Fluctuation data determination method and device of virtual power plant, equipment and storage medium

Abstract

The invention relates to a fluctuation data determination method and device of a virtual power plant, equipment and a storage medium. The method comprises the following steps: acquiring historical operation data and an initial topological graph of a virtual power plant; distributing weights for nodes in the initial topological graph and communication links among the nodes to obtain a target topological graph; and inputting the historical operation data and the target topological graph into a graph neural network model to perform operation simulation of the virtual power plant to obtain fluctuation data of the virtual power plant. According to the scheme, before the target topological graph is input into the graph neural network model, the weights are allocated to the nodes in the initial topological graph and the communication links between the nodes, so that the graph neural network model can conveniently capture important feature data of the target topological graph, the model focusing capability is improved, and the user experience is improved. And thus, the accuracy of fluctuation simulation of the virtual power plant is improved.

Description

Technical Field

[0001] The present application relates to the technical field of virtual power plants, and in particular to a method, device, equipment and storage medium for determining fluctuation data of a virtual power plant. Background Art

[0002] A virtual power plant (VPP) is a system that aggregates distributed energy resources (such as solar energy, wind energy, energy storage systems, electric vehicle charging facilities, and controllable loads) through advanced information technology and software systems to form a unified, dispatchable "virtual" power plant. In order to flexibly respond to load changes and distributed energy fluctuations in virtual power plants, it is necessary to simulate the volatility of virtual power plants.

[0003] In traditional technology, the volatility of virtual power plants is usually simulated through models. However, in traditional technology, when simulating the volatility of virtual power plants, due to the large amount of input data, the model's focusing ability is poor, thereby reducing the accuracy of the volatility simulation of virtual power plants. Summary of the invention

[0004] Based on this, it is necessary to provide a method, device, equipment and storage medium for determining the fluctuation data of a virtual power plant in response to the above-mentioned technical problems, which can improve the focusing ability of the model and thus improve the accuracy of simulating the volatility of the virtual power plant.

[0005] In a first aspect, the present application provides a method for determining fluctuation data of a virtual power plant, comprising:

[0006] Acquire historical operation data and an initial topology map of a virtual power plant; wherein the historical operation data at least includes load data, energy storage data and power data of distributed energy of the virtual power plant; the initial topology map includes nodes and communication links between nodes, wherein the nodes include distributed energy, power consumption equipment and contact points;

[0007] Assigning weights to nodes in the initial topology map and communication links between nodes to obtain a target topology map;

[0008] The historical operation data and the target topology map are input into a graph neural network model to perform operation simulation of the virtual power plant and obtain fluctuation data of the virtual power plant.

[0009] In one embodiment, assigning weights to nodes in the initial topology map and communication links between nodes to obtain a target topology map includes:

[0010] Assigning weights to nodes in the initial topology graph according to the degree of influence of each node on the virtual power plant;

[0011] For each communication link, a weight is assigned to the communication link according to the attribute information of the nodes at both ends of the communication link to obtain a target topology map; wherein the attribute information of the node includes at least power data, voltage data and location information.

[0012] In one embodiment, the method further comprises:

[0013] For each node, determining the importance score of each dimensional feature data corresponding to the node for the stability of the virtual power plant; wherein the each dimensional feature data includes at least two of the output feature data, reliability, response speed, topological relationship and load data of the node;

[0014] For each dimensional feature data, weighted processing is performed on the importance score of the dimensional feature data for the stability of the virtual power plant to obtain a weighted score of the dimensional feature data for the stability of the virtual power plant;

[0015] The sum of the weighted scores of the characteristic data of each dimension for the stability of the virtual power plant is taken as the influence of the node on the virtual power plant.

[0016] In one embodiment, allocating a weight to the communication link according to the attribute information of the nodes at both ends of the communication link includes:

[0017] Determine a target attention score of the first node to the second node according to the attribute information of the first node, the attribute information of each adjacent node adjacent to the first node, and the attribute information of the second node; wherein the nodes at both ends of the communication link include the first node and the second node;

[0018] The target attention score of the first node to the second node is used as the weight of the communication link.

[0019] In one embodiment, determining the target attention score of the first node to the second node according to the attribute information of the first node, the attribute information of each adjacent node adjacent to the first node, and the attribute information of the second node includes:

[0020] Based on the attention function, determining an initial attention score of the first node to the second node according to the attribute information of the first node and the attribute information of the second node;

[0021] For each neighboring node adjacent to the first node, based on the attention function, according to the attribute information of the first node and the attribute information of the neighboring node, determine the attention score of the first node to the neighboring node;

[0022] The ratio of the initial attention score of the first node to the second node to the sum of the attention scores of the first node to each adjacent node is used as the target attention score of the first node to the second node.

[0023] In one embodiment, the graph neural network model includes a basic model and a graph convolutional neural network model; the inputting of the historical operation data and the target topology map into the graph neural network model to perform operation simulation of the virtual power plant to obtain the fluctuation data of the virtual power plant includes:

[0024] Inputting the historical operation data and the target topology map into the basic model to extract basic features, thereby obtaining basic operation features corresponding to the historical operation data and a basic feature map corresponding to the target topology map;

[0025] The basic operating characteristics and basic characteristic graph are input into the graph convolutional neural network model to perform operation simulation of the virtual power plant and obtain fluctuation data of the virtual power plant.

[0026] In a second aspect, the present application further provides a device for determining fluctuation data of a virtual power plant, comprising:

[0027] An acquisition module, used to acquire historical operation data and an initial topology map of a virtual power plant; wherein the historical operation data at least includes load data, energy storage data and power data of distributed energy of the virtual power plant; the initial topology map includes nodes and communication links between nodes, wherein the nodes include distributed energy, power consumption equipment and contact points;

[0028] An allocation module is used to allocate weights to nodes in the initial topology map and communication links between nodes to obtain a target topology map;

[0029] A simulation module is used to input the historical operation data and the target topology map into a graph neural network model to perform operation simulation of the virtual power plant and obtain fluctuation data of the virtual power plant.

[0030] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0031] Acquire historical operation data and an initial topology map of a virtual power plant; wherein the historical operation data at least includes load data, energy storage data and power data of distributed energy of the virtual power plant; the initial topology map includes nodes and communication links between nodes, wherein the nodes include distributed energy, power consumption equipment and contact points;

[0032] Assigning weights to nodes in the initial topology map and communication links between nodes to obtain a target topology map;

[0033] The historical operation data and the target topology map are input into a graph neural network model to perform operation simulation of the virtual power plant and obtain fluctuation data of the virtual power plant.

[0034] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0035] Acquire historical operation data and an initial topology map of a virtual power plant; wherein the historical operation data at least includes load data, energy storage data and power data of distributed energy of the virtual power plant; the initial topology map includes nodes and communication links between nodes, wherein the nodes include distributed energy, power consumption equipment and contact points;

[0036] Assigning weights to nodes in the initial topology map and communication links between nodes to obtain a target topology map;

[0037] The historical operation data and the target topology map are input into a graph neural network model to perform operation simulation of the virtual power plant and obtain fluctuation data of the virtual power plant.

[0038] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0039] Acquire historical operation data and an initial topology map of a virtual power plant; wherein the historical operation data at least includes load data, energy storage data and power data of distributed energy of the virtual power plant; the initial topology map includes nodes and communication links between nodes, wherein the nodes include distributed energy, power consumption equipment and contact points;

[0040] Assigning weights to nodes in the initial topology map and communication links between nodes to obtain a target topology map;

[0041] The historical operation data and the target topology map are input into a graph neural network model to perform operation simulation of the virtual power plant and obtain fluctuation data of the virtual power plant.

[0042] The above-mentioned method, device, equipment and storage medium for determining the fluctuation data of the virtual power plant obtain the historical operation data and initial topology map of the virtual power plant; wherein the historical operation data at least includes the load data, energy storage data and power data of the distributed energy of the virtual power plant; the initial topology map includes nodes and communication links between nodes, and the nodes include distributed energy, power equipment and contact points; weights are assigned to the nodes in the initial topology map and the communication links between the nodes to obtain the target topology map; the historical operation data and the target topology map are input into the graph neural network model to simulate the operation of the virtual power plant and obtain the fluctuation data of the virtual power plant. The above-mentioned scheme, before inputting the target topology map into the graph neural network model, assigns weights to the nodes in the initial topology map and the communication links between the nodes, which can facilitate the graph neural network model to capture the important feature data of the target topology map, thereby improving the model's focusing ability, and further improving the accuracy of the volatility simulation of the virtual power plant. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 A schematic flow chart of a method for determining fluctuation data of a virtual power plant in one embodiment;

[0045] Figure 2 A schematic diagram of a process flow of a target topology diagram in an embodiment;

[0046] Figure 3 A schematic diagram of a process for determining the degree of influence of a node on a virtual power plant in one embodiment;

[0047] Figure 4 A schematic diagram of a process for determining a communication link weight in one embodiment;

[0048] Figure 5 A schematic diagram of a process for determining a target attention score of a first node to a second node in one embodiment;

[0049] Figure 6 A schematic diagram of a process for obtaining fluctuation data of a virtual power plant in one embodiment;

[0050] Figure 7 It is a flow chart of a method for determining fluctuation data of a virtual power plant in another embodiment;

[0051] Figure 8It is a structural block diagram of a fluctuation data determination device of a virtual power plant in one embodiment;

[0052] Fig. 9 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0054] The method for determining the fluctuation data of a virtual power plant provided in the embodiment of the present application can be applied to an application scenario of simulating the fluctuation of a virtual power plant. The method can be executed by a server or by a terminal.

[0055] The server may be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. The terminal may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, IoT devices, and portable wearable devices. IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car devices, and projection devices. Portable wearable devices may be smart watches, smart bracelets, head-mounted devices, and the like. Head-mounted devices may be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, and the like.

[0056] In an exemplary embodiment, Figure 1 As shown, a method for determining fluctuation data of a virtual power plant is provided, and the method is applied to a server as an example for explanation, and includes the following steps:

[0057] S101, obtaining historical operation data and an initial topology diagram of the virtual power plant.

[0058] Among them, the historical operation data of the virtual power plant includes at least the load data, energy storage data and power data of distributed energy of the virtual power plant.

[0059] The load data includes the electricity consumption data of industrial load, commercial load, residential load, etc. under the virtual power plant, for example, the electricity consumption curve of industrial load, the air-conditioning load of commercial buildings, the charging load of electric vehicles, etc.

[0060] Energy storage data includes data such as the charging and discharging power, energy storage capacity, and charging and discharging status of energy storage equipment under the virtual power plant. Energy storage equipment is used to balance power fluctuations in virtual power plants, and its historical operating data is very important for volatility simulation.

[0061] The power data of distributed energy includes the power output data of distributed energy such as photovoltaic power generation, wind power generation, and small hydropower stations. These data are usually recorded in time series, reflecting the distributed energy power generation under the virtual power plant in different time periods.

[0062] The initial topology includes nodes and communication links between nodes. Nodes include distributed energy, power equipment and contact points. The initial topology is a virtual power plant topology constructed based on the geographical location information, equipment installation location and electrical connection relationship of the virtual power plant. Distributed energy includes power generation devices and energy storage devices, such as photovoltaic power generation, wind power generation, small hydropower stations, etc. The contact point is the interface point for the virtual power plant to connect and interact with the external power grid or other power systems, which can also be called a "grid connection point" or "access point".

[0063] Exemplarily, meteorological data corresponding to the historical operating data of the virtual power plant can also be obtained. The meteorological data include wind speed, light intensity, temperature, humidity, etc. These data have a direct impact on the power output of renewable energy and are important inputs for virtual power plant operation and volatility simulation.

[0064] Exemplarily, the historical operation data and meteorological data of the virtual power plant may be preprocessed, wherein the preprocessing includes data cleaning, data normalization, etc. Data cleaning includes outlier removal, missing value filling, etc. Outlier removal can be performed by using Z-score standardization to identify and remove outliers.

[0065] Since the amount of historical operation data of the virtual power plant is large and includes data collected by different collection nodes, missing values ​​are inevitable, and the data cleaning process will also cause data missing. Therefore, interpolation methods or the average value based on adjacent data points can be used to fill in the gaps; the interpolation method is one of linear interpolation and time series interpolation. Data normalization can use the maximum and minimum normalization method to scale the historical operation data to the (0, 1) interval.

[0066] S102, assigning weights to nodes in the initial topology map and communication links between nodes to obtain a target topology map.

[0067] Furthermore, weights can be assigned to the nodes in the initial topology map and the communication links between the nodes. For example, weights can be assigned to the nodes according to the degree of influence of the nodes on the volatility of the virtual power plant; and the degree of influence of the communication links connecting the nodes on the node operating data or the stable operation of the virtual power plant can be determined based on the direct topological relationship between the nodes, and then weights can be assigned to the communication links according to the degree of influence of the communication links on the node operating data or the stable operation of the virtual power plant to obtain the target topology map.

[0068] The target topology graph includes the nodes in the initial topology, the communication links between the nodes, and the weights of the nodes and the weights of the communication links between the nodes.

[0069] S103, inputting the historical operation data and the target topology map into the graph neural network model to simulate the operation of the virtual power plant and obtain the fluctuation data of the virtual power plant.

[0070] Furthermore, the historical operation data and the target topology map can be input into the graph neural network model to simulate the operation of the virtual power plant and obtain the fluctuation data of the virtual power plant. The graph neural network model is pre-trained based on the sample operation data and sample topology map, as well as the actual fluctuation data of the virtual power plant.

[0071] For example, by simulating the volatility of virtual power plants, flexible scheduling solutions can be provided to ensure stable operation of the power grid when facing fluctuations in power demand or renewable energy generation. For example, when wind power, photovoltaic and other renewable energy sources are unstable due to weather and other factors, the virtual power plant can adjust the charging and discharging strategies of energy storage equipment or the output power of other distributed power sources in a timely manner based on the simulation results, balance the load of the power grid, reduce the phenomenon of "abandoning wind and light", and improve the utilization rate of renewable energy.

[0072] The above-mentioned method for determining the fluctuation data of the virtual power plant obtains the historical operation data and the initial topology map of the virtual power plant; wherein the historical operation data at least includes the load data, energy storage data and power data of the distributed energy of the virtual power plant; the initial topology map includes nodes and communication links between nodes, and the nodes include distributed energy, power equipment and contact points; weights are assigned to the nodes in the initial topology map and the communication links between the nodes to obtain the target topology map; the historical operation data and the target topology map are input into the graph neural network model to simulate the operation of the virtual power plant to obtain the fluctuation data of the virtual power plant. The above-mentioned scheme, before inputting the target topology map into the graph neural network model, assigns weights to the nodes in the initial topology map and the communication links between the nodes, which can facilitate the graph neural network model to capture the important characteristic data of the target topology map, thereby improving the model's focusing ability, and further improving the accuracy of the volatility simulation of the virtual power plant.

[0073] In some optional implementations, see Figure 2 , Figure 2 A schematic diagram of a target topology diagram is provided, which specifically includes the following steps:

[0074] S201, assigning weights to nodes in the initial topology map according to the degree of influence of each node on the virtual power plant.

[0075] Exemplarily, the degree of influence of each node on the virtual power plant can be determined based on empirical data, or by performing a correlation analysis on the node's operating data and the stability of the virtual power plant to determine the degree of influence of each node on the virtual power plant.

[0076] Furthermore, weights can be assigned to nodes in the initial topology map according to the degree of influence of each node on the virtual power plant. For example, if the node has a greater influence on the virtual power plant, a greater weight can be assigned to the node, and if the node has a smaller influence on the virtual power plant, a smaller weight can be assigned to the node. For example, the distributed energy, power equipment and contact points in the initial topology map can be assigned weights of 60%, 30% and 10% respectively.

[0077] S202: For each communication link, a weight is assigned to the communication link according to the attribute information of the nodes at both ends of the communication link.

[0078] Among them, the attribute information of the node includes at least power data, voltage data and location information. Exemplarily, for each communication link, the importance of the communication link to the stability of the virtual power plant can be determined based on the power data, voltage data and location information of the nodes at both ends of the communication link; for example, based on the power data, voltage data and location information of the nodes at both ends of the communication link, the output of the node, the voltage stability and topological relationship can be determined, and then the importance of the communication link to the stability of the virtual power plant can be determined.

[0079] Furthermore, a weight may be assigned to the communication link according to the importance of the communication link to the stability of the virtual power plant. For example, if the communication link is more important to the virtual power plant, a larger weight may be assigned to the communication link, and if the communication link is less important to the virtual power plant, a smaller weight may be assigned to the communication link.

[0080] In an embodiment of the present application, weights are assigned to the nodes in the initial topology map according to the degree of influence of each node on the virtual power plant, and weights are assigned to the communication links according to the attribute information of the nodes at both ends of the communication links. This allows the weights to be assigned to the nodes in the initial topology map and the communication links between the nodes, which can more accurately reflect the influence of the nodes in the initial topology map and the communication links between the nodes on the stability of the virtual power plant, thereby facilitating a more accurate simulation of the volatility of the virtual power plant.

[0081] In some optional implementations, the degree of influence of a node on the corresponding virtual power plant may be determined based on the multi-dimensional characteristic data of the node.

[0082] Based on this, see Figure 3 , Figure 3A flow chart of determining the influence of a node on a virtual power plant is provided, which specifically includes the following steps:

[0083] S301, for each node, determining the importance score of the characteristic data of each dimension corresponding to the node for the stability of the virtual power plant.

[0084] The characteristic data of each dimension includes at least two of the output characteristic data, reliability, response speed, topological relationship and load data of the node. The characteristic data of each dimension includes the output characteristic data, reliability, response speed, topological relationship and load data as an example for explanation.

[0085] For example, for each node, the importance score of the output characteristic data, reliability, response speed, topological relationship and load data corresponding to the node for the stability of the virtual power plant can be determined respectively. For example, the importance score corresponding to the characteristic data of each dimension is A1, A2, A3, A4 and A5.

[0086] S302: For each dimension feature data, weighted processing is performed on the importance score of the dimension feature data for the stability of the virtual power plant to obtain a weighted score of the dimension feature data for the stability of the virtual power plant.

[0087] Furthermore, for each dimensional feature data, the importance score of the dimensional feature data for the stability of the virtual power plant can be weighted to obtain the weight score of the dimensional feature data for the stability of the virtual power plant. For example, the weights corresponding to the output feature data, reliability, response speed, topological relationship and load data are B1, B2, B3, B4 and B5, and the importance score of the dimensional feature data for the stability of the virtual power plant is weighted to obtain the weight score of the dimensional feature data for the stability of the virtual power plant, namely A1×B1, A2×B2, A3×B3, A4×B4 and A5×B5.

[0088] S303, taking the sum of the weighted scores of the characteristic data of each dimension for the stability of the virtual power plant as the influence of the node on the virtual power plant.

[0089] Furthermore, the sum of the weighted scores of the characteristic data of each dimension on the stability of the virtual power plant is taken as the influence of the node on the virtual power plant, that is, the value of A1×B1+A2×B2+A3×B3+A4×B4+A5×B5 is taken as the influence of the node on the virtual power plant.

[0090] In an embodiment of the present application, the influence degree of a node on a virtual power plant is determined based on the multi-dimensional characteristic data of the node, so that the influence degree of the determined node on the virtual power plant is more reliable.

[0091] In some optional implementations, the weight of the communication link may be determined based on attribute information of nodes at both ends of the communication link.

[0092] Based on this, see Figure 4 , Figure 4 A flow chart of determining a communication link weight is provided, which specifically includes the following steps:

[0093] S401, determining a target attention score of the first node to the second node based on the attribute information of the first node, the attribute information of each adjacent node to the first node, and the attribute information of the second node.

[0094] The nodes at both ends of the communication link include a first node and a second node, and the first node and the second node are the nodes at both ends of the communication link.

[0095] Exemplarily, the attribute information of the first node may be processed by encoding or other means to form a feature vector of the first node; and the attribute information of each adjacent node adjacent to the first node may be processed by encoding or other means to form a feature vector of each adjacent node; and the attribute information of the second node may be processed by encoding or other means to form a feature vector of the second node.

[0096] Furthermore, the target attention score of the first node to the second node can be determined based on the attention score prediction model or the attention score calculation function. For example, the feature vector of the first node, the feature vectors of each adjacent node, and the feature vector of the second node can be input into the attention score prediction model to obtain the target attention score of the first node to the second node.

[0097] Alternatively, the feature vector of the first node, the feature vectors of each adjacent node, and the feature vector of the second node may be substituted into an attention score calculation function to calculate a target attention score of the first node to the second node.

[0098] S402: Use the target attention score of the first node to the second node as the weight of the communication link.

[0099] Furthermore, the target attention score of the first node to the second node can reflect the importance of the first node to the second node. The target attention score of the first node to the second node can be used as the weight of the communication link, that is, the weight of the communication link can be represented by the importance between the nodes at both ends of the communication link.

[0100] In the embodiment of the present application, the weight of the communication link is determined by the importance between the nodes at both ends of the communication link, so that the determined weight of the communication link is more reasonable, and thus the determined fluctuation data of the simulated power plant is more accurate.

[0101] In some optional implementations, see Figure 5 , Figure 5 A flow chart of determining a target attention score of a first node to a second node is provided, which specifically includes the following steps:

[0102] S501, based on the attention function, according to the attribute information of the first node and the attribute information of the second node, determine the initial attention score of the first node to the second node.

[0103] For example, the initial attention score of the first node to the second node is determined according to the attention function. The attribute information of the first node can be encoded and processed to form the feature vector h of the first node. i ; and the attribute information of the second node is processed by encoding to form the feature vector h of the second node j .

[0104] The feature vector h of the first node i and the eigenvector h of the second node j Substitute it into the attention function to get the initial attention score of the first node to the second node:

[0105]

[0106] Among them, e ij is the initial attention score of the first node to the second node, a is the attention function, and W is the weight matrix used to perform linear transformation on the feature vector.

[0107] S502: For each neighboring node adjacent to the first node, based on the attention function, according to the attribute information of the first node and the attribute information of the neighboring node, determine the attention score of the first node to the neighboring node.

[0108] Furthermore, the attribute information of each neighboring node adjacent to the first node can also be processed by encoding to form the feature vector of each neighboring node. i and the feature vectors h of the adjacent nodes k Substitute into the attention function to get the initial attention score e of the first node to the adjacent node k ik :

[0109]

[0110] S503: The ratio of the initial attention score of the first node to the second node to the sum of the attention scores of the first node to each adjacent node is used as the target attention score of the first node to the second node.

[0111] Furthermore, the sum of the attention scores of the first node to each adjacent node can be calculated first, and then the ratio of the initial attention score of the first node to the second node to the sum of the attention scores of the first node to each adjacent node can be used as the target attention score of the first node to the second node. The specific calculation formula is as follows:

[0112]

[0113] in, is the target attention score of the first node to the second node, N(i) is the set of neighboring nodes of node i, is the sum of the attention scores of the first node to each adjacent node.

[0114] In an embodiment of the present application, an initial attention score of a first node to a second node is determined based on an attention function, and a target attention score of the first node to the second node is determined according to the ratio of the initial attention score of the first node to the second node to the sum of the attention scores of the first node to each adjacent node. On the one hand, the target attention score of the first node to the second node can be determined through a simple calculation process, and on the other hand, the target attention score of the first node to the second node can be expressed more standardizedly.

[0115] In some optional implementations, the graph neural network model may include a basic model and a graph convolutional neural network model, thereby using the basic model and the graph convolutional neural network model to process historical operating data and target topology maps to obtain fluctuation data of the virtual power plant.

[0116] Based on this, see Figure 6 , Figure 6 A schematic diagram of a process for obtaining fluctuation data of a virtual power plant is provided, which specifically includes the following steps:

[0117] S601, inputting historical operation data and a target topology map into a basic model respectively to extract basic features, thereby obtaining basic operation features corresponding to the historical operation data and a basic feature map corresponding to the target topology map.

[0118] Exemplarily, the historical operation data and the target topology map may be respectively input into the basic model to extract basic features, so as to obtain basic operation features corresponding to the historical operation data and a basic feature map corresponding to the target topology map.

[0119] Among them, the basic model can be a long short-term memory network model, and the basic model can include an input layer, a hidden layer and an output layer, wherein the input layer is used to input historical operation data and a target topology map; the hidden layer is used to extract features from the data in the input layer, and the output layer is used to output basic operation features corresponding to the historical operation data and a basic feature map corresponding to the target topology map.

[0120] S602, input the basic operation characteristics and the basic characteristic graph into the graph convolutional neural network model to simulate the operation of the virtual power plant and obtain the fluctuation data of the virtual power plant.

[0121] Furthermore, the basic operation characteristics and basic feature graphs can be input into the graph convolutional neural network model to simulate the operation of the virtual power plant and obtain the fluctuation data of the virtual power plant. The graph convolutional neural network model can include a convolutional layer, a linear layer, and a nonlinear activation layer. The convolutional layer updates the feature representation of the current node by aggregating the information of neighboring nodes based on the features extracted by the basic model. Its core operation is feature propagation based on the normalized adjacency matrix. The linear layer further processes the features output by the convolutional layer through linear transformation after the convolutional layer operation. The nonlinear layer is a nonlinear activation function used to output the fluctuation data of the virtual power plant. The nonlinear activation function can be a ReLU function.

[0122] In an embodiment of the present application, by using a basic model and a graph convolutional neural network model to simulate the operation of a virtual power plant, more dimensional feature data can be extracted, and the accuracy of the simulated virtual power plant fluctuation data can be improved.

[0123] In some optional implementations, see Figure 7 , Figure 7 A flow chart of another method for determining fluctuation data of a virtual power plant is provided, which specifically includes the following steps:

[0124] S701, obtaining historical operation data and an initial topology diagram of the virtual power plant.

[0125] S702: assign weights to nodes in the initial topology map according to the degree of influence of each node on the virtual power plant.

[0126] S703, for each communication link, assign a weight to the communication link according to the attribute information of the nodes at both ends of the communication link, and obtain a target topology map.

[0127] S704, inputting the historical operation data and the target topology map into the basic model respectively to extract basic features, and obtaining basic operation features corresponding to the historical operation data and a basic feature map corresponding to the target topology map.

[0128] S705, input the basic operation characteristics and the basic characteristic graph into the graph convolutional neural network model to simulate the operation of the virtual power plant and obtain the fluctuation data of the virtual power plant.

[0129] The specific process of S701 to S705 can refer to the description of the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0130] Furthermore, the execution order of the above steps is only an exemplary description and is not used to limit the execution steps. Other execution orders of the steps are within the protection scope of the embodiments of the present application.

[0131] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0132] Based on the same inventive concept, the embodiment of the present application also provides a device for determining fluctuation data of a virtual power plant for implementing the method for determining fluctuation data of a virtual power plant involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of one or more devices for determining fluctuation data of virtual power plants provided below can refer to the limitations of the method for determining fluctuation data of virtual power plants above, and will not be repeated here.

[0133] In an exemplary embodiment, Figure 8 As shown, a device for determining fluctuation data of a virtual power plant is provided, comprising:

[0134] The acquisition module 10 is used to acquire the historical operation data and the initial topology of the virtual power plant; wherein the historical operation data at least includes the load data, energy storage data and power data of the distributed energy of the virtual power plant; the initial topology includes nodes and communication links between nodes, and the nodes include distributed energy, power equipment and contact points;

[0135] An allocation module 20 is used to allocate weights to nodes in the initial topology map and communication links between nodes to obtain a target topology map;

[0136] The simulation module 30 is used to input the historical operation data and the target topology map into the graph neural network model to simulate the operation of the virtual power plant and obtain the fluctuation data of the virtual power plant.

[0137] The above-mentioned device for determining the fluctuation data of the virtual power plant obtains the historical operation data and the initial topology map of the virtual power plant; wherein the historical operation data at least includes the load data, energy storage data and power data of the distributed energy of the virtual power plant; the initial topology map includes nodes and communication links between nodes, and the nodes include distributed energy, power equipment and contact points; weights are assigned to the nodes in the initial topology map and the communication links between the nodes to obtain the target topology map; the historical operation data and the target topology map are input into the graph neural network model to simulate the operation of the virtual power plant and obtain the fluctuation data of the virtual power plant. The above-mentioned scheme, before inputting the target topology map into the graph neural network model, assigns weights to the nodes in the initial topology map and the communication links between the nodes, which can facilitate the graph neural network model to capture the important characteristic data of the target topology map, thereby improving the model's focusing ability, and further improving the accuracy of the volatility simulation of the virtual power plant.

[0138] In one embodiment, the allocation module 20 specifically includes:

[0139] A first allocation unit is used to allocate weights to nodes in the initial topology map according to the influence of each node on the virtual power plant;

[0140] The second allocation unit is used to allocate weights to each communication link according to the attribute information of the nodes at both ends of the communication link to obtain a target topology map; wherein the attribute information of the node includes at least power data, voltage data and location information.

[0141] In one of the embodiments, the apparatus further comprises a determining module, configured to:

[0142] For each node, determine the importance score of each dimensional feature data corresponding to the node for the stability of the virtual power plant; wherein each dimensional feature data includes at least two of the node's output feature data, reliability, response speed, topological relationship and load data; for each dimensional feature data, perform weighted processing on the importance score of the dimensional feature data for the stability of the virtual power plant to obtain a weighted score of the dimensional feature data for the stability of the virtual power plant; and take the sum of the weighted scores of each dimensional feature data for the stability of the virtual power plant as the degree of influence of the node on the virtual power plant.

[0143] In one embodiment, the second allocation unit is specifically used for:

[0144] Determine a target attention score of the first node to the second node based on attribute information of the first node, attribute information of each adjacent node to the first node, and attribute information of the second node; wherein the nodes at both ends of the communication link include the first node and the second node; and use the target attention score of the first node to the second node as the weight of the communication link.

[0145] In one embodiment, the second allocation unit is specifically used for:

[0146] Based on the attention function, an initial attention score of the first node to the second node is determined according to the attribute information of the first node and the attribute information of the second node; for each adjacent node adjacent to the first node, based on the attention function, an attention score of the first node to the adjacent node is determined according to the attribute information of the first node and the attribute information of the adjacent node; the ratio of the initial attention score of the first node to the second node to the sum of the attention scores of the first node to each adjacent node is used as the target attention score of the first node to the second node.

[0147] In one embodiment, the graph neural network model includes a basic model and a graph convolutional neural network model; the simulation module 30 is specifically used for:

[0148] The historical operation data and the target topology map are respectively input into the basic model for basic feature extraction to obtain the basic operation characteristics corresponding to the historical operation data and the basic feature map corresponding to the target topology map; the basic operation characteristics and the basic feature map are input into the graph convolutional neural network model to simulate the operation of the virtual power plant and obtain the fluctuation data of the virtual power plant.

[0149] Each module in the above-mentioned fluctuation data determination device of the virtual power plant can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0150] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Fig. 9As shown. The computer device includes a processor, a memory, an input / output (I / O) interface and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the operating data and structural data of the virtual power plant. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for determining fluctuation data of a virtual power plant is implemented.

[0151] Those skilled in the art will understand that Fig. 9 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0152] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of the method for determining fluctuation data of a virtual power plant described in any of the above embodiments are implemented.

[0153] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for determining fluctuation data of a virtual power plant described in any of the above embodiments are implemented.

[0154] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the steps of the method for determining fluctuation data of a virtual power plant described in any of the above embodiments.

[0155] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0156] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0157] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0158] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for determining fluctuation data of a virtual power plant, characterized in that: The method comprises: Acquire historical operation data and an initial topology map of a virtual power plant; wherein the historical operation data at least includes load data, energy storage data and power data of distributed energy of the virtual power plant; the initial topology map includes nodes and communication links between nodes, wherein the nodes include distributed energy, power consumption equipment and contact points; Assigning weights to nodes in the initial topology map and communication links between nodes to obtain a target topology map; The historical operation data and the target topology map are input into a graph neural network model to perform operation simulation of the virtual power plant and obtain fluctuation data of the virtual power plant.

2. The method according to claim 1, characterized in that The step of assigning weights to the nodes in the initial topology map and the communication links between the nodes to obtain a target topology map includes: Assigning weights to nodes in the initial topology graph according to the degree of influence of each node on the virtual power plant; For each communication link, a weight is assigned to the communication link according to the attribute information of the nodes at both ends of the communication link to obtain a target topology map; wherein the attribute information of the node includes at least power data, voltage data and location information.

3. The method according to claim 2, characterized in that The method further comprises: For each node, determining the importance score of each dimensional feature data corresponding to the node for the stability of the virtual power plant; wherein the each dimensional feature data includes at least two of the output feature data, reliability, response speed, topological relationship and load data of the node; For each dimensional feature data, weighted processing is performed on the importance score of the dimensional feature data for the stability of the virtual power plant to obtain a weighted score of the dimensional feature data for the stability of the virtual power plant; The sum of the weighted scores of the characteristic data of each dimension for the stability of the virtual power plant is taken as the influence of the node on the virtual power plant.

4. The method according to claim 2, characterized in that: The allocating a weight to the communication link according to the attribute information of the nodes at both ends of the communication link comprises: Determine a target attention score of the first node to the second node according to the attribute information of the first node, the attribute information of each adjacent node adjacent to the first node, and the attribute information of the second node; wherein the nodes at both ends of the communication link include the first node and the second node; The target attention score of the first node to the second node is used as the weight of the communication link.

5. The method according to claim 4, characterized in that The determining, according to the attribute information of the first node, the attribute information of each adjacent node adjacent to the first node, and the attribute information of the second node, a target attention score of the first node to the second node includes: Based on the attention function, determining an initial attention score of the first node to the second node according to the attribute information of the first node and the attribute information of the second node; For each neighboring node adjacent to the first node, based on the attention function, according to the attribute information of the first node and the attribute information of the neighboring node, determine the attention score of the first node to the neighboring node; The ratio of the initial attention score of the first node to the second node to the sum of the attention scores of the first node to each adjacent node is used as the target attention score of the first node to the second node.

6. The method according to claim 1, characterized in that The graph neural network model includes a basic model and a graph convolutional neural network model; the historical operation data and the target topology map are input into the graph neural network model to perform operation simulation of the virtual power plant to obtain the fluctuation data of the virtual power plant, including: Inputting the historical operation data and the target topology map into the basic model to extract basic features, thereby obtaining basic operation features corresponding to the historical operation data and a basic feature map corresponding to the target topology map; The basic operating characteristics and basic characteristic graph are input into the graph convolutional neural network model to perform operation simulation of the virtual power plant and obtain fluctuation data of the virtual power plant.

7. A device for determining fluctuation data of a virtual power plant, characterized in that: The device comprises: An acquisition module, used to acquire historical operation data and an initial topology map of a virtual power plant; wherein the historical operation data at least includes load data, energy storage data and power data of distributed energy of the virtual power plant; the initial topology map includes nodes and communication links between nodes, wherein the nodes include distributed energy, power consumption equipment and contact points; An allocation module is used to allocate weights to nodes in the initial topology map and communication links between nodes to obtain a target topology map; A simulation module is used to input the historical operation data and the target topology map into a graph neural network model to perform operation simulation of the virtual power plant and obtain fluctuation data of the virtual power plant.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images