DTU safety monitoring method and device, equipment and storage medium

Abstract

The invention relates to a DTU safety monitoring method. The method comprises the following steps: acquiring electrical parameters of a ring main unit through a DTU; verifying the electrical parameters to detect whether the electrical parameters are within a preset electrical parameter safety range; if the electrical parameter is not in the preset electrical parameter safety range, performing multi-scale decomposition on the electrical parameter to obtain a potential hazard characteristic signal; performing frequency domain feature extraction on the potential hazard feature signal to obtain a fault frequency spectrum feature; calling a preset fault mode database, and matching the fault frequency spectrum features with frequency spectrum features in the fault mode database to obtain a fault problem area and a fault type of the ring main unit; giving a weight to the fault problem area, giving a score to the fault type, and obtaining an abnormal risk level of the ring main unit based on the weight of the fault problem area and the score of the fault type; the technical problem that diagnosis accuracy is limited in the face of complex working conditions and weak signal anomalies is solved.

Description

Technical Field

[0001] The present invention relates to the field of security monitoring technology, and in particular to a DTU security monitoring method, device, equipment and storage medium. Background Art

[0002] With the rapid development of smart grids, the automation level of distribution systems continues to improve, and the demand for intelligent monitoring and maintenance of distribution terminal equipment (such as DTU, i.e. Distribution Terminal Unit) is increasing. As a key component of the distribution network, the stable operation of the ring main unit is directly related to the safety of the power grid and the quality of power supply. However, traditional ring main unit monitoring methods mostly rely on manual inspections and simple numerical threshold alarms, lacking the ability to conduct in-depth analysis of electrical parameters and early warning of faults, especially in the face of complex working conditions and weak signal anomalies, the diagnostic accuracy is limited. In addition, the fault location is not accurate enough during the fault diagnosis process, making it difficult to respond quickly and effectively dispatch resources for maintenance, which affects the operating efficiency and safety level of the power system. Summary of the invention

[0003] The main purpose of the present invention is to provide a DTU safety monitoring method, device, equipment and storage medium, which solves the technical problem of limited diagnostic accuracy when facing complex working conditions and weak signal anomalies.

[0004] To achieve the above object, the present invention provides a DTU security monitoring method, comprising the following steps: Obtain electrical parameters of the ring main unit through the DTU; wherein the electrical parameters include voltage parameters, current parameters and power factor; The electrical parameters are verified using a preset field-by-field verification technology to detect whether the electrical parameters are within the preset electrical parameter safety range; if the electrical parameters are not within the preset electrical parameter safety range, the electrical parameters are decomposed at multiple scales using wavelet analysis to obtain a potential hidden danger characteristic signal; wherein the wavelet analysis is used to identify mutation points and abnormal fluctuations of the electrical parameters; Extracting the frequency domain features of the potential hidden danger characteristic signal to obtain the fault spectrum features; Calling a preset fault mode database, matching the fault spectrum feature with the spectrum feature in the fault mode database, and obtaining the fault problem area and fault type of the ring main unit; A weight is assigned to the fault problem area, a score is assigned to the fault type, and the abnormal risk level of the ring main unit is obtained based on the weight of the fault problem area and the score of the fault type. When the abnormal risk level is not within the preset risk level range, an alarm is issued.

[0005] As a further solution of the present invention, the electrical parameters are verified by using a preset field-by-field verification technology to detect whether the electrical parameters are within the preset electrical parameter safety range; if the electrical parameters are not within the preset electrical parameter safety range, wavelet analysis is used to perform multi-scale decomposition on the electrical parameters to obtain a potential hidden danger characteristic signal, including: Performing digital filtering on the electrical parameters to obtain digital filtering parameters; Calculate the root mean square value of the digital filter parameters to obtain the standardized electrical parameters; The standardized electrical parameters are verified field by field using a preset field-by-field verification technology to check whether the standardized electrical parameters are within the preset electrical parameter safety range. If any standardized electrical parameter exceeds the preset electrical parameter safety range, the standardized electrical parameters exceeding the electrical parameter safety range are decomposed at multiple scales using a preset wavelet basis function to obtain frequency components at different scales; wherein the frequency components include high-frequency components and low-frequency components; The frequency components are analyzed to identify abnormal fluctuations and mutation points, and obtain characteristic signals of potential hidden dangers.

[0006] As a further solution of the present invention, the frequency components are analyzed to identify abnormal fluctuations and mutation points to obtain characteristic signals of potential hidden dangers, including: Normalizing the frequency components to obtain standardized frequency component signals; Based on a preset beamforming technology, focusing processing is performed on the standardized frequency component signal to obtain an enhanced target frequency component signal; Performing a change trend analysis on the target frequency component signal to obtain a fluctuation signal; Acquire a set abnormality detection threshold, and detect the fluctuation signal by using the abnormality detection threshold to obtain a preliminary abnormal fluctuation signal; Through the preset recursive neural network model, the preliminary abnormal fluctuation signal is deeply learned to obtain a high-dimensional feature vector; Performing time series analysis on the high-dimensional feature vector to obtain a time series high-dimensional feature vector; Hidden dangers are identified on the time series high-dimensional feature vector to obtain potential hidden danger feature signals.

[0007] As a further solution of the present invention, frequency domain feature extraction is performed on the potential hidden danger characteristic signal to obtain fault spectrum features, including: Performing time-frequency conversion on the potential hidden danger characteristic signal through a preset fast Fourier transform to obtain a preliminary spectrum diagram; Using a preset Hilbert-Huang transform algorithm, the spectrum density of the preliminary spectrum graph is calculated to generate a first spectrum density graph; wherein the first spectrum density graph is used to quantify the energy distribution of each frequency component; Using an independent component analysis algorithm, identifying key frequency components in the first spectral density graph; Using an autoencoder, identifying abnormal peaks in the first spectral density map; Combining the key frequency components and the abnormal peaks, and locating the fault spectrum feature set in the first spectrum density graph; The fault spectrum feature set is subjected to dimension reduction and screening to obtain the fault spectrum feature.

[0008] As a further solution of the present invention, a preset Hilbert-Huang transform algorithm is used to calculate the spectrum density of the preliminary spectrum map to generate a first spectrum density map, including: Using a preset Hilbert-Huang transform algorithm, preprocessing the preliminary spectrum graph to obtain a preprocessed spectrum graph; Decomposing the preprocessed spectrum to obtain a plurality of intrinsic mode functions, and extracting each of the intrinsic mode functions one by one to obtain each independent intrinsic mode function; Performing Hilbert transform on each independent eigenmode function to obtain the instantaneous frequency and instantaneous amplitude of each eigenmode function; Reconstructing the instantaneous frequency and instantaneous amplitude to obtain a reconstructed time-frequency signal; Calculating the spectrum density of the reconstructed time-frequency signal to obtain spectrum density data, wherein the spectrum density data is the energy distribution of the reconstructed time-frequency signal; Based on the spectrum density data, the preliminary spectrum graph is plotted to obtain a first spectrum density graph.

[0009] As a further solution of the present invention, a preset fault mode database is called, and the fault spectrum characteristics are matched with the spectrum characteristics in the fault mode database to obtain the fault problem area and fault type of the ring main unit, including: Analyze the fault spectrum characteristics and the spectrum characteristics in the fault mode database respectively, and obtain a fault spectrum characteristic vector and a first spectrum characteristic vector accordingly; Performing Mahalanobis distance calculation on the fault spectrum feature vector and the first spectrum feature vector to obtain the Mahalanobis distance of the fault spectrum; If the Mahalanobis distance is within a preset distance range, correlation matching is performed on the fault spectrum feature vector and the first spectrum feature vector to obtain a matching result; wherein the matching result is a matching degree report; The fault spectrum characteristics are analyzed based on the matching report to obtain the fault problem area and fault type of the ring main unit.

[0010] As a further solution of the present invention, a weight is assigned to the fault problem area, a score is assigned to the fault type, and the abnormal risk level of the ring main unit is obtained based on the weight of the fault problem area and the score of the fault type. When the abnormal risk level is not within a preset risk level range, an alarm is issued, including: Assigning weights to each of the fault problem areas, and performing vector conversion on the weights of each of the fault problem areas to obtain a vector of the fault problem area; Assigning scores to each of the fault types, and performing vector conversion on the scores of each of the fault types to obtain a vector of the fault type; Multiplying the vector of the fault problem area by the vector of the fault type to obtain a risk value of the fault; The abnormal risk level of the ring main unit is obtained based on the risk value of the fault, and an alarm is issued when the abnormal risk level is not within a preset risk level range.

[0011] The present invention also provides a safety monitoring device for a ring main unit, comprising: An acquisition module is used to acquire electrical parameters of the ring main unit through the DTU; wherein the electrical parameters include voltage parameters, current parameters and power factor; A verification module, used to verify the electrical parameters using a preset field-by-field verification technology to detect whether the electrical parameters are within a preset electrical parameter safety range; if the electrical parameters are not within the preset electrical parameter safety range, the electrical parameters are decomposed at multiple scales using wavelet analysis to obtain a potential hidden danger characteristic signal; wherein the wavelet analysis is used to identify mutation points and abnormal fluctuations of the electrical parameters; An extraction module is used to extract the frequency domain features of the potential hidden danger characteristic signal to obtain the fault spectrum features; A matching module is used to call a preset fault mode database, match the fault spectrum characteristics with the spectrum characteristics in the fault mode database, and obtain the fault problem area and fault type of the ring main unit; An assignment module is used to assign weights to the fault problem areas, assign scores to the fault types, and obtain the abnormal risk level of the ring network cabinet based on the weights of the fault problem areas and the scores of the fault types. When the abnormal risk level is not within a preset risk level range, an alarm is issued.

[0012] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0013] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0014] The DTU safety monitoring method, device, equipment and storage medium provided by the present invention include the following steps: obtaining the electrical parameters of the ring network cabinet through the DTU; wherein the electrical parameters include voltage parameters, current parameters and power factor; using a preset field-by-field verification technology to verify the electrical parameters to detect whether the electrical parameters are within the preset electrical parameter safety range; if the electrical parameters are not within the preset electrical parameter safety range, using wavelet analysis to perform multi-scale decomposition on the electrical parameters to obtain potential hidden danger characteristic signals; wherein the wavelet analysis is used to identify mutation points and abnormal fluctuations of electrical parameters; performing frequency domain analysis on the potential hidden danger characteristic signals Feature extraction is performed to obtain fault spectrum features; a preset fault mode database is called to match the fault spectrum features with the spectrum features in the fault mode database to obtain the fault problem area and fault type of the ring network cabinet; a weight is assigned to the fault problem area, a score is assigned to the fault type, and the abnormal risk level of the ring network cabinet is obtained based on the weight of the fault problem area and the score of the fault type. When the abnormal risk level is not within the preset risk level range, an alarm is issued; through the above technical solution, the technical problem of limited diagnostic accuracy when facing complex working conditions and weak signal anomalies is solved, and the accuracy and efficiency of diagnosis are greatly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a schematic diagram of the steps of a DTU security monitoring method in one embodiment of the present invention; Figure 2 is a structural block diagram of a safety monitoring device of a DTU in one embodiment of the present invention; Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0016] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] like Figure 1 As shown, Figure 1 Schematic diagram of the steps of a DTU safety monitoring method in one embodiment of the present invention; An embodiment of the present invention provides a DTU security monitoring method, comprising the following steps: Step S1, obtaining electrical parameters of the ring main unit through the DTU; wherein the electrical parameters include voltage parameters, current parameters and power factor.

[0019] Specifically, DTU (remote terminal unit): DTU is a key component in the distribution automation system. It is installed near the field equipment of the power system, such as the ring main unit, and is responsible for collecting real-time operating data of the field electrical equipment, and can remotely control the switchgear to monitor and control the operation status of the power grid. It acts as a bridge between the field equipment and the master station system, enabling the dispatch center to grasp and remotely manage the power grid status in real time.

[0020] Ring main unit: Ring main unit is a switchgear used in medium voltage distribution network, mainly used for segmentation, connection and ring network power supply of urban distribution network, with the characteristics of compact structure, flexible operation, safety and reliability. It contains a variety of electrical components, such as load switches, circuit breakers, protection devices, etc., which are crucial to the safe and stable operation of power system.

[0021] Electrical parameters: Specifically for this solution, the electrical parameters mentioned include: Voltage parameters: refers to the voltage value of each point in the ring main unit, including line voltage and phase voltage, which is an important indicator for measuring the stability and power quality of the power system. Abnormal voltage fluctuations may indicate system overload, short circuit or equipment failure. Current parameters: reflects the actual flow of current in the ring main unit, including phase current and total current, which is the basis for evaluating equipment load and detecting overcurrent faults. Excessive current may cause equipment damage or fire risk. Power factor: The power factor is the ratio of actual power to apparent power, reflecting the efficiency of electrical energy use in the circuit. Low power factor indicates that there is a large amount of reactive power in the system, which may lead to problems such as reduced equipment efficiency and increased line loss. Through real-time monitoring and collection of these electrical parameters by DTU, the operating status of the ring main unit can be continuously monitored, and signs such as voltage instability, current abnormality or power factor decline can be discovered in time, providing valuable data support for subsequent fault analysis, early warning and processing. This process is a key link in implementing smart grid management and improving the overall efficiency and safety of the power system.

[0022] Step S2, using a preset field-by-field verification technology to verify the electrical parameters to detect whether the electrical parameters are within the preset electrical parameter safety range; if the electrical parameters are not within the preset electrical parameter safety range, wavelet analysis is used to perform multi-scale decomposition on the electrical parameters to obtain potential hidden danger characteristic signals; wherein the wavelet analysis is used to identify mutation points and abnormal fluctuations of electrical parameters.

[0023] Specifically, field-by-field verification technology: This is a preset verification method that performs independent checks on each electrical parameter (including voltage, current, and power factor) collected from the DTU. It determines whether each parameter is normal by comparing these real-time measurements with the preset safety range or standard value. If all parameters fall within the preset safety limits, it indicates that the current system is operating well; conversely, if any parameter exceeds the safety range, further analysis and processing is triggered to explore potential problems in depth. Wavelet analysis and multi-scale decomposition: Once it is detected that the electrical parameters deviate from the safety range, the solution turns to wavelet analysis technology. Wavelet analysis is a powerful signal processing tool that is particularly suitable for analyzing local features and time-frequency characteristics in non-stationary signals. It can effectively identify mutation points (i.e. sudden large changes) and abnormal fluctuations (periodic or non-periodic changes that are different from normal patterns) in electrical parameters by decomposing the original signal at different scales (frequency and time resolution). This multi-scale decomposition process helps to extract "noise" from complex electrical signals and highlight those potential hidden danger characteristic signals that may indicate equipment failure or system abnormality. Through this technical path, the solution can not only achieve basic safety monitoring of the electrical parameters of the ring main unit, but also use wavelet analysis to deeply explore the characteristics of potential problems when abnormalities are found, providing accurate information for subsequent fault diagnosis and preventive maintenance. This method improves the sensitivity and accuracy of fault detection, and helps to discover and solve hidden dangers that may affect the stability and reliability of the power system in advance.

[0024] Step S3, extracting the frequency domain features of the potential hidden danger characteristic signal to obtain the fault spectrum features.

[0025] Specifically, potential hidden danger characteristic signal: This is the signal obtained after wavelet analysis, which contains key information in electrical parameters that may indicate faults or abnormal behavior. These signals represent specific patterns of electrical parameter changes at different scales, especially those that deviate significantly from normal operating conditions. Frequency domain feature extraction: In order to further analyze these potential hidden danger signals, they are then converted from the time domain to the frequency domain. In the field of signal processing, frequency domain analysis can reveal the component frequency components of the signal and their relative strength, which is crucial for identifying specific types of fault characteristics. Commonly used conversion methods are Fourier transform or more advanced ones such as wavelet packet transform, which can decompose time series data into components of different frequencies. In this way, regularities and anomalies that were originally difficult to detect in the time domain become clearly visible in the frequency domain. Fault spectrum characteristics: The features extracted in the frequency domain are the so-called "fault spectrum characteristics". These features reflect the energy distribution of the fault signal at different frequencies. Specific frequency responses are often associated with specific types of electrical equipment faults, such as harmonics, frequency offsets, oscillation modes, etc. By comparing these features with the spectrum database of known fault modes, the type and location of the fault can be identified, providing a basis for subsequent maintenance decisions. In short, this step converts the potential hidden danger characteristic signal obtained after wavelet analysis into the frequency domain and extracts key features to form the "fingerprint" of fault diagnosis - the fault spectrum characteristics. This is a key step in accurately locating and identifying fault types, greatly improving the accuracy and efficiency of diagnosis.

[0026] Step S4, calling a preset fault mode database, matching the fault spectrum feature with the spectrum feature in the fault mode database, and obtaining the fault problem area and fault type of the ring main unit.

[0027] Specifically, the preset fault mode database: This is a pre-established database that contains typical fault spectrum features corresponding to various known fault conditions. These features are summarized through historical data analysis, physical modeling or actual fault cases, covering the manifestations of different fault types (such as insulation damage, short circuit, overload, etc.) in different links (such as switchgear, cable terminals, etc.). The existence of the database provides a reference standard template for fault diagnosis. Fault spectrum feature matching: The fault spectrum features extracted from the electrical parameters of the ring network cabinet are carefully compared with the features in the fault mode database. This process may involve correlation analysis, distance metrics (such as Euclidean distance, cosine similarity) or more complex machine learning classification algorithms. The purpose is to find the known fault mode that is most similar or consistent with the current fault feature, so as to identify the fault type. Fault problem area and fault type determination: Through the above matching process, not only can the specific fault type (for example, local short circuit or insulation aging) be identified, but also the specific area or component where the fault occurs can be roughly located based on the specificity of the feature and the equipment location information associated with the fault mode recorded in the database. In this way, we not only know "what type of fault" but also "where the fault is", providing a clear direction for subsequent maintenance actions. In summary, this solution achieves rapid and accurate diagnosis of ring network cabinet faults through intelligent matching of fault spectrum characteristics with the preset database, including fault type identification and problem area location, which is a key link in realizing fault warning and rapid response mechanism.

[0028] Step S5, assigning weights to the fault problem areas, assigning scores to the fault types, and obtaining the abnormal risk level of the ring main unit based on the weights of the fault problem areas and the scores of the fault types. When the abnormal risk level is not within the preset risk level range, an alarm is issued.

[0029] Specifically, weights are assigned to fault problem areas: a weight value is assigned to each possible fault area based on factors such as the importance of each area or component in the ring main unit, the degree of impact on the stability of the entire power grid operation, and the difficulty of repair. Critical areas or parts that have a great impact on power transmission will be assigned higher weights, indicating that the potential impact of their faults is more serious. Scoring of fault types: a score is assigned to each fault type based on the severity, commonness, and possible consequences of the fault type (such as power outage scope, repair cost, safety risk, etc.). Severe fault types such as direct short circuits will receive higher scores, indicating that such faults are more urgent and harmful. Abnormal risk level calculation: The abnormal risk level of the ring main unit is calculated by combining the weight of the fault problem area and the score of the fault type through a predefined algorithm (such as weighted average or a more complex evaluation model). This level comprehensively reflects the urgency and potential impact of the current fault and is a key indicator for determining whether immediate measures need to be taken. Risk level range judgment and alarm triggering: If the calculated abnormal risk level exceeds the preset safety threshold or risk level range, it means that the fault has reached the level that requires intervention, and the system will automatically issue an alarm. The alarm information usually includes the preliminary diagnosis of the fault, risk level and recommended countermeasures, so that the operation and maintenance personnel can respond quickly and take necessary maintenance or emergency measures to prevent the fault from escalating and ensure the safe and stable operation of the power system. This series of operations builds a closed-loop process from fault identification to risk assessment to emergency response, ensuring efficient management and control of ring network cabinet faults.

[0030] In a specific embodiment, the electrical parameters are verified using a preset field-by-field verification technology to detect whether the electrical parameters are within a preset electrical parameter safety range; if the electrical parameters are not within the preset electrical parameter safety range, wavelet analysis is used to perform multi-scale decomposition on the electrical parameters to obtain a potential hidden danger characteristic signal, including: Performing digital filtering on the electrical parameters to obtain digital filtering parameters; Calculate the root mean square value of the digital filter parameters to obtain the standardized electrical parameters; The standardized electrical parameters are verified field by field using a preset field-by-field verification technology to check whether the standardized electrical parameters are within the preset electrical parameter safety range. If any standardized electrical parameter exceeds the preset electrical parameter safety range, the standardized electrical parameters exceeding the electrical parameter safety range are decomposed at multiple scales using a preset wavelet basis function to obtain frequency components at different scales; wherein the frequency components include high-frequency components and low-frequency components; The frequency components are analyzed to identify abnormal fluctuations and mutation points, and obtain characteristic signals of potential hidden dangers.

[0031] Specifically, the preset field-by-field verification technology is used for initial verification: First, the preset verification rules are applied to the original electrical parameters (such as voltage, current, and power factor) to check whether these parameters are within the predefined safety range. This stage mainly involves preliminary screening of the data to ensure the validity and security of the data. Digital filtering and parameter standardization: When certain electrical parameters are found to be beyond the safety range, these parameters are digitally filtered to eliminate noise interference and extract the real information in the signal. Subsequently, the root mean square value of the filtered parameters is calculated and converted into standardized electrical parameters for subsequent unified processing and comparison. This process enhances the comparability of parameters and the accuracy of analysis. Re-verification of standardized electrical parameters: The field-by-field verification technology is used again on the standardized electrical parameters to check whether they meet safety requirements with more refined standards. This is done to ensure that any potential anomalies can still be identified even after filtering and standardization. Wavelet analysis and multi-scale decomposition: For standardized electrical parameters that still fail to pass the verification, wavelet analysis technology is used for multi-scale decomposition. Wavelet basis functions are used to decompose the signal and separate different frequency components, including high-frequency components (usually related to mutations and spikes) and low-frequency components (related to trends and slow changes). This process is a key step in identifying and locating abnormal fluctuations and mutation points. Frequency component analysis and hidden danger identification: In-depth analysis of the high-frequency and low-frequency components obtained by decomposition to identify abnormal fluctuations and mutation points. These characteristics often indicate potential problems or hidden dangers in the system. By analyzing these characteristic signals, it can be determined which parts or types of problems are most likely to exist, providing clues for subsequent fault diagnosis and maintenance. In summary, the solution uses digital filtering, standardization processing, multi-scale wavelet analysis and other technical means to conduct in-depth analysis of the electrical parameters of the ring network cabinet, aiming to accurately identify and locate any hidden dangers that may affect the safe operation of the power grid, and take timely measures to ensure the stability and reliability of the power supply.

[0032] In a specific embodiment, the frequency components are analyzed to identify abnormal fluctuations and mutation points to obtain characteristic signals of potential hidden dangers, including: Normalizing the frequency components to obtain standardized frequency component signals; Based on a preset beamforming technology, focusing processing is performed on the standardized frequency component signal to obtain an enhanced target frequency component signal; Performing a change trend analysis on the target frequency component signal to obtain a fluctuation signal; Acquire a set abnormality detection threshold, and detect the fluctuation signal by using the abnormality detection threshold to obtain a preliminary abnormal fluctuation signal; Through the preset recursive neural network model, the preliminary abnormal fluctuation signal is deeply learned to obtain a high-dimensional feature vector; Performing time series analysis on the high-dimensional feature vector to obtain a time series high-dimensional feature vector; Hidden dangers are identified on the time series high-dimensional feature vector to obtain potential hidden danger feature signals.

[0033] Specifically, the preset field-by-field verification technology is used for initial verification: First, the preset verification rules are applied to the original electrical parameters (such as voltage, current, and power factor) to check whether these parameters are within the predefined safety range. This stage mainly involves preliminary screening of the data to ensure the validity and security of the data. Digital filtering and parameter standardization: When certain electrical parameters are found to be beyond the safety range, these parameters are digitally filtered to eliminate noise interference and extract the real information in the signal. Subsequently, the root mean square value of the filtered parameters is calculated and converted into standardized electrical parameters for subsequent unified processing and comparison. This process enhances the comparability of parameters and the accuracy of analysis. Re-verification of standardized electrical parameters: The field-by-field verification technology is used again on the standardized electrical parameters to check whether they meet safety requirements with more refined standards. This is done to ensure that any potential anomalies can still be identified even after filtering and standardization. Wavelet analysis and multi-scale decomposition: For standardized electrical parameters that still fail to pass the verification, wavelet analysis technology is used for multi-scale decomposition. Wavelet basis functions are used to decompose signals and separate different frequency components, including high-frequency components (usually related to mutations and spikes) and low-frequency components (related to trends and slow changes). This process is a key step in identifying and locating abnormal fluctuations and mutation points. Frequency component analysis and hidden danger identification: The high-frequency and low-frequency components obtained by decomposition are deeply analyzed to identify abnormal fluctuations and mutation points. These characteristics often indicate potential problems or hidden dangers in the system. By analyzing these characteristic signals, it can be clear which parts or types of problems are most likely to exist, providing clues for subsequent fault diagnosis and maintenance. In summary, the scheme uses digital filtering, standardization processing, multi-scale wavelet analysis and other technical means to conduct in-depth analysis of the electrical parameters of the ring network cabinet, aiming to accurately identify and locate any hidden dangers that may affect the safe operation of the power grid, take timely measures, and ensure the stability and reliability of the power supply. In addition, the abnormal detection threshold includes an abnormal fluctuation threshold and a mutation point, the abnormal detection threshold is obtained, and the fluctuation signal is detected by the abnormal detection threshold to obtain a preliminary abnormal fluctuation signal and a mutation point.

[0034] In a specific embodiment, frequency domain feature extraction is performed on the potential hidden danger characteristic signal to obtain fault spectrum features, including: Performing time-frequency conversion on the potential hidden danger characteristic signal through a preset fast Fourier transform to obtain a preliminary spectrum diagram; Using a preset Hilbert-Huang transform algorithm, the spectrum density of the preliminary spectrum graph is calculated to generate a first spectrum density graph; wherein the first spectrum density graph is used to quantify the energy distribution of each frequency component; Using an independent component analysis algorithm, identifying key frequency components in the first spectral density graph; Using an autoencoder, identifying abnormal peaks in the first spectral density map; Combining the key frequency components and the abnormal peaks, and locating the fault spectrum feature set in the first spectrum density graph; The fault spectrum feature set is subjected to dimension reduction and screening to obtain the fault spectrum feature.

[0035] Specifically, time-frequency conversion and preliminary spectrum generation: First, the preset fast Fourier transform (FFT) is used to convert the potential hidden danger characteristic signal from the time domain to the frequency domain to form a preliminary spectrum. This step reveals the frequency composition of the signal, that is, which frequency components dominate the signal. Spectral density calculation and first spectrum density map generation: Then, the Hilbert-Huang transform (HHT) algorithm is used to further analyze the preliminary spectrum, calculate the spectrum density of each frequency component, and generate the first spectrum density map. HHT is particularly suitable for nonlinear and non-stationary signals, and can more accurately reflect the distribution of frequency energy in the signal, which helps to identify which frequency ranges have abnormally increased energy, which may be a sign of fault. Key frequency component identification: On this basis, the independent component analysis (ICA) algorithm is used to identify key frequency components in the first spectrum density map. ICA is a blind source separation technology that can separate independent signal components from mixed signals. Here, we help us find those frequency components that represent fault modes and have important diagnostic value. Abnormal peak identification: At the same time, the first spectrum density map is analyzed by an autoencoder (a deep learning method) to identify abnormal peaks. The autoencoder detects abnormal points that deviate greatly from the normal state by learning an efficient representation of the data. These abnormal peaks may indicate potential fault events. Positioning the fault spectrum feature set: By combining the key frequency components identified by ICA with the abnormal peaks identified by the autoencoder, the fault spectrum feature set can be accurately located in the first spectrum density map. These feature sets directly point to the root cause or significant features of the fault. Dimensionality reduction and screening to obtain fault spectrum features: Finally, the fault spectrum feature set is subjected to dimensionality reduction processing, such as using principal component analysis (PCA) and other methods to reduce the dimension of the data while retaining key information, and then the most representative fault features are screened out. These features will serve as inputs to the fault diagnosis and prediction models to achieve accurate assessment and timely intervention of the status of electrical equipment. The entire process uses a progressive analysis strategy to unravel the complex signals and gradually lock in the specific features of the fault, providing a scientific and accurate methodology for power system fault diagnosis.

[0036] In a specific embodiment, a preset Hilbert-Huang transform algorithm is used to calculate the spectrum density of the preliminary spectrum map to generate a first spectrum density map, including: Using a preset Hilbert-Huang transform algorithm, preprocessing the preliminary spectrum graph to obtain a preprocessed spectrum graph; Decomposing the preprocessed spectrum to obtain a plurality of intrinsic mode functions, and extracting each of the intrinsic mode functions one by one to obtain each independent intrinsic mode function; Performing Hilbert transform on each independent eigenmode function to obtain the instantaneous frequency and instantaneous amplitude of each eigenmode function; Reconstructing the instantaneous frequency and instantaneous amplitude to obtain a reconstructed time-frequency signal; Calculating the spectrum density of the reconstructed time-frequency signal to obtain spectrum density data, wherein the spectrum density data is the energy distribution of the reconstructed time-frequency signal; Based on the spectrum density data, the preliminary spectrum graph is plotted to obtain a first spectrum density graph.

[0037] Specifically, preprocessing spectrogram generation: First, the preliminary spectrogram is preprocessed using the preset Hilbert-Huang transform algorithm. This step may involve operations such as smoothing noise and removing insignificant details, with the goal of optimizing the original data so that subsequent analysis can focus more on the core features of the signal. Intrinsic mode function decomposition: Subsequently, the preprocessed spectrogram is decomposed. This process decomposes the complex signal into a series of intrinsic mode functions (IMFs) based on the principle of empirical mode decomposition (EMD). IMFs are a series of simple signals with local time-frequency characteristics, each of which represents a different vibration mode in the signal. Hilbert transform of IMF: Hilbert transform is applied to each independent IMF. Hilbert transform can extract instantaneous frequency and instantaneous amplitude information from each IMF. The instantaneous frequency reflects the change of the signal frequency over time, while the instantaneous amplitude shows the change trend of the signal strength. Together, they provide details of the dynamic behavior of the signal. Reconstructing the time-frequency signal: The instantaneous frequency and instantaneous amplitude information of all IMFs are combined to obtain a new time-frequency signal through the reconstruction process. This reconstructed signal not only retains the time-frequency characteristics of the original signal, but also improves the resolution of signal analysis by separating vibration modes of different scales. Spectral density calculation: Next, the spectral density of the reconstructed time-frequency signal is calculated. The spectral density data depicts the distribution of signal energy in the frequency domain. High-density areas mean that the energy of the corresponding frequency components is higher, which is crucial for identifying fault frequency characteristics. Draw the first spectral density graph: Finally, based on the above spectral density data, redraw the first spectral density graph. This figure intuitively shows the characteristics of the signal in the frequency domain, especially those frequency components related to potential faults. The energy distribution is clearly visible, providing intuitive and accurate information for fault diagnosis. The entire process is carefully analyzed by the HHT algorithm, which not only improves the accuracy of spectral analysis, but also enhances the understanding of the complex dynamic characteristics inside the signal, which is of great value for identifying and locating fault spectral characteristics.

[0038] In a specific embodiment, a preset fault mode database is called, and the fault spectrum feature is matched with the spectrum feature in the fault mode database to obtain the fault problem area and fault type of the ring main unit, including: Analyze the fault spectrum characteristics and the spectrum characteristics in the fault mode database respectively, and obtain a fault spectrum characteristic vector and a first spectrum characteristic vector accordingly; Performing Mahalanobis distance calculation on the fault spectrum feature vector and the first spectrum feature vector to obtain the Mahalanobis distance of the fault spectrum; If the Mahalanobis distance is within a preset distance range, correlation matching is performed on the fault spectrum feature vector and the first spectrum feature vector to obtain a matching result; wherein the matching result is a matching degree report; The fault spectrum characteristics are analyzed based on the matching report to obtain the fault problem area and fault type of the ring main unit.

[0039] Specifically, fault spectrum feature extraction and database preparation: First, the detected ring network cabinet signal is analyzed to extract the fault spectrum feature. At the same time, a preset fault mode database is called, which stores spectrum feature samples of various known fault types. These samples can be regarded as the manifestation of typical faults. Feature vectorization: In order to facilitate comparison, the extracted fault spectrum features and the spectrum features in the database need to be converted into feature vectors. The "fault spectrum feature vector" and "first spectrum feature vector" mentioned here represent the spectrum representation of the actual fault signal and the spectrum representation of a reference fault mode in the database, respectively. Both are a set of numerical sequences formed by quantizing spectrum data points. Mahalanobis distance calculation: Then, the statistical measurement method of Mahalanobis distance is used to evaluate the degree of difference between the fault spectrum feature vector and each first spectrum feature vector in the database. Mahalanobis distance takes into account the covariance of variables and can effectively measure the similarity between different feature vectors, especially suitable for comparison in multi-dimensional feature space. Matching judgment and correlation matching: If the calculated Mahalanobis distance falls within the preset distance range, it indicates that the current fault spectrum is close enough to a known fault mode and is worth further analysis. At this point, a more detailed correlation matching will be performed, that is, the correlation between the fault spectrum feature vector and the specific first spectrum feature vector in each dimension is compared to generate a matching report. This report summarizes the matching degree between the two and provides a direct basis for fault diagnosis. Fault location and type determination: Finally, based on the matching report, analysts can infer the fault problem area and fault type in the ring network cabinet. The fault mode corresponding to the feature vector with a high matching degree is very likely to be the actual cause of the fault. This diagnostic method combines the objective data of spectrum analysis with the empirical knowledge accumulated in the database to improve the accuracy and efficiency of fault identification. The entire process converts complex fault signals into comparable data through quantitative analysis and pattern matching technology, thereby realizing the rapid location and qualitative analysis of ring network cabinet faults.

[0040] In a specific embodiment, a weight is assigned to the fault problem area, a score is assigned to the fault type, and an abnormal risk level of the ring main unit is obtained based on the weight of the fault problem area and the score of the fault type. When the abnormal risk level is not within a preset risk level range, an alarm is issued, including: Assigning weights to each of the fault problem areas, and performing vector conversion on the weights of each of the fault problem areas to obtain a vector of the fault problem area; Assigning scores to each of the fault types, and performing vector conversion on the scores of each of the fault types to obtain a vector of the fault type; Multiplying the vector of the fault problem area by the vector of the fault type to obtain a risk value of the fault; The abnormal risk level of the ring main unit is obtained based on the risk value of the fault, and an alarm is issued when the abnormal risk level is not within a preset risk level range.

[0041] Specifically, weight assignment and vectorization of fault problem areas: First, a weight value is assigned to each fault area manually or based on historical data according to the severity of the impact of the fault problem area on the normal operation of the ring main unit. This weight reflects the impact that the fault in the area may have. The larger the value, the more serious the potential impact. The weights of all fault problem areas are integrated into a vector, and each component corresponds to the weight of a specific area, which helps with subsequent mathematical operations. Fault type scoring and vectorization: Similarly, for each identified fault type, a score is assigned based on the severity of the consequences it may cause (such as repair cost, power outage time, safety risk, etc.). The higher the score, the greater the potential harm of the fault type. These scores are also organized into a vector, in which each element corresponds to the score of a fault type. Risk value calculation: Next, the weight vector of the fault problem area is multiplied element by element by the fault type score vector. This operation is essentially to evaluate the comprehensive risk impact of each fault type when it occurs in different fault areas. The result of the multiplication is a new vector, in which each component represents the risk contribution value of a specific fault type in a specific area. Abnormal risk level assessment and alarm triggering: The risk value vector obtained in the previous step is aggregated (such as summation or weighted average) to obtain a single value - the abnormal risk level of the ring main unit. This level reflects the overall risk level after comprehensive consideration of all fault areas and fault types. If the calculated risk level exceeds the pre-set safety range (meaning that the risk is unacceptable or requires immediate attention), the system will automatically issue an alarm to remind relevant personnel to take necessary maintenance or emergency measures. Through this series of steps, not only the accurate positioning and classification of faults are achieved, but also the risk assessment is standardized through quantitative methods, ensuring efficient monitoring and early warning of the health status of the ring main unit, which helps to prevent major accidents.

[0042] The above describes the DTU security monitoring method in the embodiment of the present invention. The following describes the DTU security monitoring device in the embodiment of the present invention. Figure 2 , an embodiment of the safety monitoring device of DTU in the embodiment of the present invention includes: The acquisition module 21 is used to acquire the electrical parameters of the ring main unit through the DTU; wherein the electrical parameters include voltage parameters, current parameters and power factor; The verification module 22 is used to verify the electrical parameters using a preset field-by-field verification technology to detect whether the electrical parameters are within a preset electrical parameter safety range; if the electrical parameters are not within the preset electrical parameter safety range, the electrical parameters are decomposed at multiple scales using wavelet analysis to obtain a potential hidden danger characteristic signal; wherein the wavelet analysis is used to identify mutation points and abnormal fluctuations of the electrical parameters; An extraction module 23 is used to extract the frequency domain features of the potential hidden danger characteristic signal to obtain the fault spectrum features; A matching module 24 is used to call a preset fault mode database, match the fault spectrum feature with the spectrum feature in the fault mode database, and obtain the fault problem area and fault type of the ring main unit; The assignment module 25 is used to assign weights to the fault problem areas, assign scores to the fault types, and obtain the abnormal risk level of the ring network cabinet based on the weights of the fault problem areas and the scores of the fault types. When the abnormal risk level is not within the preset risk level range, an alarm is issued.

[0043] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to the above method embodiment, which will not be repeated here.

[0044] Reference Figure 3 The present invention also provides a computer device in an embodiment, wherein the internal structure of the computer device can be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a display screen, an input device system, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0045] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0046] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0047] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0048] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0049] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A DTU security monitoring method, characterized in that: The following steps are involved: Obtain electrical parameters of the ring main unit through the DTU; wherein the electrical parameters include voltage parameters, current parameters and power factor; The electrical parameters are verified by using a preset field-by-field verification technology to detect whether the electrical parameters are within a preset electrical parameter safety range; if the electrical parameters are not within the preset electrical parameter safety range, wavelet analysis is used to perform multi-scale decomposition on the electrical parameters to obtain a potential hidden danger characteristic signal; wherein the wavelet analysis is used to identify mutation points and abnormal fluctuations of electrical parameters; Extracting the frequency domain features of the potential hidden danger characteristic signal to obtain the fault spectrum features; Calling a preset fault mode database, matching the fault spectrum feature with the spectrum feature in the fault mode database, and obtaining the fault problem area and fault type of the ring main unit; A weight is assigned to the fault problem area, a score is assigned to the fault type, and the abnormal risk level of the ring main unit is obtained based on the weight of the fault problem area and the score of the fault type. When the abnormal risk level is not within the preset risk level range, an alarm is issued.

2. The DTU security monitoring method according to claim 1, characterized in that: Verifying the electrical parameters using a preset field-by-field verification technique to detect whether the electrical parameters are within a preset electrical parameter safety range; If the electrical parameters are not within the preset electrical parameter safety range, wavelet analysis is used to perform multi-scale decomposition on the electrical parameters to obtain potential hidden danger characteristic signals, including: Performing digital filtering on the electrical parameters to obtain digital filtering parameters; Calculate the root mean square value of the digital filter parameters to obtain the standardized electrical parameters; The standardized electrical parameters are verified field by field using a preset field-by-field verification technology to check whether the standardized electrical parameters are within the preset electrical parameter safety range. If any standardized electrical parameter exceeds the preset electrical parameter safety range, the standardized electrical parameters exceeding the electrical parameter safety range are decomposed at multiple scales using a preset wavelet basis function to obtain frequency components at different scales; wherein the frequency components include high-frequency components and low-frequency components; The frequency components are analyzed to identify abnormal fluctuations and mutation points, and obtain characteristic signals of potential hidden dangers.

3. The DTU security monitoring method according to claim 2, characterized in that: The frequency components are analyzed to identify abnormal fluctuations and mutation points, and obtain characteristic signals of potential hidden dangers, including: Normalizing the frequency components to obtain standardized frequency component signals; Based on a preset beamforming technology, focusing processing is performed on the standardized frequency component signal to obtain an enhanced target frequency component signal; Performing a change trend analysis on the target frequency component signal to obtain a fluctuation signal; Acquire a set abnormality detection threshold, and detect the fluctuation signal through the abnormality detection threshold to obtain a preliminary abnormal fluctuation signal; Through the preset recursive neural network model, the preliminary abnormal fluctuation signal is deeply learned to obtain a high-dimensional feature vector; Performing time series analysis on the high-dimensional feature vector to obtain a time series high-dimensional feature vector; Hidden dangers are identified on the time series high-dimensional feature vector to obtain potential hidden danger feature signals.

4. The DTU security monitoring method according to claim 1, characterized in that: The frequency domain feature extraction is performed on the potential hidden danger characteristic signal to obtain the fault spectrum feature, including: Performing time-frequency conversion on the potential hidden danger characteristic signal through a preset fast Fourier transform to obtain a preliminary spectrum diagram; Using a preset Hilbert-Huang transform algorithm, the spectrum density of the preliminary spectrum graph is calculated to generate a first spectrum density graph; wherein the first spectrum density graph is used to quantify the energy distribution of each frequency component; Using an independent component analysis algorithm, identifying key frequency components in the first spectral density graph; Using an autoencoder, identifying abnormal peaks in the first spectral density map; Combining the key frequency component and the abnormal peak, and locating the fault spectrum feature set in the first spectrum density map; The fault spectrum feature set is subjected to dimension reduction and screening to obtain the fault spectrum feature.

5. The DTU security monitoring method according to claim 4, characterized in that: Using a preset Hilbert-Huang transform algorithm to calculate the spectrum density of the preliminary spectrum map to generate a first spectrum density map, including: Using a preset Hilbert-Huang transform algorithm, preprocessing the preliminary spectrum graph to obtain a preprocessed spectrum graph; Decomposing the preprocessed spectrum to obtain a plurality of intrinsic mode functions, and extracting each of the intrinsic mode functions one by one to obtain each independent intrinsic mode function; Performing Hilbert transform on each independent eigenmode function to obtain the instantaneous frequency and instantaneous amplitude of each eigenmode function; Reconstructing the instantaneous frequency and instantaneous amplitude to obtain a reconstructed time-frequency signal; Calculating the spectrum density of the reconstructed time-frequency signal to obtain spectrum density data, wherein the spectrum density data is the energy distribution of the reconstructed time-frequency signal; Based on the spectrum density data, the preliminary spectrum graph is plotted to obtain a first spectrum density graph.

6. The DTU security monitoring method according to claim 1, characterized in that: Calling a preset fault mode database, matching the fault spectrum characteristics with the spectrum characteristics in the fault mode database, and obtaining the fault problem area and fault type of the ring main unit, including: Analyze the fault spectrum characteristics and the spectrum characteristics in the fault mode database respectively, and obtain a fault spectrum characteristic vector and a first spectrum characteristic vector accordingly; Performing Mahalanobis distance calculation on the fault spectrum feature vector and the first spectrum feature vector to obtain the Mahalanobis distance of the fault spectrum; If the Mahalanobis distance is within a preset distance range, correlation matching is performed on the fault spectrum feature vector and the first spectrum feature vector to obtain a matching result; wherein the matching result is a matching degree report; The fault spectrum characteristics are analyzed based on the matching report to obtain the fault problem area and fault type of the ring main unit.

7. The DTU security monitoring method according to claim 1, characterized in that: Assigning a weight to the fault problem area, assigning a score to the fault type, and obtaining an abnormal risk level of the ring main unit based on the weight of the fault problem area and the score of the fault type. When the abnormal risk level is not within a preset risk level range, an alarm is issued, including: Assigning weights to each of the fault problem areas, and performing vector conversion on the weights of each of the fault problem areas to obtain a vector of the fault problem area; Assigning scores to each of the fault types, and performing vector conversion on the scores of each of the fault types to obtain a vector of the fault type; Multiplying the vector of the fault problem area by the vector of the fault type to obtain a risk value of the fault; The abnormal risk level of the ring main unit is obtained based on the risk value of the fault, and an alarm is issued when the abnormal risk level is not within a preset risk level range.

8. A safety monitoring device for a DTU, characterized in that: include: An acquisition module is used to acquire electrical parameters of the ring main unit through the DTU; wherein the electrical parameters include voltage parameters, current parameters and power factor; A verification module, used to verify the electrical parameters using a preset field-by-field verification technology to detect whether the electrical parameters are within a preset electrical parameter safety range; if the electrical parameters are not within the preset electrical parameter safety range, the electrical parameters are decomposed at multiple scales using wavelet analysis to obtain a potential hidden danger characteristic signal; wherein the wavelet analysis is used to identify mutation points and abnormal fluctuations of the electrical parameters; An extraction module is used to extract the frequency domain features of the potential hidden danger characteristic signal to obtain the fault spectrum features; A matching module is used to call a preset fault mode database, match the fault spectrum characteristics with the spectrum characteristics in the fault mode database, and obtain the fault problem area and fault type of the ring main unit; An assignment module is used to assign weights to the fault problem areas, assign scores to the fault types, and obtain the abnormal risk level of the ring network cabinet based on the weights of the fault problem areas and the scores of the fault types. When the abnormal risk level is not within a preset risk level range, an alarm is issued.

9. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images