Switch cabinet double-frequency coupling partial discharge type dynamic identification method and system and storage medium

Abstract

The invention provides a switch cabinet dual-frequency coupling partial discharge type dynamic identification method and system and a storage medium, and the method comprises the steps: obtaining a potential partial discharge point position set of a target switch cabinet, and presetting a dual-frequency monitoring array according to the potential partial discharge point position set; deploying a synchronous acquisition system; executing dual-frequency coupling cooperative acquisition through the dual-frequency monitoring array and the acquisition system to obtain a dual-frequency signal pair array; performing space-time-frequency domain multi-dimensional feature dynamic fusion on the dual-frequency signal pair array; inputting the partial discharge characteristic parameter array fused with the space-time-frequency domain multi-dimensional characteristics into a preset partial discharge characteristic decision forest for discharge type discrimination; and carrying out dynamic weight adjustment on a discharge type judgment result. According to the invention, the space-time correlated dual-frequency monitoring array and acquisition system are deployed, and a three-level dynamic trigger mechanism is constructed, so that the cooperative capture capability of transient pulses and intermittent discharge is significantly enhanced, and the problem of leak detection of a single sensor on weak discharge signals is effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of partial discharge identification, and particularly to a method, system and storage medium for dynamically identifying the types of dual-frequency coupled partial discharges in switchgear cabinets. Background Art

[0002] As a key device in the power system, the insulation state of the switchgear cabinet is directly related to the power supply reliability. Partial discharge (PD) is an early sign of insulation deterioration. Accurately identifying the discharge types (such as surface discharge of insulators, internal air gap discharge, floating potential discharge, etc.) is crucial for fault warning and precise operation and maintenance.

[0003] Currently, the partial discharge monitoring of switchgear cabinets mainly relies on two types of sensing technologies: transient earth voltage (TEV) and ultra-high frequency (UHF). However, a single sensor (TEV or UHF) is vulnerable to electromagnetic interference and signal attenuation, and has insufficient ability to capture low-energy discharges or transient pulses. The traditional fixed-threshold triggering mechanism is difficult to adapt to the randomness and intermittency of discharge signals, resulting in a significant increase in the missed detection rate of weak discharge events. Moreover, existing methods mostly extract time-domain (such as pulse amplitude, phase) or frequency-domain (such as spectral centroid) features based on a single sensor, and it is difficult to comprehensively describe the physical essence of the discharge.

[0004] For example, TEV signals are sensitive to internal discharges but have low positioning accuracy, while UHF signals can be spatially located but are easily shielded by the cabinet structure. The lack of cross-modal features leads to insufficient discrimination of similar discharge types, such as a high misjudgment rate between surface discharge of insulators and metal tip discharge. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides a method, system and storage medium for dynamically identifying the types of dual-frequency coupled partial discharges in switchgear cabinets to solve the problems raised in the above background art.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A method for dynamically identifying the types of dual-frequency coupled partial discharges in switchgear cabinets includes the following steps:

[0007] Obtain a set of potential partial discharge points of the target switchgear cabinet, and preset a dual-frequency monitoring array according to the set of potential partial discharge points;

[0008] Deploy a synchronous acquisition system;

[0009] Among them, the acquisition system includes a reference clock, acquisition channels and an FPGA trigger logic unit;

[0010] Perform dual-frequency coupled collaborative acquisition through the dual-frequency monitoring array and the acquisition system to obtain a dual-frequency signal array;

[0011] Perform spatio-temporal and frequency-domain multi-dimensional feature dynamic fusion on the dual-frequency signal array;

[0012] Input the partial discharge feature parameter array integrating spatio-temporal and frequency-domain multi-dimensional features into a preset partial discharge feature decision forest for discharge type discrimination;

[0013] Perform dynamic weight adjustment on the result of discharge type discrimination to obtain the final recognition result of the discharge type;

[0014] Output the final recognition result of the discharge type.

[0015] As a further preference, the process of obtaining the set of potential partial discharge points of the target switchgear and presetting the dual-frequency monitoring array according to the set of potential partial discharge points includes:

[0016] Based on the three-dimensional structure model of the switchgear, the historical fault database, and electromagnetic field simulation analysis, interactively determine the set of potential positions where partial discharge occurs inside the target switchgear;

[0017] Design the dual-frequency sensor layout based on the set of potential discharge points, deploy a transient earth voltage sensor array at key points on the cabinet ground wire and metal surface, and deploy a UHF sensor array at electromagnetic wave leakage points such as cabinet gaps and observation windows;

[0018] Establish a sensor spatial position mapping model to ensure that each transient earth voltage monitoring point has a corresponding UHF monitoring point to form a spatial correlation pair;

[0019] Among them, when determining the set of potential positions of partial discharge, high-risk areas such as the surface of insulators, busbar connections, and cable terminals are key marked.

[0020] As a further preference, the process of obtaining the dual-frequency signal array includes:

[0021] Execute dual-frequency coupling collaborative acquisition, and the specific process of obtaining the dual-frequency signal array includes:

[0022] Perform dynamic discharge monitoring on the target switchgear, and adopt a three-level trigger strategy to obtain the dual-frequency signal time series array:

[0023] Among them, the three-level trigger strategy includes: start the main trigger when the pulse front of the UHF sensor detects that it exceeds the threshold;

[0024] Start the auxiliary trigger when the TEV sensor detects that the energy exceeds the threshold;

[0025] Start the joint trigger when the UHF and TEV signals satisfy the logical AND relationship;

[0026] Generate a dual - frequency signal array with time - scale alignment based on the dual - frequency signal time - series array, where each signal pair contains a TEV signal and a UHF signal that are time - synchronized.

[0027] As a further preference, the specific process of constructing an equilibrium model of switching cost and production efficiency and obtaining the optimal production path based on the equilibrium model includes:

[0028] Perform wavelet packet decomposition on the TEV signal in each signal pair, calculate its energy entropy, construct a Hankel matrix for singular value decomposition, extract the first 8 singular values as features, perform an S - transform on the UHF signal to obtain the time - frequency spectrum, and calculate the main frequency offset and bandwidth factor;

[0029] Calculate the cross - correlation function between the TEV and UHF signals in the same signal pair, extract the cross - correlation peak position as the propagation time - delay difference, calculate the logarithmic ratio of the UHF signal energy to the TEV signal energy, and use the time - delay difference information of several sensor pairs to establish a system of equations to solve for the three - dimensional space coordinates of the power source;

[0030] For consecutive multiple discharge pulses, statistically analyze the power - frequency phase and propagation time - delay difference of each pulse, construct a phase - time - delay joint histogram, and extract the phase distribution entropy and time - delay fluctuation coefficient from it;

[0031] After dynamically weighted fusion, obtain a local discharge characteristic parameter array that fuses multi - dimensional characteristics of space - time and frequency - domain.

[0032] As a further preference, the specific process of the dynamic feature weighted fusion includes:

[0033] Combine the single - modality feature, cross - modality joint feature, and dynamic PRPD feature into an 8 - dimensional feature vector;

[0034] According to the signal quality (such as the singular value stability of TEV and the signal - to - noise ratio of UHF), perform weighted adjustment on each feature in the feature vector;

[0035] Perform a sliding window process on the adjusted feature vector, scan the signal stream (feature vector) with a window length of 100 ms and a step size of 20 ms, and trigger processing when the number of valid pulses in the window ≥ 5;

[0036] Perform PCA on the feature matrix of all pulses in the window (dimension is n×8, n is the number of pulses), retain the first 5 - dimensional features with 95% energy information, form a feature parameter array after dimensionality reduction, and obtain a local discharge characteristic parameter array that fuses multi - dimensional characteristics of space - time and frequency - domain.

[0037] As a further preference, the decision - making root node of the local discharge characteristic decision - making forest is divided into a stable discharge branch and an unstable discharge branch using the standard deviation feature of the time - delay difference.

[0038] As a further preference, the specific process of dynamically adjusting the weights of the discharge type discrimination results includes:

[0039] Calculate the weight coefficients of UHF and TEV based on the signal quality of the dual - frequency monitoring array;

[0040] Among them, the calculation formulas for the weight coefficients of UHF and TEV are:

[0041] ;

[0042] ;

[0043] Among them, represents the UHF weight, represents the UHF signal - to - noise ratio, represents the normalized UHF noise energy, represents the TEV channel stability, represents the power - frequency harmonic distortion rate of the TEV channel, represents the TEV weight;

[0044] Fuse the weight coefficients of UHF and TEV with the results of the discharge type discrimination of the discharge feature decision forest probabilistically to obtain the final recognition result of the discharge type;

[0045] The probability fusion formula is:

[0046] ;

[0047] ;

[0048] Among them, represents the fusion probability result, represents the type probability output by the decision forest based on UHF features, represents the type probability output by the decision forest based on TEV features, represents the final fusion result.

[0049] As a further preference, the content of the final recognition result of the discharge type includes the discharge type, confidence level, weight assignment, key feature values, and signal quality indicators.

[0050] As a further preference, a dynamic recognition system for the dual - frequency coupled partial discharge type of switchgear, which is used to implement the above - mentioned dynamic recognition method for the dual - frequency coupled partial discharge type of switchgear, includes:

[0051] A three - dimensional discharge point positioning and array deployment module, which is used to obtain the set of potential partial discharge points of the target switchgear, and preset a dual - frequency monitoring array and deploy a synchronous acquisition system according to the set of potential partial discharge points;

[0052] A dual - frequency collaborative signal processing module is used to perform dual - frequency coupled collaborative acquisition through the dual - frequency monitoring array and the acquisition system, obtain a dual - frequency signal array, and perform spatio - temporal - frequency domain multi - dimensional feature dynamic fusion on the dual - frequency signal array.

[0053] A hierarchical discharge type decision module is used to input the partial discharge feature parameter array that fuses spatio - temporal - frequency domain multi - dimensional features into a preset partial discharge feature decision forest for discharge type discrimination.

[0054] A confidence level fusion and result output module is used to dynamically adjust the weights of the results of discharge type discrimination, obtain the final recognition result of the discharge type and output it.

[0055] As a further preference, a computer - readable storage medium stores a computer program thereon. When the computer program is executed by a processor, the steps of the above - mentioned method for dynamically identifying the dual - frequency coupled partial discharge type of a switchgear are implemented.

[0056] The present invention provides a method, a system and a storage medium for dynamically identifying the dual - frequency coupled partial discharge type of a switchgear, having the following beneficial effects: By deploying a spatio - temporally correlated dual - frequency monitoring array and an acquisition system, a three - level dynamic trigger mechanism is constructed, significantly enhancing the collaborative capture ability for transient pulses and intermittent discharges, and effectively solving the problem of missed detection of weak discharge signals by a single sensor; Based on a spatio - temporal - frequency domain multi - dimensional feature dynamic fusion mechanism, cross - modal features such as signal energy entropy, main frequency offset, and propagation time delay difference are extracted in parallel, and an adaptive fusion strategy weighted by signal quality is introduced to overcome the limitations of traditional single - feature characterization and achieve a holographic description of discharge characteristics; By constructing a hierarchical decision forest model and combining the threshold logic of key features such as propagation time delay difference stability, UHF frequency deviation range, and energy ratio, a hierarchical discrimination path for discharge types is established, significantly improving the discrimination accuracy for typical defects such as insulator surface discharge and internal air gap discharge; An innovative dual - channel probability dynamic fusion mechanism is designed to adjust the weight distribution in real - time according to the UHF signal - to - noise ratio and the TEV channel stability, and still ensure the recognition robustness in strong electromagnetic interference or signal attenuation scenarios; Combining PCA dimensionality reduction of pulse groups within a sliding time window and dynamic PRPD phase analysis, online tracking and early warning of discharge mode evolution are realized. This method breaks through the dependence of traditional partial discharge monitoring on continuously stable discharges, increases the intermittent discharge detection rate by more than 40%, and at the same time, through multi - dimensional feature fusion and dynamic decision - making mechanisms, realizes minute - level real - time discrimination of discharge types under complex working conditions, providing technical support for the active defense and precise operation and maintenance of switchgear insulation status. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a flow chart of the method for dynamically identifying the dual - frequency coupled partial discharge type of the switchgear of the present invention.

[0058] Figure 2 This is a block diagram of the switch cabinet dual-frequency coupled partial discharge type dynamic identification system of the present invention. DETAILED DESCRIPTION

[0059] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and are not to be construed as limiting the present invention.

[0060] The disclosure below provides many different embodiments or examples for realizing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below. Of course, they are merely examples and are not intended to limit the present invention. In addition, the present invention may repeat reference numbers and / or reference letters in different examples. Such repetition is for the purpose of simplicity and clarity and does not in itself indicate the relationship between the various embodiments and / or settings discussed. In addition, the present invention provides examples of various specific processes and materials, but those skilled in the art will recognize the application of other processes and / or the use of other materials.

[0061] like Figure 1 As shown, an embodiment of the present invention provides a method for dynamically identifying the type of dual-frequency coupled partial discharge in a switch cabinet, comprising the following steps:

[0062] S1: Obtain a set of potential partial discharge points of the target switchgear and preset a dual-frequency monitoring array based on the set of potential partial discharge points;

[0063] Specifically, the specific process of step S1 includes:

[0064] Based on the switchgear 3D structural model, historical fault database and electromagnetic field simulation analysis, the potential location set of partial discharge inside the target switchgear is interactively determined;

[0065] Design a dual-frequency sensor layout based on a set of potential discharge points: deploy a transient earth voltage (TEV) sensor array on key points of the cabinet grounding wire and metal surfaces; and deploy an ultra-high frequency (UHF) sensor array at electromagnetic wave leakage points such as cabinet gaps and observation windows.

[0066] A sensor spatial position mapping model is established to ensure that each transient ground voltage monitoring point has a corresponding ultra-high frequency monitoring point to form a spatial correlation pair.

[0067] Among them, when determining the potential location set of partial discharge, focus on marking high-risk areas such as insulator surfaces, busbar connections, and cable terminals.

[0068] S2: Deployment of synchronous acquisition system;

[0069] Among them, the acquisition system includes a reference clock, acquisition channels, and an FPGA trigger logic unit.

[0070] Specifically, configure a high-precision atomic clock to tame the clock source as the system reference clock, and build a dual-channel synchronous acquisition hardware architecture. Use a GPS-tamed clock as the reference clock source, transmit it to each acquisition channel through a clock distribution unit, and the FPGA trigger logic unit realizes cross-channel combined trigger control to ensure that the time synchronization accuracy of all sensor data is within 1 nanosecond.

[0071] It should be noted that the hardware configuration can be selected according to actual needs. The realization of cross-channel combined trigger control by the FPGA trigger logic unit belongs to existing technical means and will not be elaborated here.

[0072] S3: Perform dual-frequency coupling collaborative acquisition through the dual-frequency monitoring array and the acquisition system to obtain a dual-frequency signal array against the array;

[0073] Specifically, the specific process of performing dual-frequency coupling collaborative acquisition to obtain a dual-frequency signal array against the array includes:

[0074] Implement dynamic discharge monitoring on the target switchgear, and adopt a three-level trigger strategy to obtain a dual-frequency signal time series array:

[0075] Among them, the three-level trigger strategy includes: when the UHF sensor detects that the pulse front edge exceeds the threshold, start the main trigger;

[0076] When the TEV sensor detects that the energy exceeds the threshold, start the auxiliary trigger;

[0077] When the UHF and TEV signals satisfy the logical AND relationship, start the combined trigger;

[0078] Generate a dual-frequency signal array with time scale alignment based on the dual-frequency signal time series array, and each signal pair contains a TEV signal and a UHF signal with time synchronization.

[0079] S4: Perform spatio-temporal-frequency domain multi-dimensional feature dynamic fusion on the dual-frequency signal array against the array;

[0080] Specifically, the spatio-temporal-frequency domain multi-dimensional feature dynamic fusion extraction is specifically hierarchical feature fusion, and its specific process includes:

[0081] Parallel extraction of time-frequency features: Perform wavelet packet decomposition on the TEV signal in each signal pair, calculate its energy entropy, and construct a Hankel matrix for singular value decomposition, and extract the first 8 singular values as features; Perform an S transform on the UHF signal to obtain a time-frequency spectrum, and calculate the main frequency offset and bandwidth factor;

[0082] Cross-modal spatio-temporal feature fusion: Calculate the cross-correlation function of TEV and UHF signals in the same signal pair, and extract the cross-correlation peak position as the propagation time delay difference; Calculate the logarithmic ratio of the energy of the UHF signal to the energy of the TEV signal; Use the time delay difference information of several sensor pairs to establish a system of equations to solve the three-dimensional spatial coordinates of the discharge source;

[0083] Dual-frequency phase correlation analysis: For continuous multiple discharge pulses, count the power frequency phase and propagation time delay difference of each pulse, construct a phase-delay joint histogram, and extract the phase distribution entropy and time delay fluctuation coefficient from it;

[0084] Dynamic feature weighted fusion: Combine single-modal features, cross-modal joint features and dynamic PRPD features into an 8-dimensional feature vector;

[0085] According to the signal quality (such as the singular value stability of TEV and the signal-to-noise ratio of UHF), weight and adjust each feature in the feature vector;

[0086] Perform a sliding window process on the adjusted feature vector, scan the signal stream (feature vector) with a window length of 100 ms and a step size of 20 ms, and trigger processing when the number of valid pulses in the window ≥ 5;

[0087] Perform PCA on the feature matrix of all pulses in the window (dimension n×8, n is the number of pulses), retain the first 5-dimensional features with 95% energy information, form a reduced-dimensional feature parameter array, and obtain a partial discharge feature parameter array that fuses spatio-temporal and frequency-domain multi-dimensional features.

[0088] Specifically, simultaneously utilize time-domain (such as energy entropy), frequency-domain (such as main frequency shift, bandwidth factor), spatio-temporal domain (propagation time delay difference, spatial positioning) and phase statistical features to overcome the representational limitations of single features and comprehensively describe discharge characteristics; Dynamically weight the features through signal quality evaluation factors, reduce the weight of UHF when it is interfered, and reduce the weight when the TEV signal is weak to improve the robustness of the features; Different discharge types (such as internal discharge, surface discharge, corona discharge) show significant differences in features such as propagation time delay difference, energy ratio, and phase distribution, and the fused features can be effectively distinguished. The spatial positioning information not only provides the position of the discharge source, but its positioning residual can also be used as an index of the severity of insulation defects; The sliding time window process combined with the statistical features of pulses in the window can capture the dynamic evolution process of the discharge pattern.

[0089] S5: Input the partial discharge feature parameter array that fuses spatio-temporal and frequency-domain multi-dimensional features into a preset partial discharge feature decision forest for discharge type discrimination;

[0090] The decision tree root node of the partial discharge feature decision forest is divided into a stable discharge branch and an unstable discharge branch using the standard deviation feature of the time delay difference;

[0091] Exemplarily, the root node is divided using the time delay difference feature a:

[0092] When a < 2 ns, it enters the stable discharge branch;

[0093] When a ≥ 2 ns, it enters the unstable discharge branch;

[0094] Stable discharge branch (a < 2 ns):

[0095] The secondary node uses the TEV oscillation period feature (b):

[0096] If 1 ns < b < 5 ns: it enters the tertiary PRPD phase characteristic node (c);

[0097] c > 0.8: It is determined as insulator surface discharge;

[0098] c ≤ 0.8: It enters the quaternary energy ratio node (d);

[0099] d > -10 dB: It is determined as floating discharge;

[0100] d ≤ -10 dB: It is determined as internal air gap discharge;

[0101] If b ≤ 1 ns or ≥ 5 ns: It is marked as abnormal discharge;

[0102] Unstable discharge branch (a ≥ 2 ns):

[0103] The secondary node uses the UHF frequency deviation range feature (e):

[0104] 300 MHz < e < 600 MHz: It is determined as surface discharge along the surface;

[0105] e > 800 MHz: It is determined as tip discharge.

[0106] Among them, the hierarchical feature decision tree structure includes branches of typical discharge types such as insulator surface discharge, internal air gap discharge, and floating discharge.

[0107] It should be noted that each branch sets feature discrimination nodes: discrimination conditions such as the TEV oscillation period threshold (1 - 5 ns), the UHF frequency deviation range (300 - 1500 MHz), the energy ratio interval (-20 ~ 10 dB), the time delay difference standard deviation limit (< 2 ns), and the PRPD phase distribution characteristic (the surface discharge shows a double-peak characteristic).

[0108] S6: Dynamically adjust the weights of the results of the discharge type discrimination to obtain the final recognition result of the discharge type;

[0109] The specific process of dynamically adjusting the weights of the discharge type discrimination results includes:

[0110] Calculate the weight coefficients of UHF and TEV based on the signal quality of the dual-frequency monitoring array;

[0111] Among them, the calculation formulas for the weight coefficients of UHF and TEV are:

[0112] ;

[0113] ;

[0114] Among them, represents the UHF weight, represents the UHF signal-to-noise ratio, represents the normalized UHF noise energy, represents the TEV channel stability, represents the power frequency harmonic distortion rate of the TEV channel, represents the TEV weight;

[0115] Fuse the weight coefficients of UHF and TEV with the results of the discharge type discrimination of the discharge feature decision forest to obtain the final recognition result of the discharge type;

[0116] The probability fusion formula is:

[0117] ;

[0118] ;

[0119] [[ID=4,4]]Among them, represents the fusion probability result, represents the type probability output by the decision forest based on the UHF feature, represents the type probability output by the decision forest based on the TEV feature, represents the final fusion result.

[0120] It should be noted that the calculation formula for the UHF signal-to-noise ratio is: ; The calculation formula for the TEV channel stability is: ;

[0121] Among them, represents the peak amplitude of the partial discharge pulse, represents the standard deviation of the background noise signal, represents the standard deviation of the TEV baseline voltage, represents the rated voltage of the device.

[0122] S7: Output the final recognition result of the discharge type.

[0123] The content of the final recognition result of the discharge type includes: discharge type, confidence level, weight assignment, key feature values, and signal quality indicators.

[0124] The dynamic recognition method for the dual - frequency coupled partial discharge type of switchgear in this application deploys a spatio - temporally correlated dual - frequency monitoring array and acquisition system, constructs a three - level dynamic trigger mechanism, significantly enhances the collaborative capture ability for transient pulses and intermittent discharges, and effectively solves the problem of missed detection of weak discharge signals by a single sensor; based on a spatio - temporal - frequency domain multi - feature dynamic fusion mechanism, cross - modal features such as signal energy entropy, main frequency shift, and propagation time delay difference are extracted in parallel, and an adaptive fusion strategy weighted by signal quality is introduced to overcome the limitations of traditional single - feature characterization and achieve a holographic characterization of discharge characteristics; by constructing a hierarchical decision forest model, combined with the threshold logic of key features such as propagation time delay difference stability, UHF frequency deviation range, and energy ratio, a hierarchical discrimination path for the discharge type is established, significantly improving the discrimination accuracy for typical defects such as insulator surface discharge and internal air - gap discharge; an innovative dual - channel probability dynamic fusion mechanism is designed to adjust the weight assignment in real - time according to the UHF signal - to - noise ratio and the stability of the TEV channel, and still ensure the recognition robustness in strong electromagnetic interference or signal attenuation scenarios; combined with PCA dimensionality reduction of pulse groups within a sliding time window and dynamic PRPD phase analysis, online tracking and early warning of discharge mode evolution are realized. This method breaks through the dependence of traditional partial discharge monitoring on continuously stable discharges, increases the intermittent discharge detection rate by more than 40%, and at the same time, through multi - feature fusion and dynamic decision - making mechanisms, realizes minute - level real - time discrimination of discharge types under complex working conditions, providing core technical support for the active defense and precise operation and maintenance of switchgear insulation status.

[0125] As Figure 2 shown, this embodiment also provides a dynamic recognition system for the dual - frequency coupled partial discharge type of switchgear, used to implement the above - mentioned dynamic recognition method for the dual - frequency coupled partial discharge type of switchgear, including:

[0126] A three - dimensional discharge point positioning and array deployment module, used to obtain the set of potential partial discharge points of the target switchgear, and preset a dual - frequency monitoring array and deploy a synchronous acquisition system according to the set of potential partial discharge points;

[0127] A dual - frequency collaborative signal processing module, used to perform dual - frequency coupled collaborative acquisition through the dual - frequency monitoring array and acquisition system to obtain a dual - frequency signal array, and perform spatio - temporal - frequency domain multi - feature dynamic fusion on the dual - frequency signal array;

[0128] A hierarchical discharge type decision module, used to input the partial discharge feature parameter array that fuses spatio - temporal - frequency domain multi - features into a preset partial discharge feature decision forest for discharge type discrimination;

[0129] The confidence fusion and result output module is used to dynamically adjust the weights of the results of the discharge type discrimination, obtain the final recognition result of the discharge type and output it.

[0130] This embodiment also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned dynamic recognition method for the dual-frequency coupled partial discharge type of the switch cabinet are realized.

[0131] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A dynamic recognition method for the type of double - frequency coupled partial discharge in a switchgear, characterized in that, Including the following steps: Obtain the set of potential partial discharge points of the target switchgear cabinet, and preset a dual-frequency monitoring array according to the set of potential partial discharge points; Deploy a synchronous acquisition system; Among them, the acquisition system includes a reference clock, acquisition channels, and an FPGA trigger logic unit; Execute dual-frequency coupled collaborative acquisition through the dual-frequency monitoring array and the acquisition system to obtain a dual-frequency signal array; Perform spatio-temporal and frequency-domain multi-dimensional feature dynamic fusion on the dual-frequency signal array; Input the partial discharge feature parameter array that fuses spatio-temporal and frequency-domain multi-dimensional features into a preset partial discharge feature decision forest for discharge type discrimination; Perform dynamic weight adjustment on the result of the discharge type discrimination to obtain the final recognition result of the discharge type; Output the final recognition result of the discharge type.

2. The dynamic recognition method for the dual-frequency coupled partial discharge type of a switchgear cabinet according to claim 1, wherein The process of obtaining the set of potential partial discharge points of the target switchgear cabinet and presetting a dual-frequency monitoring array according to the set of potential partial discharge points includes: Based on the three-dimensional structure model of the switchgear cabinet, the historical fault database, and electromagnetic field simulation analysis, interactively determine the set of potential positions where partial discharge occurs inside the target switchgear cabinet; Design the layout of dual-frequency sensors based on the set of potential discharge points, arrange a transient earth voltage sensor array at key points on the cabinet grounding wire and metal surface, and arrange a very high frequency (VHF) sensor array at electromagnetic wave leakage points such as cabinet gaps and observation windows; Establish a sensor spatial position mapping model to ensure that each transient earth voltage monitoring point has a corresponding VHF monitoring point to form a spatial correlation pair; Among them, when determining the set of potential positions of partial discharge, high-risk areas such as the surface of insulators, bus connections, and cable terminals are key marked.

3. A dynamic recognition method for the type of dual-frequency coupled partial discharge in a switchgear cabinet according to claim 1, characterized in that, The process of obtaining the dual-frequency signal array includes: The specific process of performing dual-frequency coupled collaborative acquisition to obtain a dual-frequency signal array includes: Implement dynamic discharge monitoring on the target switchgear cabinet, and adopt a three-level trigger strategy to obtain a dual-frequency signal time series array: Among them, the three-level trigger strategy includes: start the main trigger when the UHF sensor detects that the pulse front edge exceeds the threshold; Start the auxiliary trigger when the TEV sensor detects that the energy exceeds the threshold; Start the joint trigger when the UHF and TEV signals satisfy the logical AND relationship; Generate a time-scale aligned dual-frequency signal array based on the dual-frequency signal time series array, and each signal pair includes a TEV signal and a UHF signal with time synchronization.

4. A dynamic recognition method for the type of dual-frequency coupled partial discharge in a switchgear cabinet according to claim 1, characterized in that, The specific process of constructing an equilibrium model of switching cost and production efficiency and obtaining the optimal production path based on the equilibrium model includes: Perform wavelet packet decomposition on the TEV signal in each signal pair, calculate its energy entropy, construct a Hankel matrix for singular value decomposition, extract the first 8 singular values as features, perform an S transform on the UHF signal to obtain a time-frequency spectrum, and calculate the main frequency offset and bandwidth factor; Calculate the cross-correlation function of the TEV and UHF signals in the same signal pair, extract the cross-correlation peak position as the propagation time delay difference, calculate the logarithmic ratio of the UHF signal energy to the TEV signal energy, and use the time delay difference information of several sensor pairs to establish an equation system to solve the three-dimensional spatial coordinates of the discharge source; For a series of consecutive discharge pulses, the power frequency phase and propagation time delay difference of each pulse are statistically analyzed, a phase-time delay joint histogram is constructed, and the phase distribution entropy and time delay fluctuation coefficient are extracted therefrom. After dynamically weighting and fusing the features, a partial discharge feature parameter array integrating spatio-temporal and frequency-domain multi-dimensional features is obtained.

5. A method for dynamically identifying the type of dual-frequency coupled partial discharge in a switchgear cabinet according to claim 1, characterized in that, The specific process of the dynamic feature weighting and fusion includes: Combining the single-modal features, cross-modal joint features, and dynamic PRPD features into an 8-dimensional feature vector; Weighting and adjusting each feature in the feature vector according to the signal quality; Performing a sliding window process on the adjusted feature vector, scanning the signal stream with a window length of 100 ms and a step size of 20 ms, and triggering the process when the number of valid pulses in the window ≥ 5; Performing PCA on the feature matrices of all pulses within the window, retaining the first 5 dimensions of features with 95% energy information, forming a feature parameter array after dimensionality reduction, and obtaining a partial discharge feature parameter array integrating spatio-temporal and frequency-domain multi-dimensional features.

6. A dynamic recognition method for the type of dual-frequency coupled partial discharge in a switchgear cabinet according to claim 1, characterized in that The decision tree root node of the partial discharge feature decision forest is divided into a stable discharge branch and an unstable discharge branch using the standard deviation feature of the time delay difference.

7. A dynamic recognition method for the double-frequency coupled partial discharge type of a switchgear cabinet according to claim 1, characterized in that The specific process of dynamically adjusting the weights of the results of the discharge type discrimination includes: Calculating the weight coefficients of UHF and TEV based on the signal quality of the dual-frequency monitoring array; Among them, the calculation formulas for the weight coefficients of UHF and TEV are: ; ; Among them, represents the UHF weight, represents the UHF signal-to-noise ratio and represents the normalized UHF noise energy, represents the TEV channel stability, represents the power frequency harmonic distortion rate of the TEV channel, represents the TEV weight; Performing probability fusion on the weight coefficients of UHF and TEV and the results of the discharge type discrimination by the partial discharge feature decision forest to obtain the final recognition result of the discharge type; The probability fusion formula is: ； ； Among them, represents the fusion probability result, represents the type probability output by the decision forest based on UHF features, represents the type probability output by the decision forest based on TEV features, represents the final fusion result.

8. A dynamic identification method for the type of dual-frequency coupled partial discharge in a switchgear cabinet according to claim 1, characterized in that The content of the final recognition result of the discharge type includes the discharge type, confidence level, weight distribution, key feature values, and signal quality indicators.

9. A dual-frequency coupled partial discharge type dynamic recognition system for switchgear, which is used to implement the dual-frequency coupled partial discharge type dynamic recognition method described in any one of claims 1-8, characterized in that, Including: A three-dimensional discharge point positioning and array deployment module, which is used to obtain a set of potential partial discharge points of the target switchgear cabinet, and preset a dual-frequency monitoring array and deploy a synchronous acquisition system according to the set of potential partial discharge points; A dual-frequency collaborative signal processing module, which is used to perform dual-frequency coupled collaborative acquisition through the dual-frequency monitoring array and the acquisition system to obtain a dual-frequency signal array, and perform spatio-temporal and frequency-domain multi-dimensional feature dynamic fusion on the dual-frequency signal array; A hierarchical discharge type decision module, which is used to input the partial discharge feature parameter array integrating spatio-temporal and frequency-domain multi-dimensional features into a preset partial discharge feature decision forest for discharge type discrimination; A confidence level fusion and result output module, which is used to dynamically adjust the weights of the results of the discharge type discrimination to obtain the final recognition result of the discharge type and output it.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the switchgear cabinet dual-frequency coupled partial discharge type dynamic recognition method according to any one of claims 1-8 are implemented.

Images