Full-link intelligent monitoring method and system based on power grid data

Abstract

The embodiment of the invention provides a full-link intelligent monitoring method and system based on power grid data, and belongs to the technical field of power grid data early warning. The full-link intelligent monitoring method comprises the following steps: acquiring historical power grid data on a power grid operation full link, and preprocessing the historical power grid data; analyzing the preprocessed historical power grid data to determine link nodes in the historical power grid data acquisition process; determining an influence condition and an early warning threshold range of each link node according to the historical power grid data on the link nodes; and acquiring real-time power grid data on the power grid operation full link, and performing real-time monitoring on the power grid full link according to the real-time power grid data and the influence condition and the early warning threshold range of each link node. According to the full-link intelligent monitoring method, the operation condition of the full link of the power grid can be judged in real time through various conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid data early warning technology, and in particular to a full-link intelligent monitoring method and system based on power grid data. Background Art

[0002] Power grid big data is characterized by large volumes, diverse sources, and complex formats. It covers every aspect of power grid operations, including power generation, transmission, transformation, distribution, and consumption. This data is sourced from diverse information acquisition channels, including sensors, smart devices, audio communication equipment, mobile terminals, and video surveillance equipment. Smart grid big data not only has a massive volume but also a diverse structure, including structured, semi-structured, and unstructured data, with close business connections. However, due to the current large volume of power grid data, data distortion is prone to occur during the collection process, and the conditions for fault diagnosis are relatively simple, resulting in insufficient real-time monitoring of the entire power grid operation chain. Therefore, a full-chain intelligent monitoring method based on power grid data is needed to assess the operational status of the entire power grid chain in real time using multiple conditions. Summary of the Invention

[0003] The purpose of the embodiments of the present invention is to provide a full-link intelligent monitoring method and system based on power grid data, which can judge the operating status of the entire power grid link in real time through various conditions.

[0004] To achieve the above objectives, an embodiment of the present invention provides a full-link intelligent monitoring method based on power grid data, the full-link intelligent monitoring method comprising: Acquiring historical power grid data on the entire power grid operation link and preprocessing the historical power grid data; Analyzing the pre-processed historical power grid data to determine link nodes in a process of collecting the historical power grid data; Determine the impact of each link node and the warning threshold range based on the historical power grid data on the link node; Acquire real-time grid data on the entire link of the grid operation, and monitor the entire link of the grid in real time based on the real-time grid data and the impact of each link node and the warning threshold range.

[0005] Optionally, obtaining historical power grid data on the entire power grid operation link and preprocessing the historical power grid data includes: Acquiring digital data from the historical power grid data, and arranging the digital data in chronological order to form a set arranged in chronological order; Determine whether there are missing values ​​in the set; When there is a missing value in the set, obtaining the digital data of the same historical period as the missing value; The digital data in the same historical period are averaged to serve as missing values, thereby completing the digital data and completing the preprocessing of the historical power grid data.

[0006] Optionally, performing data analysis on the pre-processed historical power grid data to determine link nodes in the historical power grid data collection process includes: Preprocessing the text data in the historical power grid data; The preprocessed text data in the historical power grid data is obtained in the LDA model, and the number of topics in the text data is selected as the number of nodes, and the hyperparameter vector ; Corresponding to each word of each text in the text data, a topic number is randomly assigned ; Rescan the text data, and for each word, update its topic number using the Gibbs sampling formula, and update the number of the word in the text; Repeat the steps of updating the topic number using the Gibbs sampling formula and updating the number of the word in the text until Gibbs sampling converges; The topics of the respective words of the respective texts in the text data are counted as link nodes.

[0007] Optionally, determining the impact of each link node and the warning threshold range according to historical power grid data on the link node includes: Obtaining the link node and the corresponding digital data, as well as the data flow direction of the link node; According to the data flow direction of the link node, constructing a vector autoregressive model of a link node and another link node through corresponding digital data; Selecting the order of the vector autoregression model to ensure that the vector autoregression model can accurately capture the dynamic relationship between the two link nodes; Performing an F test on the digital data of the two link nodes after constructing the vector autoregression model to determine whether the two link nodes have an impact; The link nodes that have an impact are screened, and the impacting nodes and the affected nodes are determined.

[0008] Optionally, determining the impact of each link node and the warning threshold range according to historical power grid data on the link node includes: Acquiring digital data from the historical power grid data of the same historical period; constructing digital data in the historical power grid data of the same historical period into a time-historical power grid data graph, thereby obtaining a scatter plot with respect to time; Fitting the scatter plot with respect to time into a fitting curve, where points on the fitting curve are standard points that best represent historical power grid data; A tolerance value is preset to set upper and lower warning lines based on the fitting curve, and the area between the upper and lower warning lines is the warning threshold range.

[0009] Optionally, obtaining real-time grid data on the entire grid operation link, and performing real-time monitoring of the entire grid link based on the real-time grid data and the impact of each link node and the warning threshold range, includes: Acquiring digital data of affected link nodes in the historical power grid data; Preprocessing the digital data of the link node; The pre-processed digital data of the influencing node and the affected node in the link node are sent to the Kalman filter network for training, thereby obtaining a prediction model for the affected node.

[0010] Optionally, obtaining real-time grid data on the entire grid operation link, and performing real-time monitoring of the entire grid link based on the real-time grid data and the impact of each link node and the warning threshold range, includes: Acquiring real-time grid data on the entire operation link of the grid; Determine the link nodes according to the real-time power grid data, and determine the influencing nodes and the affected nodes therein; Bringing the digital data corresponding to the influencing node into the prediction model to obtain the digital data of the corresponding affected node; According to the distribution of the predicted digital data and the real-time digital data in the real-time power grid data currently collected in real time within the warning threshold range, the situation of the link node is determined, thereby avoiding the occurrence of distortion of the real-time digital data collected in real time.

[0011] Optionally, determining the status of the link node according to the predicted distribution of the digital data and the real-time digital data in the real-time power grid data currently collected in real time within the warning threshold range includes: Determining whether the predicted digital data or the real-time digital data currently collected are both within the warning threshold range; When the predicted digital data or the real-time digital data currently collected are not all within the warning threshold range, it is determined that the corresponding link node is abnormal and a warning reminder is issued.

[0012] On the other hand, the present invention also provides a full-link intelligent monitoring system based on power grid data, the full-link intelligent monitoring system comprising: Data acquisition module, used to obtain historical grid data on the entire link of grid operation; A monitoring module is used to execute the above-mentioned full-link intelligent monitoring method based on power grid data according to the historical power grid data.

[0013] Through the above technical solution, the present invention provides a full-link intelligent monitoring method and system based on power grid data, which obtains historical power grid data on the entire power grid operation link and can pre-process the historical power grid data, so that the pre-processed historical power grid data can be analyzed to further determine the link nodes in the historical power grid data acquisition process. After determining the link node, the influence of each link node and the warning threshold range can be determined based on the historical power grid data on the link node. After determining the influence and warning threshold range between each link node on the entire link of the power grid, the real-time power grid data on the entire power grid operation link can be obtained, and then the entire power grid link can be monitored in real time based on the real-time power grid data, the influence of each link node and the determined warning threshold range. This full-link intelligent monitoring method can judge the operation status of the entire power grid link in real time through multiple conditions.

[0014] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following detailed description, they are used to explain the embodiments of the present invention, but do not constitute a limitation of the embodiments of the present invention. In the accompanying drawings: Figure 1 is a flow chart of a full-link intelligent monitoring method based on power grid data according to one embodiment of the present invention; Figure 2 is a flowchart of preprocessing of a full-link intelligent monitoring method based on power grid data according to one embodiment of the present invention; Figure 3 This is a flowchart of determining link nodes in a full-link intelligent monitoring method based on power grid data according to one embodiment of the present invention; Figure 4 This is a flowchart of determining an influencing link node according to a full-link intelligent monitoring method based on power grid data according to one embodiment of the present invention; Figure 5is a flowchart of determining a warning threshold range of a full-link intelligent monitoring method based on power grid data according to an embodiment of the present invention; Figure 6 is a flowchart of obtaining a prediction model of a full-link intelligent monitoring method based on power grid data according to one embodiment of the present invention; Figure 7 is a first flow chart of real-time monitoring of a full-link intelligent monitoring method based on power grid data according to an embodiment of the present invention; Figure 8 This is a second flow chart of real-time monitoring of a full-link intelligent monitoring method based on power grid data according to an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The following describes the specific implementation of the embodiment of the present invention in detail with reference to the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the embodiment of the present invention and is not used to limit the embodiment of the present invention.

[0017] In the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned. They should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0018] Figure 1 This is a flow chart of a full-link intelligent monitoring method based on power grid data according to one embodiment of the present invention. The full-link intelligent monitoring process may include: In step S1, historical grid data on the entire grid operation link is acquired and preprocessed.

[0019] In step S2, the pre-processed historical power grid data is analyzed to determine the link nodes in the historical power grid data collection process.

[0020] In step S3, the impact of each link node and the warning threshold range are determined based on the historical power grid data on the link node.

[0021] In step S4, real-time grid data on the entire power grid operation link is obtained, and the entire power grid link is monitored in real time based on the real-time grid data and the impact of each link node and the warning threshold range.

[0022] In the present invention, when full-link intelligent monitoring is performed based on the power grid data, historical power grid data on the entire power grid operation link can be obtained, and the historical power grid data can be preprocessed, so that the preprocessed historical power grid data can be analyzed to further determine the link nodes in the historical power grid data acquisition process. After determining the link nodes, the influence of each link node and the warning threshold range can be determined based on the historical power grid data on the link node. After determining the influence and warning threshold range between each link node on the entire link of the power grid, real-time power grid data on the entire power grid operation link can be obtained, and then the entire power grid link can be monitored in real time based on the real-time power grid data, the influence of each link node and the determined warning threshold range. This full-link intelligent monitoring method can judge the operation status of the entire power grid link in real time based on multiple conditions.

[0023] In one embodiment of the present invention, Figure 2 As shown, the pre-processing process may include: In step S5 , digital data in the historical power grid data is acquired, and the digital data is arranged in chronological order to form a set arranged in chronological order.

[0024] In step S6, it is determined whether there are missing values ​​in the set.

[0025] In step S7, when there is a missing value in the set, the digital data of the same historical period as the missing value is obtained.

[0026] In step S8, the digital data in the same historical period are averaged to serve as missing values, thereby completing the digital data and completing the preprocessing of the historical power grid data.

[0027] In the present invention, when preprocessing the digital data of historical power grid data, the digital data can be first arranged in chronological order, so that a set arranged in chronological order can be formed. After the sorting is completed, it can be determined whether there are missing values ​​in the set. In the case where it is determined that there are missing values, it is necessary to supplement the missing values. At this time, digital data in the same historical period as the missing values ​​can be obtained. The same historical period can be data collected on the same day of the previous week, data collected on the same day of the previous month, data collected on the same day of previous years, etc. After obtaining the digital data of the same historical period, the average value of the numbers in the same historical period can be calculated, and then the average value can be used as the missing value, so that the missing digital data can be supplemented to complete the preprocessing of the historical power grid data.

[0028] In one embodiment of the present invention, Figure 3 As shown, the process of determining the link node may include: In step S9, the text data in the historical power grid data is preprocessed.

[0029] In step S10, the pre-processed text data in the historical power grid data is obtained in the LDA model, and the number of topics in the text data is selected as the number of nodes, and the hyperparameter vector .

[0030] In step S11, a topic number is randomly copied for each word of each text in the text data. .

[0031] In step S12, the text data is rescanned, and for each word, its topic number is updated by the Gibbs sampling formula, and the number of the word in the text is also updated.

[0032] In step S13, the steps of updating the topic number using the Gibbs sampling formula and updating the number of the word in the text are repeated until the Gibbs sampling converges.

[0033] In step S14, the topics of the respective words of the respective texts in the text data are counted as link nodes.

[0034] In the present invention, when historical power grid data is obtained, each node on the entire link can be determined based on the historical power grid data. When determining the link nodes, the topic in the historical power grid data can be found through the LDA model, and then the topic can be used as the link node. The text data in the historical power grid data can be preprocessed to remove redundant words in the text data. After preprocessing, the preprocessed text data in the historical power grid data can be fed into the LDA model, and the number of topics in the text data can be selected as the number of nodes, and the hyperparameter vector . A topic number can be assigned to each word in each text in the text data, and then the text data can be rescanned. For each word in the text, its topic number can be updated using the Gibbs sampling formula, and the number of the word can be updated in the text. The above updating steps can be repeated until Gibbs sampling converges. Through the above repeated sampling, the topic of each word in each text in the text data can be obtained, and then the topic of each word can be used as a link node.

[0035] In one embodiment of the present invention, Figure 4 As shown, the process of determining the affected link node may include: In step S15 , the link nodes and the corresponding digital data, as well as the data flow direction of the link nodes are obtained.

[0036] In step S16, according to the data flow direction of the link node, a vector autoregressive model of one link node and another link node is constructed using the corresponding digital data.

[0037] In step S17 , the order of the vector autoregression model is selected to ensure that the vector autoregression model can accurately capture the dynamic relationship between the two link nodes.

[0038] In step S18, an F test is performed on the digital data of the two link nodes after the vector autoregression model is constructed to determine whether the two link nodes have an influence.

[0039] In step S19, the affected link nodes are screened, and the influencing nodes and the affected nodes are determined.

[0040] In the present invention, when determining the influence of one link node on another link node on the entire link, the link node and the corresponding digital data, as well as the data flow direction of each link node, can be first obtained. According to the data flow direction of each link node, the affected link node can be preliminarily determined. According to the data flow direction of the link node, a vector autoregression model of one link node and another link node can be constructed using the corresponding digital data. After constructing the vector autoregression model, the order of the vector autoregression model can be selected to ensure that the vector autoregression model can accurately capture the dynamic relationship between the two link nodes. After constructing the vector autoregression model, the vector autoregression model can be subjected to an F test to determine whether there is mutual influence between the two link nodes. The above method can be used to screen link nodes that have mutual influence, and the influencing nodes and affected nodes can be determined.

[0041] In one embodiment of the present invention, Figure 5 As shown, the process of determining the warning threshold range may include: In step S20 , digital data in historical power grid data of the same historical period is obtained.

[0042] In step S21 , digital data in historical power grid data of the same historical period are constructed into a time-historical power grid data graph, thereby obtaining a scatter plot with respect to time.

[0043] In step S22 , the scatter plot with respect to time is fitted into a fitting curve, and the points on the fitting curve are standard points that best represent the historical power grid data.

[0044] In step S23, a tolerance value is preset to set upper and lower warning lines based on the fitting curve, and the area between the upper and lower warning lines is the warning threshold range.

[0045] In the present invention, when determining the warning threshold range, digital data from historical power grid data from the same historical period can be first obtained. The digital data from the historical power grid data from the same historical period can then be constructed into a time-historical power grid data graph. The graph can contain the historical power grid data from the same historical period, and each data point can be represented as a point, thereby obtaining a scatter plot of each historical power grid data over time. After obtaining the scatter plot, the scatter plot over time can be fitted into a fitting curve, and the points on the fitting curve can be the standard points that best represent the historical power grid data. After obtaining the fitting curve, a tolerance value can be preset, and then upper and lower warning lines can be set based on the fitting curve based on the tolerance value. The points on the upper and lower warning lines can be obtained by adding and subtracting the points on the fitting curve and the tolerance value. After obtaining the upper and lower warning lines, it can be determined that the area between the upper and lower warning lines is the warning threshold range.

[0046] In one embodiment of the present invention, Figure 6 As shown, the process of obtaining a prediction model may include: In step S24 , digital data of the affected link nodes in the historical power grid data are obtained.

[0047] In step S25 , the digital data of the link node is pre-processed.

[0048] In step S26, the pre-processed digital data of the influencing nodes and the affected nodes in the link nodes are fed into a Kalman filter network for training, thereby obtaining a prediction model for the affected nodes.

[0049] In the present invention, after mutually influencing link nodes are obtained, digital data corresponding to the mutually influencing link nodes can be obtained and preprocessed. After the preprocessing is completed, the digital data of the influencing and affected nodes in the preprocessed link nodes can be fed into a Kalman filter network for training, thereby obtaining a prediction model for the affected node.

[0050] In one embodiment of the present invention, Figure 7 As shown, the first process of real-time monitoring may include: In step S27, real-time grid data on the entire grid operation link is obtained.

[0051] In step S28 , link nodes are determined according to the real-time power grid data, and influencing nodes and affected nodes therein are determined.

[0052] In step S29, the corresponding digital data of the influencing node is brought into the prediction model to obtain the corresponding digital data of the affected node.

[0053] In step S30, the status of the link node is determined based on the distribution of the predicted digital data and the real-time digital data in the real-time power grid data currently collected in real time within the warning threshold range, thereby avoiding the occurrence of distortion of the real-time digital data collected in real time.

[0054] In the present invention, when monitoring the entire power grid in Jining in real time, real-time power grid data on the entire power grid operation link can be obtained, and then the link nodes therein can be determined based on the real-time power grid data, and the image nodes and affected nodes therein can be determined. The corresponding digital data of the influencing node is brought into the prediction model, and the digital data of the corresponding affected node can be obtained. According to the distribution of the digital data of the affected node obtained by the prediction and the real-time digital data in the real-time power grid data currently collected in real time within the warning threshold range, the situation of the link node can be determined. It can be known that in the entire link, the affected node of a link node may also be the influencing node of another link node. Because the present application may have two corresponding digital data for a link node, one is predicted and the other is collected in real time, dual judgment is performed through the two digital data, and the link node can be monitored more accurately. Moreover, if the real-time digital data collected in real time is distorted, the digital data obtained by the prediction can also be monitored. This dual guarantee makes the monitoring of the entire power grid link more accurate.

[0055] In one embodiment of the present invention, Figure 8 As shown, the second process of real-time monitoring may include: In step S31 , it is determined whether the predicted digital data or the real-time digital data currently collected are both within the warning threshold range.

[0056] In step S32, when the predicted digital data or the real-time digital data currently collected are not all within the warning threshold range, it is determined that the corresponding link node is abnormal and a warning reminder is issued.

[0057] In the present invention, when the predicted digital data or the real-time digital data currently being collected is obtained, it can be determined whether the predicted digital data or the real-time digital data currently being collected is within the warning threshold range. If the predicted digital data or the real-time digital data currently being collected is not both within the warning threshold range, it can be determined that the corresponding link node has an abnormality, and a warning alert can be issued. This approach allows for a judgment to be made regarding the corresponding link node even if only one of the predicted digital data or the real-time digital data is obtained.

[0058] In another aspect, the present invention further provides a full-link intelligent monitoring system based on power grid data, comprising a data acquisition module and a monitoring module. The data acquisition module can be used to acquire historical power grid data across the entire power grid operation chain. The monitoring module can be used to execute the above-described full-link intelligent monitoring method based on power grid data based on the historical power grid data.

[0059] Through the above technical solution, the present invention provides a full-link intelligent monitoring method and system based on power grid data, which obtains historical power grid data on the entire power grid operation link and can pre-process the historical power grid data, so that the pre-processed historical power grid data can be analyzed to further determine the link nodes in the historical power grid data acquisition process. After determining the link node, the influence of each link node and the warning threshold range can be determined based on the historical power grid data on the link node. After determining the influence and warning threshold range between each link node on the entire link of the power grid, the real-time power grid data on the entire power grid operation link can be obtained, and then the entire power grid link can be monitored in real time based on the real-time power grid data, the influence of each link node and the determined warning threshold range. This full-link intelligent monitoring method can judge the operation status of the entire power grid link in real time through multiple conditions.

[0060] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0061] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0062] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0063] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0064] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0065] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0066] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can be implemented using any method or technology for information storage. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change RAM (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0067] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0068] The above are merely embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A full-link intelligent monitoring method based on power grid data, characterized in that: The full-link intelligent monitoring method includes: Acquire historical grid data on the entire link of grid operation and preprocess the historical grid data; Analyzing the pre-processed historical power grid data to determine link nodes in a process of collecting the historical power grid data; Determine the impact of each link node and the warning threshold range based on the historical power grid data on the link node; Acquire real-time grid data on the entire link of the grid operation, and monitor the entire link of the grid in real time based on the real-time grid data and the impact of each link node and the warning threshold range.

2. The full-link intelligent monitoring method according to claim 1, characterized in that: Acquiring historical power grid data on the entire power grid operation link and preprocessing the historical power grid data, including: Acquiring digital data from the historical power grid data, and arranging the digital data in chronological order to form a set arranged in chronological order; Determine whether there are missing values ​​in the set; When there is a missing value in the set, obtaining the digital data of the same historical period as the missing value; The digital data in the same historical period are averaged to serve as missing values, thereby completing the digital data and completing the preprocessing of the historical power grid data.

3. The full-link intelligent monitoring method according to claim 1, characterized in that: Performing data analysis on the pre-processed historical power grid data to determine link nodes in the historical power grid data collection process includes: Preprocessing the text data in the historical power grid data; The preprocessed text data in the historical power grid data is obtained in the LDA model, and the number of topics in the text data is selected as the number of nodes, and the hyperparameter vector ; Corresponding to each word of each text in the text data, a topic number is randomly assigned ; Rescan the text data, and for each word, update its topic number using the Gibbs sampling formula, and update the number of the word in the text; Repeat the steps of updating the topic number using the Gibbs sampling formula and updating the number of the word in the text until Gibbs sampling converges; The topics of the respective words of the respective texts in the text data are counted as link nodes.

4. The full-link intelligent monitoring method according to claim 3, characterized in that: Determining the impact of each link node and the warning threshold range according to the historical power grid data on the link node includes: Obtaining the link node and the corresponding digital data, as well as the data flow direction of the link node; According to the data flow direction of the link node, constructing a vector autoregressive model of a link node and another link node through corresponding digital data; Selecting the order of the vector autoregression model to ensure that the vector autoregression model can accurately capture the dynamic relationship between the two link nodes; Performing an F test on the digital data of the two link nodes after constructing the vector autoregression model to determine whether the two link nodes have an impact; The link nodes that have an impact are screened, and the impacting nodes and the affected nodes are determined.

5. The full-link intelligent monitoring method according to claim 1, characterized in that: Determining the impact of each link node and the warning threshold range according to the historical power grid data on the link node includes: Acquiring digital data from the historical power grid data of the same historical period; constructing digital data in the historical power grid data of the same historical period into a time-historical power grid data graph, thereby obtaining a scatter plot with respect to time; Fitting the scatter plot with respect to time into a fitting curve, where points on the fitting curve are standard points that best represent historical power grid data; A tolerance value is preset to set upper and lower warning lines based on the fitting curve, and the area between the upper and lower warning lines is the warning threshold range.

6. The full-link intelligent monitoring method according to claim 5, characterized in that: Acquire real-time grid data on the entire link of the grid operation, and monitor the entire link of the grid in real time based on the real-time grid data and the impact of each link node and the warning threshold range, including: Acquiring digital data of affected link nodes in the historical power grid data; Preprocessing the digital data of the link node; The pre-processed digital data of the influencing node and the affected node in the link node are sent to the Kalman filter network for training, thereby obtaining a prediction model for the affected node.

7. The full-link intelligent monitoring method according to claim 6, characterized in that: Acquire real-time grid data on the entire link of the grid operation, and monitor the entire link of the grid in real time based on the real-time grid data and the impact of each link node and the warning threshold range, including: Acquiring real-time grid data on the entire operation link of the grid; Determine the link nodes according to the real-time power grid data, and determine the influencing nodes and the affected nodes therein; Bringing the digital data corresponding to the influencing node into the prediction model to obtain the digital data of the corresponding affected node; According to the distribution of the predicted digital data and the real-time digital data in the real-time power grid data currently collected in real time within the warning threshold range, the situation of the link node is determined, thereby avoiding the occurrence of distortion of the real-time digital data collected in real time.

8. The full-link intelligent monitoring method according to claim 7, characterized in that: Determining the status of the link node according to the predicted distribution of the digital data and the real-time digital data in the real-time power grid data currently collected in real time within the warning threshold range includes: Determining whether the predicted digital data or the real-time digital data currently collected are both within the warning threshold range; When the predicted digital data or the real-time digital data currently collected are not all within the warning threshold range, it is determined that the corresponding link node is abnormal and a warning reminder is issued.

9. A full-link intelligent monitoring system based on power grid data, characterized in that: The full-link intelligent monitoring system includes: Data acquisition module, used to obtain historical grid data on the entire link of grid operation; A monitoring module is used to execute a full-link intelligent monitoring method based on power grid data as described in any one of claims 1-8 according to the historical power grid data.

Images