Real-time monitoring method and device for data of electric power internet of things

Abstract

The invention discloses a real-time monitoring method and device for data of an electric power internet of things. According to the invention, three-phase voltage and current signals of power grid nodes are captured in real time by using a synchronous sampling and frequency tracking technology, time-frequency characteristics of the signals are obtained through generalized S transformation, and fault monitoring and rapid harmonic identification are realized. The power grid disturbance is analyzed in real time, and the working information is transmitted to the edge computing node. Edge nodes integrate information, and the risk is assessed through node quality calculation, data standardization, relation graph establishment and topology potential value. Monitoring and risk assessment are carried out synchronously, so that the state of a power grid is mastered in real time, potential risks are responded quickly, and the problem that the power internet of things cannot realize quick data monitoring in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the field of real-time monitoring of electric power Internet of Things data, and in particular to a method and device for real-time monitoring of electric power Internet of Things data. Background Art

[0002] The power Internet of Things is the product of the integration of smart grid and Internet of Things. The power Internet of Things provides effective support for the operation of basic grid services. Sensors, charging piles, smart substations, smart meters and other terminals in all links of the power system have achieved interconnection and interoperability, promoting the intelligent and integrated construction of the power grid. In order to improve the intelligence level of the power grid and the operating efficiency of the power system, strict requirements are put forward in terms of positioning, identification, detection, monitoring, and inspection.

[0003] Edge computing provides real-time computing and communication services to terminals by deploying edge servers near the terminals. Edge computing and cloud computing are complementary. Edge computing deploys computing and storage resources on data source servers close to the edge. Edge computing extends the cloud computing resources of the power communication network center to the edge network, enabling it to better support real-time data processing of the power information system to meet the different needs of the power system business and complete various grid distribution businesses.

[0004] With the continuous development of smart grid systems, intelligent equipment and wireless communication methods are constantly updated, and power automation and intelligence are gradually entering a new stage. The power system relies on distribution equipment for power distribution and power consumption activities. Due to the limitation of technical level, power grid failure is inevitable, and the "power quality" problem is also the focus of people's attention. Qualified power quality means that the power provided to sensitive equipment and the grounding system set are suitable for the normal operation of the equipment. However, in actual applications, there are a large number of high-power, nonlinear, and unbalanced loads, which cause a certain degree of pollution to the power grid, causing a series of power quality problems such as harmonics, interharmonics, three-phase imbalance, voltage fluctuations and flicker, voltage drops and short-term interruptions. How to effectively monitor the operation of the distribution network or the power outage and restoration has become the focus of improving the level of power supply services. At present, the data monitoring of the power Internet of Things is mainly achieved by converging all data into the cloud for calculation. With the expansion of the scale of the power grid and the increase in the number of terminals, the cloud computing model cannot effectively meet the needs of massive data transmission, calculation and storage collected by large-scale intelligent terminals, nor can it make decisions in the first time when the power grid encounters an emergency to reduce losses and restore the operation of the power grid. Summary of the invention

[0005] The present invention provides a method and device for real-time monitoring of electric power Internet of Things data, so as to solve the problem that the electric power Internet of Things cannot realize rapid data monitoring in the prior art.

[0006] In a first aspect, the present application provides a method for real-time monitoring of power Internet of Things data, comprising:

[0007] Obtain the three-phase voltage and current signals of each node in the power Internet of Things after preprocessing;

[0008] Uploading the three-phase voltage and current signals to an edge computing node;

[0009] According to the generalized S transform, the time-frequency characteristics of the three-phase voltage and current signals are extracted in the edge computing node;

[0010] Identify faults or disturbances in the power Internet of Things according to the time-frequency features, and input the time-frequency features into a pre-trained neural network so that the neural network outputs the disturbance type or fault mode in the power Internet of Things;

[0011] In the edge computing node, the topological potential value of each node in the disturbance type or fault mode area is calculated, and the risk level of the disturbance type or fault mode is determined according to the topological potential value.

[0012] This application can effectively identify faults or disturbances in the power Internet of Things by obtaining the pre-processed three-phase voltage and current signals of each node in the power Internet of Things, and uploading them to the edge computing node for generalized S transform to extract time-frequency features. Furthermore, these time-frequency features are input into a pre-trained neural network to accurately identify and classify the type of disturbance or fault mode in the power grid. In addition, the edge computing node calculates the topological potential value of each node in the fault or disturbance area, and determines the risk level based on these potential values. This application not only improves the speed and accuracy of data monitoring by the power Internet of Things, but also realizes real-time assessment and classification of power grid risks, reduces potential fault impacts, and solves the problem that the power Internet of Things in the prior art cannot realize rapid data monitoring.

[0013] As a preferred embodiment of the first aspect, the time-frequency features are input into a pre-trained neural network so that the neural network outputs a disturbance type or a fault mode in the power Internet of Things, specifically:

[0014] Acquire the time-frequency features; wherein the time-frequency features include the square of the time amplitude, the mean value and the characteristic frequency points;

[0015] Inputting the time-frequency features into a pre-trained fuzzy SOM neural network;

[0016] The fuzzy SOM neural network performs fuzzy clustering analysis according to the input time-frequency characteristics and outputs the disturbance type or fault mode in the power Internet of Things.

[0017] In this preferred embodiment, the present application inputs the pre-processed time-frequency features in the power Internet of Things, including the square of the time amplitude, the mean and the characteristic frequency points, into the pre-trained fuzzy SOM neural network. The present application utilizes the fuzzy clustering analysis capability of the neural network to accurately identify and classify different disturbance types or fault modes in the power grid. This deep analysis method based on time-frequency features improves the accuracy of identifying abnormal conditions in the power grid, because the fuzzy SOM neural network can handle the uncertainty and ambiguity in the time-frequency features, thereby outputting more accurate and reliable disturbance types or fault modes. This process not only enhances the real-time monitoring capability of the power grid status, but also provides faster and more accurate decision support for power grid operation and maintenance personnel through intelligent data analysis, effectively improving the operational safety and stability of the power grid.

[0018] As a preferred embodiment of the first aspect, in the edge computing node, the topological potential value of each node in the disturbance type or fault mode area is calculated, and the risk level of the disturbance type or fault mode is determined according to the topological potential value, specifically:

[0019] According to the actual location and interaction relationship of each node in each area of ​​the power Internet of Things, a topological map of each node in each area is constructed;

[0020] Determine the influence factor of each node in each area of ​​the topological graph according to minimizing the topological potential entropy;

[0021] Calculating the topological potential value of each node in each area of ​​the topological graph according to the quality of the impact factor;

[0022] The risk level of the disturbance type or the fault mode is determined according to the topological potential value of each node in the disturbance type or the fault mode area.

[0023] In this preferred embodiment, the present application conducts an in-depth analysis of the actual position and interaction relationship of the nodes in each region in the power Internet of Things, constructs a topological map, and realizes an accurate risk assessment of the disturbance type or fault mode area in the power grid. Specifically, the present application first simulates the complex network structure of the power grid by constructing a topological map reflecting the relationship between nodes. Then, the influence factor of each node is determined by minimizing the topological potential entropy. This step quantifies the importance of the node in the power grid and the degree of mutual influence. Then, based on the quality of these influence factors, the topological potential value of each node is calculated, and these potential values ​​reflect the stability and potential risks of the node in the power grid. Finally, the edge computing node combines these topological potential values ​​to evaluate the risk level of the disturbance type or fault mode, and realizes the quantitative analysis and hierarchical management of the potential risks of the power grid. This process not only improves the accuracy of power grid fault prediction, but also provides a scientific basis for preventive maintenance and emergency response of the power grid through quantitative risk assessment, thereby enhancing the overall stability and security of the power grid.

[0024] As a preferred embodiment of the first aspect, the calculating the topological potential value of each node in each area of ​​the topological graph according to the quality of the impact factor further includes:

[0025] The edge computing node integrates the topological potential values ​​of each node in each area of ​​the topological graph to obtain an integrated topological potential value;

[0026] Sending the integrated topological potential value to a cloud server, so that the cloud server locates the node with the highest risk among the nodes in each area in the topological map according to the received integrated topological potential value;

[0027] The cloud server also analyzes and locates the disturbance type or fault mode based on the disturbance type or fault mode in combination with the real-time operation data and historical fault database of the power Internet of Things.

[0028] In this preferred embodiment, the edge computing node of the present application calculates the topological potential value of each node by analyzing the interaction between nodes and the quality of the influencing factors, which provides a quantitative indicator for evaluating the risk of each node in the power grid. Then, these potential values ​​are integrated to form a comprehensive topological potential map, and further analyzed by the cloud server. The cloud server uses these integrated topological potential values, combined with real-time operation data and historical fault information, to conduct in-depth analysis and precise positioning of the disturbance type or fault mode in the power grid. This process not only improves the accuracy of risk assessment, but also through the powerful computing power of the cloud server, it realizes the rapid identification and response to potential problems in the power grid, thereby optimizing the operation and maintenance strategy of the power grid and enhancing the stability and security of the power grid.

[0029] As a preferred embodiment of the first aspect, according to the generalized S transform, the time-frequency characteristics of the three-phase voltage and current signals are extracted in the edge computing node; and the fault or disturbance in the power Internet of Things is identified according to the time-frequency characteristics, specifically:

[0030] Visualize the time-frequency features through a time-frequency graph, where different colors in the time-frequency graph represent different amplitudes;

[0031] When the amplitude of the frequency harmonic exceeds a preset threshold, it is marked with a warning color, and a fault or disturbance in the power Internet of Things is identified based on the warning color.

[0032] In this preferred embodiment, the present application converts complex time-frequency data into an intuitive graphical representation through the visualization of the time-frequency graph, wherein different colors in the time-frequency graph represent different amplitudes, which further enhances the intuitive understanding of signal characteristics and enables operation and maintenance personnel to quickly identify subtle changes in the state of the power grid. When the frequency harmonic amplitude in the signal exceeds the preset safety threshold, the system will automatically mark it with a warning color. This color change provides an immediate visual alert, allowing potential faults or disturbances to be immediately identified and noticed. This method not only reduces the burden of manual monitoring and reduces the risk of misjudgment and missed judgment, but also greatly improves the speed of fault location.

[0033] In a second aspect, the present application provides a real-time monitoring device for power Internet of Things data. The load balancing device includes an acquisition module, an upload module, an extraction module, an input and output module, and a calculation module;

[0034] The acquisition module is used to obtain the three-phase voltage and current signals of each node in the power Internet of Things after preprocessing;

[0035] The uploading module is used to upload the three-phase voltage and current signals to the edge computing node;

[0036] The extraction module is used to extract the time-frequency characteristics of the three-phase voltage and current signals in the edge computing node according to the generalized S transform;

[0037] The input-output module is used to identify the fault or disturbance in the power Internet of Things according to the time-frequency characteristics, and input the time-frequency characteristics into a pre-trained neural network so that the neural network outputs the disturbance type or fault mode in the power Internet of Things;

[0038] The calculation module is used to calculate the topological potential value of each node in the disturbance type or fault mode area in the edge computing node, and determine the risk level of the disturbance type or fault mode according to the topological potential value.

[0039] This device uses five modules to divide the work and coordinate work to better monitor the power Internet of Things data. This application obtains the three-phase voltage and current signals of each node in the power Internet of Things after preprocessing, and uploads them to the edge computing node for generalized S transform to extract time-frequency features, which can effectively identify faults or disturbances in the power Internet of Things. Furthermore, these time-frequency features are input into a pre-trained neural network to accurately identify and classify the type of disturbance or fault mode in the power grid. In addition, the edge computing node calculates the topological potential value of each node in the fault or disturbance area, and determines the risk level based on these potential values. This application not only improves the speed and accuracy of data monitoring by the power Internet of Things, but also realizes real-time assessment and classification of power grid risks, reduces potential fault impacts, and solves the problem that the power Internet of Things in the prior art cannot realize rapid data monitoring.

[0040] As a preferred embodiment of the second aspect, the time-frequency features are input into a pre-trained neural network so that the neural network outputs a disturbance type or a fault mode in the power Internet of Things, specifically:

[0041] Acquire the time-frequency features; wherein the time-frequency features include the square of the time amplitude, the mean value and the characteristic frequency points;

[0042] Inputting the time-frequency features into a pre-trained fuzzy SOM neural network;

[0043] The fuzzy SOM neural network performs fuzzy clustering analysis according to the input time-frequency characteristics and outputs the disturbance type or fault mode in the power Internet of Things.

[0044] In this preferred embodiment, the present application inputs the pre-processed time-frequency features in the power Internet of Things, including the square of the time amplitude, the mean and the characteristic frequency points, into the pre-trained fuzzy SOM neural network. The present application utilizes the fuzzy clustering analysis capability of the neural network to accurately identify and classify different disturbance types or fault modes in the power grid. This deep analysis method based on time-frequency features improves the accuracy of identifying abnormal conditions in the power grid, because the fuzzy SOM neural network can handle the uncertainty and ambiguity in the time-frequency features, thereby outputting more accurate and reliable disturbance types or fault modes. This process not only enhances the real-time monitoring capability of the power grid status, but also provides faster and more accurate decision support for power grid operation and maintenance personnel through intelligent data analysis, effectively improving the operational safety and stability of the power grid.

[0045] As a preferred embodiment of the second aspect, in the edge computing node, the topological potential value of each node in the disturbance type or fault mode area is calculated, and the risk level of the disturbance type or fault mode is determined according to the topological potential value, specifically:

[0046] According to the actual location and interaction relationship of each node in each area of ​​the power Internet of Things, a topological map of each node in each area is constructed;

[0047] Determine the influence factor of each node in each area of ​​the topological graph according to minimizing the topological potential entropy;

[0048] Calculating the topological potential value of each node in each area of ​​the topological graph according to the quality of the impact factor;

[0049] The risk level of the disturbance type or the fault mode is determined according to the topological potential value of each node in the disturbance type or the fault mode area.

[0050] In this preferred embodiment, the present application conducts an in-depth analysis of the actual position and interaction relationship of the nodes in each region in the power Internet of Things, constructs a topological map, and realizes an accurate risk assessment of the disturbance type or fault mode area in the power grid. Specifically, the present application first simulates the complex network structure of the power grid by constructing a topological map reflecting the relationship between nodes. Then, the influence factor of each node is determined by minimizing the topological potential entropy. This step quantifies the importance of the node in the power grid and the degree of mutual influence. Then, based on the quality of these influence factors, the topological potential value of each node is calculated, and these potential values ​​reflect the stability and potential risks of the node in the power grid. Finally, the edge computing node combines these topological potential values ​​to evaluate the risk level of the disturbance type or fault mode, and realizes the quantitative analysis and hierarchical management of the potential risks of the power grid. This process not only improves the accuracy of power grid fault prediction, but also provides a scientific basis for preventive maintenance and emergency response of the power grid through quantitative risk assessment, thereby enhancing the overall stability and security of the power grid.

[0051] As a preferred embodiment of the second aspect, the calculating the topological potential value of each node in each area of ​​the topological map according to the quality of the impact factor further includes:

[0052] The edge computing node integrates the topological potential values ​​of each node in each area of ​​the topological graph to obtain an integrated topological potential value;

[0053] Sending the integrated topological potential value to a cloud server, so that the cloud server locates the node with the highest risk among the nodes in each area in the topological map according to the received integrated topological potential value;

[0054] The cloud server also analyzes and locates the disturbance type or fault mode based on the disturbance type or fault mode in combination with the real-time operation data and historical fault database of the power Internet of Things.

[0055] In this preferred embodiment, the edge computing node of the present application calculates the topological potential value of each node by analyzing the interaction between nodes and the quality of the influencing factors, which provides a quantitative indicator for evaluating the risk of each node in the power grid. Then, these potential values ​​are integrated to form a comprehensive topological potential map, and further analyzed by the cloud server. The cloud server uses these integrated topological potential values, combined with real-time operation data and historical fault information, to conduct in-depth analysis and precise positioning of the disturbance type or fault mode in the power grid. This process not only improves the accuracy of risk assessment, but also through the powerful computing power of the cloud server, it realizes the rapid identification and response to potential problems in the power grid, thereby optimizing the operation and maintenance strategy of the power grid and enhancing the stability and security of the power grid.

[0056] As a preferred embodiment of the second aspect, according to the generalized S transform, the time-frequency characteristics of the three-phase voltage and current signals are extracted in the edge computing node; and the fault or disturbance in the power Internet of Things is identified according to the time-frequency characteristics, specifically:

[0057] Visualize the time-frequency features through a time-frequency graph, where different colors in the time-frequency graph represent different amplitudes;

[0058] When the amplitude of the frequency harmonic exceeds a preset threshold, it is marked with a warning color, and a fault or disturbance in the power Internet of Things is identified based on the warning color.

[0059] In this preferred embodiment, the present application converts complex time-frequency data into an intuitive graphical representation through the visualization of the time-frequency graph, wherein different colors in the time-frequency graph represent different amplitudes, which further enhances the intuitive understanding of signal characteristics and enables operation and maintenance personnel to quickly identify subtle changes in the state of the power grid. When the frequency harmonic amplitude in the signal exceeds the preset safety threshold, the system will automatically mark it with a warning color. This color change provides an immediate visual alert, allowing potential faults or disturbances to be immediately identified and noticed. This method not only reduces the burden of manual monitoring and reduces the risk of misjudgment and missed judgment, but also greatly improves the speed of fault location. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A flowchart of an embodiment of a method for real-time monitoring of power Internet of Things data provided by the present application;

[0061] Figure 2 Signal harmonic analysis flow chart provided for this application;

[0062] Figure 3 Node risk assessment flow chart provided for this application;

[0063] Figure 4A schematic diagram of the structure of an embodiment of the real-time monitoring device for power Internet of Things data provided in this application. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] Embodiment 1

[0066] Please refer to Figure 1 , which is a real-time monitoring method for power Internet of Things data provided by an embodiment of the present invention.

[0067] In this embodiment, the process of the real-time monitoring method of power Internet of Things data in this application is described in detail through steps S01-S05.

[0068] S01: Obtain the three-phase voltage and current signals of each node in the power Internet of Things after preprocessing;

[0069] More specifically, a node in each area of ​​the power Internet of Things is selected for monitoring. The node can be a working point or a device in the power grid. For a certain circuit in the power equipment or circuit system that needs to be monitored, a sensor is connected, and the three-phase voltage and current signals of the node are collected through synchronous sampling and frequency tracking technology, and the analog quantity is converted into a digital signal.

[0070] S02: Uploading the three-phase voltage and current signals to the edge computing node;

[0071] More specifically, edge computing nodes refer to business platforms built at the edge of the network. These platforms provide storage, computing, and network resources to sink some key business applications to the edge of the access network. These nodes are located between users and cloud centers. Compared with traditional cloud centers, they are closer to users (data sources) and call the trained model sent from the cloud server when performing disturbance analysis on the current node when the system finds a fault or the current node is at high risk during risk analysis.

[0072] S03: extracting the time-frequency characteristics of the three-phase voltage and current signals in the edge computing node according to the generalized S transform;

[0073] More specifically, the three-phase voltage and current signals are represented as discrete sequences a(p) (p = 0, T, 2T, ..., (N-1)T), where T is the sampling period, which is involved in synchronous sampling. Directly calculate the S transform S of the discrete a(p) T[m,n] is:

[0074]

[0075] Use the fast FT algorithm to directly calculate the S transform S F [m,n] is:

[0076]

[0077] Where: m,n=0,1,...,N-1; A(l) is the discrete Fourier transform of a(m). Discrete Fourier transform (DFT) is a transformation method in which Fourier transform is presented in discrete form in both time domain and frequency domain. It realizes the conversion of signal from time domain to frequency domain by transforming the sampling of time domain signal into sampling in discrete time Fourier transform (DTFT) frequency domain, which is convenient for studying the spectral structure and change law of signal. Therefore, the steps of discrete fast S transform can be obtained as follows:

[0078] 1) Calculate the first column S[jT,0] of the S matrix when n=0.

[0079] 2) Calculate the Fourier transform A[n / (NT)] of a[kT] when n=1, 2, ..., N-1.

[0080] 3) Expand A[n / (NT)] in the frequency domain to A[(n+m) / (NT)].

[0081] 4) Calculate the frequency domain Fourier transform of the Gaussian window function.

[0082] 5) Calculate the product of the frequency domain expanded dimension array A[(n+m) / (NT)] with m=0,1,...,N-1 and the Fourier transform of the frequency domain Gaussian window function.

[0083] 6) Perform a discrete inverse Fourier transform on the product.

[0084] 7) Repeat steps 5) and 6) until all frequency points are calculated.

[0085] After the signal has been transformed by S, it is easy to determine the type of fault in the power grid, including power grid short circuit, short circuit and impact fault. It can accurately detect faulty equipment in the power grid and actively report the fault to the edge computing node. If there is a disturbance signal in the power grid, the S transform can accurately extract the frequency of the disturbance signal. In the process of extracting the frequency characteristics of the disturbance signal, when the frequencies of the two disturbance signals are similar, frequency overlap will occur when the frequency characteristics are directly identified using the S transform two-dimensional contour map, which is not conducive to the extraction of frequency characteristics. In order to improve the time-frequency resolution of the signal, adjustable factors α and β are introduced in the S transform. The generalized S transform is defined as:

[0086]

[0087] α is the Gaussian window amplitude stretch factor; β is the frequency scale stretch factor. When and only when α=1, β=1, the generalized S transform is the S transform; when α>1 or β>1, the signal time resolution is improved while the frequency resolution is reduced; when α<1 or β<1, the signal frequency resolution is improved while the time resolution is reduced. By adjusting the values ​​of the two, the signal harmonic distribution can be more comprehensively reflected. In the subsequent result visualization process, the results of different factors can be presented in the same spectrum diagram. The Gaussian window amplitude stretch factor (α) is an important parameter introduced in the field of signal processing, especially in time-frequency analysis methods such as generalized S transform. It is mainly used to control the width change of the Gaussian window function in the time domain, thereby affecting the time resolution and frequency resolution in time-frequency analysis.

[0088] More specifically, after the S transform, a set of two-dimensional vectors is obtained, with the dimensions of time and frequency. For the time-frequency diagram, different colors are used to represent the amplitude. Therefore, if a certain frequency harmonic exceeds the preset value, its color is set to a warning color for warning. For the spectrum diagram, in the generalized S transform, the adjusted value can display the transformation results under different time resolutions and frequency resolutions, thereby largely avoiding the phenomenon of poor feature extraction caused by frequency overlap. Help decision makers locate abnormal harmonic frequencies.

[0089] S04: Identify faults or disturbances in the power Internet of Things according to the time-frequency features, and input the time-frequency features into a pre-trained neural network so that the neural network outputs the disturbance type or fault mode in the power Internet of Things;

[0090] More specifically, the time-frequency features are input into a pre-trained neural network so that the neural network outputs the disturbance type or fault mode in the power Internet of Things, specifically:

[0091] Acquire the time-frequency features; wherein the time-frequency features include the square of the time amplitude, the mean value and the characteristic frequency points;

[0092] Inputting the time-frequency features into a pre-trained fuzzy SOM neural network;

[0093] The fuzzy SOM neural network performs fuzzy clustering analysis according to the input time-frequency characteristics and outputs the disturbance type or fault mode in the power Internet of Things.

[0094] More specifically, the fuzzy SOM neural network can qualitatively and quantitatively classify the voltage multi-disturbance superposition situation and accurately reflect the real situation of the circuit disturbance. The fuzzy SOM neural network uses the following rules to determine the competition layer structure and fuzzy clustering rules respectively.

[0095] The SOM neural network consists of an input layer, a competitive layer, a fuzzy recognition layer, and an output layer. The determination of the competitive layer is very important for the recognition of the perturbation features of the input. Even if the number of times the network is trained is increased, the classification performance of the network cannot be completely changed.

[0096] Specifically, there are two empirical formulas for determining the number of competitive layer structures:

[0097] 1) p 1 ×p 2 >k, where k is the total number of sample categories, p 1 is the number of row units in the competition layer, p 2 is the number of cells in the competition layer.

[0098] 2) or Where m is the number of output neurons, n is the number of input units, and z is a constant between [1,10].

[0099] Optionally, an exhaustive number of structures is constructed, the perturbation sample data of the test array is performed on the competing structures in all cases, and the structure with the best performance is selected.

[0100] For fuzzy clustering rules, the fuzzy SOM neural network needs to perform perturbation classification on the perturbation data. Assume that the ideal output of the SOM network is G, and the actual output of the test data is g. When the distance between the two outputs is the smallest, the perturbation classification result can be output, that is:

[0101]

[0102] Among them, g i is the clustering result closest to the ideal output G. N is the total number of clusters, which refers to the total number of cluster centers (or clusters, groups) defined or generated during the clustering analysis process. In clustering algorithms, data points are grouped into different clusters in order to reveal the inherent structure or pattern in the data. Each cluster consists of a group of data points that are more similar than data points in other clusters under certain metrics (such as distance, similarity, etc.).

[0103] For the superposition of two adjacent disturbances, the primary and secondary disturbances can be determined by the minimum distance, that is:

[0104]

[0105] For more interference situations, clustering rules can be determined similarly.

[0106] The edge service node collects data, cleans, filters, and labels the data, and uploads it to the cloud server for model training. After the training is completed, the model parameters are sent to the edge computing node. During the application process, the cloud server still trains and updates the model parameters, and regularly sends new parameters for use.

[0107] Edge computing nodes focus on real-time data processing and local computing, while edge service nodes focus more on data collection, preprocessing and uploading, as well as the distribution of model parameters and system updates. The two usually work together in application scenarios such as the Internet of Things and edge computing to achieve efficient and real-time data processing and service provision.

[0108] According to the actual scenario, the fuzzy SOM neural network can determine the types of disturbances existing in the circuit and the percentage of disturbances they respectively occupy, helping decision makers locate the source of the problem more quickly.

[0109] The time-frequency diagram and spectrum diagram after the generalized S transform help locate abnormal time-frequency points, and the disturbance analysis results help locate the source of the problem. Combining the two, real-time analysis and monitoring of circuit harmonics can be achieved, and decision makers can control the power grid, eliminate the source of the problem, and ensure the safety of the power grid.

[0110] This application uses the fuzzy clustering analysis capability of the neural network to accurately identify and classify different disturbance types or fault modes in the power grid by inputting the pre-processed time-frequency features in the power Internet of Things, including the square of the time amplitude, the mean, and the characteristic frequency points. This application improves the recognition accuracy of abnormal conditions in the power grid, because the fuzzy SOM neural network can handle the uncertainty and ambiguity in the time-frequency features, thereby outputting more accurate and reliable disturbance types or fault modes. This process not only enhances the real-time monitoring capability of the power grid status, but also provides faster and more accurate decision support for power grid operation and maintenance personnel through intelligent data analysis, effectively improving the operational safety and stability of the power grid.

[0111] Specifically, the process of using the generalized S transform in the edge node and inputting the time series features after the generalized S transform into the fuzzy SOM neural network to analyze the disturbance, and using the time-frequency diagram and percentage diagram to analyze the results is as follows: Figure 2 shown.

[0112] S05: In the edge computing node, a topological potential value of each node in the disturbance type or fault mode area is calculated, and a risk level of the disturbance type or fault mode is determined according to the topological potential value.

[0113] More specifically, in the edge computing node, the topological potential value of each node in the disturbance type or fault mode area is calculated, and the risk level of the disturbance type or fault mode is determined according to the topological potential value, such as Figure 3 As shown, specifically:

[0114] According to the actual location and interaction relationship of each node in each area of ​​the power Internet of Things, a topological map of each node in each area is constructed;

[0115] Specifically, according to the location and interaction relationship of each node in the actual power grid, the relationship between nodes is abstracted into a topological graph to obtain a graph G, that is, G = {V, E} represents the power system node risk graph, E is the set of edges, V = {v 1 ,v 2 ,…,v n} is a graph node.

[0116] Determine the influence factor of each node in each area of ​​the topological graph according to minimizing the topological potential entropy;

[0117] Specifically, the topological potential field node is an important concept in the network topology structure. It combines the ideas of data field and network topology to describe the interaction between network nodes and their position and importance in the network. The interaction between topological potential field nodes is expressed as:

[0118]

[0119] in, is the inter-node effect; m i is the node quality, which can be represented by the working status of each node. The healthier the working status of the node, the higher the quality. The description process of the present invention adopts the difference between the actual mathematical expectation and the normal expectation of the node working voltage or other parameters, and the variance to jointly represent it; is the path length between nodes, which can be the actual distance between nodes, or the distance between edge nodes in each area, and can also be represented by the number of hops between nodes; σ is the node influence factor. Topological potential is a concept that combines topology in mathematics and field theory in physics. It is mainly used to describe the interaction between nodes in complex networks and their position and importance in the network.

[0120] The topological potential of different nodes can be expressed as follows:

[0121]

[0122] The topological potential entropy H of the graph is:

[0123]

[0124] The more concentrated the number of nodes, the greater the influence of the node; the smaller the distance between nodes, the stronger their correlation and the greater the mutual influence between nodes.

[0125] To calculate the topological potential value of each node, it is necessary to first determine the node influence factor σ and obtain the influence factor by minimizing the potential entropy. Specifically, by traversal or other search algorithms, find the influence factor value corresponding to the minimum topological potential entropy as the optimal σ value.

[0126] Calculating the topological potential value of each node in each area of ​​the topological graph according to the quality of the impact factor;

[0127] The risk level of the disturbance type or the fault mode is determined according to the topological potential value of each node in the disturbance type or the fault mode area.

[0128] Specifically, after obtaining the optimal σ value, the topological potential value of each node is calculated according to formula (7). The larger the topological potential value, the greater the risk of the node. It is necessary to pay special attention to the nodes with higher risks or rectify them to avoid failures.

[0129] More specifically, the calculating the topological potential value of each node in each area in the topological graph according to the quality of the influencing factor also includes:

[0130] The edge computing node integrates the topological potential values ​​of each node in each area of ​​the topological graph to obtain an integrated topological potential value;

[0131] Sending the integrated topological potential value to a cloud server, so that the cloud server locates the node with the highest risk among the nodes in each area in the topological map according to the received integrated topological potential value;

[0132] Specifically, the edge node is responsible for the data integration and calculation of several nodes in a region. Several regions serve as new nodes to form the next level of regions, until the next level of the current region is the cloud server. For any region, if the data collected by the node is multi-source heterogeneous, the data needs to be integrated and normalized before it can be sent to the next level for the next step of calculation.

[0133] When testing multi-source heterogeneous data, multiple sensors need to be set between each parallel device group. Sensor data collection and management are very difficult. It is necessary to integrate the sensor differential data of each parallel device group to improve the efficiency of synchronous testing. Assume that the number of sensors is n and the sensor data set is {d 1 ,d 2 ,…,d n}, the sensor acquisition time is t, based on the data fusion theory of probability, {λ 1 (t),λ 2(t),…,λ n (t)} is the collection data set. The sensor collection accuracy is improved through weighted calculation. The weighted calculation result is:

[0134]

[0135] Among them, ω i For sensor weighting, it is important to mention that due to the rich types of nodes, the scale of data collected by sensors may be inconsistent, and mathematical methods are needed to determine the weighting values ​​to scale the data to the same scale for subsequent calculations.

[0136] Based on the theory of multivariate probability, which is an important branch of probability theory, it mainly studies the probability distribution and properties of multiple random variables changing simultaneously. 1 (t),λ 2 (t),…,λ n (t)} should satisfy the n-dimensional normal distribution, and its distribution function is:

[0137]

[0138] A combined sensor is a sensor composed of different single conversion devices. By integrating multiple conversion devices, this sensor can measure multiple physical quantities or integrate multiple functions, thereby improving the performance and application range of the sensor. The data set {λ 1 (t),λ 2 (t),…,λ n (t)} Available weighted results The distribution density function of is expressed as:

[0139]

[0140] After completing sensor data fusion, the mathematical expectation value of multi-source heterogeneous data is:

[0141]

[0142] That is, the weighted sum of the mathematical expectations of each sensor.

[0143] The standard deviation is:

[0144]

[0145] Furthermore, the cloud server also analyzes and locates the disturbance type or fault mode based on the disturbance type or fault mode in combination with the real-time operation data and historical fault database of the power Internet of Things.

[0146] The edge computing nodes of this application calculate the topological potential value of each node by analyzing the interaction between nodes and the quality of influencing factors, which provides a quantitative indicator for assessing the risk of each node in the power grid. Then, these potential values ​​are integrated to form a comprehensive topological potential map, which is further analyzed by the cloud server. The cloud server uses these integrated topological potential values, combined with real-time operation data and historical fault information, to conduct in-depth analysis and precise positioning of the disturbance type or fault mode in the power grid. This process not only improves the accuracy of risk assessment, but also through the powerful computing power of the cloud server, it realizes the rapid identification and response to potential problems in the power grid, thereby optimizing the operation and maintenance strategy of the power grid and enhancing the stability and security of the power grid.

[0147] More specifically, the topological potential values ​​of all nodes in the power Internet of Things are sent up by the edge nodes one level at a time and summarized to the cloud server. The cloud server proposes rectification suggestions for the higher-risk areas below, and finds the higher-risk areas or nodes under the corresponding areas, so that the risk nodes can be quickly located. If a power grid fails, the located risk node is the node with the highest probability of failure. The generalized S transform result and disturbance analysis result of the edge computing node corresponding to the node are called to analyze its fault type. If the node does not have a fault, the node with the second highest fault probability is found to perform the same operation. This method can quickly find high-risk nodes, thereby avoiding power grid failures, and can achieve rapid fault location and fault analysis after the failure occurs.

[0148] This application can effectively identify faults or disturbances in the power Internet of Things by obtaining the pre-processed three-phase voltage and current signals of each node in the power Internet of Things, and uploading them to the edge computing node for generalized S transform to extract time-frequency features. Furthermore, these time-frequency features are input into a pre-trained neural network to accurately identify and classify the type of disturbance or fault mode in the power grid. In addition, the edge computing node calculates the topological potential value of each node in the fault or disturbance area, and determines the risk level based on these potential values. This application not only improves the speed and accuracy of data monitoring by the power Internet of Things, but also realizes real-time assessment and classification of power grid risks, reduces potential fault impacts, and solves the problem that the power Internet of Things in the prior art cannot realize rapid data monitoring.

[0149] Embodiment 2

[0150] Please refer to Figure 4 , which is a real-time monitoring device for electric power Internet of Things data provided in an embodiment of the present application.

[0151] In this embodiment, the real-time monitoring device for power Internet of Things data includes an acquisition module 10, an upload module 20, an extraction module 30, an input and output module 40 and a calculation module 50.

[0152] The acquisition module 10 is used to obtain the three-phase voltage and current signals of each node in the power Internet of Things after preprocessing;

[0153] More specifically, a node in each area of ​​the power Internet of Things is selected for monitoring. The node can be a working point or a device in the power grid. For a certain circuit in the power equipment or circuit system that needs to be monitored, a sensor is connected, and the three-phase voltage and current signals of the node are collected through synchronous sampling and frequency tracking technology, and the analog quantity is converted into a digital signal.

[0154] The uploading module 20 is used to upload the three-phase voltage and current signals to the edge computing node;

[0155] More specifically, edge computing nodes refer to business platforms built at the edge of the network. These platforms provide storage, computing, and network resources to sink some key business applications to the edge of the access network. These nodes are located between users and cloud centers. Compared with traditional cloud centers, they are closer to users (data sources) and call the trained model sent from the cloud server when performing disturbance analysis on the current node when the system finds a fault or the current node is at high risk during risk analysis.

[0156] The extraction module 30 is used to extract the time-frequency characteristics of the three-phase voltage and current signals in the edge computing node according to the generalized S transform;

[0157] More specifically, the three-phase voltage and current signals are represented as discrete sequences a(p) (p = 0, T, 2T, ..., (N-1)T), where T is the sampling period, which is involved in synchronous sampling. Directly calculate the S transform S of the discrete a(p) T [m,n] is:

[0158]

[0159] Use the fast FT algorithm to directly calculate the S transform S F [m,n] is:

[0160]

[0161] Where: m,n=0,1,...,N-1; A(l) is the discrete Fourier transform of a(m). Discrete Fourier transform (DFT) is a transformation method in which Fourier transform is presented in discrete form in both time domain and frequency domain. It realizes the conversion of signal from time domain to frequency domain by transforming the sampling of time domain signal into sampling in discrete time Fourier transform (DTFT) frequency domain, which is convenient for studying the spectral structure and change law of signal. Therefore, the steps of discrete fast S transform can be obtained as follows:

[0162] 1) Calculate the first column S[jT,0] of the S matrix when n=0.

[0163] 2) Calculate the Fourier transform A[n / (NT)] of a[kT] when n=1, 2, ..., N-1.

[0164] 3) Expand A[n / (NT)] in the frequency domain to A[(n+m) / (NT)].

[0165] 4) Calculate the frequency domain Fourier transform of the Gaussian window function.

[0166] 5) Calculate the product of the frequency domain expanded dimension array A[(n+m) / (NT)] with m=0,1,...,N-1 and the Fourier transform of the frequency domain Gaussian window function.

[0167] 6) Perform a discrete inverse Fourier transform on the product.

[0168] 7) Repeat steps 5) and 6) until all frequency points are calculated.

[0169] After the signal has been transformed by S, it is easy to determine the type of fault in the power grid, including power grid short circuit, short circuit and impact fault. It can accurately detect faulty equipment in the power grid and actively report the fault to the edge computing node. If there is a disturbance signal in the power grid, the S transform can accurately extract the frequency of the disturbance signal. In the process of extracting the frequency characteristics of the disturbance signal, when the frequencies of the two disturbance signals are similar, frequency overlap will occur when the frequency characteristics are directly identified using the S transform two-dimensional contour map, which is not conducive to the extraction of frequency characteristics. In order to improve the time-frequency resolution of the signal, adjustable factors α and β are introduced in the S transform. The generalized S transform is defined as:

[0170]

[0171] α is the Gaussian window amplitude stretch factor; β is the frequency scale stretch factor. When and only when α=1, β=1, the generalized S transform is the S transform; when α>1 or β>1, the signal time resolution is improved while the frequency resolution is reduced; when α<1 or β<1, the signal frequency resolution is improved while the time resolution is reduced. By adjusting the values ​​of the two, the signal harmonic distribution can be more comprehensively reflected. In the subsequent result visualization process, the results of different factors can be presented in the same spectrum diagram. The Gaussian window amplitude stretch factor (α) is an important parameter introduced in the field of signal processing, especially in time-frequency analysis methods such as generalized S transform. It is mainly used to control the width change of the Gaussian window function in the time domain, thereby affecting the time resolution and frequency resolution in time-frequency analysis.

[0172] More specifically, after the S transform, a set of two-dimensional vectors is obtained, with the dimensions of time and frequency. For the time-frequency diagram, different colors are used to represent the amplitude. Therefore, if a certain frequency harmonic exceeds the preset value, its color is set to a warning color for warning. For the spectrum diagram, in the generalized S transform, the adjusted value can display the transformation results under different time resolutions and frequency resolutions, thereby largely avoiding the phenomenon of poor feature extraction caused by frequency overlap. Help decision makers locate abnormal harmonic frequencies.

[0173] The input-output module 40 is used to identify the fault or disturbance in the power Internet of Things according to the time-frequency characteristics, and input the time-frequency characteristics into a pre-trained neural network so that the neural network outputs the disturbance type or fault mode in the power Internet of Things;

[0174] More specifically, the time-frequency features are input into a pre-trained neural network so that the neural network outputs the disturbance type or fault mode in the power Internet of Things, specifically:

[0175] Acquire the time-frequency features; wherein the time-frequency features include the square of the time amplitude, the mean value and the characteristic frequency points;

[0176] Inputting the time-frequency features into a pre-trained fuzzy SOM neural network;

[0177] The fuzzy SOM neural network performs fuzzy clustering analysis according to the input time-frequency characteristics and outputs the disturbance type or fault mode in the power Internet of Things.

[0178] More specifically, the fuzzy SOM neural network can qualitatively and quantitatively classify the voltage multi-disturbance superposition situation and accurately reflect the real situation of the circuit disturbance. The fuzzy SOM neural network uses the following rules to determine the competition layer structure and fuzzy clustering rules respectively.

[0179] The SOM neural network consists of an input layer, a competitive layer, a fuzzy recognition layer, and an output layer. The determination of the competitive layer is very important for the recognition of the perturbation features of the input. Even if the number of times the network is trained is increased, the classification performance of the network cannot be completely changed.

[0180] Specifically, there are two empirical formulas for determining the number of competitive layer structures:

[0181] 1) p 1 ×p 2 >k, where k is the total number of sample categories, p 1 is the number of row units in the competition layer, p 2 is the number of cells in the competition layer.

[0182] 2) or Where m is the number of output neurons, n is the number of input units, and z is a constant between [1,10].

[0183] Optionally, an exhaustive number of structures is constructed, the perturbation sample data of the test array is performed on the competing structures in all cases, and the structure with the best performance is selected.

[0184] For fuzzy clustering rules, the fuzzy SOM neural network needs to perform perturbation classification on the perturbation data. Assume that the ideal output of the SOM network is G, and the actual output of the test data is g. When the distance between the two outputs is the smallest, the perturbation classification result can be output, that is:

[0185]

[0186] Among them, g i is the clustering result closest to the ideal output G. N is the total number of clusters, which refers to the total number of cluster centers (or clusters, groups) defined or generated during the clustering analysis process. In clustering algorithms, data points are grouped into different clusters in order to reveal the inherent structure or pattern in the data. Each cluster consists of a group of data points that are more similar than data points in other clusters under certain metrics (such as distance, similarity, etc.).

[0187] For the superposition of two adjacent disturbances, the primary and secondary disturbances can be determined by the minimum distance, that is:

[0188]

[0189] For more interference situations, clustering rules can be determined similarly.

[0190] The edge service node collects data, cleans, filters, and labels the data, and uploads it to the cloud server for model training. After the training is completed, the model parameters are sent to the edge computing node. During the application process, the cloud server still trains and updates the model parameters, and regularly sends new parameters for use.

[0191] Edge computing nodes focus on real-time data processing and local computing, while edge service nodes focus more on data collection, preprocessing and uploading, as well as the distribution of model parameters and system updates. The two usually work together in application scenarios such as the Internet of Things and edge computing to achieve efficient and real-time data processing and service provision.

[0192] According to the actual scenario, the fuzzy SOM neural network can determine the types of disturbances existing in the circuit and the percentage of disturbances they respectively occupy, helping decision makers locate the source of the problem more quickly.

[0193] The time-frequency diagram and spectrum diagram after the generalized S transform help locate abnormal time-frequency points, and the disturbance analysis results help locate the source of the problem. Combining the two, real-time analysis and monitoring of circuit harmonics can be achieved, and decision makers can control the power grid, eliminate the source of the problem, and ensure the safety of the power grid.

[0194] This application uses the fuzzy clustering analysis capability of the neural network to accurately identify and classify different disturbance types or fault modes in the power grid by inputting the pre-processed time-frequency features in the power Internet of Things, including the square of the time amplitude, the mean, and the characteristic frequency points. This application improves the recognition accuracy of abnormal conditions in the power grid, because the fuzzy SOM neural network can handle the uncertainty and ambiguity in the time-frequency features, thereby outputting more accurate and reliable disturbance types or fault modes. This process not only enhances the real-time monitoring capability of the power grid status, but also provides faster and more accurate decision support for power grid operation and maintenance personnel through intelligent data analysis, effectively improving the operational safety and stability of the power grid.

[0195] Specifically, the process of using the generalized S transform in the edge node and inputting the time series features after the generalized S transform into the fuzzy SOM neural network to analyze the disturbance, and using the time-frequency diagram and percentage diagram to analyze the results is as follows: Figure 2 shown.

[0196] The calculation module 50 is used to calculate the topological potential value of each node in the disturbance type or fault mode area in the edge computing node, and determine the risk level of the disturbance type or fault mode according to the topological potential value.

[0197] More specifically, in the edge computing node, the topological potential value of each node in the disturbance type or fault mode area is calculated, and the risk level of the disturbance type or fault mode is determined according to the topological potential value, such as Figure 3 As shown, specifically:

[0198] According to the actual location and interaction relationship of each node in each area of ​​the power Internet of Things, a topological map of each node in each area is constructed;

[0199] Specifically, according to the location and interaction relationship of each node in the actual power grid, the relationship between nodes is abstracted into a topological graph to obtain a graph G, that is, G = {V, E} represents the power system node risk graph, E is the set of edges, V = {v 1 ,v 2 ,…,v n} is a graph node.

[0200] Determine the influence factor of each node in each area of ​​the topological graph according to minimizing the topological potential entropy;

[0201] Specifically, the topological potential field node is an important concept in the network topology structure. It combines the ideas of data field and network topology to describe the interaction between network nodes and their position and importance in the network. The interaction between topological potential field nodes is expressed as:

[0202]

[0203] in, is the inter-node effect; m i is the node quality, which can be represented by the working status of each node. The healthier the working status of the node, the higher the quality. The description process of the present invention adopts the difference between the actual mathematical expectation and the normal expectation of the node working voltage or other parameters, and the variance to jointly represent it; is the path length between nodes, which can be the actual distance between nodes, or the distance between edge nodes in each area, and can also be represented by the number of hops between nodes; σ is the node influence factor. Topological potential is a concept that combines topology in mathematics and field theory in physics. It is mainly used to describe the interaction between nodes in complex networks and their position and importance in the network.

[0204] The topological potential of different nodes can be expressed as follows:

[0205]

[0206] The topological potential entropy H of the graph is:

[0207]

[0208] The more concentrated the number of nodes, the greater the influence of the node; the smaller the distance between nodes, the stronger their correlation and the greater the mutual influence between nodes.

[0209] To calculate the topological potential value of each node, it is necessary to first determine the node influence factor σ and obtain the influence factor by minimizing the potential entropy. Specifically, by traversal or other search algorithms, find the influence factor value corresponding to the minimum topological potential entropy as the optimal σ value.

[0210] Calculating the topological potential value of each node in each area of ​​the topological graph according to the quality of the impact factor;

[0211] The risk level of the disturbance type or the fault mode is determined according to the topological potential value of each node in the disturbance type or the fault mode area.

[0212] Specifically, after obtaining the optimal σ value, the topological potential value of each node is calculated according to formula (7). The larger the topological potential value, the greater the risk of the node. It is necessary to pay special attention to the nodes with higher risks or rectify them to avoid failures.

[0213] More specifically, the calculating the topological potential value of each node in each area in the topological graph according to the quality of the influencing factor also includes:

[0214] The edge computing node integrates the topological potential values ​​of each node in each area of ​​the topological graph to obtain an integrated topological potential value;

[0215] Sending the integrated topological potential value to a cloud server, so that the cloud server locates the node with the highest risk among the nodes in each area in the topological map according to the received integrated topological potential value;

[0216] Specifically, the edge node is responsible for the data integration and calculation of several nodes in a region. Several regions serve as new nodes to form the next level of regions, until the next level of the current region is the cloud server. For any region, if the data collected by the node is multi-source heterogeneous, the data needs to be integrated and normalized before it can be sent to the next level for the next step of calculation.

[0217] When testing multi-source heterogeneous data, multiple sensors need to be set between each parallel device group. Sensor data collection and management are very difficult. It is necessary to integrate the sensor differential data of each parallel device group to improve the efficiency of synchronous testing. Assume that the number of sensors is n and the sensor data set is {d 1 ,d 2 ,…,d n}, the sensor acquisition time is t, based on the data fusion theory of probability, {λ 1 (t),λ 2 (t),…,λ n (t)} is the collection data set. The sensor collection accuracy is improved through weighted calculation. The weighted calculation result is:

[0218]

[0219] Among them, ω i For sensor weighting, it is important to mention that due to the rich types of nodes, the scale of data collected by sensors may be inconsistent, and mathematical methods are needed to determine the weighting values ​​to scale the data to the same scale for subsequent calculations.

[0220] Based on the theory of multivariate probability, which is an important branch of probability theory, it mainly studies the probability distribution and properties of multiple random variables changing simultaneously. 1 (t),λ 2 (t),…,λ n (t)} should satisfy the n-dimensional normal distribution, and its distribution function is:

[0221]

[0222] A combined sensor is a sensor composed of different single conversion devices. By integrating multiple conversion devices, this sensor can measure multiple physical quantities or integrate multiple functions, thereby improving the performance and application range of the sensor. The data set {λ 1 (t),λ 2 (t),…,λ n (t)} Available weighted results The distribution density function of is expressed as:

[0223]

[0224] After completing sensor data fusion, the mathematical expectation value of multi-source heterogeneous data is:

[0225]

[0226] That is, the weighted sum of the mathematical expectations of each sensor.

[0227] The standard deviation is:

[0228]

[0229] Furthermore, the cloud server also analyzes and locates the disturbance type or fault mode based on the disturbance type or fault mode in combination with the real-time operation data and historical fault database of the power Internet of Things.

[0230] The edge computing nodes of this application calculate the topological potential value of each node by analyzing the interaction between nodes and the quality of influencing factors, which provides a quantitative indicator for assessing the risk of each node in the power grid. Then, these potential values ​​are integrated to form a comprehensive topological potential map, which is further analyzed by the cloud server. The cloud server uses these integrated topological potential values, combined with real-time operation data and historical fault information, to conduct in-depth analysis and precise positioning of the disturbance type or fault mode in the power grid. This process not only improves the accuracy of risk assessment, but also through the powerful computing power of the cloud server, it realizes the rapid identification and response to potential problems in the power grid, thereby optimizing the operation and maintenance strategy of the power grid and enhancing the stability and security of the power grid.

[0231] More specifically, the topological potential values ​​of all nodes in the power Internet of Things are sent up by the edge nodes one level at a time and summarized to the cloud server. The cloud server proposes rectification suggestions for the higher-risk areas below, and finds the higher-risk areas or nodes under the corresponding areas, so that the risk nodes can be quickly located. If a power grid fails, the located risk node is the node with the highest probability of failure. The generalized S transform result and disturbance analysis result of the edge computing node corresponding to the node are called to analyze its fault type. If the node does not have a fault, the node with the second highest fault probability is found to perform the same operation. This method can quickly find high-risk nodes, thereby avoiding power grid failures, and can achieve rapid fault location and fault analysis after the failure occurs.

[0232] This device uses five modules to divide the work and coordinate work to better monitor the power Internet of Things data. This application obtains the three-phase voltage and current signals of each node in the power Internet of Things after preprocessing, and uploads them to the edge computing node for generalized S transform to extract time-frequency features, which can effectively identify faults or disturbances in the power Internet of Things. Furthermore, these time-frequency features are input into a pre-trained neural network to accurately identify and classify the type of disturbance or fault mode in the power grid. In addition, the edge computing node calculates the topological potential value of each node in the fault or disturbance area, and determines the risk level based on these potential values. This application not only improves the speed and accuracy of data monitoring by the power Internet of Things, but also realizes real-time assessment and classification of power grid risks, reduces potential fault impacts, and solves the problem that the power Internet of Things in the prior art cannot realize rapid data monitoring.

[0233] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is 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 real-time monitoring of power Internet of Things data, characterized in that: include: Obtain the three-phase voltage and current signals of each node in the power Internet of Things after preprocessing; Uploading the three-phase voltage and current signals to an edge computing node; According to the generalized S transform, the time-frequency characteristics of the three-phase voltage and current signals are extracted in the edge computing node; Identify faults or disturbances in the power Internet of Things according to the time-frequency features, and input the time-frequency features into a pre-trained neural network so that the neural network outputs the disturbance type or fault mode in the power Internet of Things; In the edge computing node, the topological potential value of each node in the disturbance type or fault mode area is calculated, and the risk level of the disturbance type or fault mode is determined according to the topological potential value.

2. The method for real-time monitoring of electric power Internet of Things data according to claim 1 is characterized in that: The time-frequency features are input into a pre-trained neural network so that the neural network outputs the disturbance type or fault mode in the power Internet of Things, specifically: Acquire the time-frequency features; wherein the time-frequency features include the square of the time amplitude, the mean value and the characteristic frequency points; Inputting the time-frequency features into a pre-trained fuzzy SOM neural network; The fuzzy SOM neural network performs fuzzy clustering analysis according to the input time-frequency characteristics and outputs the disturbance type or fault mode in the power Internet of Things.

3. The method for real-time monitoring of electric power Internet of Things data according to claim 1 is characterized in that: In the edge computing node, the topological potential value of each node in the disturbance type or fault mode area is calculated, and the risk level of the disturbance type or fault mode is determined according to the topological potential value, specifically: According to the actual location and interaction relationship of each node in each area of ​​the power Internet of Things, a topological map of each node in each area is constructed; Determine the influence factor of each node in each area of ​​the topological graph according to minimizing the topological potential entropy; Calculating the topological potential value of each node in each area of ​​the topological graph according to the quality of the impact factor; The risk level of the disturbance type or the fault mode is determined according to the topological potential value of each node in the disturbance type or the fault mode area.

4. The method for real-time monitoring of electric power Internet of Things data according to claim 3 is characterized in that: The step of calculating the topological potential value of each node in each area of ​​the topological graph according to the quality of the influencing factor further includes: The edge computing node integrates the topological potential values ​​of each node in each area of ​​the topological graph to obtain an integrated topological potential value; Sending the integrated topological potential value to a cloud server, so that the cloud server locates the node with the highest risk among the nodes in each area in the topological map according to the received integrated topological potential value; The cloud server also analyzes and locates the disturbance type or fault mode based on the disturbance type or fault mode in combination with the real-time operation data and historical fault database of the power Internet of Things.

5. The method for real-time monitoring of electric power Internet of Things data according to claim 1 is characterized in that: The extracting the time-frequency characteristics of the three-phase voltage and current signals in the edge computing node according to the generalized S transform; and identifying the fault or disturbance in the power Internet of Things according to the time-frequency characteristics, specifically: Visualize the time-frequency features through a time-frequency graph, where different colors in the time-frequency graph represent different amplitudes; When the amplitude of the frequency harmonic exceeds a preset threshold, it is marked with a warning color, and a fault or disturbance in the power Internet of Things is identified based on the warning color.

6. A real-time monitoring device for power Internet of Things data, characterized in that: It includes an acquisition module, an upload module, an extraction module, an input and output module, and a calculation module; The acquisition module is used to obtain the three-phase voltage and current signals of each node in the power Internet of Things after preprocessing; The uploading module is used to upload the three-phase voltage and current signals to the edge computing node; The extraction module is used to extract the time-frequency characteristics of the three-phase voltage and current signals in the edge computing node according to the generalized S transform; The input-output module is used to identify the fault or disturbance in the power Internet of Things according to the time-frequency characteristics, and input the time-frequency characteristics into a pre-trained neural network so that the neural network outputs the disturbance type or fault mode in the power Internet of Things; The calculation module is used to calculate the topological potential value of each node in the disturbance type or fault mode area in the edge computing node, and determine the risk level of the disturbance type or fault mode according to the topological potential value.

7. The real-time monitoring device for electric power Internet of Things data according to claim 6 is characterized in that: The time-frequency features are input into a pre-trained neural network so that the neural network outputs the disturbance type or fault mode in the power Internet of Things, specifically: Acquire the time-frequency features; wherein the time-frequency features include the square of the time amplitude, the mean value and the characteristic frequency points; Inputting the time-frequency features into a pre-trained fuzzy SOM neural network; The fuzzy SOM neural network performs fuzzy clustering analysis according to the input time-frequency characteristics and outputs the disturbance type or fault mode in the power Internet of Things.

8. The real-time monitoring device for electric power Internet of Things data according to claim 6 is characterized in that: In the edge computing node, the topological potential value of each node in the disturbance type or fault mode area is calculated, and the risk level of the disturbance type or fault mode is determined according to the topological potential value, specifically: According to the actual location and interaction relationship of each node in each area of ​​the power Internet of Things, a topological map of each node in each area is constructed; Determine the influence factor of each node in each area of ​​the topological graph according to minimizing the topological potential entropy; Calculating the topological potential value of each node in each area of ​​the topological graph according to the quality of the impact factor; The risk level of the disturbance type or the fault mode is determined according to the topological potential value of each node in the disturbance type or the fault mode area.

9. The real-time monitoring device for electric power Internet of Things data according to claim 8 is characterized in that: The step of calculating the topological potential value of each node in each area of ​​the topological graph according to the quality of the influencing factor further includes: The edge computing node integrates the topological potential values ​​of each node in each area of ​​the topological graph to obtain an integrated topological potential value; Sending the integrated topological potential value to a cloud server, so that the cloud server locates the node with the highest risk among the nodes in each area in the topological map according to the received integrated topological potential value; The cloud server also analyzes and locates the disturbance type or fault mode based on the disturbance type or fault mode in combination with the real-time operation data and historical fault database of the power Internet of Things.

10. The real-time monitoring device for electric power Internet of Things data according to claim 6, characterized in that: The extracting the time-frequency characteristics of the three-phase voltage and current signals in the edge computing node according to the generalized S transform; and identifying the fault or disturbance in the power Internet of Things according to the time-frequency characteristics, specifically: Visualize the time-frequency features through a time-frequency graph, where different colors in the time-frequency graph represent different amplitudes; When the amplitude of the frequency harmonic exceeds a preset threshold, it is marked with a warning color, and a fault or disturbance in the power Internet of Things is identified based on the warning color.

Images