Analysis method and device for telemetering and telesignaling data and storage medium

Abstract

The invention discloses a telemetering and telesignalling data analysis method and device and a storage medium. The method comprises the following steps: constructing a power flow telemetering matrix and an adjacent matrix according to telemetering and telesignalling data to be analyzed; taking the power flow telemetering matrix, the adjacent matrix and a pre-constructed power grid graph model as input data, and inputting the input data into a graph neural network model; performing node feature extraction on the input data according to the graph neural network, and generating a probability matrix of each edge according to the extracted node features; and taking the average value of the probability matrixes of all the edges as the total remote signaling error probability of the power grid system, and when the total remote signaling error probability is lower than a preset threshold value, judging that the telemetering and remote signaling data to be analyzed are abnormal data. According to the method, node feature extraction is carried out on the input data through the graph neural network, the probability matrix is generated, and the telemetering and telecommand data problems such as repeated jumping and long-term invariability in a power grid system can be accurately analyzed, so that the accuracy and reliability of telemetering and telecommand data analysis can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power systems, and in particular to an analysis method, device and storage medium for telemetry and telesignaling data. Background Art

[0002] Power system state estimation is the basis of power system analysis. However, in actual power grid operation, there may be data problems such as telemetry and telesignaling data jump or remain unchanged, power data imbalance, etc., which leads to low pass rate of state estimation and inability to provide accurate system topology state and model for subsequent applications (including spot market). Therefore, accurate analysis of telemetry and telesignaling data is the key to power system state estimation.

[0003] The existing analysis method for telemetry and telesignaling data usually sets data jump / unchanged threshold parameters, judges individual data, issues an alarm if the limit is exceeded, and then manually investigates the reasons for data jump / unchanged. By setting simple threshold parameters, it is difficult to accurately analyze telemetry and telesignaling data problems such as repeated jumps and long-term invariance, resulting in low accuracy in telemetry and telesignaling data analysis. Summary of the invention

[0004] The present invention provides a method, device and storage medium for analyzing telemetry and telesignaling data, so as to solve the technical problem that the existing method for analyzing telemetry and telesignaling data is difficult to accurately analyze telemetry and telesignaling data problems such as repeated jumps and long-term invariance by setting simple threshold parameters, resulting in low accuracy of telemetry and telesignaling data analysis.

[0005] The present invention provides a method for analyzing telemetry and telesignaling data, comprising:

[0006] Construct the power flow telemetry matrix and adjacency matrix according to the telemetry and telesignaling data to be analyzed;

[0007] The power flow telemetry matrix, the adjacency matrix and the pre-constructed power grid graph model are input as input data into a graph neural network model, wherein the edge features of the power grid graph model include power flow telemetry data and telesignaling data indicating the connection status of the power grid;

[0008] Extracting node features from the input data according to the graph neural network, and generating a probability matrix for each edge according to the extracted node features;

[0009] The average value of the probability matrix of all edges is used as the total telesignaling error probability of the power grid system. When the total telesignaling error probability is lower than a preset threshold, the telemetry and telesignaling data to be analyzed are determined to be abnormal data.

[0010] Further, the telemetry and telesignaling data to be analyzed include telemetry data and telesignaling data, and the step of constructing a power flow telemetry matrix and an adjacency matrix according to the telemetry and telesignaling data to be analyzed includes:

[0011] constructing a power flow telemetry matrix based on the telemetry data;

[0012] An adjacency matrix is ​​constructed based on the remote signaling data.

[0013] Furthermore, the analysis method of the telemetry and telesignaling data also includes:

[0014] Determine a corresponding space vertex according to each device element in the power grid system;

[0015] If there is an electrical connection relationship between the spatial vertices, the electrical connection relationship is used as an edge;

[0016] Determining a corresponding time vertex according to each time stamp in the power grid system;

[0017] A power grid graph model is constructed based on the spatial vertices, the edges and the time vertices, wherein the edges store telemetry data and telesignaling data corresponding to different time nodes.

[0018] Furthermore, extracting node features from the input data according to the graph neural network includes:

[0019] Extracting node global features of the input data;

[0020] Node features in the node global features are extracted in the row direction of the matrix and the column direction of the matrix.

[0021] Furthermore, after performing node feature extraction on the input data according to the graph neural network and generating a probability matrix of each edge according to the extracted node features, the method further includes:

[0022] Calculate the loss value based on the generated probability matrix and the actual label probability matrix;

[0023] Using a back-propagation algorithm, the network parameters of the graph neural network are updated according to the loss value.

[0024] The present invention also provides a remote sensing and remote signaling data analysis device, comprising:

[0025] A matrix construction module is used to construct a power flow telemetry matrix and an adjacency matrix based on the telemetry and telesignaling data to be analyzed;

[0026] A data input module, used to input the power flow telemetry matrix, the adjacency matrix and the pre-built power grid graph model as input data into a graph neural network model, wherein the edge features of the power grid graph model include power flow telemetry data and telesignaling data indicating the power grid connection status;

[0027] A probability matrix generation module, used to extract node features from the input data according to the graph neural network, and generate a probability matrix for each edge according to the extracted node features;

[0028] The data analysis module is used to take the average value of the probability matrix of all edges as the total telesignaling error probability of the power grid system, and when the total telesignaling error probability is lower than a preset threshold, determine that the telemetry and telesignaling data to be analyzed are abnormal data.

[0029] Furthermore, the telemetry and telesignaling data to be analyzed include telemetry data and telesignaling data, and the matrix construction module is further used for:

[0030] constructing a power flow telemetry matrix based on the telemetry data;

[0031] An adjacency matrix is ​​constructed based on the remote signaling data.

[0032] Furthermore, the telemetry and telesignaling data analysis device further includes a power grid diagram model building module, which is used to:

[0033] Determine a corresponding space vertex according to each device element in the power grid system;

[0034] If there is an electrical connection relationship between the spatial vertices, the electrical connection relationship is used as an edge;

[0035] Determining a corresponding time vertex according to each time stamp in the power grid system;

[0036] A power grid graph model is constructed based on the spatial vertices, the edges and the time vertices, wherein the edges store telemetry data and telesignaling data corresponding to different time nodes.

[0037] Furthermore, the probability matrix generation module is also used for:

[0038] Extracting node global features of the input data;

[0039] Node features in the node global features are extracted in the row direction of the matrix and the column direction of the matrix.

[0040] The present invention also provides a terminal device, comprising: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and the processor implements the above-mentioned telemetry and telesignaling data analysis method when executing the computer program.

[0041] The present invention also provides a computer-readable storage medium, which includes a stored computer program; wherein, when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the telemetry and telesignaling data analysis method as described above.

[0042] The preset threshold value of the embodiment of the present invention can be dynamically adjusted according to actual needs. The embodiment of the present invention can accurately represent the active power, reactive power and topological structure of the power grid by constructing a power flow telemetry matrix and an adjacency matrix, thereby providing an accurate data basis for telemetry and telesignaling data analysis. The power flow telemetry matrix, the adjacency matrix and the pre-constructed power grid graph model are used as input data and input into the graph neural network model. The graph neural network is used to extract node features of the input data and generate a probability matrix. The telemetry and telesignaling data problems such as repeated jumps and long-term invariance in the power grid system can be accurately analyzed, thereby effectively improving the accuracy and reliability of the telemetry and telesignaling data analysis.

[0043] Furthermore, the embodiment of the present invention adopts DGCNN to effectively extract local features in point cloud data through edge convolution and dynamic graph, and GCN combines the attention mechanism to further extract global features of nodes, which can comprehensively capture local and global information in the data, so as to accurately analyze whether the telemetry and telesignaling data are abnormal based on the extracted node features, and effectively improve the accuracy of telemetry and telesignaling data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flow chart of a method for analyzing telemetry and telesignaling data provided by an embodiment of the present invention;

[0045] Figure 2 is a schematic diagram of the structure of a power grid diagram model provided by an embodiment of the present invention;

[0046] Figure 3 is a schematic diagram of a feature extraction process provided by an embodiment of the present invention;

[0047] Figure 4 It is a structural schematic diagram of a remote sensing and remote signaling data analysis device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0049] In the description of this application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the feature. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0050] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0051] See also Figure 1 The present invention provides a method for analyzing telemetry and telesignaling data, comprising:

[0052] S1. Construct a power flow telemetry matrix and an adjacency matrix based on the telemetry and telesignaling data to be analyzed;

[0053] In the embodiment of the present invention, the telemetry and telesignaling data include telemetry and telesignaling data, and the telemetry data include active power and reactive power.

[0054] S2, inputting the power flow telemetry matrix, the adjacency matrix and the pre-built power grid graph model as input data into the graph neural network model, wherein the edge features of the power grid graph model include power flow telemetry data and telesignaling data indicating the connection status of the power grid;

[0055] S3. Extract node features from the input data according to the graph neural network, and generate a probability matrix for each edge according to the extracted node features;

[0056] In an embodiment of the present invention, the input data includes a power flow telemetry matrix, an adjacency matrix, and a pre-built power grid graph model, which includes power flow telemetry data of each node in the power grid, that is, power flow information of each node in the power grid. The embodiment of the present invention inputs these data into the GCN model as node features, which enables the model to understand the power state and characteristics of each node; the connection relationship between each node in the adjacency matrix, each element can indicate whether there is an edge between two nodes, and the weight of the edge; the power grid graph model is a graph composed of nodes and edges, where nodes represent equipment in the power grid (such as transformers, transmission lines, etc.), and edges represent connections between equipment. The power grid graph model is used as the input of the GCN model, and the model can predict the telesignaling error probability of the edge by learning the structure and node features of the power grid graph model.

[0057] S4. The average value of the probability matrix of all edges is taken as the total telesignaling error probability of the power grid system. When the total telesignaling error probability is lower than a preset threshold, the telemetry and telesignaling data to be analyzed are determined to be abnormal data.

[0058] In the embodiment of the present invention, the preset threshold can be set to 95% according to actual needs. The embodiment of the present invention generates a probability matrix P for each edge, which can calculate the user-defined time period N T The total remote signaling error probability of the power grid system is obtained by taking the average value of all data of

[0059] In the embodiment of the present invention, when the telemetry and telesignaling data to be analyzed is determined to be abnormal data, a reminder message may be generated to remind relevant personnel to handle the telemetry and telesignaling data problem.

[0060] The preset threshold value of the embodiment of the present invention can be dynamically adjusted according to actual needs. The embodiment of the present invention can accurately represent the active power, reactive power and topological structure of the power grid by constructing a power flow telemetry matrix and an adjacency matrix, thereby providing an accurate data basis for telemetry and telesignaling data analysis. The power flow telemetry matrix, the adjacency matrix and the pre-constructed power grid graph model are used as input data and input into the graph neural network model. The graph neural network is used to extract node features of the input data and generate a probability matrix. The telemetry and telesignaling data problems such as repeated jumps and long-term invariance in the power grid system can be accurately analyzed, thereby effectively improving the accuracy and reliability of the telemetry and telesignaling data analysis.

[0061] In one embodiment, the telemetry and telesignaling data to be analyzed include telemetry data and telesignaling data. Step S1, constructing a power flow telemetry matrix and an adjacency matrix according to the telemetry and telesignaling data to be analyzed, includes:

[0062] S11, constructing a power flow telemetry matrix based on telemetry data;

[0063] In the embodiment of the present invention, the telemetry data includes active power and reactive power. Each element in the power flow telemetry matrix S (size: NbxNb, Nb is the number of grid nodes) contains active power and reactive power. If a corresponding power flow telemetry matrix is ​​constructed between node i and node j, the corresponding telemetry data P is input in row i and column j of the matrix. ij +jQ ij , where P ij is the active power, jQ ij is the reactive power.

[0064] S12. Construct an adjacency matrix based on the remote signaling data.

[0065] In the embodiment of the present invention, the adjacency matrix A (size: Nb x Nb, Nb is the number of grid nodes) represents the topological structure of the grid. If the line is between nodes i and j, the corresponding telesignaling data is input in the i row and j column of the adjacency matrix. The telesignaling data is 1 for the existence of a connected branch, and 0 for the disconnection of the branch or the absence of the branch. The total number of telesignals is N.

[0066] The embodiment of the present invention can accurately detect the active power and reactive power between nodes in the power grid system by constructing a power flow telemetry matrix, and the adjacency matrix can accurately represent the topological structure of the power grid. By analyzing the telemetry and telesignaling data in matrix form, the analysis efficiency of the telemetry and telesignaling data can be effectively improved.

[0067] In one embodiment, the method for analyzing telemetry and telesignaling data further includes:

[0068] Determine a corresponding space vertex according to each device element in the power grid system;

[0069] If there is an electrical connection between the spatial vertices, the electrical connection is taken as an edge;

[0070] Determine a corresponding time vertex according to each time stamp in the power grid system;

[0071] A power grid graph model is constructed based on spatial vertices, edges and time vertices, wherein the edges store telemetry data and telesignaling data corresponding to different time nodes.

[0072] In an embodiment of the present invention, each device element in the power grid system corresponds to a spatial vertex, and if each spatial vertex has an electrical connection relationship, there is an edge; each timestamp corresponds to a time vertex, and there is an abstract edge between the time vertex and each element, and its edge attributes can be the telesignaling data and telemetry data of the element at the time node.

[0073] In an embodiment of the present invention, graph data information of a single time section is first extracted, and for time T, all spatial vertices and time nodes T are extracted, and the telesignaling data and telemetering data on time node T are assigned to corresponding edges on the spatial nodes. For example, time node T stores active power, reactive power and telesignaling information of the line between node 1 and node 2, then these three pieces of information are stored in the corresponding edges, thereby constructing a power grid graph model.

[0074] The embodiment of the present invention constructs a power grid diagram model and uses the power grid diagram model as one of the input data, and can describe the association between the diagram and the diagram time according to the power grid diagram model, thereby facilitating the capture of abnormal telesignaling status.

[0075] See also Figure 2 , which is a schematic diagram of a power grid diagram model structure provided in an embodiment of the present invention.

[0076] In one embodiment, step S3, extracting node features from input data according to a graph neural network, includes:

[0077] S31, extracting node global features of input data;

[0078] In an embodiment of the present invention, DGCNN (point cloud dynamic graph convolution) can be used to extract the relationship and local features between points through edge convolution and dynamic graphs, and then GCN (graph convolutional network) is combined with the attention mechanism to further extract the global features of the nodes, taking into account the different importance of nodes in the extraction process.

[0079] S32. Extract node features from the node global features in the row direction of the matrix and the column direction of the matrix.

[0080] The design of the neural network in the embodiment of the present invention is as follows:

[0081] Using graph convolutional network (GCN) as the core, high-level feature representation of nodes is learned through multi-layer graph convolution;

[0082] The attention layer is introduced on the basis of GCN, so that the model can adaptively focus on the relevant key nodes and edges.

[0083] In an embodiment of the present invention, Feature-direction MLP can be used to extract row node features in the matrix row direction based on node global features, and Channel-direction MLP can be used to extract node features in the matrix column direction based on node global features, thereby completing the extraction of node features.

[0084] In the embodiment of the present invention, after each feature extraction, the extracted features are normalized to eliminate the difference in the feature value range, thereby accelerating the convergence of the model.

[0085] The embodiment of the present invention adopts DGCNN to effectively extract local features in point cloud data through edge convolution and dynamic graph, and GCN combines the attention mechanism to further extract the global features of the nodes, which can comprehensively capture the local and global information in the data, so as to accurately analyze whether the telemetry and telesignaling data are abnormal according to the extracted node features, and effectively improve the accuracy of telemetry and telesignaling data analysis.

[0086] See also Figure 3 , which is a schematic diagram of a feature extraction process provided by an embodiment of the present invention.

[0087] In one embodiment, after extracting node features from input data according to a graph neural network and generating a probability matrix for each edge according to the extracted node features, the method further includes:

[0088] Calculate the loss value based on the generated probability matrix and the actual label probability matrix;

[0089] Using the back-propagation algorithm, the network parameters of the graph neural network are updated according to the loss value.

[0090] In an embodiment of the present invention, supervised learning is used to update network parameters, and the correct telesignaling data that has been manually processed is used as an evaluation criterion. The correct label probability matrix is ​​recorded as L (size: Nb*Nb). The telesignaling data is 100% for the existence of a connected branch, and 0 for the disconnection of the branch or the absence of a branch. A cross entropy loss function or a mean square error loss function is used to calculate the loss based on the predicted probability matrix P and the actual label probability matrix L. The network parameters are updated through a back propagation algorithm, and a gradient descent optimization algorithm (such as Adam) is used to minimize the loss function.

[0091] The embodiment of the present invention calculates the loss value based on the generated probability matrix and the actual label probability matrix, and updates the network parameters of the graph neural network according to the loss value. It can automatically update the network parameters without manual adjustment of the parameters, and can quickly respond to changes in the state of the power grid. It can effectively improve the accuracy of telemetry and telesignaling data analysis, thereby effectively improving the accuracy and reliability of power grid state estimation.

[0092] The implementation of the embodiments of the present invention has the following beneficial effects:

[0093] The preset threshold value of the embodiment of the present invention can be dynamically adjusted according to actual needs. The embodiment of the present invention can accurately represent the active power, reactive power and topological structure of the power grid by constructing a power flow telemetry matrix and an adjacency matrix, thereby providing an accurate data basis for telemetry and telesignaling data analysis. The power flow telemetry matrix, the adjacency matrix and the pre-constructed power grid graph model are used as input data and input into the graph neural network model. The graph neural network is used to extract node features of the input data and generate a probability matrix. The telemetry and telesignaling data problems such as repeated jumps and long-term invariance in the power grid system can be accurately analyzed, thereby effectively improving the accuracy and reliability of the telemetry and telesignaling data analysis.

[0094] Furthermore, the embodiment of the present invention adopts DGCNN to effectively extract local features in point cloud data through edge convolution and dynamic graph, and GCN combines the attention mechanism to further extract global features of nodes, which can comprehensively capture local and global information in the data, so as to accurately analyze whether the telemetry and telesignaling data are abnormal based on the extracted node features, and effectively improve the accuracy of telemetry and telesignaling data analysis.

[0095] See also Figure 4 Based on the same inventive concept as the above embodiment, the present invention also provides an analysis device for telemetry and telesignaling data, comprising:

[0096] A matrix construction module 10 is used to construct a power flow telemetry matrix and an adjacency matrix according to the telemetry and telesignaling data to be analyzed;

[0097] A data input module 20 is used to input the power flow telemetry matrix, the adjacency matrix and the pre-built power grid graph model as input data into the graph neural network model, wherein the edge features of the power grid graph model include power flow telemetry data and telesignaling data indicating the power grid connection status;

[0098] A probability matrix generation module 30 is used to extract node features from input data according to the graph neural network, and generate a probability matrix for each edge according to the extracted node features;

[0099] The data analysis module 40 is used to take the average value of the probability matrix of all edges as the total telesignaling error probability of the power grid system, and when the total telesignaling error probability is lower than a preset threshold, determine that the telemetry and telesignaling data to be analyzed are abnormal data.

[0100] In one embodiment, the telemetry and telesignaling data to be analyzed include telemetry data and telesignaling data, and the matrix building module 10 is further used for:

[0101] Construct a power flow telemetry matrix based on telemetry data;

[0102] Construct an adjacency matrix based on remote signaling data.

[0103] In one embodiment, the telemetry and telesignaling data analysis device further includes a power grid diagram model building module, which is used to:

[0104] Determine a corresponding space vertex according to each device element in the power grid system;

[0105] If there is an electrical connection between the spatial vertices, the electrical connection is taken as an edge;

[0106] Determine a corresponding time vertex according to each time stamp in the power grid system;

[0107] A power grid graph model is constructed based on spatial vertices, edges and time vertices, wherein the edges store telemetry data and telesignaling data corresponding to different time nodes.

[0108] In one embodiment, the probability matrix generation module 30 is further used to:

[0109] Extract node global features of input data;

[0110] Node features in the node global features are extracted in the row direction of the matrix and the column direction of the matrix.

[0111] Accordingly, an embodiment of the present invention further provides a terminal device, comprising: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the method for analyzing telemetry and telesignaling data of any one of the above embodiments is implemented.

[0112] The terminal device of this embodiment includes: a processor, a memory, and a computer program and computer instructions stored in the memory and executable on the processor. When the processor executes the computer program, each step in the above embodiment 1 is implemented, for example Figure 1 Alternatively, when the processor executes the computer program, the functions of each module / unit in the above device embodiment, such as the probability matrix generation module 30, are implemented.

[0113] Exemplarily, the computer program can be divided into one or more modules / units, one or more modules / units are stored in a memory and executed by a processor to complete the present invention. One or more modules / units can be a series of computer program instruction segments that can perform specific functions, and the instruction segments are used to describe the execution process of the computer program in a terminal device. For example, the probability matrix generation module 30 is used to extract node features of input data according to a graph neural network, and generate a probability matrix for each edge according to the extracted node features.

[0114] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The terminal device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the schematic diagram is merely an example of a terminal device and does not constitute a limitation on the terminal device. The terminal device may include more or fewer components than shown in the diagram, or may combine certain components, or different components. For example, the terminal device may also include an input / output device, a network access device, a bus, etc.

[0115] The processor may be a central processing unit (CPU), 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. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device.

[0116] The memory can be used to store computer programs and / or modules. The processor realizes various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile terminal, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0117] Among them, if the module / unit integrated in the terminal device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electric carrier signals and telecommunication signals.

[0118] Accordingly, an embodiment of the present invention further provides a computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the telemetry and telesignaling data analysis method of any one of the above embodiments.

[0119] The above specific embodiments further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for analyzing telemetry and telesignaling data, characterized in that: include: Construct the power flow telemetry matrix and adjacency matrix according to the telemetry and telesignaling data to be analyzed; The power flow telemetry matrix, the adjacency matrix and the pre-constructed power grid graph model are input as input data into a graph neural network model, wherein the edge features of the power grid graph model include power flow telemetry data and telesignaling data indicating the connection status of the power grid; Extracting node features from the input data according to the graph neural network, and generating a probability matrix for each edge according to the extracted node features; The average value of the probability matrix of all edges is used as the total telesignaling error probability of the power grid system. When the total telesignaling error probability is lower than a preset threshold, the telemetry and telesignaling data to be analyzed are determined to be abnormal data.

2. The method for analyzing telemetry and telesignaling data according to claim 1, characterized in that: The telemetry and telesignaling data to be analyzed include telemetry data and telesignaling data, and constructing a power flow telemetry matrix and an adjacency matrix according to the telemetry and telesignaling data to be analyzed includes: constructing a power flow telemetry matrix based on the telemetry data; An adjacency matrix is ​​constructed based on the remote signaling data.

3. The method for analyzing telemetry and telesignaling data according to claim 1, characterized in that: Also includes: Determine a corresponding space vertex according to each device element in the power grid system; If there is an electrical connection relationship between the spatial vertices, the electrical connection relationship is used as an edge; Determining a corresponding time vertex according to each time stamp in the power grid system; A power grid graph model is constructed based on the spatial vertices, the edges and the time vertices, wherein the edges store telemetry data and telesignaling data corresponding to different time nodes.

4. The method for analyzing telemetry and telesignaling data according to claim 1, characterized in that: The extracting node features from the input data according to the graph neural network includes: Extracting node global features of the input data; Node features in the node global features are extracted in the row direction of the matrix and the column direction of the matrix.

5. The method for analyzing telemetry and telesignaling data according to claim 1, characterized in that: After extracting node features from the input data according to the graph neural network and generating a probability matrix for each edge according to the extracted node features, the method further includes: Calculate the loss value based on the generated probability matrix and the actual label probability matrix; Using a back-propagation algorithm, the network parameters of the graph neural network are updated according to the loss value.

6. A device for analyzing telemetry and telesignaling data, characterized in that: include: A matrix construction module is used to construct a power flow telemetry matrix and an adjacency matrix based on the telemetry and telesignaling data to be analyzed; A data input module, used to input the power flow telemetry matrix, the adjacency matrix and the pre-built power grid graph model as input data into a graph neural network model, wherein the edge features of the power grid graph model include power flow telemetry data and telesignaling data indicating the power grid connection status; A probability matrix generation module, used to extract node features from the input data according to the graph neural network, and generate a probability matrix for each edge according to the extracted node features; The data analysis module is used to take the average value of the probability matrix of all edges as the total telesignaling error probability of the power grid system, and when the total telesignaling error probability is lower than a preset threshold, determine that the telemetry and telesignaling data to be analyzed are abnormal data.

7. The telemetry and telesignaling data analysis device according to claim 6, characterized in that: The telemetry and telesignaling data to be analyzed include telemetry data and telesignaling data, and the matrix building module is further used for: constructing a power flow telemetry matrix based on the telemetry data; An adjacency matrix is ​​constructed based on the remote signaling data.

8. The telemetry and telesignaling data analysis device according to claim 6, characterized in that: The telemetry and telesignaling data analysis device further includes a power grid diagram model building module, which is used to: Determine a corresponding space vertex according to each device element in the power grid system; If there is an electrical connection relationship between the spatial vertices, the electrical connection relationship is used as an edge; Determining a corresponding time vertex according to each time stamp in the power grid system; A power grid graph model is constructed based on the spatial vertices, the edges and the time vertices, wherein the edges store telemetry data and telesignaling data corresponding to different time nodes.

9. A terminal device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the method for analyzing telemetry and telesignaling data as described in any one of claims 1 to 5 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program; wherein, when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for analyzing telemetry and telesignaling data according to any one of claims 1 to 5.

Images