Power grid dispatching optimization method based on deep learning

Abstract

The invention provides a power grid dispatching optimization method based on a deep learning network, and the method comprises the following steps: collecting power output data and spatio-temporal data of a wind driven generator according to time, and carrying out the preprocessing of the data; designing a multi-modal network structure suitable for wind power generation power prediction; designing a loss function to train the multi-modal network to obtain a trained network model; and carrying out real-time prediction on the wind power generation power by using the trained model. By adopting the deep learning method, the power grid dispatching optimization method based on the deep learning network optimizes the focusing of key information by utilizing an attention mechanism in deep learning, improves the accuracy of wind power generation power prediction, predicts the wind power generation power, can realize better power grid energy management and dispatching optimization, and improves the power grid dispatching optimization efficiency. The utilization efficiency of renewable energy sources is improved, dependence on traditional energy sources is reduced, and the stability and efficiency of a power grid are ensured.

Description

Technical Field

[0001] The present invention relates to the intersection of artificial intelligence and power grid dispatching optimization, and specifically to a power grid dispatching optimization method based on deep learning. Background Art

[0002] In modern power systems, grid dispatch optimization is a key factor in ensuring grid stability and improving energy efficiency. Traditional grid dispatch methods rely on preset rules and manual decisions, which often seem inadequate when dealing with the instability and unpredictability of renewable energy, especially wind power generation. As the proportion of renewable energy in the power system continues to increase, traditional methods often find it difficult to maximize the potential of wind power generation and effectively integrate it into the power grid due to their lack of real-time and flexibility.

[0003] With the rapid development of machine learning and artificial intelligence technology, deep learning technology has been widely studied and applied to the prediction and optimization tasks of power systems. However, existing deep learning models usually only consider time series prediction when processing wind power data, ignoring the spatial position relationship of generators, which will have an important impact on the prediction accuracy due to factors such as uneven spatial distribution of wind resources, wake effect, and interference effect. Therefore, studying how to comprehensively utilize time series and spatial position information to improve the accuracy of wind power prediction and the efficiency of power grid dispatch has become a research direction in current power system management.

[0004] The Chinese patent with publication number CN 117709502 A discloses "A wind power prediction method based on CEEMDAN-IHBA-XGBoost model", proposing a method that can effectively process and predict the power output of wind farms through multi-modal decomposition and optimization prediction of wind power. This method can use the maximum information coefficient (MIC) for feature selection, enhance the signal-to-noise ratio of the data through the CEEMDAN algorithm, and finally use XGBoost for accurate prediction. However, this method mainly relies on traditional time series prediction technology and does not make full use of the spatial location data of wind power stations, which may limit the comprehensiveness and accuracy of the prediction model in highly dynamic and complex wind scenarios. In addition, although the IHBA algorithm is introduced to optimize the hyperparameters of XGBoost, the flexibility and real-time performance of this method in processing high-dimensional data and complex network structures may not be as good as the method using deep learning and attention mechanism. Summary of the invention

[0005] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a power grid dispatch optimization method based on deep learning to solve the problems raised in the above background technology. The present invention provides a new multimodal network that can comprehensively utilize the spatiotemporal data of wind turbines, convert the spatial position data into 2D images, and combine the one-dimensional signal processing of time series information to comprehensively capture the spatial relationship and dynamic changes of each generator in the wind farm. This method uses the attention mechanism in deep learning to optimize the focus on key information and improve the accuracy of wind power generation prediction.

[0006] In order to achieve the above object, the present invention is implemented by the following technical solution: a power grid dispatch optimization method based on deep learning, comprising the following steps:

[0007] S1. Collect the historical power output data and spatiotemporal data of each wind turbine in the wind power plant and pre-process the data.

[0008] S2. Design a multimodal network structure suitable for wind power prediction, including a one-dimensional information feature extraction module, a multi-channel image space feature extraction module and a feature fusion module;

[0009] S3. Design a loss function to train the multimodal network and obtain a trained network model;

[0010] S4. Use the trained model to make real-time predictions of wind power generation to optimize grid dispatch.

[0011] Furthermore, step S1 specifically includes the following contents:

[0012] S1-1. Obtain the power generation data, wind speed, wind direction, air pressure and latitude and longitude information of each wind turbine in the wind power plant every 4 hours within one year;

[0013] S1-2. Clean the acquired data. First, use the weighted linear interpolation method to process the missing values ​​of wind speed and wind direction. The formula is:

[0014] x missing =w1·x before +w2·x after

[0015]

[0016] Among them, x missing is the predicted value for the missing time point; x before and x after are the known values ​​before and after the missing point respectively; w1 and w2 are weights, calculated based on time distance; t before , t after , tmissing is the corresponding time point.

[0017] Then the IQR rule is used to process the abnormal values ​​of wind speed, and the formula is:

[0018] IQR=Q3-Q1

[0019] LowerBound=Q1-k×IQR

[0020] UpperBound=Q3+k×IQR

[0021] Where Q1 is the lower quartile (25%) of all values ​​in the data set; Q3 is the upper quartile (75%) of all values ​​in the data set; IQR is the interquartile range; k is an adjustable parameter, which is set according to the distribution sensitivity of the data; Upper Bound and Lower Bound are the upper and lower bounds respectively, and any value below the lower bound or above the upper bound needs to be eliminated.

[0022] S1-3. Convert the acquired data to convert the wind direction from an angle (0-360 degrees) into sine and cosine values. The formula is as follows:

[0023]

[0024] Where "Wind Direction" is the wind direction angle in degrees.

[0025] S1-4. Remove the proportional effect in the data by normalization. The formula is:

[0026]

[0027] Among them, x[n] is the original data, x normalized [n] is the normalized data, min(x) is the minimum value in the data, max(x) is the maximum value in the data, and α and β are adjustable scaling factors and offsets.

[0028] S1-5. Perform sine and cosine transformation on the longitude and latitude coordinates of the wind turbine. First, convert the degrees into radians. The formula is as follows:

[0029]

[0030] Among them, Radians Lat、Long Degrees are the radians corresponding to latitude and longitude respectively; Lat、Long are latitude and longitude respectively; π is pi.

[0031] Then calculate the sine and cosine values, the formula is as follows:

[0032] Latitude cos = cos(Radians Lat )

[0033] Latitude sin =sin(Radians Lat )

[0034] Longitude cos = cos(Radians Long )

[0035] Longitude sin =sin(Radians Long )

[0036] Among them, Latitude cos is the cosine value of the dimension; Latitude sin is the sine value of the dimension; Longitude cos is the cosine value of longitude; sin is the sine of longitude.

[0037] S1-6. Arrange all data in a time series format so that it can be input into the one-dimensional feature extraction module.

[0038] Furthermore, the S1 further includes:

[0039] S1-7. Define an M×N grid and map the geographic coordinates of each wind turbine to the grid coordinates. The formula is:

[0040]

[0041] Where x and y are the grid coordinates of the generator; N and M are the number of columns and rows of the grid, respectively, which are determined by the specific distribution and number of generators; longitude and latitude are the actual longitude and latitude of the power station; max_longitude and min_latitude are the maximum values ​​of the longitude and latitude of all generators, respectively; min_longitude and min_latitude are the minimum values ​​of the longitude and latitude of all generators, respectively.

[0042] S1-8. For each generator, use the x and y calculated above as grid coordinates to map the wind speed, wind direction, atmospheric pressure, and power generation data to the corresponding grid cells.

[0043] S1-9. Generate an image based on the data values ​​on the grid. The image has 5 channels, which are respectively allocated by 5 parameters: wind speed, sine and cosine of wind direction, atmospheric pressure, and power generation.

[0044] Furthermore, the multimodal network structure in S2 specifically includes:

[0045] S2-1. Design a multi-channel image spatial feature extraction module, as follows:

[0046] S2-1-1, input multi-channel 2D image, use 3×3 convolution kernel for initial convolution, followed by batch normalization and LeakyReLU activation function, then use 1×1 convolution kernel for deep feature extraction, and then go through batch normalization and LeakyReLU activation function again;

[0047] S2-1-2, introduce the channel and spatial dual attention mechanism. The first is the channel attention mechanism, which compresses the two-dimensional feature map of each channel into a single value through a squeezing operation. The formula is as follows:

[0048]

[0049] where z c represents the global feature of the cth channel, which is the global average of all features of the channel; u c is the feature map of the cth channel, u c (i, j) is the value of the feature map at position (i, j); H and W are the height and width of the feature map; ∑ is the summation symbol, which is used to accumulate the value of the entire feature map; is the normalization factor for the averaging operation.

[0050] Then comes the excitation operation, which uses a small fully connected neural network to learn the weights of each channel to strengthen the features of important channels. The formula is as follows:

[0051] s c =σ(W2δ(W1z c ))

[0052] where s c represents the weight vector of the cth channel; z is the global feature vector from the squeeze operation, which is input into the excitation network; W1 and W2 represent the weight matrices of the two fully connected layers, the first fully connected layer is used for dimensionality reduction, and the second fully connected layer is used to restore the dimension; δ is the ReLU activation function, which is used to increase nonlinearity; σ is the Sigmoid activation function, which is used to limit the weight between 0 and 1.

[0053] The last step is the recalibration operation, which is used to multiply the calculated channel weights by the original feature map channel by channel to emphasize the features of important channels. The formula is:

[0054]

[0055] where s c is the weight of channel c, is the weighted feature map.

[0056] Then comes the spatial attention mechanism, the formula is as follows:

[0057]

[0058] Among them A s is the generated spatial attention map; Conv 3×3 is a 3×3 convolution kernel, used to aggregate features across channels; σ is the Sigmoid function; It is the feature map processed by the channel attention mechanism.

[0059] Finally, the spatial attention mechanism is applied, and the formula is as follows:

[0060]

[0061] represents the feature map after spatial attention adjustment; ⊙ is the element multiplication operation, representing A s Each spatial weight in Multiply the values ​​at corresponding positions in .

[0062] S2-1-3. Add a global average pooling layer, a Dropout layer, and a fully connected layer after the attention mechanism, apply an adaptive activation function, and output the final feature vector; the adaptive parameterized ReLU (APReLU) activation function is expressed as follows:

[0063]

[0064] Where x is the input of the activation function, and α, β, γ, δ are learnable parameters that can be dynamically adjusted according to the input features during training.

[0065] Furthermore, the S2 also includes the following contents:

[0066] S2-2. Design a one-dimensional information feature extraction module, as follows:

[0067] S2-2-1. Input the preprocessed time series data, first use a 3×1 convolution kernel for convolution to capture short-term dependencies, then pass it through the ReLU activation function, and then use a 5×1 convolution kernel for convolution to capture longer-term dependencies.

[0068] S2-2-2. Apply 2×2 average pooling after convolution, and then apply a batch normalization layer.

[0069] S2-2-3. Introduce the temporal attention mechanism and use a small convolutional network to generate the temporal attention weight. The formula is as follows:

[0070] A t =σ(Conv 1×1 (F conv2 ))

[0071] Among them A t is the temporal attention map; F conv2 It is the feature map after the second convolution; Conv 1×1 is a small convolutional layer for generating temporal attention weights; σ is the Sigmoid activation function.

[0072] Then the attention weights are applied, as follows:

[0073]

[0074] represents the feature map after temporal attention adjustment; ⊙ is the element multiplication operation.

[0075] S2-2-4. After applying the temporal attention mechanism, the result Input to the LSTM layer to capture discontinuous patterns in the time series.

[0076] S2-2-5. After LSTM, the output is globally average pooled, then the ELU activation function is applied, and finally the Dropout layer and the fully connected layer are applied.

[0077] S2-3, design multi-channel image spatial feature extraction, and perform feature fusion on the outputs of the one-dimensional information feature extraction module and the multi-channel image spatial feature extraction module, as follows:

[0078] S2-3-1. Perform global average pooling on the output features of the two modules to ensure the same dimensions, and then use Z-score to standardize the features.

[0079] S2-3-2. Concatenate the processed one-dimensional and two-dimensional feature vectors. The formula is as follows:

[0080] F combined =[F 1D ; F 2D ]

[0081] where F combined represents the joint eigenvector after concatenation; F 1D and F 2D They represent the features output by the one-dimensional feature extraction module and the multi-channel image space feature extraction module respectively.

[0082] Then use a fully connected layer from F combined The feature weights are learned in the following formula:

[0083] w 1D ,w 2D =softmax(W·F combined +b)

[0084] where w 1D ,w 2D is the obtained weight; W is the weight matrix of the fully connected layer; b is the bias term; F combined represents the joint feature vector; the softmax activation function ensures that the sum of the output weights is 1.

[0085] Then perform weighted feature fusion, the formula is as follows:

[0086] F fused =w 1D ·F 1D +w 2D ·F 2D

[0087] where F fused represents the fused features; w 1D and w 2D is the weight obtained through training; F 1D and F 2D They represent the features output by the one-dimensional feature extraction module and the multi-channel image space feature extraction module respectively.

[0088] S2-3-3. After feature fusion is completed, the first fully connected layer is applied, using the ReLU activation function; then the Dropout layer is added; then the second fully connected layer is applied, using the ReLU activation function. Finally, the predicted value is output.

[0089] Furthermore, the loss function designed in S3 is as follows:

[0090] S3-1. The loss function consists of three parts: prediction error loss, modal coordination loss, and regularization term.

[0091] S3-1-1. The prediction error loss part measures the difference between the predicted value and the actual value through the mean square error. The formula is as follows:

[0092]

[0093] where y i is the actual power value of sample i, is the power value predicted by the model and N is the total number of samples.

[0094] S3-1-2, modality coordination loss aims to optimize the coordination between different modality features (image and time series data), and the formula is as follows:

[0095]

[0096] Among them, F 1D,j and F 2D,j They represent the extraction of one-dimensional and two-dimensional features on the j-th sample respectively; corr represents the correlation coefficient; M is the number of modes.

[0097] S3-1-3, the regularization term is used to control the complexity of the model and prevent overfitting. The formula is as follows:

[0098]

[0099] where w k represents the weight of each layer of the model, and λ is the regularization coefficient.

[0100] Furthermore, the final loss function of the three parts is expressed as follows:

[0101] L=L pred +αL modal +L reg

[0102] Here, α is an adjustable hyperparameter used to balance the importance of prediction error loss and modality coordination loss.

[0103] Furthermore, the detailed steps of performing real-time prediction and optimizing power grid dispatch in S4 are as follows:

[0104] S4-1. Before making a real-time forecast, it is necessary to ensure that all relevant real-time data sources are ready and can be accessed in real time. These data include: wind speed and direction data, air pressure and temperature data, power generation data, and latitude and longitude data of each generator.

[0105] S4-2. As in the training phase, all real-time data must go through the same preprocessing steps to ensure data consistency and model effectiveness. This includes data cleaning, data conversion, and feature extraction.

[0106] S4-3. Use the preprocessed data to perform real-time predictions through the trained multimodal network model. The model will output the predicted power generation for the next 7 days.

[0107] S4-4. The predicted wind power data will be directly input into the power grid dispatching system. The system dynamically adjusts the power generation resource allocation in the power grid according to the prediction results and the current load demand of the power grid.

[0108] Furthermore, the following aspects are optimized in S4-4:

[0109] S4-4-1. Power generation efficiency: Adjust the output of the wind farm based on the predicted results of wind power generation to improve power generation efficiency.

[0110] S4-4-2. Power supply reliability: Reduce the risk of power outages due to unstable wind power through forecast data.

[0111] S4-4-3. Cost control: Optimize the power source combination, reduce the use of expensive peak-shaving power stations, and reduce power generation costs.

[0112] Beneficial effects of the present invention:

[0113] 1. This deep learning-based power grid dispatch optimization method uses advanced data preprocessing techniques, such as sine-cosine transformation of wind direction and IQR outlier processing, to effectively improve the quality of input data. This refined processing helps the model better understand and utilize data, thereby reducing prediction errors.

[0114] 2. The deep learning-based power grid dispatch optimization method processes time series and spatial image data simultaneously, and the multimodal network of the present invention can comprehensively analyze the dynamic and spatial distribution of wind turbines. This integrated method provides richer information for the model, which helps to more accurately predict wind power generation in the next few hours to a week.

[0115] 3. By converting the spatial position of wind turbines into 2D images, the present invention can capture the relative positions and potential interactions between turbines. This consideration of spatial relationships significantly improves the model's prediction accuracy for the overall wind farm dynamics; the model's integrated channel and spatial attention mechanisms enable the model to automatically identify and focus on the most important data features for prediction, such as areas of sudden wind speed changes. This dynamic attention capability makes predictions more accurate and responses faster.

[0116] 4. The present invention designs a composite loss function including modal coordination loss, which not only optimizes the prediction error, but also strengthens the consistency between different data modes. This method ensures that the model achieves optimal coordination between different data sources, thereby improving the stability and accuracy of the overall prediction; through the application of technologies such as algorithm optimization and hardware acceleration, it can be quickly deployed and maintain efficient operation in a resource-constrained environment. This is particularly critical for power grid operations that require real-time scheduling, making the model not only accurate but also practical.

[0117] 5. The present invention can quickly respond to environmental changes and grid demand by processing and analyzing the spatiotemporal data of wind turbines in real time. This rapid data processing capability enables grid operators to adjust grid load and power generation plans in real time to cope with sudden changes in wind speed and direction, thereby reducing energy waste caused by insufficient or excessive power generation.

[0118] 6. The prediction results of the present invention can help operators allocate and utilize renewable energy, especially wind energy, more efficiently. By accurately predicting wind power generation in the next few days or even hours, the power grid can more finely adjust the ratio between different energy sources, optimize the overall energy mix, and reduce dependence on traditional fossil fuels. BRIEF DESCRIPTION OF THE DRAWINGS

[0119] Figure 1 A flowchart of a power grid dispatch optimization method based on a deep learning network provided in an embodiment of the present invention;

[0120] Figure 2 A network structure diagram of a power grid dispatch optimization method based on a deep learning network provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0121] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0122] See also Figure 1 to Figure 2 The present invention provides the following technical solution: A power grid dispatch optimization method comprises the following steps:

[0123] S1. Collect historical power output data and spatiotemporal data of wind turbines and preprocess the data.

[0124] S2. Design a multimodal network structure suitable for wind power prediction, including a one-dimensional information feature extraction module and a multi-channel image space feature extraction module;

[0125] S3. Design a loss function to train the multimodal network and obtain a trained network model;

[0126] S4. Use the trained model to make real-time predictions of wind power generation to optimize grid dispatch.

[0127] The step S1 is specifically as follows:

[0128] S1-1. Obtain the power generation data, wind speed, wind direction, air pressure and latitude and longitude information of each wind turbine in the wind power plant every 4 hours within one year. Because obtaining at least one year of data can include changes in all seasons, it helps the model learn the differences in wind power generation patterns in different seasons, and obtaining power generation data every 4 hours helps capture more detailed changes in wind speed and wind direction, which is particularly important for real-time scheduling.

[0129] S1-2. Clean the acquired data. First, use the weighted linear interpolation method to process the missing values ​​of wind speed and wind direction. In traditional linear interpolation, the value of the predicted point is based on the average of two adjacent points. Weighted linear interpolation introduces time or space distance as a weight to weight adjacent points, so that observation points closer to the predicted point have greater influence. The improved formula is:

[0130] x missing =w1·x before +w2·x after

[0131]

[0132] Among them, x missing is the predicted value for the missing time point; x before and x after are the known values ​​before and after the missing point respectively; w1 and w2 are weights, calculated based on time distance; t before , t after , t missing is the corresponding time point.

[0133] Then the IQR rule is used to process the outliers of wind speed. IQR (interquartile range) is a commonly used statistical method to identify outliers in data. IQR is based on the quartiles of data and is a very robust outlier detection method that is not affected by extreme values ​​in the data. Its formula is:

[0134] IQR=Q3-Q1

[0135] LowerBound=Q1-k×IQR

[0136] UpperBound=Q3+k×IQR

[0137] Where Q1 is the lower quartile (25%) of all values ​​in the data set; Q3 is the upper quartile (75%) of all values ​​in the data set; IQR is the interquartile range; k is an adjustable parameter, which is set according to the distribution sensitivity of the data; Upper Bound and Lower Bound are the upper and lower bounds respectively. Any value below the lower bound or above the upper bound is considered an outlier and needs to be removed.

[0138] S1-3. Perform data conversion on the obtained data. Wind direction is angular data, usually expressed in degrees (0° to 360°). In order to make the wind direction data more meaningful in the mathematical model, the wind direction is converted from an angle to sine and cosine values, which can maintain its cyclic nature and avoid mutations. For example, 359 degrees and 1 degree are very different in value, but in fact they point to almost the same direction. Using only a single angle value may cause the model to misunderstand this close relationship. By using sine and cosine conversion, the wind direction data can correctly express its periodicity mathematically, ensuring that, for example, the north wind (0° or 360°) and the direction close to the north wind (such as 1° or 359°) are appropriately represented as similar in the model. Therefore, conversion is required, and the formula is as follows:

[0139]

[0140]

[0141] Where "Wind Direction" is the wind direction angle in degrees. The sine and cosine functions provide a way to decompose an angle into two orthogonal components (on the unit circle). Wind_Cos represents the projection of the angle on the horizontal axis, reflecting the east-west component. Wind_sin represents the projection of the angle on the vertical axis, reflecting the north-south component.

[0142] S1-4. Normalize the data. Normalization is the process of adjusting the scale of the data so that it is within a small, specific range (between 0 and 1), which helps improve algorithm performance and computational efficiency. The normalization method used here introduces a scaling factor to further control the scaling range and center point of the data to adjust the final distribution of the data. The formula is:

[0143]

[0144] Among them, x[n] is the original data, x normalized [n] is the normalized data, min(x) is the minimum value in the data, max(x) is the maximum value in the data, and α and β are adjustable scaling factors and offsets.

[0145] S1-5. Longitude and latitude are strong geographic indicators in themselves, but directly inputting them into the model as features is not the most effective approach. The spatial data (longitude and latitude) needs to be converted into a format that is more useful for machine learning models. This method performs sine and cosine transformations on the longitude and latitude coordinates of wind turbines to help the model capture the continuity and periodicity of the geographic location and avoid discontinuities caused by sudden changes, such as from 180 degrees to -180 degrees. First, convert degrees to radians. Longitude and latitude are usually in degrees and need to be converted to radians for trigonometric calculations. The formula is as follows:

[0146]

[0147] Among them, Radians Lat、Long Degrees are the radians corresponding to latitude and longitude respectively; Lat、Long are latitude and longitude respectively; π is pi.

[0148] Then calculate the sine and cosine values ​​of the latitude to represent the north-south position information, and the formula is as follows:

[0149] Latitude cos = cos(Radians Lat )

[0150] Latitude sin =sin(Radians Lat )

[0151] Then calculate the sine and cosine values ​​of the longitude to represent the position information in the east-west direction. The formula is as follows:

[0152] Longitude cos = cos(Radians Long )

[0153] Longitude sin =sin(Radians Long )

[0154] Among them, Latitude cos is the cosine value of the dimension; Latitude sin is the sine value of the dimension; Longitude cos is the cosine value of longitude; sin is the sine of longitude.

[0155] S1-6. Arrange all data in a time series format so that it can be input into the one-dimensional feature extraction module.

[0156] S1-7. The multi-channel image spatial feature extraction module requires a preprocessing process that maps the wind speed, wind direction, air pressure, and power generation data of the wind turbine to a two-dimensional grid and converts it into a multi-channel image processing. First, define an M×N grid based on the spatial coverage and actual distribution accuracy of the generator, and map the geographic coordinates of each wind turbine to the grid coordinates. The formula is:

[0157]

[0158] Where x and y are the grid coordinates of the generator; N and M are the number of columns and rows of the grid, respectively, which are determined by the specific distribution and number of generators; longitude and latitude are the actual longitude and latitude of the power station; max_longitude and min_latitude are the maximum values ​​of the longitude and latitude of all generators, respectively; min_longitude and min_latitude are the minimum values ​​of the longitude and latitude of all generators, respectively.

[0159] S1-8. For each generator, use the x and y calculated above as grid coordinates to map the wind speed, wind direction, atmospheric pressure, and power generation data to the corresponding grid cells.

[0160] S1-9. Generate an image based on the data values ​​on the grid. The image has 5 channels. Each parameter (wind speed, sine and cosine of wind direction, air pressure, power generation) can be assigned to a channel of the image, similar to the RGB channels of a color image.

[0161] As an embodiment of the present application, the multi-channel image spatial feature extraction module in step S2 specifically includes:

[0162] S2-1-1. Input a multi-channel 2D image and use a 3×3 convolution kernel for initial convolution, followed by batch normalization and LeakyReLU activation function to enhance the learning stability and nonlinear expression ability of the model. Then use a 1×1 convolution kernel for deep feature extraction, and then go through batch normalization and LeakyReLU activation function again;

[0163] S2-1-2. Introduce the dual attention mechanism of channel and space. First, apply channel attention to weight certain channels, and then use spatial attention to further refine the important areas of the feature map. The first is the channel attention mechanism, which compresses the two-dimensional feature map of each channel into a single value through a squeezing operation. The formula is as follows:

[0164]

[0165] where z c represents the global feature of the cth channel, which is the global average of all features of the channel; u c is the feature map of the cth channel, u c (i, j) is the value of the feature map at position (i, j); H and W are the height and width of the feature map; ∑ is the summation symbol, which is used to accumulate the value of the entire feature map; is the normalization factor for the averaging operation.

[0166] Then comes the excitation operation, which uses a small fully connected neural network to learn the weights of each channel to strengthen the features of important channels. The formula is as follows:

[0167] s c =σ(W2δ(W1z c ))

[0168] where s c represents the weight vector of the cth channel; z is the global feature vector from the squeeze operation, which is input into the excitation network; W1 and W2 represent the weight matrices of the two fully connected layers, the first fully connected layer is used for dimensionality reduction, and the second fully connected layer is used to restore the dimension; δ is the ReLU activation function, which is used to increase nonlinearity; σ is the Sigmoid activation function, which is used to limit the weight between 0 and 1.

[0169] The last step is the recalibration operation, which is used to multiply the calculated channel weights by the original feature map channel by channel to emphasize the features of important channels. The formula is:

[0170]

[0171] where s c is the weight of channel c, is the weighted feature map.

[0172] Then comes the spatial attention mechanism, which requires generating a spatial attention map. Specifically, a 3×3 convolution kernel is used to aggregate features across channels, and then the importance of each position is obtained through the Sigmoid function. The formula is as follows:

[0173]

[0174] Among them A s The generated spatial attention map is a two-dimensional matrix of the same size as the input feature map, where the value of each element is between 0 and 1, representing the importance weight of the corresponding position; Conv 3×3 is a 3×3 convolution kernel used to aggregate input feature maps across channels The convolution operation does not change the spatial dimension of the feature map because padding is used to keep the dimension unchanged. σ is the Sigmoid function, which is used to normalize the convolution output to the range of [0,1] to ensure that the result can be used as a multiplication weight. It is the feature map processed by the channel attention mechanism.

[0175] Finally, the spatial attention mechanism is applied to the feature map after channel adjustment to strengthen the features of important spatial areas. The formula is as follows:

[0176]

[0177] represents the feature map after spatial attention adjustment, which is the data to be passed to the next processing stage; ⊙ is the element multiplication operation, indicating A s Each spatial weight in This step is performed element by element, which means that each pixel is weighted by its corresponding spatial attention weight to emphasize the more important parts of the feature map; The adjusted feature map output from the channel attention mechanism has been preliminarily adjusted by the channel weights, and this step further refines it by the spatial weights.

[0178] S2-1-3. A global average pooling layer is added after the attention mechanism to reduce parameters and prepare for classification or regression tasks; then a Dropout layer is introduced to prevent overfitting; all features are subsequently integrated through a fully connected layer; finally, an adaptive parameterized ReLU activation function (APReLU) is applied for final activation, the parameters of which (slope and intercept) are dynamically adjusted according to each feature (regarded as an independent channel) to adapt to the characteristics of different features, and the final feature vector is output; the adaptive parameterized ReLU (APReLU) activation function is expressed as follows:

[0179]

[0180] Where x is the input of the activation function, and α, β, γ, δ are learnable parameters that can be dynamically adjusted according to the input features during training.

[0181] As an embodiment of the present application, the one-dimensional information feature extraction module in step S2 specifically includes:

[0182] S2-2-1. Input the preprocessed time series data, first use a 3×1 convolution kernel for convolution to capture short-term dependencies, then pass it through the ReLU activation function to increase nonlinearity to help the model learn more complex data patterns, and then use a 5×1 convolution kernel for a second convolution to capture longer-term time dependencies.

[0183] S2-2-2. Apply a 2×2 average pooling operation after convolution to reduce the temporal dimension of the data while retaining important information; then apply a batch normalization layer to stabilize network training, speed up convergence, and reduce internal covariate shift.

[0184] S2-2-3. Introduce the temporal attention mechanism to enhance the model's attention to key time points. Specifically, a small convolutional network is used to generate temporal attention weights, which are implemented through a simple 1x1 convolutional layer followed by a Sigmoid activation function. The size of the attention weight is the same as the feature map, and the formula is as follows:

[0185] A t =σ(Conv 1×1 (F conv2 ))

[0186] Among them A t is the temporal attention map; F conv2 It is the feature map that has been convolved for the second time; Conv 1×1 is a small convolutional layer for generating temporal attention weights; σ is the Sigmoid activation function.

[0187] Then the attention weights are applied, as follows:

[0188]

[0189] represents the feature map after time attention adjustment; ⊙ is the element-wise multiplication of the time attention weight and the feature map to strengthen the model's attention to important time segments.

[0190] S2-2-4. After applying the temporal attention mechanism, the result Input to the LSTM layer. LSTM can effectively retain information in long time series, which is very important for prediction. In time series analysis, many important events or changes are often short-lived and discontinuous. LSTM can not only capture these discontinuous patterns, but also pre-adjust the feature maps through the attention mechanism. The data provided has already emphasized these key time points, allowing LSTM to focus more on those parts that may contain important events or state changes, thereby improving the accuracy and timeliness of predictions.

[0191] S2-2-5. After LSTM, the output is globally averaged pooled to compress each feature dimension of the time series into a single value, simplifying subsequent processing; then the ELU activation function is applied. ELU is an activation function that is particularly useful when processing time series data. Due to its exponential form, ELU can maintain a certain gradient when processing negative values, which is more effective than ReLU and other variants when training deep networks, helping to accelerate the learning process and increase the robustness of the network. The Dropout layer is then applied to further prevent overfitting and enhance the generalization ability of the model; finally, the fully connected layer is applied to integrate the learned features through one or more fully connected layers and output the final prediction results. .

[0192] As an embodiment of the present application, the feature fusion module in step S2 specifically includes:

[0193] S2-3-1. Perform global average pooling on the output features of the two modules respectively. Global average pooling can map multi-dimensional features to one-dimensional vectors, which can unify the dimensions of the outputs of different modules and simplify the subsequent fusion steps. Whether it is two-dimensional image data or one-dimensional time series data, global average pooling can provide a compact global feature description; then use Z-score to standardize the features to ensure that the features of different sources are at the same order of magnitude for easy fusion.

[0194] S2-3-2. After flattening and normalizing the features, the outputs of the two modules are concatenated to form a longer feature vector, as shown below:

[0195] F combined =[F 1D ; F 2D ]

[0196] where F combined represents the joint eigenvector after concatenation; F 1D and F 2D They represent the features output by the one-dimensional feature extraction module and the multi-channel image space feature extraction module respectively.

[0197] Because different features contribute differently to the final task, their importance should be reflected by different weights. To this end, this module introduces a weighted feature fusion method. Weighted fusion enables the model to consider the relative importance of each feature when making decisions, thereby improving the prediction accuracy and the generalization ability of the model. The steps are as follows:

[0198] First, a fully connected layer is used from F combined The role of the fully connected layer here is to combine the features F extracted from different modalities. combined This layer performs linear transformation on the input combined features through the weight matrix W and the bias term b, and then uses the softmax function to ensure that the output weight sum is 1. Such output can be directly used as the coefficient of each feature in the subsequent weighted sum. The formula is as follows:

[0199] w 1D ,w 2D =softmax(W·F combined +b)

[0200] where w 1D ,w 2D is the obtained weight; W is the weight matrix of the fully connected layer; b is the bias term; F combined represents the joint feature vector; the softmax activation function ensures that the sum of the output weights is 1.

[0201] Then perform weighted feature fusion, the formula is as follows:

[0202] F fused =w 1D ·F 1D +w 2D ·F 2D

[0203] where F fused represents the fused features; w 1D and w 2D is the weight obtained through training; F 1D and F 2D They represent the features output by the one-dimensional feature extraction module and the multi-channel image space feature extraction module respectively.

[0204] This weighted fusion approach enables the model to automatically adjust the contribution of different features in the final decision based on their importance. This is especially important when dealing with multimodal data, as different modalities may contribute differently to the final task.

[0205] S2-3-3. After feature fusion, the first fully connected layer is applied, using the ReLU activation function to increase the expressiveness of the model; then the Dropout layer is added to reduce overfitting; then the second fully connected layer is applied, using the ReLU activation function. This step uses two fully connected layers to further process the fused features, helping the model capture the complex relationship between different feature combinations. Finally, the predicted value is output.

[0206] As an embodiment of the present application, the loss function designed in step 3 is designed as follows:

[0207] S3-1. In order to improve the accuracy of wind power prediction and optimize the performance of the network when processing different data modes, a loss function combining prediction error, modal weight adjustment and regularization is proposed. This function considers both prediction accuracy and feature coordination, and improves the model's adaptability to complex data. The details are as follows:

[0208] S3-1-1. The prediction error loss part measures the difference between the predicted value and the actual value through the mean square error, which is a common loss function form in regression tasks. The formula is as follows:

[0209]

[0210] where y i is the actual power value of sample i, is the power value predicted by the model and N is the total number of samples.

[0211] S3-1-2. In order to promote information coordination between different modalities (image spatial features and time series features), a modality weight adjustment loss function based on the internal correlation of the two features is designed. The formula is as follows:

[0212]

[0213] Among them, F 1D,j and F 2D,j They represent the extraction of one-dimensional and image features on the jth sample; corr represents the correlation coefficient, which is used to measure F 2D,j ,F 1D,j A linear relationship exists between them; M is the number of modes.

[0214] Specifically, if F 2D,j and F 1D,j If the two features are completely positively correlated, the correlation coefficient will be 1, and the modal coordination loss part of the loss function will be 0, indicating that the two features are completely coordinated in terms of information. If the two are uncorrelated or negatively correlated, the loss value increases, pushing the model to adjust parameters to enhance the correlation between the two modal features. In this way, modal coordination loss helps the model learn how to better integrate information from different data sources and improve the performance of prediction tasks.

[0215] S3-1-3. Introduce the L2 regularization term to control the model complexity and prevent overfitting. The formula is as follows:

[0216]

[0217] where w k Represents the weights of each layer of the model, which are learned through the Adam optimization algorithm during the training process. λ is the regularization coefficient, which controls the strength of regularization. Increasing λ will increase the penalty for the size of the weight, forcing the weight value to tend to a smaller number, which helps prevent the model from overfitting.

[0218] S3-1-4. The final loss function combining the three parts is expressed as follows:

[0219] L=L pred +αL modal +L reg

[0220] Here, α is an adjustable hyperparameter used to balance the importance of prediction error loss and modality coordination loss.

[0221] Specifically, the loss function is designed to promote the model to learn how to integrate information from different data sources and improve the consistency and accuracy of the overall prediction.

[0222] As an embodiment of the present application, the real-time prediction and power grid dispatch optimization in step S4 specifically includes:

[0223] Before making real-time predictions, S4-1 first ensures that all key real-time data sources can be accessed and processed by the system in real time. These data sources include wind speed, wind direction, air pressure, temperature and real-time power generation data of each wind turbine, as well as the precise geographic location (latitude and longitude) of each turbine.

[0224] S4-2 As in the training phase, all real-time data must undergo a standardized preprocessing process to maintain data consistency and ensure that the accuracy of the model is not affected. The preprocessing steps include data cleaning (eliminating errors and outliers), data conversion (converting wind direction into sine and cosine values), and appropriate feature extraction.

[0225] S4-3 uses pre-processed real-time data to make predictions through a trained multimodal network model. The model uses an integrated attention mechanism and advanced deep learning technology to output predicted power generation every hour for the next 7 days, providing accurate data support for power grid dispatch.

[0226] The wind power data predicted by S4-4 will be transmitted to the power grid dispatching system in real time. The system will dynamically adjust according to these data and the current load demand of the power grid to optimize the allocation of power generation resources. Specific optimization measures include:

[0227] S4-4-1 Power generation efficiency optimization: adjust the power generation output of each wind farm according to the forecast results of wind power generation, optimize the operating status of the generator set and the power generation scheduling plan, so as to maximize the wind power utilization efficiency and overall power generation efficiency.

[0228] S4-4-2 Improved power supply reliability: By using accurate forecasting data, the system is able to identify potential power shortages or surpluses in advance, thereby taking preventive measures such as starting or shutting down auxiliary power generation facilities to reduce the risk of power outages caused by unstable wind power.

[0229] S4-4-3 optimizes the power mix and reduces overall power generation costs by reducing reliance on expensive peak-shaving power stations. At the same time, it achieves both economic and environmental benefits by precisely dispatching renewable energy to reduce carbon emissions and environmental impacts.

[0230] Through the above steps, the present invention not only improves the prediction accuracy of wind power generation, but also significantly improves the operation efficiency and reliability of the power grid through intelligent power grid scheduling, while reducing operating costs, demonstrating its practical application value in modern power grid management.

[0231] The basic principles and main features of the present invention and the advantages of the present invention are shown and described above. For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention.

[0232] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A power grid dispatch optimization method based on deep learning, Features: The steps include: S1. Collect historical power output data and spatiotemporal data of each wind turbine in the wind power plant and pre-process the data; S2. Design a multimodal network structure suitable for wind power prediction, including a one-dimensional information feature extraction module, a multi-channel image space feature extraction module and a feature fusion module; S3. Design a loss function to train the multimodal network and obtain a trained network model; S4. Use the trained model to make real-time predictions of wind power generation to optimize grid dispatch.

2. According to the deep learning-based power grid dispatch optimization method of claim 1, It is characterized in that Step S1 specifically includes the following contents: S1-1. Obtain the power generation data, wind speed, wind direction, air pressure and latitude and longitude information of each wind turbine in the wind power plant every 4 hours within one year; S1-2. Clean the acquired data. First, use the weighted linear interpolation method to process the missing values ​​of wind speed and wind direction. The formula is: x missing =w 1 ·x before +w 2 ·x after Among them, x missing is the predicted value for the missing time point; x before and x after are the known values ​​before and after the missing point respectively; w 1 and w 2 is the weight, calculated based on the time distance; t before , t after , t missing is the corresponding time point; Then the IQR rule is used to process the abnormal values ​​of wind speed, and the formula is: IQR=Q3-Q1 LowerBound=Q1-k×IQR UpperBound=Q3+k×IQR Where Q1 is the lower quartile (25%) of all values ​​in the data set; Q3 is the upper quartile (75%) of all values ​​in the data set; IQR is the interquartile range; k is an adjustable parameter, which is set according to the distribution sensitivity of the data; Upper Bound and Lower Bound are the upper and lower bounds respectively, and any value below the lower bound or above the upper bound needs to be eliminated; S1-3. Convert the acquired data to convert the wind direction from an angle (0-360 degrees) into sine and cosine values. The formula is as follows: Where "Wind Direction" is the wind direction angle in degrees; S1-4. Remove the proportional effect in the data by normalization. The formula is: Among them, x[n] is the original data, x normalized [n] is the normalized data, min(x) is the minimum value in the data, max(x) is the maximum value in the data, α and β are adjustable scaling factors and offsets; S1-5. Perform sine and cosine transformation on the longitude and latitude coordinates of the wind turbine. First, convert the degrees into radians. The formula is as follows: Among them, Radians Lat、Long Degrees are the radians corresponding to latitude and longitude respectively; Lat、Long are latitude and longitude respectively; π is pi; Then calculate the sine and cosine values, the formula is as follows: Latitude cos =cos(Radians Lat ) Latitude sin =sin(Radians Lat ) Longitude cos =cos(Radians Long ) Longitude sin =sin(Radians Long ) Among them, Latitude cos is the cosine value of the dimension; Latitude sin is the sine value of the dimension; Longitude cos is the cosine value of longitude; sin is the sine of longitude; S1-6. Arrange all data in a time series format so that it can be input into the one-dimensional feature extraction module.

3. According to the deep learning-based power grid dispatch optimization method of claim 2, It is characterized in that The S1 further comprises: S1-7. Define an M×N grid and map the geographic coordinates of each wind turbine to the grid coordinates. The formula is: Where x and y are the grid coordinates of the generator; N and M are the number of columns and rows of the grid, respectively, which are determined by the specific distribution and number of generators; longitude and latitude are the actual longitude and latitude of the power station; max_longitude and min_latitude are the maximum values ​​of the longitude and latitude of all generators, respectively; min_longitude and min_latitude are the minimum values ​​of the longitude and latitude of all generators, respectively; S1-8. For each generator, use the x and y calculated above as grid coordinates to map the wind speed, wind direction, atmospheric pressure, and power generation data to the corresponding grid cells; S1-9. Generate an image based on the data values ​​on the grid. The image has 5 channels, which are respectively allocated by 5 parameters: wind speed, sine and cosine of wind direction, atmospheric pressure, and power generation.

4. According to the deep learning-based power grid dispatch optimization method of claim 1, It is characterized in that The multimodal network structure in S2 specifically includes: S2-1. Design a multi-channel image spatial feature extraction module, as follows: S2-1-1, input multi-channel 2D image, use 3×3 convolution kernel for initial convolution, followed by batch normalization and LeakyReLU activation function, then use 1×1 convolution kernel for deep feature extraction, and then go through batch normalization and LeakyReLU activation function again; S2-1-2, introduce the channel and spatial dual attention mechanism. The first is the channel attention mechanism, which compresses the two-dimensional feature map of each channel into a single value through a squeezing operation. The formula is as follows: where z c represents the global feature of the cth channel, which is the global average of all features of the channel; u c is the feature map of the cth channel, u c (i, j) is the value of the feature map at position (i, j); H and W are the height and width of the feature map; ∑ is the summation symbol, which is used to accumulate the value of the entire feature map; is the normalization factor for the averaging operation; Then comes the excitation operation, which uses a small fully connected neural network to learn the weights of each channel to strengthen the features of important channels. The formula is as follows: s c =σ(W 2 δ(W 1 z c )) where s c represents the weight vector of the cth channel; z is the global feature vector from the squeeze operation, which is input to the excitation network; W 1 and W 2 Represents the weight matrices of two fully connected layers, the first fully connected layer is used for dimensionality reduction, and the second fully connected layer is used for dimensionality recovery; δ is the ReLU activation function, which is used to increase nonlinearity; σ is the Sigmoid activation function, which is used to limit the weights between 0 and 1; The last step is the recalibration operation, which is used to multiply the calculated channel weights by the original feature map channel by channel to emphasize the features of important channels. The formula is: where s c is the weight of channel c, is the weighted feature map; Then comes the spatial attention mechanism, the formula is as follows: Among them A s is the generated spatial attention map; Conv 3×3 is a 3×3 convolution kernel, used to aggregate features across channels; σ is the Sigmoid function; It is the feature map processed by the channel attention mechanism; Finally, the spatial attention mechanism is applied, and the formula is as follows: represents the feature map after spatial attention adjustment; ⊙ is the element multiplication operation, representing A s Each spatial weight in Multiply the values ​​at corresponding positions in ; S2-1-3. Add a global average pooling layer, a Dropout layer, and a fully connected layer after the attention mechanism, apply an adaptive activation function, and output the final feature vector; the adaptive parameterized ReLU (APReLU) activation function is expressed as follows: Where x is the input of the activation function, and α, β, γ, δ are learnable parameters that can be dynamically adjusted according to the input features during training.

5. According to the deep learning-based power grid dispatch optimization method of claim 4, It is characterized in that The S2 also includes the following contents: S2-2. Design a one-dimensional information feature extraction module, as follows: S2-2-1. Input the preprocessed time series data, first use a 3×1 convolution kernel for convolution to capture short-term dependencies, then pass through the ReLU activation function, and then use a 5×1 convolution kernel for convolution to capture longer-term dependencies; S2-2-2, apply 2×2 average pooling after convolution, and then apply batch normalization layer; S2-2-3. Introduce the temporal attention mechanism and use a small convolutional network to generate the temporal attention weight. The formula is as follows: A t =σ(Conv 1×1 (F conv2 )) Among them A t is the temporal attention map; F conv2 It is the feature map after the second convolution; Conv 1×1 is a small convolutional layer for generating temporal attention weights; σ is the Sigmoid activation function; Then the attention weights are applied, as follows: Represents the feature map after temporal attention adjustment; ⊙ is the element multiplication operation; S2-2-4. After applying the temporal attention mechanism, the result Input to the LSTM layer to capture discontinuous patterns in the time series; S2-2-5. After LSTM, the output is globally averaged pooled, then the ELU activation function is applied, and finally the Dropout layer and the fully connected layer are applied; S2-3, design multi-channel image spatial feature extraction, and perform feature fusion on the outputs of the one-dimensional information feature extraction module and the multi-channel image spatial feature extraction module, as follows: S2-3-1. Perform global average pooling on the output features of the two modules to ensure the same dimensions, and then use Z-score to standardize the features; S2-3-2. Concatenate the processed one-dimensional and two-dimensional feature vectors. The formula is as follows: F combined =[F 1D ;F 2D ] Among them, F combined represents the joint eigenvector after concatenation; F 1D and F 2D They represent the features output by the one-dimensional feature extraction module and the multi-channel image space feature extraction module respectively; Then use a fully connected layer from F combined The feature weights are learned in the following formula: w 1D ,w 2D =softmax(W·F combined +b) where w 1D ,w 2D is the obtained weight; W is the weight matrix of the fully connected layer; b is the bias term; F combined represents the joint feature vector; the softmax activation function ensures that the sum of the output weights is 1; Then perform weighted feature fusion, the formula is as follows: F fused =w 1D ·F 1D +w 2D ·F 2D Among them, F fused Represents the fused features; w 1D and w 2D is the weight obtained through training; F 1D and F 2D They represent the features output by the one-dimensional feature extraction module and the multi-channel image space feature extraction module respectively; S2-3-3. After feature fusion is completed, the first fully connected layer is applied, using the ReLU activation function; then the Dropout layer is added; then the second fully connected layer is applied, using the ReLU activation function; finally, the predicted value is output.

6. According to the deep learning-based power grid dispatch optimization method of claim 1, It is characterized in that The loss function designed in S3 is as follows: S3-1, the loss function consists of three parts: prediction error loss, modal coordination loss, and regularization term; S3-1-1. The prediction error loss part measures the difference between the predicted value and the actual value through the mean square error. The formula is as follows: where y i is the actual power value of sample i, is the power value predicted by the model, and N is the total number of samples; S3-1-2, modality coordination loss aims to optimize the coordination between different modality features (image and time series data), and the formula is as follows: Among them, F 1D,j and F 2D,j Respectively represent the extraction of one-dimensional and two-dimensional features on the j-th sample; corr represents the correlation coefficient; M is the number of modes; S3-1-3, the regularization term is used to control the complexity of the model and prevent overfitting. The formula is as follows: where w k represents the weight of each layer of the model, and λ is the regularization coefficient.

7. A power grid dispatch optimization method based on deep learning according to claim 6, Features: The final loss function of the three parts is expressed as follows: L=L pred +αL modal +L reg Here, α is an adjustable hyperparameter used to balance the importance of prediction error loss and modality coordination loss.

8. According to the deep learning-based power grid dispatch optimization method of claim 1, Features: The detailed steps of performing real-time prediction and power grid dispatch optimization in S4 are as follows: S4-1. Before making a real-time forecast, it is necessary to ensure that all relevant real-time data sources are ready and can be accessed in real time; these data include: wind speed and direction data, air pressure and temperature data, power generation data, and latitude and longitude data of each generator; S4-2. As in the training phase, all real-time data must go through the same preprocessing steps to ensure data consistency and model effectiveness; this includes data cleaning, data conversion, and feature extraction. S4-3. Use the preprocessed data to perform real-time prediction through the trained multimodal network model; the model will output the predicted power generation for the next 7 days; S4-4. The predicted wind power data will be directly input into the power grid dispatching system; the system will dynamically adjust the allocation of power generation resources in the power grid according to the prediction results and the current load demand of the power grid.

9. A power grid dispatch optimization method based on deep learning according to claim 8, Features: The S4-4 optimizes the following aspects: S4-4-1. Power generation efficiency: adjust the output of the wind farm according to the forecast results of wind power generation to improve power generation efficiency; S4-4-2. Power supply reliability: Reduce the risk of power outages caused by unstable wind power through forecast data; S4-4-3. Cost control: Optimize the power source combination, reduce the use of expensive peak-shaving power stations, and reduce power generation costs.

Images