Cooperative control analysis system and method of distributed wind power generation system

Abstract

The invention relates to the field of wind power generation control, and particularly discloses a cooperative control analysis system and method of a distributed wind power generation system. The artificial intelligence technology based on the deep learning field is used for calculating wind intensity values of a plurality of preset time points in a preset time period of the wind power plant, wind frequency values of the plurality of preset time points in the preset time period and wind frequency values of the plurality of preset time points in the preset time period. And performing feature extraction and coding on the wind speed values at the plurality of preset time points in the preset time period and the wind direction values at the plurality of preset time points in the preset time period to obtain a classification result of the type of the wind driven generator which is suitable for running. Therefore, wind energy can be effectively utilized by intelligently selecting the type of the wind driven generator which runs properly, and the stability and reliability of power generation of the wind power plant are improved.

Description

Technical Field

[0001] The present application relates to the field of wind power generation control, and in particular to a coordinated control analysis system and method for a distributed wind power generation system. Background Art

[0002] A wind farm is a facility that converts wind energy into electricity. Typically, a wind farm consists of multiple large wind turbines (also called wind generators) mounted on tall towers to capture the energy of high-altitude winds. The rotation of the wind turbine blades drives the generator to generate electricity. This electricity can be transmitted to the power grid and supplied to residents, industries, and other users. Wind power is a clean, renewable form of energy with a relatively low impact on the environment, and is therefore widely used worldwide. However, wind energy is volatile, meaning that wind speed and direction vary with time and location. This volatility can affect the stability and reliability of wind power generation.

[0003] Therefore, a coordinated control and analysis system for distributed wind power generation systems is expected. Summary of the Invention

[0004] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a coordinated control analysis system and method for a distributed wind power generation system, which uses artificial intelligence technology based on the field of deep learning to extract and encode the wind intensity values at multiple predetermined time points within a predetermined time period of a wind farm, the wind frequency values at multiple predetermined time points within the predetermined time period, the wind speed values at multiple predetermined time points within the predetermined time period, and the wind direction values at multiple predetermined time points within the predetermined time period, so as to obtain classification results of suitable wind turbine types. In this way, by intelligently selecting suitable wind turbine types, wind energy can be effectively utilized and the stability and reliability of wind farm power generation can be improved.

[0005] According to one aspect of the present application, a coordinated control and analysis system for a distributed wind power generation system is provided, comprising:

[0006] a wind farm data acquisition module, configured to acquire wind intensity values at a plurality of predetermined time points within a predetermined time period of the wind farm, wind frequency values at a plurality of predetermined time points within the predetermined time period, wind speed values at a plurality of predetermined time points within the predetermined time period, and wind direction values at a plurality of predetermined time points within the predetermined time period;

[0007] a wind power generation vector construction module, configured to arrange the wind intensity values at a plurality of predetermined time points within the predetermined time period, the wind frequency values at a plurality of predetermined time points within the predetermined time period, the wind speed values at a plurality of predetermined time points within the predetermined time period, and the wind direction values at a plurality of predetermined time points within the predetermined time period into a wind intensity input vector, a wind frequency input vector, a wind speed input vector, and a wind speed input vector, respectively, according to a time dimension;

[0008] a wind power generation feature extraction module, configured to extract and encode features of the wind intensity input vector, the wind frequency input vector, the wind speed input vector, and the wind velocity input vector to obtain a two-dimensional wind power generation feature matrix;

[0009] The wind turbine type selection module is used to analyze the two-dimensional wind power generation characteristic matrix to obtain the result of the wind turbine type that is suitable for operation.

[0010] According to another aspect of the present application, a coordinated control analysis method for a distributed wind power generation system is provided, which includes:

[0011] Obtaining wind intensity values at multiple predetermined time points within a predetermined time period of the wind farm, wind frequency values at multiple predetermined time points within the predetermined time period, wind speed values at multiple predetermined time points within the predetermined time period, and wind direction values at multiple predetermined time points within the predetermined time period;

[0012] Arrange the wind intensity values at multiple predetermined time points within the predetermined time period, the wind frequency values at multiple predetermined time points within the predetermined time period, the wind speed values at multiple predetermined time points within the predetermined time period, and the wind direction values at multiple predetermined time points within the predetermined time period into a wind intensity input vector, a wind frequency input vector, a wind speed input vector, and a wind speed input vector, respectively, according to the time dimension;

[0013] Performing feature extraction and encoding on the wind intensity input vector, the wind frequency input vector, the wind speed input vector, and the wind velocity input vector to obtain a two-dimensional wind power generation feature matrix;

[0014] The two-dimensional wind power generation characteristic matrix is analyzed to obtain the results of the wind turbine type that is suitable for operation.

[0015] In summary, the coordinated control and analysis system and method for a distributed wind power generation system provided in this application utilizes deep learning-based artificial intelligence technology to extract and encode features from wind intensity values, wind frequency values, wind speed values, and wind direction values at multiple predetermined time points within a predetermined time period at a wind farm, thereby obtaining classification results for suitable wind turbine types. This intelligent selection of suitable wind turbine types effectively utilizes wind energy and improves the stability and reliability of wind farm power generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without inventive effort. In the drawings, the same reference numerals generally represent the same components or steps.

[0017] Figure 1 4 is a block diagram of a coordinated control and analysis system for a distributed wind power generation system according to an embodiment of the present application.

[0018] Figure 2 4 is a block diagram of a wind power generation feature extraction module in a coordinated control analysis system of a distributed wind power generation system according to an embodiment of the present application.

[0019] Figure 3 4 is a block diagram of an enhanced wind power generation feature extraction unit in a coordinated control analysis system of a distributed wind power generation system according to an embodiment of the present application.

[0020] Figure 4 Flowchart of a coordinated control analysis method for a distributed wind power generation system according to an embodiment of the present application. DETAILED DESCRIPTION

[0021] Below, the exemplary embodiments of the present application will be described in detail with reference to the accompanying drawings, clearly and completely describing the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0022] Figure 1 FIG is a block diagram of a coordinated control analysis system for a distributed wind power generation system according to an embodiment of the present application. Figure 1As shown, the coordinated control and analysis system 100 of the distributed wind power generation system according to the embodiment of the present application includes: a wind farm data acquisition module 110, which is used to obtain wind intensity values at multiple predetermined time points within a predetermined time period of the wind farm, wind frequency values at multiple predetermined time points within the predetermined time period, wind speed values at multiple predetermined time points within the predetermined time period, and wind direction values at multiple predetermined time points within the predetermined time period; a wind power generation vector construction module 120, which is used to construct the wind intensity values at multiple predetermined time points within the predetermined time period, wind frequency values at multiple predetermined time points within the predetermined time period, and wind direction values at multiple predetermined time points within the predetermined time period. The wind speed values at multiple predetermined time points within a predetermined time period and the wind direction values at multiple predetermined time points within the predetermined time period are respectively arranged according to the time dimension as a wind intensity input vector, a wind frequency input vector, a wind speed input vector and a wind speed input vector; a wind power generation feature extraction module 130 is used to extract and encode the wind intensity input vector, the wind frequency input vector, the wind speed input vector and the wind speed input vector to obtain a two-dimensional wind power generation feature matrix; and a wind turbine type selection module 140 is used to analyze the two-dimensional wind power generation feature matrix to obtain a result of a suitable wind turbine type for operation.

[0023] In the above-mentioned coordinated control and analysis system 100 for a distributed wind power generation system, the wind farm data acquisition module 110 is used to obtain wind intensity values at multiple predetermined time points within a predetermined time period of the wind farm, wind frequency values at multiple predetermined time points within the predetermined time period, wind speed values at multiple predetermined time points within the predetermined time period, and wind direction values at multiple predetermined time points within the predetermined time period. As mentioned in the above-mentioned background technology, wind power generation is a clean and renewable form of energy with relatively little impact on the environment, and therefore is widely used worldwide. However, wind energy is volatile, that is, wind speed and wind direction vary with time and location. This volatility may affect the stability and reliability of wind power generation. Therefore, a coordinated control and analysis system for a distributed wind power generation system is desired.

[0024] To address the above technical issues, a coordinated control and analysis system for distributed wind power generation systems has been proposed. This system uses deep learning-based artificial intelligence to extract and encode features from wind intensity, wind frequency, wind speed, and wind direction at multiple predetermined time points within a predetermined time period at a wind farm, thereby classifying suitable wind turbine types. This intelligent selection of suitable wind turbine types effectively utilizes wind energy and improves the stability and reliability of wind farm power generation.

[0025] Currently, deep learning and neural networks are widely used in fields such as computer vision, natural language processing, and speech signal processing. Furthermore, deep learning and neural networks have demonstrated capabilities approaching or even surpassing those of humans in areas such as image classification, object detection, semantic segmentation, and text translation.

[0026] In recent years, the development of deep learning and neural networks has provided new solutions and plans for the coordinated control and analysis system of distributed wind power generation systems.

[0027] Specifically, first, wind intensity values at multiple predetermined time points within a predetermined time period of the wind farm, wind frequency values at multiple predetermined time points within the predetermined time period, wind speed values at multiple predetermined time points within the predetermined time period, and wind direction values at multiple predetermined time points within the predetermined time period are obtained. By obtaining wind data at different time points of the wind farm, wind energy resources can be evaluated and analyzed, and the intensity and variation patterns of wind power within different time periods can be understood, which helps to determine the optimal generator set configuration. At the same time, different types of wind turbines have different adaptability requirements for wind speed, wind direction, and wind frequency parameters. By obtaining wind speed, wind direction, and wind frequency data at multiple time points, it is possible to select the type of wind turbine that best suits the current wind energy conditions, thereby improving power generation efficiency.

[0028] In the above-mentioned coordinated control and analysis system 100 for a distributed wind power generation system, the wind power generation vector construction module 120 is configured to arrange the wind intensity values, wind frequency values, wind speed values, and wind direction values at multiple predetermined time points within the predetermined time period into a wind intensity input vector, a wind frequency input vector, a wind speed input vector, and a wind speed input vector, respectively, according to the time dimension. To effectively organize and process wind farm data, the wind intensity values, wind frequency values, wind speed values, and wind direction values at multiple time points within the predetermined time period are arranged according to the time dimension into a wind intensity input vector, a wind frequency input vector, a wind speed input vector, and a wind speed input vector, respectively, to facilitate subsequent feature extraction and analysis. Arranging data according to the time dimension can better preserve time series information, facilitating subsequent analysis and extraction of temporal features of wind data. This helps discover characteristics such as periodicity and trends in wind data.

[0029] In the above-mentioned coordinated control and analysis system 100 of the distributed wind power generation system, the wind power generation feature extraction module 130 is used to extract and encode the wind intensity input vector, wind frequency input vector, wind speed input vector and wind speed input vector to obtain a two-dimensional wind power generation feature matrix.

[0030] Figure 2FIG. 1 is a block diagram of a wind power generation feature extraction module in a coordinated control analysis system of a distributed wind power generation system according to an embodiment of the present application. Figure 2 As shown, the wind power generation feature extraction module 130 includes: a wind power generation multi-scale feature extraction unit 131, which is used to perform multi-scale feature extraction on the wind intensity input vector, the wind frequency input vector, the wind speed input vector and the wind speed input vector to obtain a wind intensity feature vector, a wind frequency feature vector, a wind speed feature vector and a wind direction feature vector; and a two-dimensional matrixing unit 132, which is used to perform two-dimensional arrangement on the wind intensity feature vector, the wind frequency feature vector, the wind speed feature vector and the wind direction feature vector to obtain a two-dimensional wind power generation feature matrix.

[0031] Specifically, in an embodiment of the present application, the wind power generation multi-scale feature extraction unit 131 is used to: pass the wind intensity input vector, the wind frequency input vector, the wind speed input vector and the wind speed input vector through a multi-scale neighborhood feature extraction module to obtain a wind intensity feature vector, a wind frequency feature vector, a wind speed feature vector and a wind direction feature vector.

[0032] More specifically, to more comprehensively and accurately describe the wind energy characteristics of a wind farm, the wind intensity input vector, wind frequency input vector, wind speed input vector, and wind direction input vector are passed through a multi-scale neighborhood feature extraction module to obtain wind intensity feature vectors, wind frequency feature vectors, wind speed feature vectors, and wind direction feature vectors, thereby improving the data representation and feature extraction results. Wind energy data from wind farms often exhibit multi-scale characteristics, including short-term fluctuations and long-term trends. Multi-scale neighborhood feature extraction can capture wind energy characteristics at different time scales and better reflect the spatiotemporal characteristics of wind energy data. Multi-scale neighborhood feature extraction can obtain richer information from different scales, including local details and overall trends, which helps to improve the richness and representation of features. Fusion of multi-scale features can comprehensively consider information at different scales, improve the robustness and generalization of features, and help better describe the complexity of wind energy data.

[0033] Specifically, in this embodiment of the present application, the first-scale convolutional encoding subunit is configured to use the first convolutional layer of the multi-scale neighborhood feature extraction module to perform one-dimensional convolution encoding on the wind intensity input vector using the following first convolution formula to obtain the first-scale wind intensity feature vector; wherein the first convolution formula is:

[0034]

[0035] Wherein, a is the width of the first convolution kernel in the x direction, F(a) is the first convolution kernel parameter vector, G(xa) is the local vector matrix operated with the first convolution kernel function, w is the size of the first convolution kernel, X represents the wind intensity input vector, and Cov(X) represents the one-dimensional convolution encoding of the wind intensity input vector respectively;

[0036] The second-scale convolutional encoding subunit is configured to use the second convolutional layer of the multi-scale neighborhood feature extraction module to perform one-dimensional convolution encoding on the wind intensity input vector using the following second convolution formula to obtain the second-scale wind intensity feature vector; wherein the second convolution formula is:

[0037]

[0038] Wherein, b is the width of the second convolution kernel in the x direction, F(b) is the second convolution kernel parameter vector, G(xb) is the local vector matrix operated with the second convolution kernel function, m is the size of the second convolution kernel, X represents the wind strength input vector, Cov(X) represents the one-dimensional convolution encoding of the wind strength input vector respectively; and a cascade subunit is used to cascade the first-scale wind strength feature vector and the second-scale wind strength feature vector to obtain the data sub-segment semantic association feature vector.

[0039] More specifically, arranging the wind intensity, wind frequency, wind speed, and wind direction eigenvectors in two dimensions helps integrate the relationships between these different features, forming a more complete feature description. This integration can more comprehensively express the characteristic information of a wind farm.

[0040] In the above-mentioned coordinated control analysis system 100 for distributed wind power generation systems, the wind turbine type selection module 140 is used to analyze the two-dimensional wind power generation characteristic matrix to obtain the result of the appropriate wind turbine type for operation.

[0041] Specifically, in an embodiment of the present application, the wind turbine type selection module 140 includes: a wind power generation hybrid feature extraction unit, which is used to pass the two-dimensional wind power generation feature matrix through a wind power generation feature extraction module containing multiple hybrid convolutional layers to obtain a wind power generation interaction feature matrix; an enhanced wind power generation feature extraction unit, which is used to pass the wind power generation interaction feature matrix through a wind power generation feature bidirectional attention enhancement module to obtain an enhanced wind power generation feature matrix; a matrix expansion unit, which is used to expand the enhanced wind power generation feature matrix into an enhanced wind power generation feature vector; a compensation optimization unit, which is used to perform subspace projection reconstruction based on orthogonal basis regression on the enhanced wind power generation feature vector to obtain a compensated enhanced wind power generation feature vector; and a wind turbine type selection unit, which is used to pass the compensated enhanced wind power generation feature vector through a classifier to obtain a classification result, and the classification result is used to represent the type of wind turbine suitable for operation.

[0042] Specifically, in an embodiment of the present application, the wind power generation feature extraction module is a convolutional neural network model including multiple hybrid convolutional layers.

[0043] More specifically, the two-dimensional wind power generation feature matrix is obtained by a wind power generation feature extraction module comprising multiple hybrid convolutional layers to obtain a wind power generation interaction feature matrix. The wind power generation feature extraction module is a convolutional neural network model comprising multiple hybrid convolutional layers. Convolutional neural networks (CNNs) perform well in image and spatial data processing, and features can be effectively extracted through convolutional layers. The hybrid convolutional layer combines convolution kernels of different scales to capture features at different scales, which helps to extract spatial information and interaction relationships in the wind power generation feature matrix. The hybrid convolutional layer can fuse features of different scales, comprehensively consider feature information at different scales, and improve the expressiveness and robustness of the features. This helps to better capture the complex interaction relationships in the wind power generation feature matrix. By learning the interaction features in the wind power generation feature matrix through the convolutional neural network model, the associations and patterns between features can be automatically learned, thereby improving the model's ability to understand wind power generation data.

[0044] Specifically, in an embodiment of the present application, the wind power generation hybrid feature extraction unit includes: a first convolution branch subunit, used to use a first convolution kernel with a first size to perform convolution encoding on the two-dimensional wind power generation feature matrix to obtain a first wind power generation feature matrix; a second convolution branch subunit, used to use a second convolution kernel with a first void rate to perform convolution encoding on the S-transform time-frequency map to obtain a second wind power generation feature matrix; a third convolution branch subunit, used to use a third convolution kernel with a second void rate to perform convolution encoding on the two-dimensional wind power generation feature matrix to obtain a third wind power generation feature matrix. feature matrix; a fourth convolution branch subunit, used to use a fourth convolution kernel with a third void rate to perform convolution encoding on the two-dimensional wind power generation feature matrix to obtain a fourth wind power generation feature matrix; a multi-scale feature fusion subunit, used to aggregate the first wind power generation feature matrix, the second wind power generation feature matrix, the third wind power generation feature matrix and the fourth wind power generation feature matrix along the channel dimension to obtain a wind power generation feature map; and a dimension adjustment subunit, used to perform global mean pooling processing on the feature map along the channel dimension to obtain the wind power generation interaction feature matrix.

[0045] Specifically, in an embodiment of the present application, the first convolution kernel, the second convolution kernel, the third convolution kernel and the fourth convolution kernel have the same size, and the second convolution kernel, the third convolution kernel and the fourth convolution kernel have different void rates.

[0046] More specifically, the wind power generation interaction feature matrix is passed through the wind power generation feature bidirectional attention enhancement module to obtain an enhanced wind power generation feature matrix. The wind power generation feature bidirectional attention enhancement module is a bidirectional attention mechanism model. The bidirectional attention mechanism can dynamically learn the importance weights between features based on the correlation between different features, thereby achieving feature weighting. This helps to strengthen the association between different features in the wind power generation interaction feature matrix, highlight important features, and improve the expressiveness of features. At the same time, the bidirectional attention mechanism can simultaneously consider the previous and next associations between features, that is, contextual information. Through this mechanism, the long-range dependencies and interactions between features in the wind power generation feature matrix can be better captured, improving the expressiveness of features and the prediction accuracy of the model. The bidirectional attention mechanism can promote information interaction and transmission between different features, help to establish closer connections between features, and improve the expressiveness of features and the learning effect of the model.

[0047] Figure 3 FIG. 1 is a block diagram of an enhanced wind power generation feature extraction unit in a coordinated control analysis system of a distributed wind power generation system according to an embodiment of the present application. Figure 3As shown, the enhanced wind power generation feature extraction unit 10 includes: a bidirectional pooling subunit 11, used to pool the wind power generation interaction feature matrix along the horizontal direction and the vertical direction respectively to obtain a first-direction pooling vector and a second-direction pooling vector; an association coding subunit 12, used to association encode the first-direction pooling vector and the second-direction pooling vector to obtain a bidirectional association matrix; an activation subunit 13, used to input the bidirectional association matrix into a Sigmoid activation function to obtain a bidirectional association weight matrix; and an attention application subunit 14, used to calculate the position point multiplication between the bidirectional association weight matrix and the wind power generation interaction feature matrix to obtain the enhanced wind power generation feature matrix.

[0048] More specifically, the enhanced wind power generation feature matrix is expanded into enhanced wind power generation feature vectors. Expanding the feature matrix into feature vectors can better represent the characteristic information of each sample. Feature vectors are easier to process and understand, helping to extract key features and reduce unnecessary information redundancy. Furthermore, expanding the feature matrix into feature vectors can reduce the dimensionality of the data, thereby reducing model complexity and computational cost. In practical applications, high-dimensional feature matrices may lead to overly complex models, and expanding them into feature vectors can better meet model requirements.

[0049] Considering that although the multi-scale neighborhood feature extraction module and the wind power generation feature extraction module containing multiple hybrid convolutional layers can effectively capture the spatial characteristics of the input data and utilize the bidirectional attention mechanism to enhance key features, these steps primarily focus on improving feature representation capabilities and highlighting the role of important features. In this complex transformation process, there may be insufficient attention paid to some subtle but important internal structural information of features. This is mainly because with the increase in feature dimensions and the complexity of the feature transformation process, some local or detailed feature information may be weakened or lost, especially when high-dimensional data is mapped to a low-dimensional representation space. In addition, while the bidirectional attention mechanism helps to enhance key features, it may also tend to amplify already significant feature patterns and ignore potential internal structural information that is unique but less prominent. Based on this, the enhanced wind power generation feature vector is reconstructed through subspace projection based on orthogonal basis regression to obtain a compensated enhanced wind power generation feature vector.

[0050] Specifically, in the embodiment of the present application, the compensation optimization unit is used to: first, construct a fine-grained correlation matrix of the enhanced wind power generation feature vector, which is expressed as follows:

[0051]

[0052] Wherein, V represents the enhanced wind power generation characteristic vector, v iand v j represent the eigenvalues of the i-th and j-th positions of the enhanced wind power generation eigenvector, d(v i ,v j ) represents the calculation of Euclidean distance, D i,j Represents the eigenvalue at the (i, j) position of the fine-grained incidence matrix.

[0053] That is, due to the nonlinear spatiotemporal correlation between wind parameters, traditional feature encoding methods may not be able to effectively capture their cross-dimensional coordinated change patterns. Therefore, this application constructs a fine-grained correlation matrix, which is essentially to quantify and strengthen the local response relationship between the units within the wind power generation characteristic vector, and establish an implicit dynamic interaction network in the feature space, thereby converting discrete time series data into a resolvable topological structure. In this way, the model's ability to decouple the dynamic characteristics of the wind field can be improved, so that the subsequent orthogonal basis projection can more accurately separate the characteristic modes with physical significance.

[0054] Secondly, the subspace feature analysis of the fine-grained correlation matrix is performed based on the convolution layer to obtain the enhanced wind power generation feature subspace activation response matrix, which is expressed as follows:

[0055] M=Conv(D)

[0056] Wherein, D represents the fine-grained correlation matrix, Conv represents the convolutional layer, and M represents the enhanced wind power generation feature subspace activation response matrix.

[0057] Specifically, by using convolution kernels targeting correlation patterns, we can exploit the dynamic cross-dimensional coupling patterns implicit in wind parameter sequences. For example, we can identify the synergistic patterns between wind frequency fluctuations and wind intensity changes within a specific time window. This allows us to map the high-dimensional, fine-grained correlation matrix onto an activation response matrix for the enhanced wind power generation characteristic subspace. This essentially creates a heat map of the contribution of dynamic wind parameters to generator type selection. This improves the physical interpretability of feature analysis, thereby enhancing the accuracy of eigenmode separation using subsequent orthogonal basis projection.

[0058] Then, the fine-grained correlation matrix is modally decomposed to obtain a set of orthogonal basis encoding vectors for enhanced wind power generation characteristics, which can be expressed as follows:

[0059]

[0060] Where T represents the transpose of the vector, Λ represents the diagonal matrix, λ1 and λ m They represent the first and mth eigenvalues of the diagonal matrix respectively, U represents the set of orthogonal basis encoding vectors of enhanced wind power generation characteristics, x1, x2, x m represent the first, second and mth enhanced wind power generation feature orthogonal basis encoding vectors respectively.

[0061] That is, through modal decomposition, the fine-grained correlation moments are decomposed into a set of orthogonal basis encoding vectors that enhance the characteristics of wind power generation, and the complex multi-dimensional correlations are decoupled into independent dynamic modes with physical significance. By eliminating redundant correlation noise, the key driving factors of generator type switching under different wind conditions can be explicitly characterized, so that the subsequent classification model can still make rapid decisions based on stable modes when the wind conditions switch rapidly, thereby reducing the oscillation amplitude of power generation caused by modal confusion and improving the robust response of the coordinated control system to intermittent wind energy.

[0062] Next, each enhanced wind power generation feature orthogonal basis encoding vector in the set of enhanced wind power generation feature orthogonal basis encoding vectors is input into a feature selection unit based on the self-attention mechanism to obtain a set of wind power generation feature saliency encoding vectors, which is expressed as follows:

[0063] Y=Transformer{[x1,x2,…,x m ]}=[y1,y2,…,y m ]

[0064] Among them, Transformer represents the feature selection unit based on the self-attention mechanism, Y represents the set of wind power generation feature saliency encoding vectors, y1, y2, y m represent the first, second and m-th wind power generation feature significance encoding vectors respectively.

[0065] Specifically, the attention mechanism dynamically evaluates the contextual relevance of each enhanced wind power generation feature orthogonal basis encoding vector, capturing the implicit mapping rules between wind mode and generator type adaptation within a specific time window. Specifically, by reconstructing the feature weight distribution through global context awareness, the model can self-organize feature activation paths based on real-time wind farm status, thereby improving the spatiotemporal specificity of classification decisions and avoiding control command ambiguity caused by the superposition of multimodal weights.

[0066] Then, each wind power generation feature saliency coding vector in the set of wind power generation feature saliency coding vectors is projected onto the enhanced wind power generation feature subspace activation response matrix to obtain a set of wind power generation feature subspace masking coding vectors, which can be expressed as follows:

[0067]

[0068] in, represents matrix multiplication, S represents the characteristic scale of the activation response matrix of the enhanced wind power generation characteristic subspace, and y i represents the i-th wind power generation feature significance coding vector, L represents the length of the initial feature eigencomponent significant modulation coding vector, z irepresents the shielding coding vector of the i-th wind power generation feature subspace.

[0069] Specifically, through projection, the wind power feature saliency encoding vectors interact across domains with the enhanced wind power feature subspace activation response matrix from convolutional parsing, forming a physically constrained feature masking mechanism. Specifically, by dynamically masking redundant or conflicting response channels, the generated set of wind power feature subspace masking encoding vectors can embed compensation information for local transient disturbances, improving the critical state decision-making accuracy of the control system.

[0070] Finally, the set of wind power generation characteristic subspace masking coding vectors is cascaded to obtain the compensated enhanced wind power generation characteristic vector, which is expressed as follows:

[0071] V′=Concat{z1,z2,…,z m}

[0072] Among them, Concat represents the cascade function, z1, z2, z m They represent the first, second and m-th wind power generation characteristic subspace shielding coding vectors respectively, and V' represents the enhanced wind power generation characteristic vector after compensation.

[0073] In other words, through cascading operations rather than compression fusion, the independence of each shielding channel in the dynamic coordination of wind parameters is preserved, avoiding the dilution of key response patterns caused by feature superposition. By constructing a multidimensional feature topology with spatiotemporal resolution, the generated compensated enhanced wind power generation feature vector can explicitly distinguish the differential driving mechanisms of wind parameters on generator control strategies at different time scales, thereby improving the timing matching accuracy of the coordinated control of generator sets.

[0074] More specifically, the compensated enhanced wind power generation feature vector is passed through a classifier to generate classification results. These classification results are used to identify suitable wind turbine types for operation. This identification of suitable wind turbine types based on the classification results provides intelligent decision support for operators and maintenance personnel. Based on the classification results, they can select the appropriate wind turbine type for the current environment and conditions, optimizing power generation efficiency and equipment operational stability.

[0075] Specifically, in the embodiment of the present application, the wind turbine type selection unit is configured to: use the classifier to process the compensated enhanced wind power generation feature vector using the following classification formula to obtain the classification result; wherein the classification formula is:

[0076] O=softmax{(W n ,B n ):…:(W1,B1)|V'}

[0077] Among them, W1 to W n is the weight matrix, B1 to B n is the bias vector, V' is the compensated enhanced wind power generation feature vector, softmax represents the softmax function, and O represents the classification result.

[0078] In summary, the coordinated control and analysis system for a distributed wind power generation system according to an embodiment of the present application has been described. It uses artificial intelligence technology based on deep learning to extract and encode features of wind intensity values, wind frequency values, wind speed values, and wind direction values at multiple predetermined time points within a predetermined time period at a wind farm, thereby obtaining classification results for suitable wind turbine types. Thus, by intelligently selecting the appropriate wind turbine type, wind energy can be effectively utilized, improving the stability and reliability of wind farm power generation.

[0079] As described above, the coordinated control and analysis system 100 for a distributed wind power generation system according to an embodiment of the present application can be implemented in various terminal devices, such as a coordinated control and analysis server for a distributed wind power generation system. In one example, the coordinated control and analysis system 100 for a distributed wind power generation system according to an embodiment of the present application can be integrated into a terminal device as a software module and / or a hardware module. For example, the coordinated control and analysis system 100 for a distributed wind power generation system can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the coordinated control and analysis system 100 for a distributed wind power generation system can also be one of the many hardware modules of the terminal device.

[0080] Alternatively, in another example, the collaborative control and analysis system 100 of the distributed wind power generation system and the terminal device may also be separate devices, and the collaborative control and analysis system 100 of the distributed wind power generation system may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0081] Based on the same inventive concept, an embodiment of the present application further provides a coordinated control analysis method for a distributed wind power generation system, which can be used to implement the system described in the above embodiment, as described in the following embodiment.

[0082] Figure 4 FIG. 1 is a flow chart of a coordinated control analysis method for a distributed wind power generation system according to an embodiment of the present application. Figure 4As shown, the collaborative control analysis method of the distributed wind power generation system according to the embodiment of the present application includes the steps of: S110, obtaining wind intensity values at multiple predetermined time points within a predetermined time period of the wind farm, wind frequency values at multiple predetermined time points within the predetermined time period, wind speed values at multiple predetermined time points within the predetermined time period, and wind direction values at multiple predetermined time points within the predetermined time period; S120, arranging the wind intensity values at multiple predetermined time points within the predetermined time period, the wind frequency values at multiple predetermined time points within the predetermined time period, the wind speed values at multiple predetermined time points within the predetermined time period, and the wind direction values at multiple predetermined time points within the predetermined time period into a wind intensity input vector, a wind frequency input vector, a wind speed input vector, and a wind speed input vector according to the time dimension; S130, performing feature extraction and encoding on the wind intensity input vector, the wind frequency input vector, the wind speed input vector, and the wind speed input vector to obtain a two-dimensional wind power generation feature matrix; and S140, analyzing the two-dimensional wind power generation feature matrix to obtain a result of a suitable wind turbine generator type.

[0083] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other. Here, for the coordinated control analysis method of the distributed wind power generation system disclosed in the embodiment, those skilled in the art will understand that the specific operations of each step in the coordinated control analysis method of the distributed wind power generation system have been referred to above. Figures 1 to 4 The collaborative control analysis system of the distributed wind power generation system has been introduced in detail, so the description is relatively simple. For relevant details, please refer to the description of the collaborative control analysis system of the distributed wind power generation system, and therefore, its repeated description will be omitted.

[0084] In summary, the coordinated control analysis method for a distributed wind power generation system according to an embodiment of the present application has been described. It uses artificial intelligence technology based on deep learning to extract and encode features of wind intensity values, wind frequency values, wind speed values, and wind direction values at multiple predetermined time points within a predetermined time period at a wind farm, thereby obtaining classification results for suitable wind turbine types. Thus, by intelligently selecting the appropriate wind turbine type, wind energy can be effectively utilized, improving the stability and reliability of wind farm power generation.

[0085] In summary, after reading this detailed disclosure, those skilled in the art will appreciate that the foregoing detailed disclosure may be presented by way of example only and may not be limiting. Although not explicitly stated herein, those skilled in the art will understand that this application is intended to encompass various reasonable changes, improvements, and modifications to the embodiments. Such changes, improvements, and modifications are intended to be proposed by this application and are within the spirit and scope of the exemplary embodiments of this application.

[0086] In addition, certain terms in this application have been used to describe embodiments of the present application. For example, "one embodiment," "an embodiment," and / or "some embodiments" mean that a particular feature, structure, or characteristic described in conjunction with that embodiment may be included in at least one embodiment of the present application. Therefore, it is emphasized and should be understood that two or more references to "an embodiment," "one embodiment," or "an alternative embodiment" in various parts of this specification do not necessarily refer to the same embodiment. Furthermore, particular features, structures, or characteristics may be appropriately combined in one or more embodiments of the present application.

[0087] Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to encompass non-exclusive inclusion, such that an article or device comprising a list of elements includes not only those elements but also other elements not explicitly listed or inherent to such article or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of additional identical elements in the article or device comprising the aforementioned elements.

[0088] It should be understood that in the foregoing description of the embodiments of this application, in order to facilitate understanding of a feature and to simplify this application, this application combines various features into a single embodiment, figure, or description thereof. However, this does not mean that the combination of these features is required. When reading this application, it is entirely possible for those skilled in the art to extract some of the features and understand them as separate embodiments. In other words, the embodiments of this application can also be understood as the integration of multiple secondary embodiments. This also applies when the content of each secondary embodiment is less than all the features of a single aforementioned disclosed embodiment.

[0089] Finally, it should be understood that the embodiments of the application disclosed herein are illustrations of the principles of the embodiments of the present application. Other modified embodiments are also within the scope of the present application. Therefore, the embodiments disclosed in this application are merely examples and not limitations. Those skilled in the art can adopt alternative configurations to implement the application in this application based on the embodiments in this application.

[0090] Therefore, the embodiments of the present application are not limited to the precise embodiments described in the application.

Claims

1. A coordinated control and analysis system for a distributed wind power generation system, characterized in that: include: a wind farm data acquisition module, configured to acquire wind intensity values at a plurality of predetermined time points within a predetermined time period of the wind farm, wind frequency values at a plurality of predetermined time points within the predetermined time period, wind speed values at a plurality of predetermined time points within the predetermined time period, and wind direction values at a plurality of predetermined time points within the predetermined time period; a wind power generation vector construction module, configured to arrange the wind intensity values at a plurality of predetermined time points within the predetermined time period, the wind frequency values at a plurality of predetermined time points within the predetermined time period, the wind speed values at a plurality of predetermined time points within the predetermined time period, and the wind direction values at a plurality of predetermined time points within the predetermined time period into a wind intensity input vector, a wind frequency input vector, a wind speed input vector, and a wind speed input vector, respectively, according to a time dimension; a wind power generation feature extraction module, configured to extract and encode features of the wind intensity input vector, the wind frequency input vector, the wind speed input vector, and the wind velocity input vector to obtain a two-dimensional wind power generation feature matrix; The wind turbine type selection module is used to analyze the two-dimensional wind power generation characteristic matrix to obtain the result of the wind turbine type that is suitable for operation.

2. The coordinated control and analysis system for distributed wind power generation systems according to claim 1, characterized in that: The wind power generation feature extraction module includes: a wind power generation multi-scale feature extraction unit, configured to perform multi-scale feature extraction on the wind intensity input vector, the wind frequency input vector, the wind speed input vector, and the wind speed input vector, respectively, to obtain a wind intensity feature vector, a wind frequency feature vector, a wind speed feature vector, and a wind direction feature vector; The two-dimensional matrix unit is used to arrange the wind intensity characteristic vector, the wind frequency characteristic vector, the wind speed characteristic vector and the wind direction characteristic vector in two dimensions to obtain a two-dimensional wind power generation characteristic matrix.

3. The coordinated control and analysis system for distributed wind power generation systems according to claim 2, characterized in that: The wind power generation multi-scale feature extraction unit is used to: The wind intensity input vector, the wind frequency input vector, the wind speed input vector and the wind direction input vector are respectively passed through a multi-scale neighborhood feature extraction module to obtain a wind intensity feature vector, a wind frequency feature vector, a wind speed feature vector and a wind direction feature vector.

4. The coordinated control and analysis system for distributed wind power generation systems according to claim 3, characterized in that: The wind turbine type selection module includes: a wind power generation hybrid feature extraction unit, configured to pass the two-dimensional wind power generation feature matrix through a wind power generation feature extraction module comprising a plurality of hybrid convolutional layers to obtain a wind power generation interaction feature matrix; an enhanced wind power generation feature extraction unit, configured to pass the wind power generation interaction feature matrix through a wind power generation feature bidirectional attention enhancement module to obtain an enhanced wind power generation feature matrix; a matrix expansion unit, configured to expand the enhanced wind power generation characteristic matrix into an enhanced wind power generation characteristic vector; a compensation optimization unit, configured to perform subspace projection reconstruction based on orthogonal basis regression on the enhanced wind power generation characteristic vector to obtain a compensated enhanced wind power generation characteristic vector; The wind turbine type selection unit is used to pass the compensated enhanced wind power generation feature vector through a classifier to obtain a classification result, and the classification result is used to indicate the type of wind turbine that is suitable for operation.

5. The coordinated control and analysis system for distributed wind power generation systems according to claim 4, characterized in that: The compensation optimization unit is used to: Constructing a fine-grained correlation matrix of the enhanced wind power generation feature vector; Performing subspace feature analysis on the fine-grained correlation matrix based on a convolutional layer to obtain an enhanced wind power generation feature subspace activation response matrix; Performing modal decomposition on the fine-grained correlation matrix to obtain a set of orthogonal basis encoding vectors for enhancing wind power generation characteristics; Inputting each enhanced wind power generation feature orthogonal basis encoding vector in the set of enhanced wind power generation feature orthogonal basis encoding vectors into a feature selection unit based on a self-attention mechanism to obtain a set of wind power generation feature saliency encoding vectors; Projecting each wind power generation feature saliency coding vector in the set of wind power generation feature saliency coding vectors onto the enhanced wind power generation feature subspace activation response matrix to obtain a set of wind power generation feature subspace masking coding vectors; The set of wind power generation characteristic subspace masking coding vectors is cascaded to obtain the compensated enhanced wind power generation characteristic vector.

6. The coordinated control and analysis system for distributed wind power generation systems according to claim 5, characterized in that: The wind turbine type selection unit is configured to: Using the classifier to process the compensated enhanced wind power generation feature vector using the following classification formula to obtain the classification result; Wherein, the classification formula is: O=softmax{(W n ,B n ):…:(W1,B1)|V'} Among them, W1 to W n is the weight matrix, B1 to B n is the bias vector, V' is the compensated enhanced wind power generation feature vector, softmax represents the softmax function, and O represents the classification result.

7. A coordinated control analysis method for a distributed wind power generation system, characterized in that: include: Obtaining wind intensity values at multiple predetermined time points within a predetermined time period of the wind farm, wind frequency values at multiple predetermined time points within the predetermined time period, wind speed values at multiple predetermined time points within the predetermined time period, and wind direction values at multiple predetermined time points within the predetermined time period; Arrange the wind intensity values at multiple predetermined time points within the predetermined time period, the wind frequency values at multiple predetermined time points within the predetermined time period, the wind speed values at multiple predetermined time points within the predetermined time period, and the wind direction values at multiple predetermined time points within the predetermined time period into a wind intensity input vector, a wind frequency input vector, a wind speed input vector, and a wind speed input vector, respectively, according to the time dimension; Performing feature extraction and encoding on the wind intensity input vector, the wind frequency input vector, the wind speed input vector, and the wind velocity input vector to obtain a two-dimensional wind power generation feature matrix; The two-dimensional wind power generation characteristic matrix is analyzed to obtain the results of the wind turbine type that is suitable for operation.

Images