system according to an embodiment of the present application. As shown in fig. 6, in the yaw angle estimation system of the wind turbine generator system according to the embodiment of the present application, first, actual yaw angles of a plurality of other wind turbine generator systems located in the same power generation plant as a target wind turbine generator system at a plurality of predetermined time points in a predetermined time period are obtained; then, respectively arranging the obtained actual yaw angles of a plurality of preset time points into input vectors, and then obtaining a plurality of yaw angle feature vectors through a multi-scale neighborhood feature extraction module; the distance between the target wind generating set and each wind generating set in the other wind generating sets is obtained, so that a distance input vector is obtained; then, the obtained distance input vector passes through a sequence encoder comprising a one-dimensional convolution layer to obtain a distance characteristic vector, wherein the activation function of the last layer of the sequence encoder is a Softmax activation function; respectively weighting the corresponding yaw angle feature vectors by taking the feature values of each position in the obtained distance feature vectors as weights so as to obtain a plurality of weighted yaw angle feature vectors; calculating local measurement factors based on context statistics of each weighted yaw angle feature vector in the weighted yaw angle feature vectors, and weighting each weighted yaw angle feature vector by taking the local measurement factors based on context statistics of each weighted yaw angle feature vector as weights to obtain a plurality of re-weighted yaw angle feature vectors; two-dimensionally arranging the obtained yaw angle feature vectors after re-weighting into a feature matrix, and then obtaining a decoding feature matrix through a convolution feature network serving as a feature extractor; and finally, the decoding characteristic matrix passes through a decoder to obtain a decoding value for representing the yaw angle estimated value of the target wind generating set at the current time point.

Referring to fig. 7, fig. 7 is a schematic diagram of a yaw angle estimation system of a wind turbine generator set according to an embodiment of the present application. As shown in fig. 7, the system 700 includes: a first obtaining module 701, configured to obtain a plurality of distance values between a target wind turbine in a wind farm and a plurality of other wind turbine; a second obtaining module 702, configured to obtain a plurality of yaw angles of a plurality of wind generating sets at a plurality of predetermined time points, and a first processing module 703, configured to obtain a first distance feature vector based on a plurality of distance values; a second processing module 704, configured to obtain a plurality of first yaw angle feature vectors based on a plurality of yaw angles; the weighting module 705 is configured to weight the plurality of first yaw angle feature vectors with the first feature values of each position in the first distance feature vector as weights, to obtain a plurality of second yaw angle feature vectors; the third processing module 706 obtains yaw angle estimates of the target wind turbine generator set at the current point in time based on the plurality of second yaw angle feature vectors.

In one implementation, the first processing module 703 is specifically configured to: acquiring a distance input vector based on a plurality of distance values; the distance input vector is input to a sequence encoder to obtain a first distance feature vector.

In an alternative implementation, the sequence encoder includes a full-concatenated layer, a one-dimensional convolution layer, and a normalized index Softmax activation function, and the first processing module 703 is specifically configured to: inputting the distance input vector into the full connection layer to obtain high-dimensional implicit features of second feature values of all positions in the distance input vector; inputting the distance input vector into a one-dimensional convolution layer to obtain high-dimensional implicit correlation features among second feature values; acquiring a second distance feature vector based on the high-dimensional implicit correlation feature and the high-dimensional implicit correlation feature; the second distance feature vector is input into a Softmax activation function to obtain a first distance feature vector.

Optionally, the formula for obtaining the high-dimensional implicit feature is:

In one implementation, the second processing module 704 is specifically configured to: obtaining a plurality of yaw angle input vectors based on the plurality of yaw angles; a plurality of yaw angle input vectors are input to a multi-scale neighborhood feature extractor to obtain a plurality of first yaw angle feature vectors.

In an alternative implementation, the multi-scale neighborhood feature extractor includes a first convolution layer and a second convolution layer, and the second processing module 704 is specifically configured to: inputting a plurality of yaw angle input vectors into a first convolution layer to obtain a plurality of yaw angle feature vectors of a first neighborhood scale, wherein the first convolution layer has a first one-dimensional convolution kernel of a first length; inputting a plurality of yaw angle input vectors into a second convolution layer to obtain a plurality of yaw angle feature vectors of a second neighborhood scale, wherein the second convolution layer is provided with a second one-dimensional convolution kernel of a second length; and cascading the plurality of first neighborhood scale yaw angle feature vectors and the plurality of second neighborhood scale yaw angle feature vectors to obtain a plurality of first yaw angle feature vectors.

/>

In one implementation, the third processing module 706 is specifically configured to: obtaining a plurality of local metric factors based on context statistics for a plurality of second yaw angle feature vectors; weighting the second yaw angle feature vectors by taking the local measurement factors as weights to obtain third yaw angle feature vectors; two-dimensional arrangement is carried out on a plurality of third yaw angle feature vectors to obtain a feature matrix; inputting the feature matrix into a convolution feature network to obtain a decoding feature matrix; the convolution feature network has learned to obtain the capability of generating a decoding feature matrix based on the feature matrix; the decoding feature matrix is input into a decoder to obtain a yaw angle estimation value.

In an alternative implementation, the local metric factor is calculated as:

According to the device provided by the embodiment of the application, the yaw angles of the target wind generating sets can be estimated according to the time sequence dynamic correlation characteristics among the yaw angles of the wind generating sets at a plurality of preset time points, and the first distance characteristic vector is obtained based on a plurality of distance values between the target wind generating sets and other wind generating sets, so that the weighting processing is carried out based on the first distance characteristic vector, the accuracy of the estimation result is improved, and the normal operation of the target wind generating sets is ensured.

Based on the embodiments of the present application, the present application also provides a computer-readable storage medium, in which computer instructions are used to make a computer execute the yaw angle estimation method of the wind turbine generator set according to any of the foregoing embodiments provided by the embodiments of the present application.

Referring now to fig. 8, shown in fig. 8 is a schematic block diagram of an example electronic device that may be used to implement an embodiment of the present application. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the applications described and/or claimed herein.

As shown in fig. 8, the apparatus 800 includes a computing unit 801 that can perform various appropriate actions and processes according to a computer program stored in a Read-Only Memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access Memory (Random Access Memory, RAM) 803. In the RAM 803, various programs and data required for the operation of the device 800 can also be stored. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other by a bus 804. An Input/Output (I/O) interface 805 is also connected to bus 804.

Various components in device 800 are connected to I/O interface 805, including: an input unit 806 such as a keyboard, mouse, etc.; an output unit 807 such as various types of displays, speakers, and the like; a storage unit 808, such as a magnetic disk, optical disk, etc.; and a communication unit 809, such as a network card, modem, wireless communication transceiver, or the like. The communication unit 809 allows the device 800 to exchange information/data with other devices via a computer network such as the internet and/or various telecommunication networks.

The computing unit 801 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 801 include, but are not limited to, a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU), various dedicated Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DIGITAL SIGNAL processes, DSPs), and any suitable processors, controllers, microcontrollers, etc. The calculation unit 801 performs the above-described respective methods and processes, such as a yaw angle estimation method of a wind turbine. For example, in some embodiments, the yaw angle estimation method of the wind park may be implemented as a computer software program, which is tangibly embodied on a machine-readable medium, such as the storage unit 808. In some embodiments, part or all of the computer program may be loaded and/or installed onto device 800 via ROM 802 and/or communication unit 809. When the computer program is loaded into the RAM 803 and executed by the computing unit 801, one or more steps of the yaw angle estimation method of a wind turbine generator set described above may be performed. Alternatively, in other embodiments, the computing unit 801 may be configured to perform the yaw angle estimation method of the wind park by any other suitable means (e.g. by means of firmware).

Various implementations of the systems and techniques described here above can be implemented in digital electronic circuitry, integrated circuitry, field programmable gate arrays (Field Programmable GATE ARRAY, FPGA), application specific integrated circuits (Application SpecificIntegrated Circuit, ASIC), application specific standard products (Application SPECIFIC STANDARD PARTS, ASSP), system On Chip (SOC), load programmable logic devices (Complex Programmable Logic Device, CPLD), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present application may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of the present application, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-Only Memory (ROM), an erasable programmable read-Only Memory (EPROM) or flash Memory, an optical fiber, a portable compact disc read-Only Memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a Cathode Ray Tube (CRT) or an LCD (Liquid CRYSTAL DISPLAY) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local area networks (Local Area Network, LANs), wide area networks (Wide Area Network, WANs), the internet, and blockchain networks.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so as to solve the defects of high management difficulty and weak service expansibility in the traditional physical host and Virtual private server (VPS PRIVATE SERVER) service. The server may also be a server of a distributed system or a server that incorporates a blockchain.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present application may be performed in parallel, sequentially, or in a different order, provided that the desired results of the technical solution of the present application are achieved, and are not limited herein.

The above embodiments do not limit the scope of the present application. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the scope of the present application.

Claims

1. A yaw angle estimation method of a wind turbine generator system, comprising:

acquiring a plurality of distance values between a target wind generating set and other wind generating sets in a wind power plant;

acquiring a plurality of yaw angles of the wind generating sets at a plurality of preset time points;

acquiring a first distance feature vector based on the plurality of distance values;

acquiring a plurality of first yaw angle feature vectors based on the plurality of yaw angles;

Taking the first characteristic value of each position in the first distance characteristic vector as a weight, and weighting the plurality of first yaw angle characteristic vectors to obtain a plurality of second yaw angle characteristic vectors;

And acquiring yaw angle estimated values of the target wind generating set at the current time point based on the second yaw angle feature vectors.

2. The method of claim 1, wherein the obtaining a first distance feature vector based on the plurality of distance values comprises:

acquiring a distance input vector based on the plurality of distance values;

and inputting the distance input vector to a sequence encoder to obtain the first distance characteristic vector.

3. The method of claim 2, wherein the sequence encoder comprises a full-concatenated layer, a one-dimensional convolutional layer, and a normalized exponential Softmax activation function, wherein the inputting the distance input vector to the sequence encoder to obtain the first distance feature vector comprises:

inputting the distance input vector into the full connection layer to obtain high-dimensional implicit features of second feature values of all positions in the distance input vector;

Inputting the distance input vector into the one-dimensional convolution layer to obtain high-dimensional implicit correlation features among the second feature values;

Acquiring a second distance feature vector based on the high-dimensional implicit feature and the high-dimensional implicit associated feature;

and inputting the second distance feature vector into the Softmax activation function to obtain the first distance feature vector.

4. The method of claim 3, wherein the acquisition formula for the high-dimensional implicit feature is:

Wherein Y is the high-dimensional implicit characteristic, W is a preset weight matrix, X is the distance input vector and is a preset offset vector, Representing a matrix multiplication;

The acquisition formula of the high-dimensional implicit association features is as follows:

Wherein a is the width of the convolution kernel of the one-dimensional convolution layer in the X direction, F (a) is a convolution kernel parameter vector, G (X-a) is a local vector matrix operated by a convolution kernel function, w is the size of the convolution kernel, and X is the distance input vector.

5. The method of claim 1, wherein the obtaining a plurality of first yaw angle feature vectors based on the plurality of yaw angles comprises:

arranging the yaw angles to obtain a plurality of yaw angle input vectors;

inputting the plurality of yaw angle input vectors to a multi-scale neighborhood feature extractor to obtain the plurality of first yaw angle feature vectors.

6. The method of claim 5, wherein the multi-scale neighborhood feature extractor comprises a first convolution layer and a second convolution layer, wherein the inputting the plurality of yaw angle input vectors into the multi-scale neighborhood feature extractor to obtain the plurality of first yaw angle feature vectors comprises:

Inputting the yaw angle input vectors into the first convolution layer to obtain a plurality of yaw angle feature vectors of a first neighborhood scale, wherein the first convolution layer has a first one-dimensional convolution kernel of a first length;

inputting the yaw angle input vectors into the second convolution layer to obtain a plurality of yaw angle feature vectors with a second neighborhood scale, wherein the second convolution layer has a second one-dimensional convolution kernel with a second length;

and cascading the yaw angle feature vectors of the first neighborhood scale and the yaw angle feature vectors of the second neighborhood scale to obtain the yaw angle feature vectors of the first neighborhood scale.

7. The method of claim 6, wherein the first neighborhood scale yaw angle feature vector is obtained by the formula:

Wherein a is the width of the first one-dimensional convolution kernel in the X direction, F (a) is a parameter vector of the first one-dimensional convolution kernel, G (X-a) is a first local vector matrix operated with a convolution kernel function, w is the size of the first one-dimensional convolution kernel, and X is the yaw angle input vector;

the obtaining formula of the yaw angle characteristic vector of the second neighborhood scale is as follows:

8. The method of claim 1, wherein the obtaining yaw angle estimates for the target wind turbine generator set at the current point in time based on the plurality of second yaw angle feature vectors comprises:

Obtaining a plurality of local metric factors based on context statistics for the plurality of second yaw angle feature vectors;

Weighting the second yaw angle feature vectors by taking the local measurement factors as weights to obtain third yaw angle feature vectors;

Two-dimensionally arranging the plurality of third yaw angle feature vectors to obtain a feature matrix;

Inputting the feature matrix into a convolution feature network to obtain a decoding feature matrix; wherein the convolutional feature network has learned the ability to generate the decoding feature matrix based on the feature matrix;

inputting the decoding characteristic matrix into a decoder to obtain the yaw angle estimated value.

9. The method of claim 8, wherein the local metric factor is calculated as:

10. The method of claim 8, wherein the yaw angle estimate is obtained by the formula:

11. A yaw angle estimation system of a wind turbine generator system, comprising:

The first acquisition module is used for acquiring a plurality of distance values between a target wind generating set and other wind generating sets in the wind power plant;

the second acquisition module is used for acquiring a plurality of yaw angles of the wind generating sets at a plurality of preset time points;

The first processing module is used for acquiring a first distance characteristic vector based on the plurality of distance values;

The second processing module is used for acquiring a plurality of first yaw angle feature vectors based on the yaw angles;

The weighting module is used for weighting the first yaw angle feature vectors by taking the first feature values of all the positions in the first distance feature vector as weights to obtain a plurality of second yaw angle feature vectors;

and the third processing module is used for acquiring yaw angle estimated values of the target wind generating set at the current time point based on the second yaw angle feature vectors.

12. An electronic device, comprising:

At least one processor;

and a memory communicatively coupled to the at least one processor;

Wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of yaw angle estimation of a wind park according to any one of claims 1 to 10.

13. A computer readable storage medium storing instructions which, when executed, cause the method of any one of claims 1 to 10 to be implemented.