Comprehensive energy intelligent management and control platform data analysis method and device, equipment and medium

Abstract

The invention provides a comprehensive energy intelligent management and control platform data analysis method and device, equipment and a medium, and belongs to the technical field of data processing, and the method comprises the steps: constructing a graph data model corresponding to each moment according to the operation data of power grid equipment at N moments; according to the graph data model corresponding to each moment, determining fusion space-time sequence features; dynamically predicting the operation data of the power grid equipment after the N moments according to the fused space-time sequence features to obtain a predicted value of the power grid equipment; and determining a power equipment control strategy according to the predicted value of the operation data. According to the comprehensive energy intelligent management and control platform data analysis method and device, the equipment and the medium provided by the invention, accurate prediction and intelligent decision making are realized, and the accuracy of real-time prediction can be improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of data processing, and more particularly relates to a comprehensive energy intelligent management and control platform data analysis method and device, equipment and medium. BACKGROUND

[0002] In the power system, multi-source data such as power grid operation data, user power consumption records and power generation equipment data lack deep mining, and hidden correlations between data cannot be accurately found. The power load prediction is greatly affected by external factors, and the existing prediction method cannot realize real-time dynamic prediction based on complex scenarios. Therefore, an innovative dynamic prediction method is needed to improve the efficiency of the comprehensive energy intelligent management and control platform. SUMMARY

[0003] The application aims to provide a comprehensive energy intelligent management and control platform data analysis method and device, equipment and medium to realize real-time dynamic prediction of the operation data of power equipment.

[0004] The first aspect of the embodiment of the application provides a comprehensive energy intelligent management and control platform data analysis method, comprising: constructing a graph data model corresponding to each time according to the operation data of the power grid equipment at N times; the nodes in the graph data model represent the power grid equipment, the edges in the graph data model represent the connection relationship between different nodes, the direction of the edges in the graph data model represents the current transmission direction between different nodes, and the nodes in the graph data model contain node attributes and attribute values, which are one-to-one corresponding to the node attributes; determining a fused spatio-temporal sequence feature according to the graph data model corresponding to each time; dynamically predicting the operation data of the power grid equipment after the N times according to the fused spatio-temporal sequence feature to obtain a predicted value of the power grid equipment; determining the power equipment control strategy according to the predicted value of the power grid equipment.

[0005] The second aspect of the embodiment of the application provides a comprehensive energy intelligent management and control platform data analysis device, comprising: a construction module configured to construct a graph data model corresponding to each time according to the operation data of the power grid equipment at N times; the nodes in the graph data model represent the power grid equipment, the edges in the graph data model represent the connection relationship between different nodes, the direction of the edges in the graph data model represents the current transmission direction between different nodes, and the nodes in the graph data model contain node attributes and attribute values, which are one-to-one corresponding to the node attributes; a determination module configured to determine a fused spatio-temporal sequence feature according to the graph data model corresponding to each time; a prediction module configured to dynamically predict operation data of the power grid device after the N time points according to the fused spatio-temporal sequence features, to obtain predicted values of the power grid device; a control module configured to determine the power device control strategy according to the predicted values of the power grid device.

[0006] In a third aspect, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of the comprehensive energy smart management and control platform data analysis method described above are implemented.

[0007] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the steps of the comprehensive energy smart management and control platform data analysis method described above are implemented.

[0008] The comprehensive energy smart management and control platform data analysis method and device, equipment, and medium provided by the embodiments of the present application have the following beneficial effects. By structurally modeling the complex power grid devices and current flow directions in the power grid, a graph data model at each time point is constructed, which can realize the conversion from scattered data to structured data, facilitating the quick positioning of key nodes. By using the graph data models corresponding to different time points, the fused spatio-temporal sequence features are determined, which can integrate the spatial topology information and time dimension information of power grid operation, capture the dynamic evolution of device states over time and the mutual influence between nodes in the power grid operation process. By combining the fused spatio-temporal sequence features, the future operation state of the power grid device can be accurately predicted, and the prediction accuracy is improved. By determining the targeted power grid device control strategy based on the accurate predicted values, the regulation and control of the power grid can be realized. BRIEF DESCRIPTION OF DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0010] Figure 1 A flowchart of the comprehensive energy smart management and control platform data analysis method provided by an embodiment of the present application is shown in the figure. Figure 2 A structural block diagram of the comprehensive energy smart management and control platform data analysis device provided by an embodiment of the present application is shown in the figure. Figure 3 A schematic block diagram of the electronic device provided by an embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0011] In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular architectures, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known methods, devices, circuits, and

[0012] In order to make the purposes, technical solutions and advantages of the present application clearer, the following will be described by specific embodiments in conjunction with the accompanying drawings.

[0013] Reference will be made to Figure 1 , Figure 1 The flowchart of the data analysis method of the comprehensive energy smart management and control platform provided by an embodiment of the present application, the method comprises: S101, constructing a graph data model corresponding to each time according to the operation data of the power grid equipment at N time points.

[0014] In the present embodiment, the operation data of the power grid system in each power supply area can be monitored in real time, the power grid equipment can include each device in the power generation equipment, power transmission equipment and power transformation equipment in each power supply area, and the operation data includes the operation data of the power generation equipment, the operation data of the power transmission equipment and the operation data of the power transformation equipment in each power supply area.

[0015] Among them, the power generation data can include the output power of the power generation equipment such as wind power generation equipment, solar photovoltaic panel power generation equipment and thermal power generation equipment in each first time period, the total amount of power generation in the first time period, the device operation parameters such as the working temperature, voltage, current of the power generation equipment, the maintenance record of the power generation equipment, the historical fault information of the power generation equipment, etc.

[0016] The power transmission equipment can include line loss information, power information and frequency information of the power transmission line in the running process, the line loss information represents the power loss generated by the line resistance in the power transmission process. The power information includes active power and reactive power, the active power is used to represent the actual consumed power, and the reactive power represents the exchange power between the power supply and the load. The frequency information is used to represent the frequency of the power grid; the device operation state includes transformer oil temperature, winding temperature, gas content in oil, on-off state of circuit breaker, action times, energy storage state, and other parameters such as vibration, noise, insulation resistance of the device; The power transformation equipment can include transformer, distribution box and other equipment, the operation data of the transformer can include the data corresponding to the operation parameters such as capacity, transformation ratio and temperature, and the operation parameters of the distribution box can include the data corresponding to branch current, switch state, load condition, etc.

[0017] In this embodiment, for each power supply area, the operation data of the power grid devices in the power supply area at N time points can be obtained, and a graph data model corresponding to each time point can be constructed according to the operation data. N is a positive integer, which can be determined according to actual conditions. The N time points also include the operation data of the current time point and the previous N-1 time points. Each power grid device can be abstracted as a node in the graph data model. The attribute value of the node attribute of each node corresponding to the power grid device at a certain time point can be determined according to the location information of the power grid device, the physical identifier of the power grid device, and the operation data of the power grid device at the time point. The graph data model can be determined according to the nodes, node attributes, and direct connection relationship between nodes. The direct connection relationship is used to represent the direct data transmission relationship between two nodes, such as direct connection between two nodes in the physical world, direct communication channel between two nodes, etc. The nodes with direct connection relationship can be connected to obtain the edges in the graph data model. The nodes in the graph data model represent power grid devices, the edges in the graph data model represent the direct connection relationship between different nodes, the direction of the edges in the graph data model represents the current transmission direction between different nodes, and the information of the nodes in the graph data model includes node attributes and attribute values, which are one-to-one corresponding.

[0018] In this embodiment, by mapping the power grid from the physical world to the digital space and structurally expressing the relationship between the power grid devices, the overall situation of the power grid operation can be mastered through the graph data model, and the accuracy of the power grid data processing can be improved.

[0019] S102, determine the spatio-temporal sequence features according to the graph data model corresponding to each time point.

[0020] In this embodiment, the collection time points can be determined according to a pre-set time interval, such as 15 minutes, 30 minutes, or 1 hour. The operation data of the power system can be periodically collected. A corresponding graph data model can be generated for each time point. Each graph data model can record the operation data of all power grid devices at the corresponding time point, such as the power generation power, voltage, current, and device temperature of the power generation device, the transformer oil temperature and tap position of the power transformation device, and the current, voltage, and power loss of the power transmission line at the time point. In addition, the graph data model also retains the connection relationship between nodes and the power transmission path information. The graph data model corresponding to each time point in the N time points can be obtained.

[0021] In this embodiment, the spatial features corresponding to each node can be determined based on the spatial position and connection relationship of each node in each graph data model.

[0022] In this embodiment, the time sequence features of each node dynamically changing over time can be determined based on the graph data model corresponding to each time point.

[0023] In the embodiment, the time sequence features of each time and the spatial features of each time can be fused and processed to construct the spatio-temporal sequence features of the graph data model corresponding to each time. By obtaining the spatial features and the time sequence features of the nodes in the power grid and obtaining the spatio-temporal sequence according to the spatial features and the time sequence features, the spatio-temporal correlation of the nodes can be captured, and the accuracy of data processing can be improved.

[0024] S103, dynamically predicting the operation data of the power grid equipment after the N time according to the fused spatio-temporal sequence features to obtain the predicted value of the power grid equipment.

[0025] In the embodiment, a deep learning prediction model can be trained based on a spatio-temporal graph convolution network, and the fused spatio-temporal sequence features are taken as the model input. Each input sample contains attribute values of node attributes in N time sequences and corresponding external factors. The node features cover the operation data of the power grid equipment, such as power generation power, voltage, current, temperature and other information. The input data can be normalized to map the data to a suitable interval, such as [0, 1] or [-1, 1], to speed up the model training speed and improve the training stability.

[0026] In the embodiment, the deep learning prediction model after training can be used to quickly output the prediction result according to the latest fused spatio-temporal sequence features in real time through forward propagation, so as to realize real-time dynamic prediction of the operation data of the power grid equipment, and predict the operation data of the power equipment in the power system at one or more future times, such as voltage and current fluctuation of the power grid equipment, and probability of equipment failure.

[0027] S104, determining the power equipment control strategy according to the predicted value of the power grid equipment.

[0028] In the embodiment, the user's power consumption data can be obtained, which can include the power consumption of various types of users at each time in the N time, such as the power consumption of residents, commercial users and industrial users.

[0029] In the embodiment, the future power consumption load can be predicted according to the user's power consumption data. As an example, the time sequence features of each time corresponding to the power consumption load can be obtained based on the user's power consumption data at each time, the meteorological data corresponding to the time and the calendar features of the time, such as peak and valley period, working day or holiday, etc. The power load prediction model corresponding to different user types can be trained by combining machine learning or time sequence method, and the power consumption load corresponding to each type of user in the future can be predicted according to the power load prediction model.

[0030] In the embodiment, the power equipment control strategy can be determined according to the predicted value of the power consumption load and the power grid equipment, and the control strategy can include at least one of the start-stop control strategy of the power generation equipment in the power grid equipment, the switching control strategy of the power transmission equipment, the tap adjustment control strategy of the transformer equipment, and the user power limiting control strategy. For example, if the attribute value of the predicted failure probability attribute of a certain power grid equipment is greater than the preset probability threshold of the power grid equipment, it is determined that the power grid equipment has a failure, and the power grid equipment is shut down, the associated power transmission equipment is switched, and load transfer is performed; if the predicted value of the power consumption load exceeds the power generation threshold of the corresponding power generation equipment, the standby power generation equipment can be started; if the predicted load rate of a certain line is greater than the first load rate threshold and the predicted load rate of the adjacent line is less than the second load rate threshold according to the power consumption load, the power transmission path of the line can be switched; if the voltage offset of the low-voltage side of a certain transformer equipment exceeds the preset voltage threshold, the tap position can be automatically adjusted; if the predicted value of the power load exceeds the maximum power generation of the corresponding power generation equipment, the power consumption limiting control can be performed.

[0031] In an embodiment of the present application, the graph data model corresponding to each time point is constructed according to the operation data of the power grid equipment at N time points, which can include: For each node in the graph data model, the node attribute of the node and the first attribute range of each node attribute are obtained; For each node attribute, if the attribute value corresponding to the node attribute is within the first attribute range corresponding to the node attribute, the failure probability attribute of the node is determined as the first value; if the attribute value corresponding to the node attribute of the node is not within the first attribute range corresponding to the node attribute, the second value of the failure probability attribute corresponding to the node is determined according to the failure probability model; According to the failure probability attribute, the first value and the second value, the nodes in the graph data model are updated; The first attribute range includes the attribute value corresponding to the node attribute of each node in the normal operation state, and the first attribute range corresponds to the node attribute one by one.

[0032] In the embodiment, for each graph data model corresponding to each time point, an attribute field of a fault probability value can be added to each power grid device in the graph data model. Information of the power grid device can be obtained from the graph data model. For each node in the graph model, the node attribute of the node and the first attribute range of each node attribute can be obtained. Each power grid device corresponds to a node in the graph data model. For example, a power transformation device can include a high-voltage transformer, a low-voltage transformer, and the like. The high-voltage transformer corresponds to a node in the graph model, and the low-voltage transformer corresponds to a node in the graph model. Each node can include at least one node attribute. For example, the node corresponding to the high-voltage transformer can include a voltage, a capacity, an oil temperature, an insulation resistance, and the like.

[0033] In the embodiment, for each node attribute of each node, it can be determined whether the attribute value corresponding to the node attribute of the node is within the first attribute range corresponding to the node attribute. If the attribute value corresponding to the node attribute is within the first attribute range corresponding to the node, the fault probability attribute of the node can be determined as a first value. If the attribute value corresponding to the node attribute is not within the first attribute range corresponding to the node attribute, a second value of the fault probability attribute corresponding to the node can be determined according to a fault probability model. The first value can be set to a very small value, such as 0, indicating that the fault probability value of the device corresponding to the node is very small. The fault probability model is obtained by training a neural network in combination with actual data.

[0034] In the embodiment, after the fault probability value is calculated by using the fault probability model, the newly added fault probability attribute, the first value and the second value of the fault probability attribute are updated to each node in the graph data model, and the new attribute value is added to the graph data model. As an example, assume that there is a graph data model corresponding to all power grid devices in a power supply area, which includes a node A representing a transformer. First, the type of the device to which the node A belongs can be obtained, such as a high-voltage transformer in a power transformation device. Then, the first attribute range corresponding to the "high-voltage transformer" can be obtained from the pre-configured device type attribute range configuration, which includes a temperature range and a voltage level range. Then, the actual attribute value of the node A is checked. If it is found that the temperature exceeds the upper limit of the temperature range, 65°C, the second value of the fault probability attribute corresponding to the node can be further determined based on the fault probability model, and then the fault probability value is added to the second node attribute of the node A in the graph data model. Subsequently, the fault probability information of the node can be directly obtained when the node is analyzed, which helps to more accurately predict the transformer and take corresponding control strategies in advance. The fault detection model can be constructed based on historical fault data of each device type and operating characteristics of all devices in the device type.

[0035] In an embodiment of the present application, the spatio-temporal sequence feature is determined according to the graph data model corresponding to each time point, including: determining a first path in the graph data model corresponding to each time point; determining a change amount of the attribute value of each node at each time point according to the attribute value of the node attribute of each node in the first path corresponding to each two adjacent time points, and determining the time sequence feature of each node according to the change amount; determining the spatial feature corresponding to each node according to the spatial position of each node in the graph data model corresponding to each time point; fusing the time sequence feature and the spatial feature corresponding to each node to obtain the fused spatio-temporal sequence feature; wherein the first path represents each connected graph in the graph data model.

[0036] In the embodiment, all connected graphs in the graph data model can be identified through a depth-first search or a breadth-first search algorithm, and each connected graph can be determined as the first path.

[0037] In the embodiment, the attribute value of the node attribute of the node in the first path corresponding to each two adjacent time points can be determined, the change amount of the attribute value of each node at each two adjacent time points can be determined, and the time sequence feature of each node can be determined according to the change amount of the attribute value of each node at each two time points. For example, if a same node is included in multiple first paths, the average value of the change amount corresponding to the node in all paths at each two adjacent time points can be determined as the change amount of the attribute value corresponding to the node; each change amount in the sequence corresponding to each node can be normalized, and the sequence obtained after the normalization is determined as the time sequence feature corresponding to each node.

[0038] In the embodiment, the spatial position of each node in the graph data model corresponding to each time point can be obtained, which can include the longitude and latitude position coordinates of the power grid device and the topological position in the graph data model, i.e., the spatial position can include the physical geographical position of the power grid device and the abstract position in the graph data model. As an example, for one of the power grid devices, which is No. 1 device, the longitude and latitude position coordinates of the power grid device can be obtained through GPS positioning or a GIS system, such as the longitude and latitude position coordinates of No. 1 device (116.42°E, 39.90°N), and the topological position of the power grid device in the graph data model can be obtained, for example: in the graph data model, the starting node in the first path where No. 1 device is located can be reached through at least 2 hops to No. 1 device, and the topological position of No. 1 device is 2.

[0039] In the embodiment, the spatial features of each node at each time point can be determined according to the spatial positions of each node in each graph data model corresponding to each time point. As an example, the spatial feature of any power grid device at any time point can be represented as [longitude, latitude, physical distance set, topology distance aggregation, azimuth angle set]. In the vector corresponding to the spatial feature, the longitude represents the longitude of the power grid device, the latitude represents the latitude of the power grid device, the physical distance set represents the set of physical geographical distances from the power grid device to other power grid devices, the topology distance set represents the set of topology distances from the power grid device to other power grid devices, and the azimuth angle set represents the set of azimuth angles from the power grid device to other power grid devices.

[0040] In the embodiment, the spatial features of each node at all time points can be determined based on the spatial features of each node at each time point. As an example, each feature in the spatial feature can be calculated by weighting based on the weight of each time point to obtain a weighted spatial feature. The weights of each time point can be equal or unequal. The weight can be determined according to the time interval between each time point and the current time point. The greater the time interval, the smaller the weight.

[0041] In the embodiment, the time sequence feature corresponding to each node can be fused with the spatial feature to form a fused time-space sequence feature. The time sequence feature and the spatial feature can be spliced to obtain the fused time-space sequence feature. Alternatively, a first feature in the time sequence feature and a second feature in the spatial feature can be obtained, the first feature and the second feature have relevance, the first feature is a feature in the time sequence feature, the second feature is a feature in the spatial feature, and the first feature and the second feature have relevance. For example, the first feature is a current change rate, and the second feature is a distance. The greater the distance, the greater the line impedance, and the weaker the adjustment effect of the power supply side on the current fluctuation, and the smaller the change rate. Therefore, the current change rate can have a negative correlation with the geographical distance. The first feature and the second feature can be multiplied to obtain a new feature. The time sequence feature and the spatial feature can be spliced, and then the new feature can be added to the spliced feature to obtain the fused time sequence feature.

[0042] In an embodiment of the present application, the operation data of the power grid device after N time points is dynamically predicted according to the fused time-space sequence feature to obtain a predicted value of the power grid device, including: determining a topology connection coefficient between each two power grid devices according to the switch state of each power grid device at the current time point; performing attention calculation processing on the topology connection coefficient and the fused time-space sequence feature according to the first neural network trained to obtain a weighted fused time-space vector; and determining the predicted value of the power grid device after the N time points according to the weighted fused time-space vector.

[0043] In the embodiment, the state value of the switch state of each device in the power grid device at the current time can be obtained. As an example, if the line L1 and the line L2 are in a normal connection state, the state value of the switch state is 1, and if the line L1 and the line L2 are in a disconnected state, the state value of the switch state is 0.

[0044] In the embodiment, the node corresponding to each edge can be determined according to the graph data model at the current time, and the node corresponding to each edge is constructed into an adjacency matrix according to the switch state of each device. Among them, the element in the adjacency matrix represents that the power grid device i and the power grid device j have a direct connection relationship, and if = 1, it represents that the power grid device i and the power grid device j are in a normal connection state, and if = 0, it represents that the power grid device i and the power grid device j are in a disconnected state.

[0045] In the embodiment, the topological connection coefficient between each two power grid devices can be determined according to the switch state of each power grid device at the current time.

[0046] In the embodiment, the first neural network trained can include a topological encoding layer, a space-time feature fusion layer and a full connection layer. Among them, the topological encoding layer can perform a deformable graph convolution operation on the topological connection coefficient matrix to generate a corresponding topological correlation feature vector; the space-time fusion layer can determine the attention weight using the topological correlation feature vector, and perform weighted calculation on the space-time sequence feature based on the attention weight; the full connection layer can determine the predicted value of the future power grid device after the N time from the fusion space-time vector obtained after the weighted calculation.

[0047] In an embodiment of the present application, the topological connection coefficient between each two power grid devices can be determined according to the switch state of each power grid device at the current time, which can be determined by using a target formula; the target formula includes:

[0048] Among them, represents the topological connection coefficient between the power grid device i and the power grid device j, represents the direct connection state between the power grid device i and the intermediate node k, , represents a damping coefficient, controls the propagation strength, and can be 0.85, m represents the total number of power grid devices, represents a random jump probability, which can represent a random jump term, which can prevent calculation faults caused by isolated nodes.

[0049] In an embodiment of the present application, the power equipment control strategy comprises: obtaining the use data of the user at each of the N time points; determining the power equipment control strategy according to the predicted value of the power grid equipment and the use data of the user; the power equipment control strategy can comprise at least one of the start-stop control strategy of the power generation equipment, the switching control strategy of the power transmission equipment, the tap adjusting control strategy of the power transformation equipment, and the user power cut control strategy.

[0050] In the embodiment, the use data of the user at each of the N time points in each power supply area can be obtained based on the smart meter acquisition system, and the types of the user can comprise residents, commercial users and industrial users; the use data can comprise data such as power consumption.

[0051] In the embodiment, the power consumption load characteristics of different time scales can be extracted from the user use data at each time point. The power consumption load characteristics of different time scales can comprise peak and valley periods of daily load curve, weekday / weekend characteristics of weekly load curve, and load abnormal characteristics under special events such as holidays and extreme weather, and the power consumption load at each future time point can be determined according to the power consumption load characteristics.

[0052] In the embodiment, the predicted value of the power grid equipment in each power supply area can be associated with the power consumption load in the power supply area, the analysis result can be determined, and at least one of the start-stop control strategy of the power generation equipment, the switching control strategy of the power transmission equipment, the tap adjusting control strategy of the power transformation equipment, and the user power cut control strategy in the power supply area can be determined according to the analysis result.

[0053] In an embodiment of the present application, the power equipment control strategy is determined according to the predicted value of the power grid equipment and the use data of the user, comprising: performing time series decomposition processing on the use data of the user to obtain the load demand of the user at each time period; determining the matching result between the power generation equipment and the load demand according to the predicted value of the power grid equipment and the load demand of the user at each time scale; determining the power equipment control strategy according to the matching result.

[0054] In the embodiment, the power consumption load in each power supply area can be decomposed into characteristics of different time scales by using the empirical mode decomposition algorithm, and the power consumption demand of the user corresponding to different time scales can be determined according to the characteristics of different time scales; the predicted value of the power equipment corresponding to different scales can be determined according to the predicted value of the power equipment in each power supply area.

[0055] In the embodiment, for the same power supply area, the power equipment control strategy can be determined according to the power consumption demand corresponding to different time scales and the predicted value of the power equipment corresponding to the different time scales.

[0056] In the embodiment, for each time scale, the power consumption demand and the predicted value of the power equipment in the time scale can be matched to obtain a matching result. The matching of the power consumption demand and the predicted value of the power equipment in the time scale can include matching the predicted power generation of the power generation equipment in the power equipment and the power consumption demand in the time scale, or matching the predicted load rate of the power transmission line in the power equipment and the line load rate in the time scale to obtain the matching result.

[0057] In the embodiment, if the power consumption demand in the time scale is less than or equal to the power generation of the power generation equipment in the power equipment, it can be determined that the matching result is that the power generation of the power equipment can meet the power consumption demand, and it can be further determined that the difference between the power consumption demand in the time scale and the power generation of the power generation equipment in the power equipment. If the difference is greater than a preset first threshold, it is determined whether the power generation equipment in the power equipment needs to be shut down. As an example, the power generation cost of each power generation equipment can be determined and sorted according to the power generation cost, and the power generation equipment with a higher power generation cost is preferentially shut down. Before shutting down, it needs to be verified whether the remaining power generation equipment can meet the power consumption demand in the time scale, i.e., whether the power generation corresponding to the remaining power generation equipment is greater than or equal to the power consumption demand in the time scale. If not, the number of shut-down power generation equipment is adjusted until the requirement is met.

[0058] In the embodiment, if the power consumption demand in the time scale is greater than the power generation of the power generation equipment in the power equipment, it is determined whether there is a standby power generation equipment that is not started in the current power generation equipment. If there is, the state of the standby power generation equipment is further determined, and the standby power generation equipment is sorted in order from low to high according to the power generation cost of the standby power generation equipment. The standby power generation equipment with a lower standby power generation cost is preferentially selected, and the selected power generation equipment can meet the power consumption demand in the time scale, i.e., the power generation of the selected power generation equipment is greater than or equal to the power consumption demand in the time scale. If not, the number of standby power generation equipment is adjusted until the requirement is met, and the standby power generation equipment is started.

[0059] If the standby power generation equipment does not exist or even all the standby power generation equipment cannot meet the power demand in the time scale, the user who needs to be power limited can be selected according to the power type in the time scale, and the user is subjected to power limiting processing. For example, in the industrial user, the non-critical load other than the continuous production process can be subjected to limiting processing; in the commercial user, the unnecessary lighting and air conditioning are turned off; in the residential user, the limiting can be based on the priority of the user, and the power supply reliability of important loads such as hospitals and transportation hubs is higher than a preset second threshold value, and the single power limiting time is not longer than a preset time threshold value.

[0060] In the embodiment, if the load rate of the first line in the line of the time scale is greater than a preset first load threshold value, and the load rate of the adjacent line of the first line is less than a preset second load threshold value, a switching path can be determined based on the graph data model to reduce the load rate of the first line to a preset load range; wherein the first load threshold value is the maximum load threshold value of the power transmission line, the second load threshold value is the minimum load threshold value of the power transmission line, and the preset load range is between the minimum load threshold value and the maximum load threshold value.

[0061] In the embodiment, the voltage prediction value of the voltage transformation device can also be determined, and the adjustment strategy of the tapping of the voltage transformation device can be determined according to the deviation of the voltage prediction value from the rated value. If the deviation between the voltage prediction value and the rated value is not within a preset voltage threshold value, the tapping of the voltage transformation device can be adjusted based on the deviation.

[0062] The comprehensive energy smart management and control platform data analysis method corresponding to the above embodiment, Figure 2 The structural block diagram of the comprehensive energy smart management and control platform data analysis device provided by an embodiment of the present application is shown. For the convenience of description, only the parts related to the embodiments of the present application are shown. For reference Figure 2 The comprehensive energy smart management and control platform data analysis device 20 includes a construction module 21, a determination module 22, a prediction module 23, and a control module 24.

[0063] The construction module 21 is configured to construct a graph data model corresponding to each time according to the operation data of the power grid equipment at N time points. The nodes in the graph data model represent the power grid equipment, the edges in the data model represent the connection relationship between different nodes, the direction of the edges in the data model represents the current transmission direction between different nodes, and the nodes in the graph data model include node attributes and attribute values, which are one-to-one corresponding. The determination module 22 is configured to determine the fused time-space sequence features according to the graph data model corresponding to each time. The prediction module 23 is configured to dynamically predict the operation data of the power grid equipment at future N time points according to the fused time-space sequence features to obtain the prediction value of the power grid equipment. The control module 24 is configured to determine the power equipment control strategy according to the predicted value of the power grid equipment.

[0064] The construction module 21 is further configured to: for each node in the graph data model, acquire a node attribute of the node and a first attribute range of each node attribute; for each node attribute, if an attribute value corresponding to the node attribute is within a first attribute range corresponding to the node attribute, determine a failure probability attribute of the node as a first value; if the attribute value corresponding to the node attribute of the node is not within the first attribute range corresponding to the node attribute, determine a second value of the failure probability attribute corresponding to the node according to a failure probability model; update each node in the graph data model according to the failure probability attribute, the first value and the second value; and wherein the first attribute range contains an attribute value corresponding to the node attribute of each node in a normal operating state, and the first attribute range corresponds to the node attribute one by one.

[0065] The determination module 22 is further configured to: determine a first path in the graph data model corresponding to each time; determine a variation of an attribute value of a node attribute of each node in the first path corresponding to each adjacent time, according to the attribute value of the node attribute of each node, determine a time sequence feature of each node in the graph data model according to the variation; determine a spatial feature corresponding to each node according to a spatial position of each node in the graph data model corresponding to each time; and fuse the time sequence feature and the spatial feature corresponding to each node to obtain a fused space-time sequence feature; wherein the first path represents each connected graph in the graph data model.

[0066] The prediction module 23 is further configured to: determine a topological connection coefficient between each two power grid equipment according to a switch state of each power grid equipment at a current time; perform attention calculation processing on the topological connection coefficient and the fused space-time sequence feature according to the trained first neural network to obtain a weighted fused space-time vector; and determine a predicted value of the power grid equipment after the N times according to the weighted fused space-time vector. The determination of the topological connection coefficient between each two power grid equipment according to the switch state of each power grid equipment at the current time includes: determination by using a target formula; the target formula includes: ; wherein, denotes a topological connection coefficient between the power grid equipment i and the power grid equipment j, denotes a direct connection state between the power grid equipment i and the intermediate node k, denotes a damping coefficient, and m denotes a total number of power grid equipments, denotes a random jump probability, denotes a topological connection coefficient between the intermediate node k and the power grid equipment j; denotes a state value of a switch state between the power grid equipment i and the power grid equipment k.

[0067] The control module 24 is further configured to: obtain user usage data at each of N moments; determine a power equipment control strategy based on the predicted value of the power grid equipment and the user usage data; the control strategy includes at least one of a start-stop control strategy for power generation equipment in the power grid equipment, a switching control strategy for power transmission equipment, a tap adjustment control strategy for transformer equipment, and a user power restriction control strategy.

[0068] The control module 24 is also used to: perform time-series decomposition processing on the user's usage data to obtain the user's load demand at different time scales; determine the matching results between the power generation equipment and the load demand based on the predicted values ​​of the power grid equipment and the user's load demand in each time period; and determine the power equipment control strategy based on the matching results.

[0069] See also Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided in one embodiment of the present application. Figure 3 The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 are used to store computer programs, which include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. The processor 301 is configured to invoke the program instructions to execute the functions of the modules / units described in the aforementioned apparatus embodiments.

[0070] It should be understood that in the embodiment of the present application, the processor 301 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0071] The input device 302 may include a touchpad, a fingerprint collection sensor (for collecting user fingerprint information and fingerprint direction information), a microphone, etc. The output device 303 may include a display (LCD, etc.), a speaker, etc.

[0072] The memory 304 can include read-only memory and random access memory, and provide instructions and data to the processor 301. A portion of the memory 304 can also include non-volatile random access memory. For example, the memory 304 can also store device type information.

[0073] In specific implementations, the processor 301, the input device 302, and the output device 303 described in the embodiments of the present application can perform the implementation manners described in the first and second embodiments of the data analysis method of the comprehensive energy wisdom management and control platform provided by the embodiments of the present application, and can also perform the implementation manners of the electronic device described in the embodiments of the present application, which will not be described here.

[0074] In another embodiment of the present application, a computer readable storage medium is provided, which stores a computer program. The computer program includes program instructions, which are executed by a processor to implement all or part of the processes of the above-mentioned embodiments. The computer program can also be used to instruct related hardware to complete the implementation. The computer program can be stored in a computer readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned method embodiments can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0075] The computer readable storage medium can be an internal storage unit of the electronic device of any of the above-mentioned embodiments, such as a hard disk or a memory of the electronic device. The computer readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the computer readable storage medium can include both the internal storage unit and the external storage device of the electronic device. The computer readable storage medium is used to store the computer program and other programs and data required by the electronic device. The computer readable storage medium can also be used to temporarily store data that has been output or will be output.

[0076] Those skilled in the art can clearly understand that, for the convenience and conciseness of description, the specific working process of the electronic device and the units described above can refer to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0077] Those skilled in the art can clearly understand that, for the convenience and conciseness of description, the specific working process of the electronic device and the units described above can refer to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0078] In several embodiments provided in the present application, it should be understood that the disclosed electronic device and method can be implemented in other ways. For example, the above-described device embodiments are only schematic, for example, the division of units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface or unit, and can also be electrical, mechanical or other forms of connection.

[0079] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0080] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0081] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A data analysis method for an integrated energy intelligent management and control platform, characterized in that: include: Build a graph data model corresponding to each moment based on the operating data of power grid equipment at N moments; The nodes in the graph data model represent power grid devices, the edges in the graph data model represent the connection relationship between different nodes, the directions of the edges in the graph data model represent the current transmission direction between different nodes, and the nodes in the graph data model include node attributes and attribute values, and the attribute values ​​correspond to the node attributes in a one-to-one manner; Determine the fused spatiotemporal sequence features according to the graph data model corresponding to each moment; Dynamically predicting the operating data of the power grid equipment after the N moments according to the fused spatiotemporal sequence features to obtain predicted values ​​of the power grid equipment; A power equipment control strategy is determined based on the predicted value of the power grid equipment.

2. The data analysis method of the integrated energy intelligent management and control platform according to claim 1, characterized in that: The step of constructing a graph data model corresponding to each moment based on the operating data of the power grid equipment at N moments includes: For each node in the graph data model, obtaining a node attribute of the node and a first attribute range of each node attribute; For each node attribute, if the attribute value corresponding to the node attribute is within the first attribute range corresponding to the node attribute, the failure probability attribute of the node is determined to be a first value; if the attribute value corresponding to the node attribute of the node is not within the first attribute range corresponding to the node attribute, a second value of the failure probability attribute corresponding to the node is determined according to the failure probability model; updating each node in the graph data model according to the fault probability attribute, the first value, and the second value; The first attribute range includes attribute values ​​corresponding to the node attributes of each node in a normal operating state, and the first attribute range corresponds to the node attributes in a one-to-one manner.

3. The data analysis method of the integrated energy intelligent management and control platform according to claim 1, characterized in that: The determining of the fused spatiotemporal sequence features according to the graph data model corresponding to each moment includes: Determine the first path in the graph data model corresponding to each moment; Determining a change in the attribute value of each node at each moment based on the attribute value of each node in the first path corresponding to each two adjacent moments, and determining a time series feature of each node in the graph data model based on the change; Determine the spatial features corresponding to each node based on the spatial position of each node in the graph data model corresponding to each moment; The temporal features and spatial features corresponding to each node are fused to obtain the fused spatiotemporal sequence features; wherein the first path represents each connected graph in the graph data model.

4. The data analysis method of the integrated energy intelligent management and control platform according to claim 1, characterized in that: The dynamically predicting the operation data of the power grid equipment after the N moments according to the fused spatiotemporal sequence features to obtain the predicted value of the power grid equipment includes: Determine the topological connection coefficient between every two grid devices according to the switching status of each grid device at the current moment; According to the trained first neural network, the topological connection coefficient and the fused spatiotemporal sequence feature are subjected to attention calculation processing to obtain a weighted fused spatiotemporal vector; based on the weighted fused spatiotemporal vector, the predicted value of the power grid equipment after the N moments is determined.

5. The data analysis method of the integrated energy intelligent management and control platform according to claim 4 is characterized in that: The determining of the topological connection coefficient between each two grid devices according to the switching state of each grid device at the current moment includes: Determined using the target formula; The target formula includes: in, represents the topological connection coefficient between grid equipment i and grid equipment j, Indicates the direct connection status between grid device i and intermediate node k, represents the damping coefficient, m represents the total number of grid equipment, represents the random jump probability, represents the topological connection coefficient between the intermediate node k and the grid equipment j; The status value indicating the switch status between grid device i and grid device k.

6. The data analysis method of the integrated energy intelligent management and control platform according to claim 1, characterized in that: The power equipment control strategy includes: Obtaining user usage data at each of the N moments; The power equipment control strategy is determined based on the predicted value of the power grid equipment and the user's usage data; the control strategy includes at least one of the start-stop control strategy of the power generation equipment in the power grid equipment, the switching control strategy of the transmission equipment, the tap adjustment control strategy of the transformer equipment and the user power restriction control strategy.

7. The data analysis method of the integrated energy intelligent management and control platform according to claim 6, characterized in that: The method determines the power equipment control strategy based on the predicted value of the power grid equipment and the user's usage data, including: Performing time series decomposition processing on the user's usage data to obtain the user's load demand at different time scales; Determining a matching result between the power generation equipment and the load demand based on the predicted value of the power grid equipment and the load demand of the user at various time scales; The power equipment control strategy is determined according to the matching result.

8. A data analysis device for an integrated energy intelligent management and control platform, characterized in that: include: A construction module is used to construct a graph data model corresponding to each moment based on the operating data of the power grid equipment at N moments; The nodes in the graph data model represent power grid devices, the edges in the graph data model represent the connection relationship between different nodes, the directions of the edges in the graph data model represent the current transmission direction between different nodes, and the nodes in the graph data model include node attributes and attribute values, and the attribute values ​​correspond to the node attributes in a one-to-one manner; A determination module, configured to determine the fused spatiotemporal sequence features based on the graph data model corresponding to each moment; A prediction module, configured to dynamically predict the operation data of the power grid equipment after the N moments based on the fused spatiotemporal sequence features to obtain a predicted value of the power grid equipment; The control module is used to determine the power equipment control strategy according to the predicted value of the power grid equipment.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing 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 7 are implemented.

Images