Electric power equipment abnormal discharge detection method and system based on discharge area segmentation, medium and processor

Abstract

The invention provides a power equipment abnormal discharge detection method based on discharge region segmentation. The method comprises the steps of obtaining time domain pulse information, pulse frequency domain spectral lines, corresponding space coordinates and holographic image data; performing feature processing to obtain a pulse waveform code and a frequency domain map; performing space field intensity mapping to obtain three-dimensional discharge field intensity distribution; constructing a first three-dimensional model by using the holographic image data, and obtaining a plurality of nodes after segmentation; establishing a first incidence matrix, a second incidence matrix and a third incidence matrix, and determining corresponding weights; performing joint processing on each matrix and the corresponding weight to obtain a preliminary segmentation result of discharge; determining a target node with abnormal discharge based on the preliminary segmentation result and the information of each node; and verification is performed based on the fault information of the preliminary segmentation result and the target node, and if the verification is passed, the preliminary segmentation result is taken as the target segmentation result, so that the accuracy and efficiency of abnormal discharge detection are improved, and the position of abnormal discharge can be accurately positioned.

Description

Technical Field

[0001] This invention relates to the field of discharge detection technology, and in particular to a method, system, medium, and processor for detecting abnormal discharge in power equipment based on discharge region segmentation. Background Technology

[0002] Abnormal discharges in electrical equipment, such as partial discharge, corona discharge, and arcing, are the main causes of insulation degradation and even failure. Traditional detection techniques mainly rely on the following methods, but all have limitations, as detailed below: 1) Pulse current method: Discharge pulses are detected by high-frequency current sensors, but it is susceptible to electromagnetic interference and cannot locate the discharge area.

[0003] 2) Ultrasonic detection: It is sensitive to mechanical vibration, but has low spatial resolution and is difficult to distinguish multi-source discharges.

[0004] 3) Ultraviolet imaging method: Visualizes the discharge location, but relies on manual interpretation and cannot automatically quantify the discharge intensity and range.

[0005] Existing technologies lack the ability to accurately segment and dynamically analyze discharge regions, resulting in high false detection rates, insufficient automation, and inadequate accuracy in locating abnormal positions.

[0006] Therefore, there is a need for a method, system, medium, and processor for detecting abnormal discharge of power equipment based on discharge region segmentation. Summary of the Invention

[0007] To address the problems of insufficient accuracy, high false detection rate, and inadequate automation in existing technologies for locating abnormal discharges, this invention provides a method, system, medium, and processor for detecting abnormal discharges in power equipment based on discharge region segmentation. This improves the accuracy and efficiency of abnormal discharge detection and enables precise location of the abnormal discharge. The specific technical solution is as follows: A method for detecting abnormal discharge in power equipment based on discharge region segmentation, comprising: S1: Obtain the time-domain pulse information, pulse frequency-domain spectrum, and spatial coordinates of the pulse corresponding to the power equipment; S2: Acquire holographic image data corresponding to the power equipment; S3: Perform feature extraction and feature encoding on the received time-domain pulse information to obtain the pulse waveform code; S4: Extract features from the received pulse frequency domain spectrum to obtain a frequency domain spectrum; S5: Perform spatial field intensity mapping on the received spatial coordinates and holographic image data to obtain the three-dimensional discharge field intensity distribution; S6: Construct a first three-dimensional model of the power equipment using the holographic image data, and segment the first three-dimensional model into nodes according to a preset segmentation precision to obtain multiple nodes of the power equipment; S7: Establish a first correlation matrix between the pulse waveform encoding and the frequency domain spectrum, a second correlation matrix between the frequency domain spectrum and the three-dimensional discharge field strength distribution, and a third correlation matrix between the pulse waveform encoding and the three-dimensional discharge field strength distribution; and determine the first weight of the first correlation matrix, the second weight of the second correlation matrix, and the third weight of the third correlation matrix. S8: Using a pulsed convolutional neural network, the first correlation matrix, the first weight, the second correlation matrix, the second weight, the third correlation matrix, and the third weight are jointly processed to obtain the preliminary segmentation result of the power equipment discharge; based on the location information in the preliminary segmentation result and the location information of each node, the target node where the abnormal discharge occurs is determined. S9: Based on the abnormal discharge fault information in the preliminary segmentation result and the target node, a rationality check is performed using the historical fault database. If the check passes, the preliminary segmentation result is used as the target segmentation result containing discharge information.

[0008] Furthermore, in step S9, the target segmentation result includes the location information of the discharge, the discharge intensity, the discharge development trend, and the risk level; Furthermore, in step S3, the process of extracting and encoding features from the received time-domain pulse information to obtain pulse waveform encoding includes the following steps: Based on the received time-domain pulse information, the rising and / or falling edge characteristics of the discharge pulse are converted into chain-like codes as pulse waveform codes.

[0009] Furthermore, in step S4, the step of extracting features from the received pulse frequency domain spectral lines to obtain a frequency domain spectrum includes the following steps: A frequency band importance weighting model is used to identify multiple characteristic frequency points in the pulse frequency domain spectrum within the range of 0.1Hz-10THz, and frequency domain maps are generated using the identified characteristic frequency points.

[0010] Furthermore, in step S2, the holographic image data includes electromagnetic field signals, optical signals, quantum state signals, and spatial topological signals.

[0011] Furthermore, the electromagnetic field signal includes terahertz waves; the optical signal includes visible light signals and ultraviolet light signals; the quantum state signal includes spin-polarized electron signals; and the spatial topological signal includes muon penetration imaging signals.

[0012] Furthermore, in step S5, the spatial field intensity mapping of the received spatial coordinates and holographic image data to obtain the three-dimensional discharge field intensity distribution includes the following steps: The distribution of internal discharge plasma in power equipment was determined by using muon penetration imaging signals; Terahertz waves were used to determine the rate of change of the spatial magnetic field. Using spin-polarized electron signals, the electron spin-sensitive field strength interface of power equipment can be determined; Using the visible light and ultraviolet light signals, a second three-dimensional model of the power equipment is constructed; By combining the internal discharge plasma distribution, the rate of change of the spatial magnetic field, the electron spin-sensitive field strength interface, and the second three-dimensional model, the three-dimensional discharge field strength distribution of the power equipment is determined.

[0013] A power equipment abnormal discharge detection system based on discharge region segmentation, applied to the above-described power equipment abnormal discharge detection method based on discharge region segmentation, includes: A quantum sensing array is used to acquire time-domain pulse information, pulse frequency-domain spectral lines, and spatial coordinates corresponding to pulses from power equipment. Holographic discharge imaging module, which is used to acquire holographic image data corresponding to power equipment; An edge computing module is used to extract and encode features from received time-domain pulse information to obtain pulse waveform encoding; extract features from received pulse frequency-domain spectra to obtain frequency-domain maps; map spatial field strengths between received spatial coordinates and holographic image data to obtain a three-dimensional discharge field strength distribution; construct a first three-dimensional model of the power equipment using the holographic image data, and segment the first three-dimensional model into nodes according to a preset segmentation precision to obtain multiple nodes of the power equipment; establish a first correlation matrix between the pulse waveform encoding and the frequency-domain map, a second correlation matrix between the frequency-domain map and the three-dimensional discharge field strength distribution, and a third correlation matrix between the pulse waveform encoding and the three-dimensional discharge field strength distribution; and determine the first weight of the first correlation matrix, the second weight of the second correlation matrix, and the third weight of the third correlation matrix. The cloud service platform is used to jointly process the first correlation matrix, the first weight, the second correlation matrix, the second weight, the third correlation matrix, and the third weight using a pulsed convolutional neural network to obtain a preliminary segmentation result of power equipment discharge; based on the location information in the preliminary segmentation result and the location information of each node, the target node where abnormal discharge occurs is determined; based on the abnormal discharge fault information in the preliminary segmentation result and the target node, a rationality check is performed using a historical fault database; if the check passes, the preliminary segmentation result is used as the target segmentation result containing discharge information.

[0014] A computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform the above-described power equipment abnormal discharge detection method based on discharge region segmentation.

[0015] A processor for running a program, wherein the program executes the above-described method for detecting abnormal discharge of power equipment based on discharge region segmentation.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. Improve the accuracy and efficiency of abnormal discharge detection, and accurately locate the position of abnormal discharge.

[0017] 2. The scheme disclosed in this application constructs a three-dimensional analysis framework of "time-domain dynamic characteristics - frequency-domain fingerprint spectrum - spatial field strength distribution" to achieve sub-millimeter-level precise segmentation of the discharge region.

[0018] 3. The solution disclosed in this application adopts a distributed architecture and utilizes edge computing modules and cloud services to process data collaboratively, which not only improves detection efficiency but also saves communication resources. Attached Figure Description

[0019] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0020] Figure 1 This is a flowchart illustrating a method for detecting abnormal discharge in power equipment based on discharge region segmentation. Figure 2 This is a schematic diagram of a power equipment abnormal discharge detection system based on discharge region segmentation. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] It should be understood that, when used in this application, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or collections thereof.

[0023] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0024] It should also be further understood that the term “and / or” as used in this application refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes these combinations.

[0025] This application addresses the shortcomings of current abnormal discharge detection methods for power equipment, such as low accuracy and inaccurate location positioning. It provides a method and device for detecting abnormal discharge in power equipment based on discharge region segmentation. This application utilizes a quantum sensing array and a holographic discharge imaging module to collect multi-dimensional information from the power equipment. An edge computing module processes this information to determine the pulse waveform encoding, frequency domain spectrum, and three-dimensional discharge field strength distribution, generating a three-dimensional model of the power equipment. The edge computing module then establishes a three-dimensional correlation matrix using the pulse waveform encoding, frequency domain spectrum, and three-dimensional discharge field strength distribution. This three-dimensional correlation matrix is ​​processed in a cloud service to obtain preliminary segmentation results. Based on the location information in the preliminary segmentation results and the location information of each node in the three-dimensional model, the target node experiencing abnormal discharge is determined. Finally, the cloud service uses a historical fault database to verify the rationality of the preliminary segmentation results. If the verification passes, the preliminary segmentation results are used as the target segmentation results. This disclosed solution constructs a three-dimensional analysis framework of "temporal dynamic features - frequency domain fingerprint spectrum - spatial field strength distribution," achieving sub-millimeter-level precise segmentation of the discharge region. In addition, this disclosure adopts a distributed architecture, utilizing edge computing modules and cloud services to collaboratively process data, which not only improves detection efficiency but also saves communication resources.

[0026] Example 1 like Figure 1 The diagram shows a flowchart of a power equipment abnormal discharge detection method based on discharge region segmentation, including: S1: The quantum sensing array is used to synchronously acquire the time-domain pulse information, pulse frequency-domain spectrum line and spatial coordinates corresponding to the power equipment, and the acquired information is transmitted to the quantum information receiving module; the quantum information receiving module transmits the received information to the edge computing module.

[0027] S2: Use the holographic discharge imaging module to acquire holographic image data corresponding to the power equipment, and transmit the holographic image data to the edge computing module.

[0028] Furthermore, in step S2, the holographic image data includes electromagnetic field signals, optical signals, quantum state signals, and spatial topological signals.

[0029] Furthermore, the electromagnetic field signal includes terahertz waves; the optical signal includes visible light signals and ultraviolet light signals; the quantum state signal includes spin-polarized electron signals; and the spatial topological signal includes muon penetration imaging signals.

[0030] Furthermore, spin-polarized electrons refer to quantum states where the electron spin direction (up or down) has a specific orientation. This polarization state can be achieved through an external magnetic field, optical pumping, or the inherent magnetism of the material. Introducing a magneto-optical crystal (such as yttrium iron garnet) into the imaging optical path converts the spin-polarized magnetic field into an optical phase difference. This difference is then acquired and converted using a holographic discharge imaging module to obtain the aforementioned quantum state signal. Alternatively, ultrafast laser pumping can be used to excite spin polarization, and its spatial distribution dynamics can be simultaneously detected through holographic imaging to obtain the aforementioned quantum state signal.

[0031] Furthermore, muon penetration imaging signals are the technical basis for imaging by utilizing the signal changes generated by the interaction of muons (a type of charged lepton) when they penetrate matter, and they have unique advantages in various detection fields.

[0032] S3: Perform feature extraction and feature encoding on the received time-domain pulse information to obtain the pulse waveform code.

[0033] Furthermore, in step S3, the process of extracting and encoding features from the received time-domain pulse information to obtain pulse waveform encoding includes the following steps: The edge computing module uses the received time-domain pulse information to convert the rising and / or falling edge characteristics of the discharge pulse into chain-like codes, which are then used as pulse waveform codes.

[0034] In addition, features can be extracted and encoded using adaptive time window segmentation. The segmentation window can also be dynamically adjusted according to the discharge intensity, for example, by adjusting the window size within the range of 1ns-1ms.

[0035] Furthermore, quantum random resonance can be used to enhance signal energy, and noise energy can be used to improve the signal-to-noise ratio of weak signals by more than 30dB.

[0036] S4: Extract features from the received pulse frequency domain spectrum to obtain the frequency domain spectrum.

[0037] Furthermore, in step S4, the step of extracting features from the received pulse frequency domain spectral lines to obtain a frequency domain spectrum includes the following steps: The frequency band importance weighting model in the edge computing module is used to identify multiple characteristic frequency points in the pulse frequency domain spectrum within the range of 0.1Hz-10THz, and the identified characteristic frequency points are used to generate a frequency domain spectrum.

[0038] Furthermore, in actual implementation, 842 feature frequency points can be extracted, and a frequency domain spectrum can be generated based on the information of the extracted 842 feature frequency points.

[0039] Furthermore, the aforementioned frequency band importance weighting model can specifically employ a neural network based on the Transformer architecture.

[0040] S5: Perform spatial field strength mapping on the received spatial coordinates and holographic image data to obtain the three-dimensional discharge field strength distribution.

[0041] Furthermore, in step S5, the spatial field intensity mapping of the received spatial coordinates and holographic image data to obtain the three-dimensional discharge field intensity distribution includes the following steps: The distribution of internal discharge plasma in power equipment was determined by using muon penetration imaging signals; Terahertz waves were used to determine the rate of change of the spatial magnetic field. Using spin-polarized electron signals, the electron spin-sensitive field strength interface of power equipment can be determined; Using the visible light and ultraviolet light signals, a second three-dimensional model of the power equipment is constructed; By combining the internal discharge plasma distribution, the rate of change of the spatial magnetic field, the electron spin-sensitive field strength interface, and the second three-dimensional model, the three-dimensional discharge field strength distribution of the power equipment is determined.

[0042] Distribution of internal discharge plasma rate of change of spatial magnetic field Field strength interface After registering to the coordinate system corresponding to the second 3D model, the 3D discharge field intensity distribution is obtained. The calculation formula is as follows: ; ; ; Let n be the dielectric constant tensor and n be the interface normal vector.

[0043] S6: Construct a first three-dimensional model of the power equipment using the holographic image data, and segment the first three-dimensional model into nodes according to a preset segmentation precision to obtain multiple nodes of the power equipment.

[0044] Furthermore, visible light signals, ultraviolet light signals, terahertz waves, etc., can be used to establish the first three-dimensional model of the power equipment.

[0045] S7: Establish a first correlation matrix between the pulse waveform encoding and the frequency domain spectrum, a second correlation matrix between the frequency domain spectrum and the three-dimensional discharge field strength distribution, and a third correlation matrix between the pulse waveform encoding and the three-dimensional discharge field strength distribution; and determine the first weight of the first correlation matrix, the second weight of the second correlation matrix, and the third weight of the third correlation matrix.

[0046] The first correlation matrix C is as follows: A = C × B; where A represents the pulse waveform code and B represents the frequency domain spectrum.

[0047] The second correlation matrix E is as follows: B = E × D; where D represents the three-dimensional discharge field strength distribution.

[0048] The third correlation matrix F has the following relationship: A = F × D.

[0049] S8: Using the pulse convolutional neural network in the cloud service platform, the first correlation matrix, the first weight, the second correlation matrix, the second weight, the third correlation matrix, and the third weight are jointly processed to obtain the preliminary segmentation result of the power equipment discharge; based on the location information in the preliminary segmentation result and the location information of each node, the target node where the abnormal discharge occurs is determined.

[0050] The above preliminary segmentation results were obtained by classification using a trained spiking convolutional neural network, which can be pre-trained.

[0051] Based on the location information in the preliminary segmentation results and the location information of each node, the target node for abnormal discharge is determined. That is, the location information in the preliminary segmentation results is matched with the location information of each node, and the node corresponding to the successfully matched location information is taken as the target node for abnormal discharge.

[0052] Furthermore, the loss function of the pulse convolutional neural network includes a regularized loss function term based on Maxwell's equations.

[0053] Furthermore, the regularization loss function term is as follows: ; Among them, ▽ conforms to Maxwell's equations; , This represents the electromagnetic field predicted by the neural network. ϵ represents the magnetic permeability; ϵ represents the dielectric constant. Indicates electrical conductivity; , This represents the trainable hyperparameters.

[0054] S9: Based on the abnormal discharge fault information in the preliminary segmentation result and the target node, a rationality check is performed using the historical fault database. If the check passes, the preliminary segmentation result is used as the target segmentation result containing discharge information.

[0055] Furthermore, in step S9, the target segmentation result includes the location information of the discharge, the discharge intensity, the discharge development trend, and the risk level.

[0056] Furthermore, the historical fault database stores historical abnormal discharge fault information for each node of each power device. If the discharge anomaly type of the target node is not stored in the historical fault database, the abnormal discharge fault information and the target node are sent to the display terminal for display. Technicians can use the display terminal to provide feedback on the verification results based on the displayed information. If the discharge anomaly type of the target node is stored in the historical fault database, the verification passes.

[0057] In specific implementation, the power equipment abnormal discharge detection method based on discharge region segmentation may further include the following steps: S10: Obtain the operating status information of the power equipment using cloud services, and determine the sampling frequency of the holographic discharge imaging module and the quantum sensing array based on the operating status information.

[0058] If the power equipment's operating status information indicates that the equipment is not in operation, a lower sampling frequency is set; if the information indicates that the equipment is operating at low power, a higher sampling frequency is set; and if the information indicates that the equipment is operating at high power, a higher sampling frequency is set. This design not only ensures the effectiveness of power equipment monitoring but also maximizes resource conservation.

[0059] In specific implementation, the power equipment abnormal discharge detection method based on discharge region segmentation may further include the following steps: S11: Generate augmented reality display data using the target segmentation results; send the augmented reality display data to the client display terminal.

[0060] AR overlays abnormal discharge detection results with the actual scene of the power equipment, marking the discharge points (such as corona and arc) on the surface of the power equipment in real time in the form of bright color blocks or 3D heat maps, thus solving the problem of "data being disconnected from the actual object" in traditional reports.

[0061] Example 2 like Figure 2The diagram shows a structural schematic of a power equipment abnormal discharge detection system based on discharge region segmentation, applied to the aforementioned power equipment abnormal discharge detection method based on discharge region segmentation, including: A quantum sensing array is used to acquire time-domain pulse information, pulse frequency-domain spectral lines, and spatial coordinates corresponding to pulses from power equipment. Furthermore, the quantum sensing array is disposed on the surface of the power equipment, for example, a 64×64 quantum sensor grid can be deployed on the surface of the power equipment.

[0062] Holographic discharge imaging module, which is used to acquire holographic image data corresponding to power equipment; An edge computing module is used to extract and encode features from received time-domain pulse information to obtain pulse waveform encoding; extract features from received pulse frequency-domain spectra to obtain frequency-domain maps; map spatial field strengths between received spatial coordinates and holographic image data to obtain a three-dimensional discharge field strength distribution; construct a first three-dimensional model of the power equipment using the holographic image data, and segment the first three-dimensional model into nodes according to a preset segmentation precision to obtain multiple nodes of the power equipment; establish a first correlation matrix between the pulse waveform encoding and the frequency-domain map, a second correlation matrix between the frequency-domain map and the three-dimensional discharge field strength distribution, and a third correlation matrix between the pulse waveform encoding and the three-dimensional discharge field strength distribution; and determine the first weight of the first correlation matrix, the second weight of the second correlation matrix, and the third weight of the third correlation matrix. The cloud service platform is used to jointly process the first correlation matrix, the first weight, the second correlation matrix, the second weight, the third correlation matrix, and the third weight using a pulsed convolutional neural network to obtain a preliminary segmentation result of power equipment discharge; based on the location information in the preliminary segmentation result and the location information of each node, the target node where abnormal discharge occurs is determined; based on the abnormal discharge fault information in the preliminary segmentation result and the target node, a rationality check is performed using a historical fault database; if the check passes, the preliminary segmentation result is used as the target segmentation result containing discharge information.

[0063] Example 3 A computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform the above-described power equipment abnormal discharge detection method based on discharge region segmentation.

[0064] Example 4 A processor for running a program, wherein the program executes the above-described method for detecting abnormal discharge of power equipment based on discharge region segmentation.

[0065] Compared with the prior art, the beneficial effects of this application are as follows: 1. Improve the accuracy and efficiency of abnormal discharge detection, and accurately locate the position of abnormal discharge.

[0066] 2. The scheme disclosed in this application constructs a three-dimensional analysis framework of "time-domain dynamic characteristics - frequency-domain fingerprint spectrum - spatial field strength distribution" to achieve sub-millimeter-level precise segmentation of the discharge region.

[0067] 3. The solution disclosed in this application adopts a distributed architecture and utilizes edge computing modules and cloud services to process data collaboratively, which not only improves detection efficiency but also saves communication resources.

[0068] This application provides a method for detecting abnormal discharge of power equipment based on discharge region segmentation, comprising: S1: acquiring time-domain pulse information, pulse frequency-domain spectrum lines, and spatial coordinates corresponding to the power equipment; S2: acquiring holographic image data corresponding to the power equipment; S3: performing feature extraction and feature encoding on the received time-domain pulse information to obtain pulse waveform encoding; S4: performing feature extraction on the received pulse frequency-domain spectrum lines to obtain a frequency domain spectrum; S5: performing spatial field strength mapping on the received spatial coordinates and holographic image data to obtain a three-dimensional discharge field strength distribution; S6: constructing a first three-dimensional model of the power equipment using the holographic image data, and performing node segmentation on the first three-dimensional model according to a preset segmentation precision to obtain multiple nodes of the power equipment; S7: establishing a first correlation matrix between the pulse waveform encoding and the frequency domain spectrum, and the frequency domain spectrum... The system defines a second correlation matrix between the domain map and the three-dimensional discharge field strength distribution, and a third correlation matrix between the pulse waveform encoding and the three-dimensional discharge field strength distribution; and determines the first weight of the first correlation matrix, the second weight of the second correlation matrix, and the third weight of the third correlation matrix; S8: using a pulse convolutional neural network to jointly process the first correlation matrix, the first weight, the second correlation matrix, the second weight, the third correlation matrix, and the third weight to obtain a preliminary segmentation result of the power equipment discharge; based on the location information in the preliminary segmentation result and the location information of each node, the target node where the abnormal discharge occurs is determined; S9: based on the abnormal discharge fault information in the preliminary segmentation result and the target node, a rationality check is performed using a historical fault database. If the check passes, the preliminary segmentation result is used as the target segmentation result containing discharge information. This improves the accuracy and efficiency of abnormal discharge detection and enables precise location of abnormal discharge.

[0069] Those skilled in the art will recognize that the units of the various examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of the invention.

[0070] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical functional division. In actual implementation, there may be other division methods, such as multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored.

[0071] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0072] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of this application.

Claims

1. A method for detecting abnormal discharge in power equipment based on discharge region segmentation, characterized in that, include: S1: Obtain the time-domain pulse information, pulse frequency-domain spectrum, and spatial coordinates of the pulse corresponding to the power equipment; S2: Acquire holographic image data corresponding to the power equipment; S3: Perform feature extraction and feature encoding on the received time-domain pulse information to obtain the pulse waveform code; S4: Extract features from the received pulse frequency domain spectrum to obtain a frequency domain spectrum; S5: Perform spatial field intensity mapping on the received spatial coordinates and holographic image data to obtain the three-dimensional discharge field intensity distribution; S6: Construct a first three-dimensional model of the power equipment using the holographic image data, and segment the first three-dimensional model into nodes according to a preset segmentation precision to obtain multiple nodes of the power equipment; S7: Establish a first correlation matrix between the pulse waveform encoding and the frequency domain spectrum, a second correlation matrix between the frequency domain spectrum and the three-dimensional discharge field strength distribution, and a third correlation matrix between the pulse waveform encoding and the three-dimensional discharge field strength distribution; and determine the first weight of the first correlation matrix, the second weight of the second correlation matrix, and the third weight of the third correlation matrix. S8: The first correlation matrix, the first weight, the second correlation matrix, the second weight, the third correlation matrix, and the third weight are jointly processed by the pulse convolutional neural network to obtain the preliminary segmentation results of the power equipment discharge. Based on the location information in the preliminary segmentation results and the location information of each node, the target node where the abnormal discharge occurred is determined. S9: Based on the abnormal discharge fault information in the preliminary segmentation result and the target node, a rationality check is performed using the historical fault database. If the check passes, the preliminary segmentation result is used as the target segmentation result containing discharge information.

2. The method for detecting abnormal discharge of power equipment based on discharge region segmentation according to claim 1, characterized in that, In step S9, the target segmentation result includes the location information of the discharge, the discharge intensity, the discharge development trend, and the risk level.

3. The method for detecting abnormal discharge of power equipment based on discharge region segmentation according to claim 1, characterized in that, In step S3, the process of extracting and encoding features from the received time-domain pulse information to obtain pulse waveform encoding includes the following steps: Based on the received time-domain pulse information, the rising and / or falling edge characteristics of the discharge pulse are converted into chain-like codes as pulse waveform codes.

4. The method for detecting abnormal discharge of power equipment based on discharge region segmentation according to claim 1, characterized in that, In step S4, the step of extracting features from the received pulse frequency domain spectral lines to obtain a frequency domain spectrum includes the following steps: A frequency band importance weighting model is used to identify multiple characteristic frequency points in the pulse frequency domain spectrum within the range of 0.1Hz-10THz, and frequency domain maps are generated using the identified characteristic frequency points.

5. The method for detecting abnormal discharge of power equipment based on discharge region segmentation according to claim 1, characterized in that, In step S2, the holographic image data includes electromagnetic field signals, optical signals, quantum state signals, and spatial topological signals.

6. The method for detecting abnormal discharge of power equipment based on discharge region segmentation according to claim 5, characterized in that, The electromagnetic field signal includes terahertz waves; the optical signal includes visible light and ultraviolet light; the quantum state signal includes spin-polarized electron signals; and the spatial topological signal includes muon penetration imaging signals.

7. The method for detecting abnormal discharge of power equipment based on discharge region segmentation according to claim 6, characterized in that, In step S5, the spatial field intensity mapping of the received spatial coordinates and holographic image data to obtain the three-dimensional discharge field intensity distribution includes the following steps: The distribution of internal discharge plasma in power equipment was determined by using muon penetration imaging signals; Terahertz waves were used to determine the rate of change of the spatial magnetic field. Using spin-polarized electron signals, the electron spin-sensitive field strength interface of power equipment can be determined; Using the visible light and ultraviolet light signals, a second three-dimensional model of the power equipment is constructed; By combining the internal discharge plasma distribution, the rate of change of the spatial magnetic field, the electron spin-sensitive field strength interface, and the second three-dimensional model, the three-dimensional discharge field strength distribution of the power equipment is determined.

8. A power equipment abnormal discharge detection system based on discharge region segmentation, characterized in that, The method for detecting abnormal discharge of power equipment based on discharge region segmentation as described in any one of claims 1 to 7 includes: A quantum sensing array is used to acquire time-domain pulse information, pulse frequency-domain spectral lines, and spatial coordinates corresponding to pulses from power equipment. Holographic discharge imaging module, which is used to acquire holographic image data corresponding to power equipment; An edge computing module is used to extract and encode features from received time-domain pulse information to obtain pulse waveform encoding; extract features from received pulse frequency-domain spectra to obtain frequency-domain maps; map spatial field strengths between received spatial coordinates and holographic image data to obtain a three-dimensional discharge field strength distribution; construct a first three-dimensional model of the power equipment using the holographic image data, and segment the first three-dimensional model into nodes according to a preset segmentation precision to obtain multiple nodes of the power equipment; establish a first correlation matrix between the pulse waveform encoding and the frequency-domain map, a second correlation matrix between the frequency-domain map and the three-dimensional discharge field strength distribution, and a third correlation matrix between the pulse waveform encoding and the three-dimensional discharge field strength distribution; and determine the first weight of the first correlation matrix, the second weight of the second correlation matrix, and the third weight of the third correlation matrix. The cloud service platform is used to jointly process the first correlation matrix, the first weight, the second correlation matrix, the second weight, the third correlation matrix, and the third weight using a pulsed convolutional neural network to obtain a preliminary segmentation result of power equipment discharge; based on the location information in the preliminary segmentation result and the location information of each node, the target node where abnormal discharge occurs is determined; based on the abnormal discharge fault information in the preliminary segmentation result and the target node, a rationality check is performed using a historical fault database; if the check passes, the preliminary segmentation result is used as the target segmentation result containing discharge information.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to perform the power equipment abnormal discharge detection method based on discharge region segmentation as described in any one of claims 1 to 7.

10. A processor, characterized in that, The processor is used to run a program, wherein the program executes the abnormal discharge detection method for power equipment based on discharge region segmentation as described in any one of claims 1 to 7.

Images