A business operation method, device and storage medium based on wind power

Abstract

The present invention discloses a business operation method, device, and storage medium based on wind power. The method includes: reconstructing multiple actual powers generated by a wind turbine generator set in history into a phase space; if the last first phase point is selected as the center point, extending the center point into multiple second phase points according to the step size; finding multiple first phase points adjacent to the center point as center neighbor points; if the first second phase point is selected as the actual prediction point, finding multiple first phase points adjacent to the actual prediction point to obtain low-dimensional neighbor points; taking the intersection of the center neighbor point and the low-dimensional neighbor points to obtain a cross-neighbor point; fusing the multiple cross-neighbor points into multiple information points; inputting the multiple information points and the center point into a power prediction model to predict the expected power generated by the wind turbine generator set within a future time range; and performing business operations on the wind turbine generator set based on the expected power. This embodiment improves the accuracy of the power prediction model in ultra-short-term wind power prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and in particular to a business operation method, device and storage medium based on wind power. Background Art

[0002] In the process of energy structure transformation, wind power generation occupies an increasingly important position in the power system due to its clean and renewable characteristics. Since wind energy resources are highly random and uncertain, wind power forecasting will be carried out in the power system in order to carry out business operations such as wind power absorption, stable operation of the power grid and optimal energy configuration.

[0003] Wind power forecasts can generally be divided into ultra-short-term (a few minutes to four hours), short-term (a few hours to three days) and medium- to long-term (a few days to several years).

[0004] Since the inherent chaotic dynamic characteristics of wind power are the main reason for its external volatility, this chaotic characteristic needs to be mapped into a high-dimensional space. Therefore, for ultra-short-term wind power prediction, wind power can be mapped into a high-dimensional phase space through reconstruction. In the phase space, the phase point at the next moment is predicted by the points in the known phase point set that are adjacent to the phase point at the next moment.

[0005] However, in some cases, the correlation between adjacent points and the phase point at the next moment is weak, resulting in low accuracy of ultra-short-term wind power forecasting, which affects the quality of business operations. Summary of the Invention

[0006] In view of this, the present invention provides a business operation method, device and storage medium based on wind power, so as to improve the accuracy of ultra-short-term wind power forecasting.

[0007] A first aspect of the present invention provides a wind power-based business operation method, comprising:

[0008] Reconstructing multiple actual powers generated by the wind turbine generator set in history into a phase space according to the delay time and the embedding dimension; the phase space has multiple first phase points; and the interval between two adjacent actual powers is a sampling time;

[0009] If the last of the first phase points is selected as the center point, the center point is extended to multiple second phase points according to the step length; the product of the step length and the sampling time is a time range whose length is less than the ultra-short time threshold;

[0010] Finding a plurality of first phase points that are adjacent to the center point as center neighbor points;

[0011] If the first second phase point is selected as the actual prediction point, then, under the condition of reducing the embedding dimension, multiple first phase points that are neighbors of the actual prediction point are searched to obtain low-dimensional neighboring points;

[0012] Taking the intersection of the central neighbor point and the low-dimensional neighbor point to obtain the cross neighbor point;

[0013] Merging a plurality of the cross-neighbor points into multiple information points;

[0014] Inputting the multiple information points and the central point into a power prediction model to predict the expected power generated by the wind turbine generator set within the future time period;

[0015] Performing business operations on the wind turbine generator set according to the expected power.

[0016] A second aspect of the present invention provides a business operation based on wind power, including:

[0017] A phase space reconstruction module is used to reconstruct multiple actual powers generated by the wind turbine generator set in history into a phase space according to the delay time and the embedding dimension; the phase space has multiple first phase points; and the interval between two adjacent actual powers is a sampling time;

[0018] a phase point extension module, configured to, if the last of the first phase points is selected as the center point, extend the center point into a plurality of second phase points according to a step length, wherein the product of the step length and the sampling time is a time range having a length less than an ultra-short-term time threshold;

[0019] a center neighbor point searching module, configured to search for a plurality of first phase points that are adjacent to the center point as center neighbor points;

[0020] a low-dimensional neighbor point search module, configured to, if the first second phase point is selected as the actual prediction point, search for a plurality of first phase points that are neighbors of the actual prediction point while reducing the embedding dimension to obtain low-dimensional neighbor points;

[0021] A neighbor point intersection module is used to take the intersection of the central neighbor point and the low-dimensional neighbor point to obtain a cross neighbor point;

[0022] A cross-neighborhood point fusion module, configured to fuse a plurality of cross-neighborhood points into multiple information points;

[0023] An expected power prediction module, configured to input the multiple information points and the central point into a power prediction model to predict the expected power generated by the wind turbine generator set within the future time period;

[0024] A business operation execution module is used to execute business operations on the wind turbine generator set according to the expected power.

[0025] A third aspect of the present invention provides an electronic device, comprising:

[0026] at least one processor; and

[0027] a memory communicatively connected to the at least one processor; wherein,

[0028] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the business operation method based on wind power as described in the first aspect above.

[0029] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the wind power-based business operation method as described in the first aspect above.

[0030] A fifth aspect of the present invention provides a computer program product, comprising a computer program, which, when executed by a processor, implements the wind power-based business operation method as described in the first aspect above.

[0031] In this embodiment, multiple actual powers generated historically by the wind turbine are reconstructed into a phase space according to the delay time and the embedding dimension; there are multiple first phase points in the phase space; the sampling time is separated by two adjacent actual powers; if the last first phase point is selected as the center point, the center point is extended into multiple second phase points according to the step size; the product of the step size and the sampling time is a time range whose length is less than the ultra-short-term time threshold; multiple first phase points that are close to the center point are found as the center neighbor points; if the first second phase point is selected as the actual prediction point, multiple first phase points that are close to the actual prediction point are found under the condition of reducing the embedding dimension to obtain low-dimensional neighbor points; the center neighbor point and the low-dimensional neighbor point are intersected to obtain a cross neighbor point; the multiple cross neighbor points are merged into multiple information points; the multiple information points and the center point are input into the power prediction model to predict the expected power generated by the wind turbine within the future time range; and business operations are performed on the wind turbine based on the power generated by the wind turbine within the future time range. In this embodiment, low-dimensional neighboring points are used to further screen the central neighboring points to ensure that the selected points are more relevant to the prediction target, improve the representativeness of the input of the power prediction model, thereby improving the accuracy of the power prediction model in ultra-short-term wind power prediction and ensuring the quality of business operations.

[0032] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0034] Figure 1 This is a flow chart of a wind power-based business operation method provided in Example 1 of the present invention.

[0035] Figure 2 It is a structural diagram of a power prediction model provided in Example 1 of the present invention.

[0036] Figure 3 This is a structural diagram of a wind power-based business operation device provided in the second embodiment of the present invention.

[0037] Figure 4 This is a structural diagram of an electronic device provided in Example 3 of the present invention. DETAILED DESCRIPTION

[0038] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0039] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can cover sequential implementations other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0040] Example 1

[0041] See also Figure 1 , shows a flow chart of a wind power-based business operation method provided by the first embodiment of the present invention. The method can be executed by a wind power-based business operation device. The wind power-based business operation device can be implemented in the form of hardware and / or software. The wind power-based business operation device can be configured in an electronic device. Figure 1 As shown, the method includes:

[0042] Step 101: Reconstruct multiple actual powers generated by the wind turbine generator set in history into a phase space according to the delay time and the embedding dimension.

[0043] In wind power plants such as offshore wind farms and onshore wind farms, multiple wind turbines can be configured. A wind turbine is a device that converts the kinetic energy of wind into mechanical kinetic energy, and then converts the mechanical energy into electrical kinetic energy. It can usually be divided into parts such as a wind wheel, a generator and a tower.

[0044] In the historical continuous period, the actual power of the wind turbine generator set is sampled at N moments according to the fixed sampling time c, and the wind power time series x(t)={x1,x2,…,x i ,…,x N}, where x i is the actual power at time i, i∈[1,N], N is the length of the wind power time series, and the sampling time interval between two adjacent actual powers.

[0045] In the specific implementation, the mutual information method can be used to select the delay time τ, and the saturated correlation dimension method can be used to select the embedding dimension m. According to the delay time τ and the embedding dimension m, the phase space of multiple actual powers generated by the wind turbine in history is reconstructed to obtain the phase space X m .

[0046] There are multiple first phase points in the phase space, the dimension of each first phase point is the embedding dimension m, and the delay time τ is between the coordinate values ​​of two adjacent dimensions in the same first phase point.

[0047] Then, the phase space X m It can be expressed as:

[0048] ;

[0049] Among them, X(i) is the i-th first phase point, i∈[1,n], n is the number of first phase points, x i+(j-1)τ is the coordinate value of the jth dimension in X(i), j∈[1,m], m is the embedding dimension, and τ is the delay time.

[0050] In general, n satisfies the condition: n=N-(m-1)τ, then the coordinate value x in the last dimension of X(n) is n+(m-1)τ The wind power time series x(t)={x1,x2,…,x i ,…,x N}The actual power of the Nth sample in (that is, the last sampling point).

[0051] Since the phase space X m If the coordinates of all the first phase points in the phase space are known, the phase space X m It is called the known m-dimensional phase space point set.

[0052] Step 102: If the last first phase point is selected as the center point, the center point is extended to multiple second phase points according to the step size.

[0053] In a known m-dimensional phase space point set X m Among the n first phase points, the last first phase point X(n) can be selected as the center point.

[0054] At this time, the center point X(n) can be extended to multiple second phase points according to the step size s of wind power prediction, and the second phase points construct a new m-dimensional phase space, which is called the prediction target phase point set X m ´, predict the target phase point set X m ´ can be regarded as a known m-dimensional phase space point set X m extension of .

[0055] Among them, the product of the step size s and the sampling time c (s×c) is a time range whose length is less than the ultra-short-term time threshold (such as 4 hours), which belongs to the ultra-short-term wind power forecast.

[0056] In a specific implementation, a variable may be determined; wherein the value range of the variable is a positive integer s between 1 and the step size, that is, [1, s].

[0057] The coordinate values ​​of each dimension in the center point are extended according to multiple variables to obtain the coordinate values ​​of each dimension in multiple second phase points; wherein the coordinate value of the last dimension in the second phase point corresponds to the power (i.e., sampling points) of the wind turbine generator set after multiple actual powers (i.e., N actual powers) and the number of which is a step size s.

[0058] Then, the predicted target phase point set X m ´ can be expressed as:

[0059] ;

[0060] Among them, X(n+i) is the i-th second phase point, i∈[1,s], s is the step size, x n+i+(j-1)τis the coordinate value of the jth dimension in X(n+i), j∈[1,m], m is the embedding dimension, and τ is the delay time.

[0061] Step 103: Find multiple first phase points that are close to the center point as the center neighbor points.

[0062] In a known m-dimensional phase space point set X m In the equation, we can traverse other first phase points except the center point and find multiple other first phase points that are close to the center point X(n), which are recorded as center neighbor points.

[0063] In the specific implementation, given the m-dimensional phase space point set X m In the embodiment, for the other first phase points except the center point, the first distances between the center point X(n) and the other first phase points are calculated.

[0064] Taking the Euclidean distance as an example of the first distance, the first distance between the center point X(n) and other first phase points can be expressed as:

[0065] ;

[0066] Where x i+(j-1)τ is the coordinate value of the first phase point X(i) in the jth dimension, x n+(j-1)τ is the coordinate value of the nth first phase point X(n) (i.e. the center point) in the jth dimension, d m is the first distance between the center point X(n) and the i-th first phase point X(i), i∈[1,n-1], n is the number of first phase points, j∈[1,m], m is the embedding dimension, and τ is the delay time.

[0067] Compare the first distances of the other first phase points and select multiple (let k) other first phase points with the smallest first distances as the central neighbor points P of the center point. center .

[0068] Step 104: If the first second phase point is selected as the actual prediction point, multiple first phase points that are close to the actual prediction point are searched under the condition of reducing the embedding dimension to obtain low-dimensional neighboring points.

[0069] In the prediction target phase point set X m Among the s second phase points of ´, the first second phase point X(n+1) can be selected as the actual prediction point.

[0070] Under the condition that both the actual prediction point X(n+1) and the first phase point X(i) reduce the embedding dimension (i.e., dimensionality reduction), find multiple first phase points X(i) that are close to the actual prediction point X(n+1) and obtain the low-dimensional neighbor point P low .

[0071] In the specific implementation, when the delay time τ is less than the sum of the actual power N and 1 (i.e., τ < N+1), the actual prediction point X(n+1) is reduced by one dimension to obtain the low-dimensional prediction target point X(n+1) m-1 , low-dimensional prediction target point X(n+1) m-1 It can be expressed as: X(n+1) m-1 =[x n+1 ,x n+1+τ ,…,x n+1+(j-1)τ ,…,x n+1+(m-2)τ ], where x n+1+(j-1)τ is the coordinate value of the jth dimension in X(n+1), j∈[1,m-1], m is the embedding dimension, and τ is the delay time.

[0072] Low-dimensional predicted target point X(n+1) m-1 The coordinate value x of the last dimension n+1+(m-2)τ It is the n+1+(m-2)τth element of the wind power time series x(t). According to n=N-(m-1)τ, the low-dimensional prediction target point X(n+1) m-1 The coordinate value x of the last dimension n+1+(m-2)τ It is the N+1-τth element of the wind power time series x(t). When τ<N+1, it satisfies N+1-τ>0, making the low-dimensional prediction target point X(n+1) m-1 It is known that the low-dimensional prediction target point X(n+1) m-1 The first m-1 dimensions of the actual prediction point X(n+1) are equal to the first m-1 dimensions of the low-dimensional prediction target point X(n+1). m-1 More relevant to the forecast target.

[0073] And, for a known m-dimensional phase space point set X m Perform low-dimensional projection and remove the coordinate value of the last dimension of each first phase point X(i) to obtain low-dimensional phase points. The low-dimensional phase points can form a new phase space, which is recorded as the known m-1-dimensional phase space point set X m-1 .

[0074] Then, we know that the m-1 dimensional phase space point set X m-1 It can be expressed as:

[0075] ;

[0076] Among them, X(i) m-1 is the i-th low-dimensional phase point, i∈[1,n], n is the number of low-dimensional phase points, x i+(j-1)τ is X(i) m-1 The coordinate value under the j-th dimension in , j∈[1,m-1], m is the embedding dimension, and τ is the delay time.

[0077] In the known m-1 dimensional phase space point set X m-1 , calculate the low-dimensional predicted target point X(n+1) m-1 With other low-dimensional phase points X(i) m-1 The second distance between them.

[0078] Taking Euclidean distance as an example of the second distance, the low-dimensional predicted target point X(n+1) m-1 With other low-dimensional phase points X(i) m-1 The second distance between can be expressed as:

[0079] ;

[0080] Among them, x i+(j-1)τ is the i-th low-dimensional phase point X(i) m-1 The coordinate value in the jth dimension, x n+1+(j-1)τ Predict the target point X(n+1) for low dimension m-1 The coordinate value in the jth dimension, d m-1 Predict the target point X(n+1) for low dimension m-1 With the i-th low-dimensional phase point X(i) m-1 The second distance between them, i∈[1,n], n is the number of low-dimensional phase points, j∈[1,m-1], m is the embedding dimension, and τ is the delay time.

[0081] Compare the second distances of each low-dimensional phase point, and select the first phase points corresponding to multiple (let k) low-dimensional phase points with the smallest second distance as the low-dimensional prediction target point X(n+1) m-1 The low-dimensional neighbor point P of the nearest neighbor low .

[0082] For example, the i-th low-dimensional phase point X(i) m-1 is one of the k low-dimensional phase points with the smallest second distance, and the i-th first phase point X(i) is the i-th low-dimensional phase point X(i) m-1 The corresponding point, then the i-th first phase point X(i) is the low-dimensional prediction target point X(n+1) m-1 The low-dimensional neighbor point P of the nearest neighbor low .

[0083] Step 105: Take the intersection of the central neighbor point and the low-dimensional neighbor point to obtain the cross neighbor point.

[0084] In this embodiment, the center neighbor point P can be center and the low-dimensional neighbor point P low Take the intersection and get the cross neighbor point P inter , that is, P inter =P center ∩P low .

[0085] Then, the intersection neighbor point P inter It can be expressed as:

[0086] ;

[0087] Among them, p i is the i-th cross neighbor point, i∈[1,N inter ],N inter is the number of cross-neighbor points, For p i The coordinate value under the j-th dimension in , j∈[1,m], m is the embedding dimension.

[0088] Step 106: Merge multiple cross-neighbor points into multiple information points.

[0089] In this embodiment, a plurality of cross-neighboring points may be fused in a linear or nonlinear manner to obtain multiple information points.

[0090] In a fusion approach, given a set of m-dimensional phase space points X m There is a central point X(n) that is more relevant to the prediction target. In the known m-1 dimensional phase space point set X m-1 There is a low-dimensional prediction target point X(n+1) that is more relevant to the prediction target. m-1 Therefore, we can predict the target point X(n+1) based on the center point X(n) and the low-dimensional m-1 Construct a multidimensional spatial distance metric.

[0091] In the specific implementation, on the one hand, each cross neighbor point p is calculated separately i The third distance between the center point X(n) and each cross neighbor point p is calculated separately. i and the low-dimensional predicted target point X(n+1) m-1 The fourth distance between them;

[0092] For the same cross-neighbor point p i , add the third distance and the fourth distance to get the original distance index.

[0093] Taking Euclidean distance as an example of the third distance and the fourth distance, the original distance index can be expressed as:

[0094] ;

[0095] in, is the i-th cross neighbor point p i The original distance indicator, For p i The coordinate value in the jth dimension, x n+(j-1)τis the coordinate value of the center point X(n) in the jth dimension, x n+1+(j-1)τ Predict the target point X(n+1) for low dimension m-1 The coordinate value in the jth dimension, i∈[1,N inter ],N inter is the number of cross-neighbor points, j∈[1,m], m is the embedding dimension, and τ is the delay time.

[0096] Then, the set F of the original distance index of each cross neighbor point can be expressed as: .

[0097] Use Min-Max (minimum - maximum), Z-Score (Z score) and other algorithms to normalize the distance indicators corresponding to multiple cross-neighbor points to obtain the target distance indicators corresponding to multiple cross-neighbor points. Then, the set of target distance indicators of each cross-neighbor point is It can be expressed as:

[0098] ,in, is the i-th cross neighbor point p i The target distance index, i∈[1,N inter ],N inter is the number of cross-neighbor points.

[0099] The cross-neighbor points are weighted according to the target distance index. The weight of the current cross-neighbor point is the ratio of the inverse index of the current cross-neighbor point to the inverse index of all cross-neighbor points. The inverse index is 1 minus the difference between the target distance index and the inverse index of all cross-neighbor points.

[0100] Then, the weight can be expressed as:

[0101] ;

[0102] Among them, w i is the weight of the i-th cross neighbor point, is the target distance index of the i-th cross neighbor point, is the target distance index of the jth cross neighbor point, i, j∈[1,N inter ],N inter is the number of cross-neighbor points.

[0103] For the same dimension, the product of the weight and the coordinate value of the cross-neighbor point in the dimension is summed to obtain the coordinate value of the multiple information points in the dimension.

[0104] Assume that multiple information points A={a1,a2,…,a j ,…,a m}, where a jis the coordinate value of multiple information points in the jth dimension.

[0105] ;

[0106] Among them, w i is the i-th cross neighbor point p i The weight of is the i-th cross neighbor point p i The coordinate value of the jth dimension, j∈[1,m], m is the number of dimensions, i∈[1,N inter ],N inter is the number of cross-neighbor points.

[0107] Step 107: Input the multiple information points and the central point into a power prediction model to predict the expected power generated by the wind turbine generator set within a future time range.

[0108] In practical applications, a power prediction model can be built in advance for ultra-short-term wind power forecasts, and the power prediction model can be supervised trained and verified using historical samples (multiple information points and center points) collected from wind turbines. When the power prediction model completes supervised training and verification, it is deployed online for operation.

[0109] At this time, the multiple information points and the central point can be input into the power prediction model individually or in combination for processing, so as to predict the multiple powers generated by the wind turbine generator set within the future time range, which is recorded as the expected power.

[0110] Among them, the power prediction model may include a machine learning model, such as a least square support vector machine classifier (LSSVM), etc., and may also include a deep learning model, such as a deep belief network (DBN), etc. This embodiment does not limit this. Historical data is used to train a high-precision power prediction model, and the parameters of the power prediction model are optimized according to different data sets to improve the stability and robustness of the power prediction model for ultra-short-term wind power prediction.

[0111] Furthermore, for deep learning models, the structure of the power prediction model is not limited to artificially designed neural networks, but can also be a neural network optimized by a model quantization method, a neural network searched for characteristics of ultra-short-term wind power prediction by a NAS (Neural Architecture Search) method, and so on. This embodiment does not impose any restrictions on this.

[0112] In one embodiment of the present invention, step 107 may include the following steps:

[0113] Step 1071: Load the power prediction model.

[0114] In this embodiment, if Figure 2 As shown, the power prediction model adopts the encoder-decoder architecture, which includes the first encoder Encoder_1, the second encoder Encoder_2, the third encoder Encoder_3, the fusion module Fusion_Module and the decoder Decoder. During training, the cross entropy loss function can be used for supervised training.

[0115] Among them, the first encoder Encoder_1, the second encoder Encoder_2 and the third encoder Encoder_3 are all parts of the complete encoder, which are respectively used for encoding multiple information points and / or center points and extracting high-level feature maps from the multiple information points and / or center points.

[0116] The fusion module Fusion_Module is used to process and fuse multiple encoded feature maps to enhance the expressiveness of the feature maps.

[0117] The decoder is used to decode the feature map with enhanced expression capability to achieve ultra-short-term wind power prediction.

[0118] Step 1072: Input the multiple information points into the first encoder and encode them into target multiple information features.

[0119] In this embodiment, if Figure 2 As shown, multiple information points are individually input into the first encoder Encoder_1 for encoding to obtain target multi-information features.

[0120] In one design, Figure 2 As shown, the first encoder includes two layers of long short-term memory (LSTM) networks, which are respectively denoted as the first long short-term memory network LSTM_1 and the second long short-term memory network LSTM_2.

[0121] In this design, multiple information points are input into the first long short-term memory network LSTM_1 and extracted as candidate multi-information features, and the candidate multi-information features are input into the second long short-term memory network LSTM_2 and extracted as target multi-information features.

[0122] Two layers of unidirectional LSTM can capture unidirectional temporal dependencies from multiple information points, achieving layer-by-layer abstraction of features. The bottom-layer LSTM captures local patterns (i.e., short-term dependencies), while the top-layer LSTM captures global patterns (i.e., long-term dependencies), enhancing the semantic hierarchy of features.

[0123] Step 1073: Splice the multiple information points and the center point into an enhanced point.

[0124] In this embodiment, if Figure 2 As shown in Figure 2, the Concat function can be used to concatenate multiple information points and the center point into enhanced points to form a richer feature representation.

[0125] Step 1074: Input the enhancement point into the second encoder and encode it into a target enhancement feature.

[0126] In this embodiment, if Figure 2 As shown, the enhanced point is input into the second encoder Encoder_2 for encoding to obtain the target enhanced feature.

[0127] In one design, Figure 2 As shown, the second encoder, Encoder_2, consists of two layers of bidirectional long short-term memory (Bi-LSTM) networks, denoted as Bi-LSTM_1 and Bi-LSTM_2. A Bi-LSTM consists of two independent LSTM networks. When a Bi-LSTM processes sequence data, the data is fed into the two LSTM networks in both forward and reverse order to extract features. The resulting output vector, formed by concatenating the features, serves as the final output for that time step.

[0128] In this design, the enhancement points are input into the first bidirectional long short-term memory network Bi-LSTM_1 to extract candidate enhancement features, and the candidate enhancement features are input into the second bidirectional long short-term memory network Bi-LSTM_2 to extract target enhancement features.

[0129] The two-layer bidirectional Bi-LSTM can perform dual temporal modeling of the enhancement points, capture the potential correlation between historical and future information, and enhance context understanding, so that the target enhancement features contain more comprehensive sequence dependencies.

[0130] Step 1075: Input the center point into the third encoder and encode it as a target center feature.

[0131] In this embodiment, if Figure 2 As shown, the center point is input into the third encoder Encoder_3 for encoding to obtain the target center feature.

[0132] In one design, Figure 2 As shown, the third encoder Encoder_3 includes two layers of LSTM, which are respectively recorded as the third long short-term memory network LSTM_3 and the fourth long short-term memory network LSTM_4.

[0133] In this design, the center point is input into the third long short-term memory network LSTM_3 to extract the candidate center feature, and the candidate center feature is input into the fourth long short-term memory network LSTM_4 to extract the target center feature.

[0134] Two layers of unidirectional LSTM can capture unidirectional temporal dependencies from the central point, achieving layer-by-layer abstraction of features. The bottom-layer LSTM captures local patterns (i.e., short-term dependencies), while the high-layer LSTM captures global patterns (i.e., long-term dependencies), enhancing the semantic hierarchy of features.

[0135] Step 1076: Input the target multi-information features, target enhancement features, and target center features into a fusion module and fuse them into multimodal features.

[0136] In this embodiment, if Figure 2 As shown in Figure 1, the target multi-information features, target enhancement features and target center features can be input into the fusion module Fusion_Module for interaction and fusion to obtain multimodal features.

[0137] In one design, Figure 2 As shown in the figure, the fusion module Fusion_Module includes a multilayer perceptron (MLP), two layers of self-attention modules (respectively denoted as the first self-attention module Self-Attention_1 and the second self-attention module Self-Attention_2) and a convolutional block attention module (CBAM).

[0138] In this design, the target enhanced features are input into the multi-layer perceptron MLP and mapped into the first encoded intermediate features as the common basic features.

[0139] Use the Concat function to concatenate the target multi-information feature and the first encoding intermediate feature into the second encoding intermediate feature.

[0140] The second encoded intermediate feature is input into the first self-attention module Self-Attention_1, and the first attention feature is generated according to the self-attention mechanism. The self-attention mechanism can mine the hierarchical dependency between multiple information points and enhancement points to achieve homomodal feature enhancement.

[0141] Use the Concat function to concatenate the target center feature and the first encoding intermediate feature into the third encoding intermediate feature.

[0142] The third encoded intermediate feature is input into the second self-attention module Self-Attention_2, and the second attention feature is generated according to the self-attention mechanism. The self-attention mechanism can mine the hierarchical dependency between the center point and the enhancement point to achieve homomodal feature enhancement.

[0143] Use the Concat function to concatenate the first attention feature and the second attention feature into the third attention feature.

[0144] The third attention feature is input into the convolutional block attention module (CBAM) to generate multimodal features. Channel attention and spatial attention are provided in the convolutional block attention module (CBAM), and high-value features are extracted from the mixed features (i.e., the third attention feature). This enables the power prediction model to pay more attention to areas and key semantic channels that are more relevant to ultra-short-term wind power prediction.

[0145] Step 1077: Input the multimodal features into a decoder for decoding to obtain the expected power generated by the wind turbine generator set within a future time range.

[0146] In this embodiment, if Figure 2 As shown in FIG, the multimodal features are input into the decoder for decoding to obtain the expected power generated by the wind turbine within the future time range.

[0147] In one design, Figure 2 As shown, the decoder includes a first decoding layer Transformer_1, a second decoding layer Transformer_2 and a third decoding layer Transformer_3, that is, the first decoding layer Transformer_1, the second decoding layer Transformer_2 and the third decoding layer Transformer_3 are all Transformers.

[0148] To reduce computational costs, the 3-layer Transformer can be lightweight designed (such as reducing the number of heads, using a bottleneck structure) or low-rank approximation optimization can be performed.

[0149] In this design, the multimodal features are input into the first decoding layer Transformer_1 for decoding to obtain the first decoding intermediate features.

[0150] The first decoding intermediate feature is input into the second decoding layer Transformer_2 for decoding to obtain the second decoding intermediate feature.

[0151] The second decoded intermediate feature is input into the third decoding layer Transformer_3 for decoding to obtain the expected power generated by the wind turbine generator set within the future time range.

[0152] The three-layer Transformer can extract features in layers. The bottom-layer Transformer is responsible for capturing local details and building short-distance dependencies. The middle-layer Transformer is responsible for learning middle-level semantics and establishing cross-regional associations. The high-layer Transformer is responsible for learning high-level semantics and modeling long-distance dependencies. It has strong anti-interference capabilities and helps improve the accuracy of sequence generation tasks (i.e., ultra-short-term wind power forecasting).

[0153] In a certain experiment, the ultra-short-term wind power prediction method in this embodiment (using the cross-neighbor point P inter The results are compared with the traditional ultra-short-term wind power forecasting method (using the forecast center neighboring point forecast).

[0154] The wind power time series selected for the experiment is the data of the entire system of a local transmission system operator from January 1, 2018 to January 31, 2018, with a data time granularity of 15 minutes. The delay time of the phase space is selected as 19 by the mutual information method, and the embedding dimension of the phase space is selected as 5 by the saturated correlation dimension method. The phase space is constructed in this way, and the center neighbor point P is set in the phase space. center , low-dimensional neighbor point P low The number of points included is 16, and the forecast time scales are 60 minutes, 90 minutes and 120 minutes in advance. These three time scales are all within the range of ultra-short-term forecasts.

[0155] In this experiment, the mean absolute error (MAE) and root mean square error (RMSE) are used as evaluation indicators to judge the prediction effect.

[0156] Among them, the mean absolute error (MAE) is:

[0157] ;

[0158] The root mean square error (RMSE) is:

[0159] ;

[0160] Where WS(i) is the actual power generated by the wind turbine at time i, WS(i)´ represents the power predicted for the wind turbine at time i, and N is the length of time.

[0161] The smaller the MAE, the higher the accuracy of the ultra-short-term wind power forecasting method, and the smaller the RMSE, the closer the power predicted by the ultra-short-term wind power forecasting method for the wind turbine is to the actual power generated by the wind turbine.

[0162] The performance comparison between the ultra-short-term wind power forecasting method and the traditional ultra-short-term wind power forecasting method is as follows:

[0163] ;

[0164] Comparing the evaluation indicators, it can be seen that the overall prediction effect of the ultra-short-term wind power prediction method in this embodiment is better than the overall prediction effect of the traditional ultra-short-term wind power prediction method, and can effectively improve the accuracy of ultra-short-term wind power prediction.

[0165] Step 108: Execute business operations on the wind turbine generator set according to the expected power.

[0166] In practical applications, corresponding business operations can be performed on wind turbines based on the expected power generated by the wind turbines in the future, such as wind power absorption, stable grid operation, energy optimization configuration, etc.

[0167] For example, for the stable operation of the power grid, if it is predicted that wind power will decrease in the next two hours, the power grid can start the backup coal-fired units in advance to fill the power gap.

[0168] In this embodiment, multiple actual powers generated historically by the wind turbine are reconstructed into a phase space according to the delay time and the embedding dimension; there are multiple first phase points in the phase space; the sampling time is separated by two adjacent actual powers; if the last first phase point is selected as the center point, the center point is extended into multiple second phase points according to the step size; the product of the step size and the sampling time is a time range whose length is less than the ultra-short-term time threshold; multiple first phase points that are close to the center point are found as the center neighbor points; if the first second phase point is selected as the actual prediction point, multiple first phase points that are close to the actual prediction point are found under the condition of reducing the embedding dimension to obtain low-dimensional neighbor points; the center neighbor point and the low-dimensional neighbor point are intersected to obtain a cross neighbor point; the multiple cross neighbor points are merged into multiple information points; the multiple information points and the center point are input into the power prediction model to predict the expected power generated by the wind turbine within the future time range; and business operations are performed on the wind turbine based on the power generated by the wind turbine within the future time range. In this embodiment, low-dimensional neighboring points are used to further screen the central neighboring points to ensure that the selected points are more relevant to the prediction target, improve the representativeness of the input of the power prediction model, thereby improving the accuracy of the power prediction model in ultra-short-term wind power prediction and ensuring the quality of business operations.

[0169] Example 2

[0170] See also Figure 3 , shows a schematic diagram of the structure of a business operation device based on wind power provided by the second embodiment of the present invention. Figure 3 As shown, the device includes:

[0171] The phase space reconstruction module 301 is used to reconstruct the multiple actual powers generated by the wind turbine generator set in history into a phase space according to the delay time and the embedding dimension; the phase space has multiple first phase points; and the interval between two adjacent actual powers is a sampling time;

[0172] A phase point extension module 302 is configured to, if the last of the first phase points is selected as the center point, extend the center point into a plurality of second phase points according to a step length, wherein the product of the step length and the sampling time is a time range having a length less than an ultra-short-term time threshold;

[0173] A center neighbor point search module 303 is configured to search for a plurality of first phase points that are adjacent to the center point as center neighbor points;

[0174] A low-dimensional neighbor point search module 304 is configured to, if the first second phase point is selected as the actual prediction point, search for multiple first phase points that are neighbors of the actual prediction point while reducing the embedding dimension to obtain low-dimensional neighbor points;

[0175] A neighbor point intersection module 305 is used to obtain an intersection of the central neighbor point and the low-dimensional neighbor point to obtain a cross neighbor point;

[0176] A cross-neighborhood point fusion module 306 is configured to fuse a plurality of cross-neighborhood points into multiple information points;

[0177] An expected power prediction module 307 is configured to input the multiple information points and the central point into a power prediction model to predict the expected power generated by the wind turbine generator set within the future time range;

[0178] The service operation execution module 308 is configured to execute a service operation on the wind turbine generator set according to the expected power.

[0179] In one embodiment of the present invention, the phase point extension module 302 includes:

[0180] A variable determination module is used to determine multiple variables; the value range of the variables is a positive integer between 1 and the step size;

[0181] A coordinate value extension module is used to extend the coordinate values ​​of each dimension in the center point according to multiple variables to obtain the coordinate values ​​of each dimension in multiple second phase points; wherein the coordinate value of the last dimension in the second phase point corresponds to the power of the wind turbine generator set after multiple actual powers, and the number is the step size.

[0182] In one embodiment of the present invention, the center neighbor point finding module 303 includes:

[0183] A first distance calculation module is configured to calculate, for the other first phase points except the central point, a first distance between the central point and the other first phase points;

[0184] A first distance screening module is configured to screen out a plurality of other first phase points having the smallest first distances as central neighbor points adjacent to the central point;

[0185] The low-dimensional neighbor point search module 304 includes:

[0186] a point dimensionality reduction module, configured to, when the delay time is less than the sum of the actual power and 1, reduce one dimension of the actual prediction point to obtain a low-dimensional prediction target point, and remove the coordinate value of the last dimension of each first phase point to obtain a low-dimensional phase point;

[0187] A second distance calculation module is used to calculate the second distance between the low-dimensional prediction target point and other low-dimensional phase points;

[0188] The second distance screening module is used to screen out the first phase points corresponding to the multiple low-dimensional phase points with the smallest second distance as low-dimensional neighboring points of the low-dimensional predicted target point.

[0189] In one embodiment of the present invention, the cross-neighborhood point fusion module 306 includes:

[0190] A third distance calculation module, configured to respectively calculate the third distance between each of the intersection neighbor points and the center point;

[0191] a fourth distance calculation module, configured to respectively calculate a fourth distance between each of the intersection neighbor points and the low-dimensional prediction target point;

[0192] a distance adding module, configured to add the third distance and the fourth distance for the same intersection neighbor point to obtain an original distance index;

[0193] a normalization module, configured to normalize the plurality of distance indicators to obtain a plurality of target distance indicators;

[0194] A weight configuration module is used to configure weights for the cross-neighbor points according to the target distance index; the weight of the current cross-neighbor point is the ratio of the inverse index of the current cross-neighbor point to the inverse index of all the cross-neighbor points, where the inverse index is 1 minus the difference between the target distance index;

[0195] The weighted summation module is used to sum the products of the weights and the coordinate values ​​of the cross-neighboring points in the same dimension to obtain the coordinate values ​​of multiple information points in the dimension.

[0196] In one embodiment of the present invention, the expected power prediction module 307 includes:

[0197] A power prediction model loading module, used to load a power prediction model; the power prediction model includes a first encoder, a second encoder, a third encoder, a fusion module and a decoder;

[0198] A first encoding module, configured to input the multiple information points into the first encoder and encode them into target multiple information features;

[0199] An enhancement point splicing module, configured to splice the multiple information points and the central point into an enhancement point;

[0200] A second encoding module, configured to input the enhancement point into the second encoder and encode it into a target enhancement feature;

[0201] A third encoding module, configured to input the center point into the third encoder and encode it into a target center feature;

[0202] A multimodal feature fusion module, configured to input the target multi-information feature, the target enhancement feature, and the target center feature into the fusion module to fuse them into a multimodal feature;

[0203] The decoding module is used to input the multimodal features into the decoder for decoding, so as to obtain the expected power generated by the wind turbine generator set within the future time range.

[0204] In one embodiment of the present invention, the first encoder includes a first long short-term memory network and a second long short-term memory network, and the first encoding module is further configured to:

[0205] Inputting the multiple information points into the first long short-term memory network to extract them as candidate multiple information features;

[0206] Inputting the candidate multi-information features into the second long short-term memory network to extract them as target multi-information features;

[0207] The second encoder includes a first bidirectional long short-term memory network and a second bidirectional long short-term memory network, and the second encoding module is further used to:

[0208] Inputting the enhancement points into the first bidirectional long short-term memory network to extract candidate enhancement features;

[0209] Inputting the candidate enhancement features into the second bidirectional long short-term memory network to extract target enhancement features;

[0210] The third encoder includes a third long short-term memory network and a fourth long short-term memory network, and the third encoding module is further used for:

[0211] Inputting the center point into the third long short-term memory network to extract candidate center features;

[0212] The candidate center features are input into the fourth long short-term memory network to extract the target center features.

[0213] In one embodiment of the present invention, the fusion module includes a multi-layer perceptron, a first self-attention module, a second self-attention module, and a convolutional block attention module; the multimodal feature fusion module is further used to:

[0214] Inputting the target enhanced feature into the multi-layer perceptron and mapping it into a first encoded intermediate feature;

[0215] splicing the target multi-information feature and the first coding intermediate feature into a second coding intermediate feature;

[0216] Inputting the second encoded intermediate feature into the first self-attention module to generate a first attention feature;

[0217] splicing the target center feature and the first coding intermediate feature into a third coding intermediate feature;

[0218] Inputting the third encoded intermediate feature into the second self-attention module to generate a second attention feature;

[0219] concatenating the first attention feature and the second attention feature into a third attention feature;

[0220] The third attention feature is input into the convolutional block attention module to generate a multimodal feature.

[0221] In one embodiment of the present invention, the decoder includes a first decoding layer, a second decoding layer, and a third decoding layer, wherein the first decoding layer, the second decoding layer, and the third decoding layer are all Transformers; and the decoding module is further configured to:

[0222] Inputting the multimodal features into the first decoding layer for decoding to obtain first decoding intermediate features;

[0223] Inputting the first decoding intermediate feature into the second decoding layer for decoding to obtain a second decoding intermediate feature;

[0224] The second decoded intermediate feature is input into the third decoding layer for decoding to obtain the expected power generated by the wind turbine generator set within the future time range.

[0225] The wind power-based business operation device provided in the embodiment of the present invention can execute the wind power-based business operation method provided in any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the wind power-based business operation method.

[0226] Example 3

[0227] See also Figure 4 , which shows a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, blade servers, mainframe computers, and other suitable computers. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0228] like Figure 4 As shown, electronic device 10 includes at least one processor 11 and memory, such as read-only memory (ROM) 12 and random access memory (RAM) 13, communicatively connected to at least one processor 11. The memory stores computer programs executable by the at least one processor. Processor 11 can perform various appropriate actions and processes based on the computer programs stored in ROM 12 or loaded from storage unit 18 into RAM 13. RAM 13 can also store various programs and data required for the operation of electronic device 10. Processor 11, ROM 12, and RAM 13 are interconnected via bus 14. An input / output (I / O) interface 15 is also connected to bus 14.

[0229] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0230] Processor 11 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any other suitable processor, controller, microcontroller, etc. Processor 11 executes the various methods and processes described above, such as the wind power-based business operation method.

[0231] In some embodiments, the wind power-based business operation method can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the wind power-based business operation method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to execute the wind power-based business operation method via any other appropriate means (e.g., via firmware).

[0232] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0233] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0234] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or apparatus. A computer-readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or apparatus, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0235] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device that has: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0236] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, 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: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0237] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within a cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0238] Example 4

[0239] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the computer program implements the wind power-based business operation method provided in any embodiment of the present invention.

[0240] During implementation, the computer program product may be written in one or more programming languages, or a combination thereof, to implement the computer program code for performing the operations of the present invention. The programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0241] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0242] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A business operation method based on wind power, characterized in that: include: Reconstruct the multiple actual powers generated by the wind turbine in history into phase space according to the delay time and embedding dimension; There are multiple first phase points in the phase space; a sampling time is separated between two adjacent actual powers; If the last of the first phase points is selected as the center point, the center point is extended to multiple second phase points according to the step length; the product of the step length and the sampling time is a time range whose length is less than the ultra-short time threshold; For other first phase points except the central point, calculating a first distance between the central point and the other first phase points; Screening out a plurality of other first phase points with the smallest first distances as central neighbor points adjacent to the central point; If the first of the second phase points is selected as the actual prediction point, when the delay time is less than the sum of the actual power and 1, the actual prediction point is reduced by one dimension to obtain a low-dimensional prediction target point, and the coordinate value of the last dimension of each of the first phase points is removed to obtain a low-dimensional phase point; Calculate a second distance between the low-dimensional prediction target point and other low-dimensional phase points; Screening out the first phase points corresponding to the plurality of the low-dimensional phase points with the smallest second distance as low-dimensional neighboring points of the low-dimensional prediction target point; Taking the intersection of the central neighbor point and the low-dimensional neighbor point to obtain the cross neighbor point; Merging a plurality of the cross-neighbor points into multiple information points; Inputting the multiple information points and the central point into a power prediction model to predict the expected power generated by the wind turbine generator set within the future time period; Performing business operations on the wind turbine generator set according to the expected power.

2. The method according to claim 1, characterized in that The step of extending the central point into a plurality of second phase points according to the step size includes: Determine multiple variables; the value range of the variables is a positive integer between 1 and the step size; The coordinate values ​​of each dimension in the center point are extended according to the multiple variables to obtain the coordinate values ​​of each dimension in the multiple second phase points; wherein the coordinate value of the last dimension in the second phase point corresponds to the power of the wind turbine generator set after the multiple actual powers, and the number is the step size.

3. The method according to claim 1, characterized in that The step of fusing the plurality of cross-neighboring points into multiple information points comprises: Calculating the third distance between each of the intersection neighbor points and the center point respectively; Calculating the fourth distance between each of the cross neighbor points and the low-dimensional prediction target point respectively; For the same intersection neighbor point, add the third distance and the fourth distance to obtain an original distance index; Normalizing the multiple distance indicators to obtain multiple target distance indicators; The cross-neighboring points are weighted according to the target distance index; the weight of the current cross-neighboring point is the ratio of the inverse index of the current cross-neighboring point to the inverse index of all the cross-neighboring points, where the inverse index is 1 minus the target distance index; For the same dimension, the product between the weight and the coordinate value of the cross neighbor point in the dimension is summed to obtain the coordinate value of the multiple information points in the dimension.

4. The method according to any one of claims 1 to 3, characterized in that Inputting the multiple information points and the central point into a power prediction model to predict the expected power generated by the wind turbine generator set within the future time period includes: Loading a power prediction model; the power prediction model includes a first encoder, a second encoder, a third encoder, a fusion module and a decoder; Inputting the multiple information points into the first encoder and encoding them into target multiple information features; Splicing the multiple information points and the central point into an enhanced point; Inputting the enhancement point into the second encoder and encoding it into a target enhancement feature; Inputting the center point into the third encoder and encoding it into a target center feature; Inputting the target multi-information feature, the target enhancement feature and the target center feature into the fusion module to fuse them into a multimodal feature; The multimodal features are input into the decoder for decoding to obtain the expected power generated by the wind turbine generator set within the future time range.

5. The method according to claim 4, characterized in that The first encoder includes a first long short-term memory network and a second long short-term memory network, and the inputting the multiple information points into the first encoder and encoding them into target multiple information features includes: Inputting the multiple information points into the first long short-term memory network to extract them as candidate multiple information features; Inputting the candidate multi-information features into the second long short-term memory network to extract them as target multi-information features; The second encoder includes a first bidirectional long short-term memory network and a second bidirectional long short-term memory network, and inputting the enhancement point into the second encoder and encoding it into a target enhancement feature includes: Inputting the enhancement points into the first bidirectional long short-term memory network to extract candidate enhancement features; Inputting the candidate enhancement features into the second bidirectional long short-term memory network to extract target enhancement features; The third encoder includes a third long short-term memory network and a fourth long short-term memory network, and inputting the center point into the third encoder to encode it into a target center feature includes: Inputting the center point into the third long short-term memory network to extract candidate center features; The candidate center features are input into the fourth long short-term memory network to extract the target center features.

6. The method according to claim 5, characterized in that The fusion module includes a multi-layer perceptron, a first self-attention module, a second self-attention module and a convolutional block attention module; the inputting the target multi-information feature, the target enhancement feature and the target center feature into the fusion module to fuse them into a multimodal feature includes: Inputting the target enhanced feature into the multi-layer perceptron and mapping it into a first encoded intermediate feature; splicing the target multi-information feature and the first coding intermediate feature into a second coding intermediate feature; Inputting the second encoded intermediate feature into the first self-attention module to generate a first attention feature; splicing the target center feature and the first coding intermediate feature into a third coding intermediate feature; Inputting the third encoded intermediate feature into the second self-attention module to generate a second attention feature; concatenating the first attention feature and the second attention feature into a third attention feature; The third attention feature is input into the convolutional block attention module to generate a multimodal feature.

7. The method according to claim 6, characterized in that The decoder includes a first decoding layer, a second decoding layer, and a third decoding layer, wherein the first decoding layer, the second decoding layer, and the third decoding layer are all Transformers; Inputting the multimodal features into the decoder for decoding to obtain the expected power generated by the wind turbine generator set within the future time range includes: Inputting the multimodal features into the first decoding layer for decoding to obtain first decoding intermediate features; Inputting the first decoding intermediate feature into the second decoding layer for decoding to obtain a second decoding intermediate feature; The second decoded intermediate feature is input into the third decoding layer for decoding to obtain the expected power generated by the wind turbine generator set within the future time range.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor. The computer program is executed by the at least one processor to enable the at least one processor to execute the wind power-based business operation method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the wind power-based business operation method according to any one of claims 1 to 7 is implemented.

Images