A scheduling method and device for hair and storage integration, equipment and storage medium

Abstract

The application provides a scheduling method and device for integrated hair and storage, equipment and storage medium, based on historical actual weather data of each weather station in a target area, the target area is divided into multiple target sub-areas; using a first hybrid neural network model to perform data prediction processing on each target sub-area based on the historical actual weather data of each target sub-area, to obtain predicted photovoltaic power generation; using a second hybrid neural network model to perform data prediction processing on each target sub-area based on the historical energy price of each target sub-area, to obtain predicted electricity demand; based on the predicted photovoltaic power generation and the predicted electricity demand, determining the predicted storage power of each target sub-area, and setting the power scheduling strategy between the multiple target sub-areas; compared with the prior art, the technical scheme of the application can realize the optimized allocation of power resources through the photovoltaic power generation and the electricity demand in the target area.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of energy management, in particular to a generation-storage-integrated scheduling method and device, equipment and a storage medium. BACKGROUND

[0002] In the prior art, photovoltaic power generation as a kind of renewable energy has randomness and volatility in power output, which poses a significant challenge to the stable operation and power balance of the power system. With the increasing proportion of new energy grid-connected, the demand for regulation capacity of the power system is increasing, however, the existing technology has deficiencies in utilizing energy storage systems and other flexible regulation resources, which limits the regulation capacity of the power grid in response to supply and demand fluctuations.

[0003] In addition, the traditional optimization scheduling method often makes decisions only according to the current electricity market price, without fully considering the prediction information of photovoltaic power generation output and user-side electricity demand, thus failing to fully realize the regulation potential and economic benefits of energy storage systems in the power grid, and also failing to achieve effective allocation and optimized utilization of power resources. SUMMARY

[0004] The technical problem to be solved by the application is to provide a generation-storage-integrated scheduling method, device, equipment and storage medium, which realizes the optimized allocation of power resources by the photovoltaic power generation and electricity demand in the target area.

[0005] To solve the above technical problems, the application provides a generation-storage-integrated scheduling method, comprising:

[0006] Obtain historical actual weather data corresponding to each meteorological station in a target area, and perform division processing on the target area based on the historical actual weather data to obtain a plurality of target sub-areas;

[0007] Obtain historical actual weather data corresponding to each target sub-area, and perform data prediction processing on each target sub-area based on the historical actual weather data using a pre-trained first hybrid neural network model to obtain predicted photovoltaic power generation corresponding to each target sub-area;

[0008] Obtain historical energy prices corresponding to each target sub-area, and perform data prediction processing on each target sub-area based on the historical energy prices using a pre-trained second hybrid neural network model to obtain predicted electricity demand corresponding to each target sub-area;

[0009] Based on the predicted photovoltaic power generation and the predicted electricity demand, determine the predicted storage power of each target sub-area, and based on the predicted storage power, set the power scheduling strategy between the plurality of target sub-areas.

[0010] In a possible implementation, the historical actual weather data corresponding to each weather station in the target region is acquired, and the target region is divided based on the historical actual weather data to obtain a plurality of target sub-regions, specifically including:

[0011] The historical actual weather data corresponding to each weather station in the target region is acquired, and a weather index corresponding to the historical actual weather data in each weather station is calculated based on a preset weather index calculation formula;

[0012] The target region is divided based on the weather index by using a rotating empirical orthogonal decomposition method to obtain a plurality of target sub-regions.

[0013] In a possible implementation, the power dispatch strategy between the plurality of target sub-regions is set based on the predicted storage power, specifically including:

[0014] The predicted storage power corresponding to each target sub-region is corrected to obtain a corrected storage power, and the plurality of target sub-regions are classified into supply sub-regions and demand sub-regions based on the corrected storage power;

[0015] When it is detected that the supply sub-regions and the demand sub-regions exist at the same time, and the total supply power corresponding to the supply sub-regions calculated is greater than the total demand power corresponding to the demand sub-regions calculated, the distance from each demand sub-region to all supply sub-regions is calculated, and the dispatch strategy between the supply sub-regions and the demand sub-regions is determined based on the distance;

[0016] When it is detected that the supply sub-regions exist and the demand sub-regions do not exist, a charging strategy is performed on the energy storage system;

[0017] When it is detected that the demand sub-regions exist and the supply sub-regions do not exist, a discharging strategy is performed on the energy storage system.

[0018] In a possible implementation, the predicted storage power corresponding to each target sub-region is corrected to obtain a corrected storage power, specifically including:

[0019] A storage power deviation rate is calculated according to a plurality of historical preset storage powers and historical actual storage powers corresponding to each historical preset storage power, and a storage power deviation rate data set is generated;

[0020] The average storage power deviation rate corresponding to the storage power deviation rate data set is calculated based on a preset confidence, and the predicted storage power corresponding to each target sub-region is corrected based on the average storage power deviation rate to obtain a corrected storage power.

[0021] In a possible implementation, the distance from each demand sub-region to all supply sub-regions is calculated respectively, and based on the distance, a scheduling strategy between the supply sub-regions and the demand sub-regions is determined, and specifically, the method comprises the following steps.

[0022] Obtaining respective demand power of all demand sub-regions, and based on the demand power, the all demand sub-regions are sorted to obtain a demand sub-region sequence;

[0023] A single target demand sub-region is selected from the demand sub-region sequence in sequence, the distance from the target demand sub-region to all supply sub-regions is calculated, and based on the distance, the all supply sub-regions are sorted in sequence from near to far to obtain a supply sub-region sequence corresponding to the target demand sub-region;

[0024] The supply sub-region sequence is detected in sequence from left to right, when a target quantity of target supply sub-regions corresponding to a target supply power is greater than a target demand power corresponding to the target demand sub-region, the target quantity of target supply sub-regions are taken as the power scheduling region of the target demand sub-region, and a supply sub-region power scheduling strategy is generated, and based on the supply sub-region power scheduling strategy, the respective supply power of the all supply sub-regions is updated, until all demand sub-regions in the demand sub-region sequence are selected, and a scheduling strategy between the supply sub-regions and the demand sub-regions is determined.

[0025] In a possible implementation, the pre-training process of the first hybrid neural network model comprises the following steps.

[0026] An initial first hybrid neural network model is constructed, wherein the first hybrid neural network model comprises a first convolutional neural network and a first graph neural network, and the first convolutional neural network is connected with the first graph neural network;

[0027] Obtaining power generation data samples and weather data samples corresponding to the same moment, inputting the weather data samples into the first hybrid neural network model, so that the first convolutional neural network in the first hybrid neural network model extracts features from the weather data samples respectively to obtain a first feature map matrix;

[0028] Based on the first convolutional neural network, the first feature map matrix is input into the first graph neural network, so that the first graph neural network performs feature fusion processing on the first feature map matrix to obtain a first feature fusion graph;

[0029] input the first feature fusion graph into the first convolutional neural network model based on the first graph neural network, so that the first convolutional neural network extracts features of the first feature fusion graph, to obtain predicted power generation data;

[0030] Compare the predicted power generation data and the power generation data sample to determine a first hybrid neural network model.

[0031] In a possible implementation, the inputting the weather data sample into the first hybrid neural network model so that the first convolutional neural network in the first hybrid neural network model respectively extracts features of the weather data sample to obtain a first feature graph matrix specifically includes:

[0032] The weather data sample includes irradiance, cloud cover, humidity, and average solar sunshine time;

[0033] The weather data sample is input into the first hybrid neural network model, so that the first convolutional neural network in the first hybrid neural network model respectively extracts features of the irradiance, the cloud cover, the humidity, and the average solar sunshine time to obtain feature weights corresponding to the irradiance, the cloud cover, the humidity, and the average solar sunshine time respectively;

[0034] Based on the feature weights, an irradiance feature graph matrix, a cloud cover feature graph matrix, a humidity feature graph matrix, and an average solar sunshine time feature graph matrix are respectively constructed, and the irradiance feature graph matrix, the cloud cover feature graph matrix, the humidity feature graph matrix, and the average solar sunshine time feature graph matrix are integrated to obtain a first feature graph matrix.

[0035] The application further provides a dispatching device integrating power generation and storage, comprising a region division module, a photovoltaic power generation amount prediction module, a power demand prediction module, and an electric quantity dispatching strategy generation module.

[0036] The region division module is configured to acquire historical actual weather data corresponding to each meteorological station in a target region, and perform division processing on the target region based on the historical actual weather data to obtain a plurality of target sub-regions.

[0037] The photovoltaic power generation amount prediction module is configured to acquire historical actual weather data corresponding to each target sub-region, and perform data prediction processing on each target sub-region based on the historical actual weather data by using a pre-trained first hybrid neural network model to obtain predicted photovoltaic power generation amounts corresponding to each target sub-region.

[0038] The electricity demand prediction module is configured to acquire historical energy prices corresponding to each target sub-region, and perform data prediction processing on each target sub-region based on the historical energy prices by using a pre-trained second hybrid neural network model to obtain predicted electricity demand of each target sub-region corresponding to the historical energy prices.

[0039] The electricity scheduling strategy generation module is configured to determine predicted storage electricity of each target sub-region based on the predicted photovoltaic power generation and the predicted electricity demand, and set an electricity scheduling strategy between the target sub-regions based on the predicted storage electricity.

[0040] In a possible implementation, the region division module is configured to acquire historical actual weather data corresponding to each weather station in a target region, and perform division processing on the target region based on the historical actual weather data to obtain a plurality of target sub-regions, and specifically includes:

[0041] The historical actual weather data corresponding to each weather station in the target region is acquired, and a weather index corresponding to the historical actual weather data in each weather station is calculated based on a preset weather index calculation formula.

[0042] The target region is divided based on the weather index by using a rotating empirical orthogonal decomposition method to obtain a plurality of target sub-regions.

[0043] In a possible implementation, the electricity scheduling strategy generation module is configured to set the electricity scheduling strategy between the target sub-regions based on the predicted storage electricity, and specifically includes:

[0044] The predicted storage electricity corresponding to each target sub-region is corrected to obtain corrected storage electricity, and the target sub-regions are classified into supply sub-regions and demand sub-regions based on the corrected storage electricity.

[0045] When it is detected that the supply sub-regions and the demand sub-regions exist at the same time, and the total supply electricity of the supply sub-regions calculated is greater than the total demand electricity of the demand sub-regions calculated, the distance from each demand sub-region to all supply sub-regions is calculated, and the scheduling strategy between the supply sub-regions and the demand sub-regions is determined based on the distance.

[0046] When it is detected that the supply sub-regions exist and the demand sub-regions do not exist, a charging strategy is performed on the energy storage system.

[0047] When it is detected that the demand sub-regions exist and the supply sub-regions do not exist, a discharging strategy is performed on the energy storage system.

[0048] In a possible implementation, the power scheduling strategy generation module is configured to respectively correct the predicted storage power corresponding to each target sub-region to obtain a corrected storage power, and specifically includes:

[0049] According to a plurality of historical preset storage powers and a historical actual storage power corresponding to each historical preset storage power, a storage power deviation rate is calculated, and a storage power deviation rate dataset is generated;

[0050] The average storage power deviation rate corresponding to the storage power deviation rate dataset is calculated based on the preset confidence, and the predicted storage power corresponding to each target sub-region is corrected based on the average storage power deviation rate to obtain a corrected storage power.

[0051] In a possible implementation, the power scheduling strategy generation module is configured to respectively calculate the distance from each demand sub-region to all supply sub-regions, and determine the scheduling strategy between the supply sub-region and the demand sub-region based on the distance, and specifically includes:

[0052] Obtain the demand power corresponding to each of all demand sub-regions, and sort all demand sub-regions based on the demand power to obtain a demand sub-region sequence;

[0053] Select a single target demand sub-region from the demand sub-region sequence in sequence, calculate the distance from the target demand sub-region to all supply sub-regions, and sort all supply sub-regions in order from near to far based on the distance to obtain a supply sub-region sequence corresponding to the target demand sub-region;

[0054] Detect the supply sub-region sequence in order from left to right, and when it is detected that the target supply power corresponding to a target number of target supply sub-regions is greater than the target demand power corresponding to the target demand sub-region, the target number of target supply sub-regions are taken as the power scheduling region of the target demand sub-region, and a supply sub-region power scheduling strategy is generated. Update the supply power corresponding to each of all supply sub-regions based on the supply sub-region power scheduling strategy until all demand sub-regions in the demand sub-region sequence are selected, and determine the scheduling strategy between the supply sub-region and the demand sub-region.

[0055] In a possible implementation, the pre-training process of the first hybrid neural network model in the photovoltaic power generation prediction module specifically includes:

[0056] An initial first hybrid neural network model is constructed, wherein the first hybrid neural network model includes a first convolutional neural network and a first graph neural network, and the first convolutional neural network is connected to the first graph neural network;

[0057] acquire a power generation data sample and a weather data sample corresponding to the same moment, input the weather data sample into the first hybrid neural network model, so that the first convolutional neural network in the first hybrid neural network model respectively extracts features of the weather data sample, and obtain a first feature map matrix;

[0058] input the first feature map matrix into the first graph neural network based on the first convolutional neural network, so that the first graph neural network performs feature fusion processing on the first feature map matrix, and obtain a first feature fusion graph;

[0059] input the first feature fusion graph into the first convolutional neural network model based on the first graph neural network, so that the first convolutional neural network extracts features of the first feature fusion graph, and obtain predicted power generation data;

[0060] compare the predicted power generation data with the power generation data sample, and determine the first hybrid neural network model.

[0061] In a possible implementation, the photovoltaic power generation amount prediction module is configured to input the weather data sample into the first hybrid neural network model, so that the first convolutional neural network in the first hybrid neural network model respectively extracts features of the weather data sample, and obtain a first feature map matrix, and specifically includes the following steps.

[0062] The weather data sample includes irradiance, cloud cover, humidity, and average solar sunshine time.

[0063] input the weather data sample into the first hybrid neural network model, so that the first convolutional neural network in the first hybrid neural network model respectively extracts features of the irradiance, the cloud cover, the humidity, and the average solar sunshine time, and respectively obtain feature weights corresponding to the irradiance, the cloud cover, the humidity, and the average solar sunshine time.

[0064] based on the feature weights, respectively construct an irradiance feature map matrix, a cloud cover feature map matrix, a humidity feature map matrix, and an average solar sunshine time feature map matrix, and integrate the irradiance feature map matrix, the cloud cover feature map matrix, the humidity feature map matrix, and the average solar sunshine time feature map matrix, to obtain a first feature map matrix.

[0065] The application further provides a terminal device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the scheduling method of the integrated generation when executing the computer program.

[0066] The application further provides a computer readable storage medium comprising a stored computer program, wherein the computer readable storage medium controls a device in which the computer readable storage medium is located to perform the integrated dispatching method of the application when the computer program is executed.

[0067] Compared with the prior art, the integrated dispatching method, device, equipment and storage medium provided by the application have the following beneficial effects:

[0068] By obtaining the historical actual weather data of each weather station in the target area and dividing the area, the photovoltaic power generation potential and electricity demand of each target sub-area can be more accurately understood, thereby realizing more reasonable resource allocation. Meanwhile, the first and second hybrid neural network models are used to predict the photovoltaic power generation and electricity demand, which can effectively improve the prediction accuracy and reduce the uncertainty caused by weather changes and energy price changes. Based on the predicted photovoltaic power generation and electricity demand, the predicted storage power of each target sub-area can be better determined, thereby optimizing the power scheduling strategy, reducing the power cost, improving the economic benefit, and realizing the optimal allocation of power resources. BRIEF DESCRIPTION OF DRAWINGS

[0069] Figure 1 is a flowchart of an embodiment of the integrated dispatching method provided by the application;

[0070] Figure 2 is a structural schematic diagram of an embodiment of the integrated dispatching device provided by the application;

[0071] Figure 3 is a structural schematic diagram of a terminal device provided by the application. DETAILED DESCRIPTION

[0072] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the application.

[0073] Embodiment 1, see Figure 1 , Figure 1 is a flowchart of an embodiment of the integrated dispatching method provided by the application, as shown in Figure 1 , the method comprises steps 101-104, specifically as follows:

[0074] Step 101: Obtain historical actual weather data corresponding to each weather station in a target region, and perform division processing on the target region based on the historical actual weather data to obtain a plurality of target sub-regions.

[0075] In an embodiment, historical actual weather data corresponding to each weather station in a target region is obtained, and a weather index corresponding to the historical actual weather data in each weather station is calculated based on a preset weather index calculation formula; the target region is divided based on the weather index using a rotated empirical orthogonal decomposition method to obtain a plurality of target sub-regions.

[0076] Specifically, the historical actual weather data corresponding to each weather station is measured weather data corresponding to a plurality of selected historical time points, and the plurality of historical time points are consecutive time points.

[0077] Specifically, the preset weather index calculation formula is as follows:

[0078]

[0079] In the formula, W is a weather index, a is a preset weather weight parameter, T n is historical actual weather data corresponding to the nth selected historical time point, and N is the total number of selected historical actual weather data.

[0080] In an embodiment, when the target region is divided based on the weather index using the rotated empirical orthogonal decomposition method, the weather index corresponding to each weather station is subjected to variance calculation processing to obtain a weather index variance matrix corresponding to each weather index, the weather index variance matrix is subjected to eigenvalue decomposition processing to obtain a plurality of eigenvalues and a feature vector corresponding to each eigenvalue, a plurality of principal components are determined according to the eigenvalues and the feature vectors corresponding to the eigenvalues, the plurality of principal components are subjected to rotation processing to obtain a feature mode corresponding to each of the plurality of principal components, and the target region is divided based on the feature mode to obtain a plurality of target sub-regions.

[0081] Specifically, when a plurality of principal components are determined based on the eigenvalues and the feature vectors, the plurality of eigenvalues are sorted according to the numerical values of the plurality of eigenvalues, the feature vectors corresponding to the first n eigenvalues with larger numerical values are selected, and a principal component corresponding to each feature vector is determined to obtain a plurality of principal components.

[0082] Step 102: Obtain historical actual weather data corresponding to each target sub-region, and perform data prediction processing on each target sub-region based on the historical actual weather data using a pre-trained first hybrid neural network model to obtain a predicted photovoltaic power corresponding to each target sub-region.

[0083] In an embodiment, the pre-training process of the first hybrid neural network model is specifically performed by constructing an initial first hybrid neural network model, wherein the first hybrid neural network model comprises a first convolutional neural network and a first graph neural network, and the first convolutional neural network is connected to the first graph neural network; obtaining power generation data samples and weather data samples corresponding to the same time point, inputting the weather data samples into the first hybrid neural network model, so that the first convolutional neural network in the first hybrid neural network model extracts features from the weather data samples respectively to obtain a first feature map matrix; inputting the first feature map matrix into the first graph neural network based on the first convolutional neural network, so that the first graph neural network performs feature fusion processing on the first feature map matrix to obtain a first feature fusion graph; inputting the first feature fusion graph into the first convolutional neural network model based on the first graph neural network, so that the first convolutional neural network extracts features from the first feature fusion graph to obtain predicted power generation data, and comparing the predicted power generation data with the power generation data samples to determine the first hybrid neural network model.

[0084] In an embodiment, the weather data samples include irradiance, cloud cover, humidity, and average solar sunshine time.

[0085] In an embodiment, when the first convolutional neural network in the first hybrid neural network model extracts features from the weather data samples to obtain a first feature map matrix, the weather data samples are input into the first hybrid neural network model, so that the first convolutional neural network in the first hybrid neural network model extracts features from the irradiance, the cloud cover, the humidity, and the average solar sunshine time respectively to obtain feature weights corresponding to the irradiance, the cloud cover, the humidity, and the average solar sunshine time respectively; based on the feature weights, irradiance feature map matrix, cloud cover feature map matrix, humidity feature map matrix, and average solar sunshine time feature map matrix are constructed respectively, and the irradiance feature map matrix, the cloud cover feature map matrix, the humidity feature map matrix, and the average solar sunshine time feature map matrix are integrated to obtain the first feature map matrix.

[0086] Preferably, the first convolutional neural network comprises a first convolutional layer, and the first convolutional layer extracts features from the irradiance, the cloud cover, the humidity, and the average solar sunshine time respectively.

[0087] Preferably, before the weather data samples are input into the first hybrid neural network model, the weather data samples are also normalized to convert the weather data samples into data in the interval [0, 1].

[0088] In an embodiment, the first graph neural network comprises a first propagation layer and a second propagation layer, wherein the first propagation layer and the second propagation layer are configured to recursively propagate node information.

[0089] In an embodiment, when performing feature fusion processing on the first feature map matrix by using the first graph neural network, the irradiance feature map matrix, the cloud cover feature map matrix, the humidity feature map matrix, and the average solar radiation time feature map matrix are respectively input into the first graph neural network, so that the first propagation layer in the first graph neural network integrates input information of a single node in each feature map matrix to obtain node input information, the second propagation layer integrates output information of the single node in each feature map matrix to obtain node output information, and the node output information and the node input information are integrated based on an attention mechanism to generate a first feature fusion graph.

[0090] In an embodiment, the first convolutional neural network further comprises a first entrance layer and a second entrance layer.

[0091] In an embodiment, for the first feature fusion graph generated by the first graph neural network, the first feature fusion graph is respectively input into the first convolutional neural network, so that the first entrance layer and the second entrance layer in the first convolutional neural network perform high-dimensional feature extraction on the first feature fusion graph to obtain high-dimensional features of the first feature fusion graph, and all high-dimensional features are connected in residual to obtain predicted power generation data corresponding to each target sub-region.

[0092] In an embodiment, after obtaining the predicted power generation data, the predicted power generation data is compared with the power generation data sample, if an error between the predicted power generation data and the power generation data sample is less than a preset threshold, it is considered that the initial first hybrid neural network model after training meets the prediction requirement, and the initial first hybrid neural network model is set as the first hybrid neural network model; if the error between the predicted power generation data and the power generation data sample is not less than the preset threshold, the first hybrid neural network model is retrained based on the power generation data sample and the weather data sample corresponding to the same time point until the error between the predicted power generation data and the power generation data sample is less than the preset threshold, and the training of the initial first hybrid neural network model is stopped.

[0093] In an embodiment, after completing the training of the first hybrid neural network model, the obtained historical actual weather data corresponding to each target sub-region is directly output to the first hybrid neural network model, so that the first hybrid neural network model performs data prediction processing on each target sub-region based on the historical actual weather data to obtain predicted photovoltaic power generation of each target sub-region.

[0094] Step 103: obtaining a historical energy price corresponding to each target sub-region, and performing data prediction processing on each target sub-region based on the historical energy price by using a pre-trained second hybrid neural network model to obtain a predicted electricity demand corresponding to each target sub-region.

[0095] In an embodiment, the pre-training process of the second hybrid neural network model is specifically performed by constructing an initial second hybrid neural network model, wherein the second hybrid neural network model comprises a second convolutional neural network and a second graph neural network, and the second convolutional neural network is connected to the second graph neural network; obtaining electricity data samples and energy price samples corresponding to the same time point, inputting the energy price samples into the second hybrid neural network model, so that the second convolutional neural network in the second hybrid neural network model extracts features from the energy price samples respectively to obtain a second feature map matrix; inputting the second feature map matrix into the second graph neural network based on the second convolutional neural network, so that the second graph neural network performs feature fusion processing on the second feature map matrix to obtain a second feature fusion graph; inputting the second feature fusion graph into the second convolutional neural network model based on the second graph neural network, so that the second convolutional neural network extracts features from the second feature fusion graph to obtain second predicted electricity data; comparing the preset electricity data with the electricity data samples to determine the second hybrid neural network model.

[0096] In an embodiment, the energy price samples include a coal-fired power generation benchmark price, a market transaction electricity price, a peak-valley electricity price, and a green electricity transaction price.

[0097] In an embodiment, when the second convolutional neural network in the second hybrid neural network model extracts features from the energy price samples to obtain a second feature map matrix, the energy price samples are input into the second hybrid neural network model, so that the second convolutional neural network in the second hybrid neural network model extracts features from the coal-fired power generation benchmark price, the market transaction electricity price, the peak-valley electricity price, and the green electricity transaction price respectively to obtain feature weights corresponding to the coal-fired power generation benchmark price, the market transaction electricity price, the peak-valley electricity price, and the green electricity transaction price respectively; based on the feature weights, a coal-fired power generation benchmark price feature map matrix, a market transaction electricity price feature map matrix, a peak-valley electricity price feature map matrix, and a green electricity transaction price feature map matrix are constructed respectively, and the coal-fired power generation benchmark price feature map matrix, the market transaction electricity price feature map matrix, the peak-valley electricity price feature map matrix, and the green electricity transaction price feature map matrix are integrated to obtain a second feature map matrix.

[0098] Preferably, before inputting the energy price sample into the second hybrid neural network model, the energy price sample is normalized to convert the energy price sample into data within the interval [0, 1].

[0099] Preferably, the second convolutional neural network includes a third convolutional layer, and the coal-fired power generation benchmark price, the market-based electricity price, the peak-valley electricity price and the green electricity transaction price are respectively extracted based on the third convolutional layer.

[0100] In an embodiment, the second graph neural network includes a third propagation layer and a fourth propagation layer, and the third propagation layer and the fourth propagation layer are used to recursively propagate node information.

[0101] In an embodiment, when the second graph neural network performs feature fusion processing on the second feature graph matrix, the coal-fired power generation benchmark price feature graph matrix, the market-based electricity price feature graph matrix, the peak-valley electricity price feature graph matrix and the green electricity transaction price feature graph matrix are input into the second graph neural network, so that the input information of a single node in each feature graph matrix is integrated based on the third propagation layer in the second graph neural network to obtain node input information, the output information of a single node in each feature graph matrix is integrated based on the fourth propagation layer to obtain node output information, and the node output information and the node input information are integrated based on an attention mechanism to generate a second feature fusion graph.

[0102] In an embodiment, the second convolutional neural network further includes a third entry layer and a fourth entry layer.

[0103] In an embodiment, for the second feature fusion graph generated by the second graph neural network, the second feature fusion graph is input into the second convolutional neural network, so that the third entry layer and the fourth entry layer in the second convolutional neural network perform high-dimensional feature extraction on the second feature fusion graph to obtain high-dimensional features of the second feature fusion graph, and all high-dimensional features are connected in residual to obtain predicted electricity consumption data corresponding to each target sub-region.

[0104] In an embodiment, after obtaining the predicted power consumption data, the predicted power consumption data is compared with the power consumption sample. If the error between the predicted power consumption data and the power consumption sample is less than a preset threshold, it is considered that the trained initial first hybrid neural network model meets the prediction requirement, and the initial second hybrid neural network model is set as the second hybrid neural network model. If the error between the predicted power consumption data and the power consumption sample is not less than the preset threshold, the second hybrid neural network model is retrained based on the power consumption sample and the energy price sample corresponding to the same time point until the error between the predicted power consumption data and the energy price sample is less than the preset threshold, and the training of the initial second hybrid neural network model is stopped.

[0105] In an embodiment, after the training of the second hybrid neural network model is completed, the obtained historical energy price corresponding to each target sub-region is directly output to the second hybrid neural network model, so that the second hybrid neural network model performs data prediction processing on each target sub-region based on the historical energy price to obtain the predicted power consumption demand corresponding to each target sub-region.

[0106] Preferably, the first hybrid neural network model and the second hybrid neural network model are the same model structure, and the selected first convolutional neural network and the second convolutional neural network are the same network, and the first graph neural network and the second graph neural network are also the same network. The difference between the first hybrid neural network model and the second hybrid neural network model is only that the data used in the model training is different.

[0107] Preferably, the first convolutional neural network and the second convolutional neural network are both CNN models, and the first graph neural network and the second graph neural network are both GNN models.

[0108] Step 104: Based on the predicted photovoltaic power generation amount and the predicted power consumption demand, the predicted storage power amount of each target sub-region is determined, and based on the predicted storage power amount, the power scheduling strategy between the plurality of target sub-regions is set.

[0109] In an embodiment, the predicted photovoltaic power generation amount and the power consumption demand corresponding to each target sub-region are obtained. When the predicted photovoltaic power generation amount is not less than the power consumption demand, the difference between the predicted photovoltaic power generation amount and the power consumption demand is calculated to obtain a first difference, and the first difference is taken as the predicted storage power amount of the target sub-region. When the predicted photovoltaic power generation amount is less than the power consumption demand, the absolute difference between the predicted photovoltaic power generation amount and the power consumption demand is calculated, and the absolute difference is taken as the predicted storage power amount corresponding to the target sub-region.

[0110] In an embodiment, the predicted storage power corresponding to each target sub-region is respectively corrected to obtain a corrected storage power, and the plurality of target sub-regions are classified into supply sub-regions and demand sub-regions based on the corrected storage power.

[0111] Specifically, a storage power deviation rate is calculated according to a plurality of historical preset storage powers and historical actual storage powers corresponding to each historical preset storage power, and a storage power deviation rate dataset is generated; an average storage power deviation rate corresponding to the storage power deviation rate dataset is calculated based on a preset confidence, and the predicted storage power corresponding to each target sub-region is respectively corrected based on the average storage power deviation rate to obtain a corrected storage power.

[0112] Specifically, when calculating the storage power deviation rate, the plurality of historical preset storage powers and the historical actual storage powers corresponding to each historical preset storage power are respectively substituted into a preset deviation rate calculation formula to obtain the storage power deviation rate corresponding to each of the plurality of historical preset storage powers, wherein the deviation rate calculation formula is as follows:

[0113] ΔQ i = Q Li - Q li / Q Li × 100%;

[0114] In the formula, ΔQ i is the storage power deviation rate corresponding to the i-th historical preset storage power, Q Li is the i-th historical actual storage power, and Q li is the i-th historical preset storage power.

[0115] Specifically, the storage power deviation rates corresponding to the plurality of historical preset storage powers are integrated to obtain a storage power deviation rate dataset.

[0116] Specifically, the average storage power deviation rate is determined by setting a preset confidence c; specifically, the preset confidence c is set and based on, wherein the preset confidence c is the value of the c-th percentile in the storage power deviation rate dataset.

[0117] Specifically, when the average storage power deviation rate corresponding to the storage power deviation rate dataset is calculated based on the preset confidence, the preset confidence and the storage power deviation rates in the storage power deviation rate dataset are respectively multiplied to obtain a plurality of storage power deviation rate multiplication values, and the plurality of storage power deviation rate multiplication values are mean value processed to obtain the average storage power deviation rate corresponding to the storage power deviation rate dataset.

[0118] Specifically, when the average storage power deviation rate is used to correct the predicted storage power of each target sub-region, the average storage power deviation rate is multiplied by the predicted storage power of each target sub-region to obtain the corrected storage power of each target sub-region.

[0119] Specifically, when the corrected storage power is used to classify the plurality of target sub-regions into supply sub-regions and demand sub-regions, it is detected whether the predicted storage power of each target sub-region is an absolute difference value. If yes, the target sub-region is classified as a demand sub-region, and the corrected storage power is used as the demand power corresponding to the demand sub-region. If not, the target sub-region is classified as a supply sub-region, and the corrected storage power is used as the supply power corresponding to the supply sub-region.

[0120] In an embodiment, when it is detected that the supply sub-region and the demand sub-region exist at the same time, and the total supply power corresponding to the supply sub-region calculated is greater than the total demand power corresponding to the demand sub-region calculated, the distance from each demand sub-region to all supply sub-regions is calculated, and based on the distance, a scheduling strategy between the supply sub-region and the demand sub-region is determined.

[0121] Specifically, the demand power corresponding to each demand sub-region is obtained, and based on the demand power, the demand sub-regions are sorted to obtain a demand sub-region sequence. A single target demand sub-region is selected from the demand sub-region sequence in turn, the distance from the target demand sub-region to all supply sub-regions is calculated, and based on the distance, the supply sub-regions are sorted in order from near to far to obtain a supply sub-region sequence corresponding to the target demand sub-region. The supply sub-region sequence is detected in order from left to right. When it is detected that the target supply power corresponding to a target number of target supply sub-regions is greater than the target demand power corresponding to the target demand sub-region, the target number of target supply sub-regions are used as the power scheduling region of the target demand sub-region, and a supply sub-region power scheduling strategy is generated. Based on the supply sub-region power scheduling strategy, the supply power corresponding to each supply sub-region is updated until all demand sub-regions in the demand sub-region sequence are selected, and a scheduling strategy between the supply sub-region and the demand sub-region is determined.

[0122] Preferably, the supply sub-region power scheduling strategy includes a target power supply type region, and the power supply from the target power supply type region to a target demand type region is aware.

[0123] In an embodiment, when it is detected that the supply sub-region exists and the demand sub-region does not exist, a charging strategy is executed on the energy storage system.

[0124] Specifically, when it is detected that the supply sub-area exists and the demand sub-area does not exist, the total supply power corresponding to all supply sub-areas is calculated, and the total supply power is used as the input power of the energy storage system to charge the energy storage system.

[0125] In one embodiment, when it is detected that the demand sub-area exists and the supply sub-area does not exist, a discharge strategy is executed on the energy storage system.

[0126] Specifically, when it is detected that the demand sub-area exists and the supply sub-area does not exist, the total power demand corresponding to all demand sub-areas is calculated, and the total power demand is used as the output power of the energy storage system to discharge the energy storage system.

[0127] In one embodiment, when it is detected that the supply sub-area and the demand sub-area exist at the same time, and the calculated total supply power corresponding to the supply sub-area is less than the calculated total demand power corresponding to the demand sub-area, the total demand power difference between the total demand power and the total supply power is calculated, and the total demand power difference is used as the output power of the energy storage system, and the energy storage system is discharged.

[0128] Example 2, see Figure 2 , Figure 2 This is a structural diagram of an embodiment of a dispatching device for integrated generation, use and storage provided by the present application. Figure 2 As shown, the device includes a region division module 201, a photovoltaic power generation prediction module 202, a power demand prediction module 203 and a power dispatch strategy generation module 204, which are specifically as follows:

[0129] The region division module 201 is configured to obtain historical actual weather data corresponding to each meteorological station in a target region, and divide the target region based on the historical actual weather data to obtain a plurality of target sub-regions.

[0130] The photovoltaic power generation prediction module 202 is used to obtain historical actual weather data corresponding to each target sub-region, and use a pre-trained first hybrid neural network model to perform data prediction processing on each target sub-region based on the historical actual weather data to obtain the predicted photovoltaic power generation corresponding to each target sub-region.

[0131] The electricity demand prediction module 203 is used to obtain the historical energy price corresponding to each target sub-area, and use the pre-trained second hybrid neural network model to perform data prediction processing on each target sub-area based on the historical energy price to obtain the predicted electricity demand corresponding to each target sub-area.

[0132] The power scheduling strategy generation module 204 is configured to determine a predicted storage power of each target sub-region based on the predicted photovoltaic power generation and the predicted power demand, and set a power scheduling strategy between the target sub-regions based on the predicted storage power.

[0133] In an embodiment, the region division module 201 is configured to obtain historical actual weather data corresponding to each weather station in a target region, and divide the target region based on the historical actual weather data to obtain a plurality of target sub-regions, specifically including: obtaining historical actual weather data corresponding to each weather station in a target region, and calculating a weather index corresponding to the historical actual weather data in each weather station based on a preset weather index calculation formula; dividing the target region based on the weather index by using a rotating empirical orthogonal decomposition method to obtain a plurality of target sub-regions.

[0134] In an embodiment, the power scheduling strategy generation module 204 is configured to set the power scheduling strategy between the target sub-regions based on the predicted storage power, specifically including: correcting the predicted storage power corresponding to each target sub-region to obtain a corrected storage power, and classifying the target sub-regions into supply sub-regions and demand sub-regions based on the corrected storage power; when it is detected that the supply sub-regions and the demand sub-regions exist at the same time, and a total supply power corresponding to the supply sub-regions calculated is greater than a total demand power corresponding to the demand sub-regions calculated, calculating a distance from each demand sub-region to all supply sub-regions, and determining a scheduling strategy between the supply sub-regions and the demand sub-regions based on the distance; when it is detected that the supply sub-regions exist and the demand sub-regions do not exist, performing a charging strategy on an energy storage system; and when it is detected that the demand sub-regions exist and the supply sub-regions do not exist, performing a discharging strategy on the energy storage system.

[0135] In an embodiment, the power scheduling strategy generation module 204 is configured to correct the predicted storage power corresponding to each target sub-region to obtain a corrected storage power, specifically including: calculating a storage power deviation rate based on a plurality of historical preset storage powers and historical actual storage powers corresponding to each historical preset storage power, and generating a storage power deviation rate data set; setting and calculating an average storage power deviation rate corresponding to the storage power deviation rate data set based on a preset confidence, and correcting the predicted storage power corresponding to each target sub-region based on the average storage power deviation rate to obtain a corrected storage power.

[0136] In an embodiment, the power scheduling strategy generation module 204 is configured to calculate the distance from each demand sub-region to all supply sub-regions, determine the scheduling strategy between the supply sub-regions and the demand sub-regions based on the distance, and specifically includes: obtaining the respective demand power of all demand sub-regions, and sorting the demand sub-regions based on the demand power to obtain a demand sub-region sequence; selecting a single target demand sub-region from the demand sub-region sequence in turn, calculating the distance from the target demand sub-region to all supply sub-regions, and sorting all supply sub-regions in order from near to far based on the distance to obtain a supply sub-region sequence corresponding to the target demand sub-region; detecting the supply sub-region sequence in order from left to right, and when the target supply power of a target number of target supply sub-regions is greater than the target demand power of the target demand sub-region, taking the target number of target supply sub-regions as the power scheduling region of the target demand sub-region, and generating a supply sub-region power scheduling strategy, updating the respective supply power of all supply sub-regions based on the supply sub-region power scheduling strategy, until all demand sub-regions in the demand sub-region sequence are selected, and the scheduling strategy between the supply sub-regions and the demand sub-regions is determined.

[0137] In an embodiment, the pre-training process of the first hybrid neural network model in the photovoltaic power generation prediction module 202 specifically includes: constructing an initial first hybrid neural network model, wherein the first hybrid neural network model includes a first convolutional neural network and a first graph neural network, and the first convolutional neural network is connected to the first graph neural network; obtaining corresponding power generation data samples and weather data samples at the same time, inputting the weather data samples into the first hybrid neural network model, so that the first convolutional neural network in the first hybrid neural network model extracts features from the weather data samples to obtain a first feature map matrix; inputting the first feature map matrix into the first graph neural network based on the first convolutional neural network, so that the first graph neural network performs feature fusion processing on the first feature map matrix to obtain a first feature fusion graph; inputting the first feature fusion graph into the first convolutional neural network model based on the first graph neural network, so that the first convolutional neural network extracts features from the first feature fusion graph to obtain predicted power generation data; comparing the predicted power generation data with the power generation data samples to determine the first hybrid neural network model.

[0138] In an embodiment, the photovoltaic power generation amount prediction module 202 is configured to input the weather data sample into the first hybrid neural network model, so that the first convolutional neural network in the first hybrid neural network model extracts features from the weather data sample respectively to obtain a first feature map matrix. Specifically, the weather data sample includes irradiance, cloud cover, humidity, and average solar sunshine time; the weather data sample is input into the first hybrid neural network model, so that the first convolutional neural network in the first hybrid neural network model extracts features from the irradiance, the cloud cover, the humidity, and the average solar sunshine time respectively to obtain feature weights corresponding to the irradiance, the cloud cover, the humidity, and the average solar sunshine time respectively; based on the feature weights, irradiance feature map matrix, cloud cover feature map matrix, humidity feature map matrix, and average solar sunshine time feature map matrix are constructed respectively, and the irradiance feature map matrix, the cloud cover feature map matrix, the humidity feature map matrix, and the average solar sunshine time feature map matrix are integrated to obtain the first feature map matrix.

[0139] The dispatching system design device for the integrated generation, transmission and storage can implement the method for the integrated generation, transmission and storage of the above-mentioned method embodiments. The optional items in the above-mentioned method embodiments are also applicable to the present embodiment, which will not be described in detail here.

[0140] Figure 3 Fig. 1 is a schematic diagram of a terminal device. As shown in Fig. 1, the terminal device 3 of the embodiment includes at least one processor 301 (only one processor is shown in the figure), a memory 302, and a computer program 303 stored in the memory 302 and executable on the at least one processor 301, and the processor 301 implements the steps in any method embodiment described above when executing the computer program 303. Figure 3 Figure 3 The terminal device 3 can be a smart phone, a notebook computer, a tablet computer, a desktop computer, and the like. The terminal device can include but is not limited to the processor 301 and the memory 302. Those skilled in the art can understand that the terminal device 3 can include more or fewer components than those shown in the figure, or combine certain components, or include different components, for example, it can also include an input / output device, a network access device, and the like.

[0141] The terminal device 3 shown in the figure is only an example of the terminal device 3 and does not constitute a limitation on the terminal device 3, and can include more or fewer components than those shown in the figure, or combine certain components, or include different components, for example, it can also include an input / output device, a network access device, and the like. Figure 3

[0142] ​​The processor 301 can be a central processing unit (CPU), and can also be other general-purpose processors, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0143] The memory 302 can be an internal storage unit of the terminal device 3, such as a hard disk or a memory of the terminal device 3 in some embodiments. The memory 302 can also be an external storage device of the terminal device 3, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal device 3 in some other embodiments. Further, the memory 302 can include both the internal storage unit and the external storage device of the terminal device 3. The memory 302 is used to store an operating system, an application program, a boot loader, data, and other programs, such as program codes of computer programs, etc. The memory 302 can also be used to temporarily store data that has been output or will be output.

[0144] In addition, the embodiments of the present application further provide a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the steps in any method embodiment described above.

[0145] In the several embodiments provided in the present application, it should be understood that each block of the flowchart or block diagram can represent a module, a segment or a portion of code, which includes one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in a different order than that noted in the figure. For example, two blocks shown in succession can in fact be executed substantially concurrently or in the reverse order, depending on the functionality involved.

[0146] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for causing an end device to execute all or part of the steps of the methods of the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0147] To sum up, the application provides a scheduling method and device for integrated power generation and storage, an apparatus, and a storage medium. Based on historical actual weather data of each weather station in a target area, the target area is divided into a plurality of target sub-areas. A first hybrid neural network model is used to perform data prediction processing on each target sub-area based on the historical actual weather data of each target sub-area, to obtain predicted photovoltaic power generation. A second hybrid neural network model is used to perform data prediction processing on each target sub-area based on historical energy prices of each target sub-area, to obtain predicted electricity demand. Based on the predicted photovoltaic power generation and the predicted electricity demand, a predicted storage power of each target sub-area is determined, and an electricity scheduling strategy between the plurality of target sub-areas is set. Compared with the prior art, the technical solution of the application can realize the optimal allocation of power resources by the photovoltaic power generation and the electricity demand in the target area.

[0148] The above only describes the preferred embodiments of the present application. It should be noted that, for those skilled in the art, without departing from the technical principles of the present application, a number of improvements and replacements can be made, which should also be considered as the protection scope of the present application.

Claims

1. A dispatching method for hair and cosmetic product integration, characterized by, The method comprises the following steps: acquiring historical actual weather data corresponding to each weather station in a target region, and performing division processing on the target region based on the historical actual weather data to obtain a plurality of target sub-regions; acquiring historical actual weather data corresponding to each target sub-region, and performing data prediction processing on each target sub-region based on the historical actual weather data by using a pre-trained first hybrid neural network model to obtain predicted photovoltaic power generation corresponding to each target sub-region; acquiring historical energy prices corresponding to each target sub-region, and performing data prediction processing on each target sub-region based on the historical energy prices by using a pre-trained second hybrid neural network model to obtain predicted electricity demand corresponding to each target sub-region; based on the predicted photovoltaic power generation and the predicted electricity demand, determining the predicted storage power of each target sub-region, and based on the predicted storage power, setting an electricity dispatching strategy between the plurality of target sub-regions; wherein the pre-training process of the first hybrid neural network model comprises: constructing an initial first hybrid neural network model, wherein the first hybrid neural network model comprises a first convolutional neural network and a first graph neural network, and the first convolutional neural network is connected to the first graph neural network; acquiring power generation data samples and weather data samples corresponding to the same time, inputting the weather data samples into the first hybrid neural network model, so that the first convolutional neural network in the first hybrid neural network model extracts features from the weather data samples to obtain a first feature map matrix; inputting the first feature map matrix into the first graph neural network based on the first convolutional neural network, so that the first graph neural network performs feature fusion processing on the first feature map matrix to obtain a first feature fusion graph; inputting the first feature fusion graph into the first convolutional neural network model based on the first graph neural network, so that the first convolutional neural network extracts features from the first feature fusion graph to obtain predicted power generation data; comparing the predicted power generation data with the power generation data samples to determine the first hybrid neural network model. The pre-training process of the second hybrid neural network model specifically comprises: constructing an initial second hybrid neural network model, wherein the second hybrid neural network model comprises a second convolutional neural network and a second graph neural network, and the second convolutional neural network is connected with the second graph neural network; obtaining power consumption data samples and energy price samples corresponding to the same moment, inputting the energy price samples into the second hybrid neural network model, so that the second convolutional neural network in the second hybrid neural network model extracts features from the energy price samples respectively to obtain a second feature map matrix; inputting the second feature map matrix into the second graph neural network based on the second convolutional neural network, so that the second graph neural network performs feature fusion processing on the second feature map matrix to obtain a second feature fusion graph; inputting the second feature fusion graph into the second convolutional neural network model based on the second graph neural network, so that the second convolutional neural network extracts features from the second feature fusion graph to obtain second predicted power consumption data; comparing the predicted power consumption data with the power consumption data samples to determine the second hybrid neural network model.

2. The dispatch method of a shampoo and conditioner integration according to claim 1, wherein, The historical actual weather data corresponding to each weather station in the target region is obtained, and the target region is divided based on the historical actual weather data to obtain a plurality of target sub-regions, specifically comprising: The historical actual weather data corresponding to each weather station in the target region is obtained, and the weather index corresponding to the historical actual weather data in each weather station is calculated based on a preset weather index calculation formula; The target region is divided based on the weather index by using a rotating empirical orthogonal decomposition method to obtain a plurality of target sub-regions.

3. The dispatch method of claim 1, wherein, The power dispatch strategy between the plurality of target sub-regions is set based on the predicted storage power, specifically comprising: The predicted storage power corresponding to each target sub-region is corrected to obtain a corrected storage power, and the plurality of target sub-regions are classified into supply sub-regions and demand sub-regions based on the corrected storage power; When it is detected that the supply sub-regions and the demand sub-regions exist at the same time, and the total supply power corresponding to the supply sub-regions calculated is greater than the total demand power corresponding to the demand sub-regions calculated, the distance from each demand sub-region to all supply sub-regions is calculated, and the dispatch strategy between the supply sub-regions and the demand sub-regions is determined based on the distance; When it is detected that the supply sub-regions exist and the demand sub-regions do not exist, a charging strategy is performed on the energy storage system; When it is detected that the demand sub-regions exist and the supply sub-regions do not exist, a discharging strategy is performed on the energy storage system.

4. The dispatch method of a shampoo and conditioner integration according to claim 3, wherein, The predicted storage power corresponding to each target sub-region is corrected to obtain a corrected storage power, specifically comprising: A storage power deviation rate is calculated according to a plurality of historical preset storage powers and historical actual storage powers corresponding to each historical preset storage power, and a storage power deviation rate data set is generated; Set and calculate the average storage power deviation rate corresponding to the storage power deviation rate data set based on the preset confidence, and correct the predicted storage power corresponding to each target sub-region based on the average storage power deviation rate, to obtain corrected storage power.

5. The dispatch method of claim 3, wherein the method further comprises: The distance from each demand sub-region to all supply sub-regions is calculated, and based on the distance, a scheduling strategy between the supply sub-region and the demand sub-region is determined, specifically including: Obtain the demand power corresponding to each demand sub-region, and sort all demand sub-regions based on the demand power to obtain a demand sub-region sequence; Select a single target demand sub-region from the demand sub-region sequence in turn, calculate the distance from the target demand sub-region to all supply sub-regions, and sort all supply sub-regions in order from near to far based on the distance to obtain a supply sub-region sequence corresponding to the target demand sub-region; In order from left to right, when the target supply power corresponding to the target number of target supply sub-regions is greater than the target demand power corresponding to the target demand sub-region, the target number of target supply sub-regions are taken as the power scheduling region of the target demand sub-region, and a supply sub-region power scheduling strategy is generated. Update the supply power corresponding to each of the all supply sub-regions based on the supply sub-region power scheduling strategy, until all demand sub-regions in the demand sub-region sequence are selected, and the scheduling strategy between the supply sub-region and the demand sub-region is determined.

6. The dispatch method of a shampoo and conditioner integration according to claim 1, wherein, The weather data sample is input into the first hybrid neural network model, so that the first convolutional neural network in the first hybrid neural network model extracts features from the weather data sample respectively to obtain a first feature map matrix, specifically including: The weather data sample includes irradiance, cloud cover, humidity and average solar sunshine time; The weather data sample is input into the first hybrid neural network model, so that the first convolutional neural network in the first hybrid neural network model extracts features from the irradiance, cloud cover, humidity and average solar sunshine time respectively to obtain feature weights corresponding to the irradiance, cloud cover, humidity and average solar sunshine time respectively. Based on the feature weights, irradiance feature map matrix, cloud cover feature map matrix, humidity feature map matrix and average solar sunshine time feature map matrix are constructed respectively, and the irradiance feature map matrix, cloud cover feature map matrix, humidity feature map matrix and average solar sunshine time feature map matrix are integrated to obtain a first feature map matrix.

7. A dispatching device for hair and cosmetic storage, characterized in that It includes: A region division module, a photovoltaic power generation prediction module, a power demand prediction module and a power scheduling strategy generation module; The region division module is configured to obtain historical actual weather data corresponding to each meteorological station in a target region, and divide the target region based on the historical actual weather data to obtain a plurality of target sub-regions; The photovoltaic power generation amount prediction module is configured to obtain historical actual weather data corresponding to each target sub-region, and perform data prediction processing on each target sub-region based on the historical actual weather data by using a pre-trained first hybrid neural network model to obtain predicted photovoltaic power generation amount corresponding to each target sub-region. The electricity demand amount prediction module is configured to obtain historical energy prices corresponding to each target sub-region, and perform data prediction processing on each target sub-region based on the historical energy prices by using a pre-trained second hybrid neural network model to obtain predicted electricity demand amount corresponding to each target sub-region. The electricity scheduling strategy generation module is configured to determine predicted storage electricity amount of each target sub-region based on the predicted photovoltaic power generation amount and the predicted electricity demand amount, and set an electricity scheduling strategy between the target sub-regions based on the predicted storage electricity amount. In the pre-training process of the first hybrid neural network model in the photovoltaic power generation amount prediction module, the initial first hybrid neural network model is constructed, wherein the first hybrid neural network model includes a first convolutional neural network and a first graph neural network, and the first convolutional neural network is connected to the first graph neural network. The power generation data sample and the weather data sample corresponding to the same time are obtained, and the weather data sample is input into the first hybrid neural network model, so that the first convolutional neural network in the first hybrid neural network model extracts features from the weather data sample to obtain a first feature map matrix. The first feature map matrix is input into the first graph neural network based on the first convolutional neural network, so that the first graph neural network performs feature fusion processing on the first feature map matrix to obtain a first feature fusion graph. The first feature fusion graph is input into the first convolutional neural network based on the first graph neural network, so that the first convolutional neural network extracts features from the first feature fusion graph to obtain predicted power generation data. The predicted power generation data and the power generation data sample are compared to determine the first hybrid neural network model. The pre-training process of the second hybrid neural network model in the electricity demand prediction module specifically comprises: constructing an initial second hybrid neural network model, wherein the second hybrid neural network model comprises a second convolutional neural network and a second graph neural network, and the second convolutional neural network is connected with the second graph neural network; obtaining electricity data samples and energy price samples corresponding to the same moment, inputting the energy price samples into the second hybrid neural network model, so that the second convolutional neural network in the second hybrid neural network model extracts features from the energy price samples respectively to obtain a second feature map matrix; inputting the second feature map matrix into the second graph neural network based on the second convolutional neural network, so that the second graph neural network performs feature fusion processing on the second feature map matrix to obtain a second feature fusion graph; inputting the second feature fusion graph into the second convolutional neural network model based on the second graph neural network, so that the second convolutional neural network extracts features from the second feature fusion graph to obtain second predicted electricity data; comparing the predicted electricity data with the electricity data samples to determine the second hybrid neural network model.

8. A terminal device, comprising: The computer readable storage medium comprises a computer program stored therein, wherein the computer program, when executed, controls a device in which the computer readable storage medium is located to perform the dispatching method integrated with power generation, power utilization and power storage as claimed in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium comprises a computer program stored therein, wherein the computer program, when executed, controls a device in which the computer readable storage medium is located to perform the dispatching method integrated with power generation, power utilization and power storage as claimed in any one of claims 1 to 6.

Images