Substation switch cabinet partial discharge abnormity identification method and device

Abstract

The invention discloses a substation switch cabinet partial discharge abnormity identification method and device. The method comprises the following steps: obtaining in-cabinet partial discharge data of a first switch cabinet collected by an inner side ultrasonic partial discharge probe of a double-ultrasonic partial discharge sensor at the current moment; when the signal intensity of the partial discharge data in the switch cabinet is greater than or equal to a preset intensity threshold value, determining that the partial discharge of the first switch cabinet is abnormal; determining a first similarity between the first in-cabinet partial discharge data and the first out-cabinet partial discharge data and a second similarity between the first in-cabinet partial discharge data and the second out-cabinet partial discharge data; and according to the first similarity and / or the second similarity, determining the source of the partial discharge abnormity of the first switch cabinet. According to the scheme, through comparing the first similarity between the first in-cabinet partial discharge data and the first in-cabinet partial discharge data and the second similarity between the first in-cabinet partial discharge data and the second out-cabinet partial discharge data, environmental interference identification and in-cabinet partial discharge fault positioning based on the in-cabinet and out-cabinet ultrasonic wave partial discharge probes are realized. And moreover, the machine is adopted to identify the abnormity, and manpower resources are saved.

Description

Technical Field

[0001] The present invention relates to the field of substation monitoring, and in particular to a method and device for identifying partial discharge anomalies in a substation switch cabinet. Background Art

[0002] Currently, high-voltage switchgear failure rates remain high during substation operation and maintenance. According to industry reports and statistics, high-voltage switchgear is one of the most prone to failures in substations, accounting for 30%-50% of the overall failure rate of substation equipment.

[0003] Currently, substation equipment status management and status monitoring operations rely heavily on manual labor. With the current severely insufficient manpower-to-station ratio, overdue work and missed inspections are inevitable. When monitoring high-voltage switchgear, abnormal monitoring data, such as partial discharge data, is detected, it's difficult to accurately determine whether the problem is caused by a fault in the switchgear itself or external interference. Any abnormal partial discharge data requires manual inspection, which is labor-intensive. Summary of the Invention

[0004] The present invention aims to at least solve the technical problems existing in the prior art. To this end, the first aspect of the present invention provides a method for identifying partial discharge anomalies in a substation switch cabinet, the method comprising: Obtaining in-cabinet partial discharge data of a first switch cabinet collected at a current moment by an inner ultrasonic partial discharge probe of a dual ultrasonic partial discharge sensor, wherein the dual ultrasonic partial discharge sensor includes an inner ultrasonic partial discharge probe and an outer ultrasonic partial discharge probe, wherein the inner ultrasonic partial discharge probe is disposed in the switch cabinet of the substation, and the outer ultrasonic partial discharge probe is disposed outside the switch cabinet; When the signal strength of the partial discharge data in the cabinet is greater than or equal to a preset strength threshold, determining that the partial discharge of the first switch cabinet is abnormal; Obtaining first in-cabinet partial discharge data and first out-cabinet partial discharge data of the first switchgear within a target time period after the current moment, and obtaining second out-cabinet partial discharge data of at least one second switchgear located in the switch room of the first switchgear within the target time period; Determining a first similarity between the first in-cabinet partial discharge data and the first out-cabinet partial discharge data, and a second similarity between the first in-cabinet partial discharge data and the second out-cabinet partial discharge data; A source of the first switchgear partial discharge anomaly is determined according to the first similarity and / or the second similarity.

[0005] Optionally, determining a source of the first switchgear partial discharge anomaly according to the first similarity and / or the second similarity includes: If at least one of the first similarity and the second similarity is greater than or equal to a preset similarity threshold, it is determined that the abnormal partial discharge of the first switchgear is caused by external interference; If both the first similarity and the second similarity are smaller than the similarity threshold, it is determined that the abnormality of the first switch cabinet originates from within the cabinet.

[0006] Optionally, determining a first similarity between the first in-cabinet partial discharge data and the first out-cabinet partial discharge data includes: Obtaining a first partial discharge spectrum of partial discharge data in the first cabinet, and extracting a first eigenvector based on the first partial discharge spectrum; Obtaining a second partial discharge spectrum of partial discharge data in the second cabinet, and extracting a second eigenvector based on the second partial discharge spectrum; respectively determining similarities of identical features in the first feature vector and the second feature vector to obtain a plurality of sub-similarity; According to the preset weights of the various features, a weighted average is performed on the multiple sub-similarity levels, and a result of the weighted average is used as the first similarity level.

[0007] Optionally, the first eigenvector is one or more of a first peak frequency, a first valley frequency, a first periodicity, and a first slope of the partial discharge spectrum; the second eigenvector is one or more of a second peak frequency, a second valley frequency, a second periodicity, and a second slope of the partial discharge spectrum, and the similarity of each identical feature in the first eigenvector and the second eigenvector is determined separately to obtain multiple sub-similarity, including: determining a similarity between the first peak frequency and the second peak frequency to obtain a first sub-similarity; Determine a similarity between the first valley frequency and the second valley frequency to obtain a second sub-similarity; determining a similarity between the first periodicity and the second periodicity to obtain a third sub-similarity; A similarity between the first slope and the second slope is determined to obtain a fourth sub-similarity.

[0008] Optionally, determining the similarity between the first peak frequency and the second peak frequency to obtain a first sub-similarity includes: A cosine similarity between the first peak frequency and the second peak frequency is determined to obtain a first sub-similarity.

[0009] Optionally, after determining that the partial discharge of the first switch cabinet is abnormal, the method further includes: Obtaining first in-cabinet partial discharge data of the first switchgear within a target time period after the current moment, and obtaining second in-cabinet partial discharge data of at least one second switchgear located in a switch room of the first switchgear within the target time period; Determining a third similarity between the first in-cabinet partial discharge data and the second in-cabinet partial discharge data; If the third similarity is greater than or equal to the similarity threshold, it is determined that the source of the abnormal partial discharge of the first switch cabinet is external interference.

[0010] Optionally, after determining that the partial discharge of the first switch cabinet is abnormal, the method further includes: Sending a DRX configuration instruction to the partial discharge probes of the plurality of dual ultrasonic partial discharge sensors, wherein the DRX configuration instruction includes a DRX cycle duration and a synchronization frame sequence number; the DRX configuration instruction is used to instruct the partial discharge probes to start collecting partial discharge data in the DRX cycle corresponding to the synchronization frame sequence number; The obtaining of first in-cabinet partial discharge data and first out-cabinet partial discharge data of the first switch cabinet within a target time period after the current moment includes: Obtain the first cabinet internal partial discharge data and the first cabinet external partial discharge data collected by the inner ultrasonic partial discharge probe set in the first switch cabinet and the outer ultrasonic partial discharge probe set outside the first switch cabinet within the target time period after the start of the DRX cycle corresponding to the synchronization frame sequence number.

[0011] A second aspect of the present invention provides a device for identifying abnormal partial discharge in a substation switch cabinet, the device comprising: a first data acquisition module, configured to acquire in-cabinet partial discharge data of a first switch cabinet collected at a current moment by an inner ultrasonic partial discharge probe of a dual ultrasonic partial discharge sensor; the dual ultrasonic partial discharge sensor comprising an inner ultrasonic partial discharge probe and an outer ultrasonic partial discharge probe, the inner ultrasonic partial discharge probe being disposed in the switch cabinet of the substation, and the outer ultrasonic partial discharge probe being disposed outside the switch cabinet; an abnormality determination module, configured to determine that the first switch cabinet has an abnormal partial discharge when the signal strength of the partial discharge data in the cabinet is greater than or equal to a preset strength threshold; A second data acquisition module is configured to acquire first in-cabinet partial discharge data and first out-cabinet partial discharge data of the first switchgear within a target time period after the current moment, and acquire second out-cabinet partial discharge data of at least one second switchgear located in the switch room of the first switchgear within the target time period; a similarity determination module, configured to determine a first similarity between the first in-cabinet partial discharge data and the first out-cabinet partial discharge data, and a second similarity between the first in-cabinet partial discharge data and the second out-cabinet partial discharge data; A source determination module is configured to determine a source of the first switch cabinet partial discharge anomaly based on the first similarity and / or the second similarity.

[0012] Optionally, the source determination module is specifically configured to: If at least one of the first similarity and the second similarity is greater than or equal to a preset similarity threshold, it is determined that the abnormal partial discharge of the first switchgear is caused by external interference; If both the first similarity and the second similarity are smaller than the similarity threshold, it is determined that the abnormality of the first switch cabinet originates from within the cabinet.

[0013] Optionally, the similarity determination module is specifically configured to: Obtaining a first partial discharge spectrum of partial discharge data in the first cabinet, and extracting a first eigenvector based on the first partial discharge spectrum; Obtaining a second partial discharge spectrum of partial discharge data in the second cabinet, and extracting a second eigenvector based on the second partial discharge spectrum; respectively determining similarities of identical features in the first feature vector and the second feature vector to obtain a plurality of sub-similarity; According to the preset weights of the various features, a weighted average is performed on the multiple sub-similarity levels, and a result of the weighted average is used as the first similarity level.

[0014] Optionally, the similarity determination module is specifically configured to: determining a similarity between the first peak frequency and the second peak frequency to obtain a first sub-similarity; Determine a similarity between the first valley frequency and the second valley frequency to obtain a second sub-similarity; determining a similarity between the first periodicity and the second periodicity to obtain a third sub-similarity; A similarity between the first slope and the second slope is determined to obtain a fourth sub-similarity.

[0015] Optionally, the similarity determination module is specifically configured to: A cosine similarity between the first peak frequency and the second peak frequency is determined to obtain a first sub-similarity.

[0016] Optionally, the device further comprises: a partial discharge data acquisition module, configured to acquire first in-cabinet partial discharge data of the first switchgear within a target time period after the current moment, and acquire second in-cabinet partial discharge data of at least one second switchgear located in the switch room of the first switchgear within the target time period; a third similarity determination module, configured to determine a third similarity between the first in-cabinet partial discharge data and the second in-cabinet partial discharge data; An interference determination module is configured to determine that the source of the partial discharge anomaly of the first switch cabinet is external interference if the third similarity is greater than or equal to the similarity threshold.

[0017] Optionally, the device further comprises: A configuration instruction sending module is used to send a DRX configuration instruction to the partial discharge probes of the plurality of dual ultrasonic partial discharge sensors, wherein the DRX configuration instruction includes a DRX cycle duration and a synchronization frame sequence number; the DRX configuration instruction is used to instruct the partial discharge probe to start collecting partial discharge data in the DRX cycle corresponding to the synchronization frame sequence number; The second data acquisition module is specifically used to: Obtain the first cabinet internal partial discharge data and the first cabinet external partial discharge data collected by the inner ultrasonic partial discharge probe set in the first switch cabinet and the outer ultrasonic partial discharge probe set outside the first switch cabinet within the target time period after the start of the DRX cycle corresponding to the synchronization frame sequence number.

[0018] A third aspect of the present invention provides a computing device, comprising: one or more processors; Memory; and One or more devices comprising instructions for executing the method according to the first aspect.

[0019] A fourth aspect of the present invention provides a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions that, when executed by a computing device, cause the computing device to perform the method described in the first aspect.

[0020] The embodiments of the present invention have the following beneficial effects: In an embodiment of the present invention, in-cabinet partial discharge data of a first switch cabinet collected by an inner ultrasonic partial discharge probe of a dual ultrasonic partial discharge sensor at a current moment is obtained; the dual ultrasonic partial discharge sensor includes an inner ultrasonic partial discharge probe and an outer ultrasonic partial discharge probe, the inner ultrasonic partial discharge probe is arranged in the switch cabinet of the substation, and the outer ultrasonic partial discharge probe is arranged outside the switch cabinet; when the signal strength of the in-cabinet partial discharge data is greater than or equal to a preset intensity threshold, it is determined that the first switch cabinet has a partial discharge abnormality; first in-cabinet partial discharge data and first external partial discharge data of the first switch cabinet within a target time period after the current moment are obtained, and second external partial discharge data of at least one second switch cabinet located in a switch room of the first switch cabinet within the target time period are obtained; a first similarity between the first in-cabinet partial discharge data and the first external partial discharge data, and a second similarity between the first in-cabinet partial discharge data and the second external partial discharge data are determined; and a source of the partial discharge abnormality of the first switch cabinet is determined based on the first similarity and / or the second similarity. This solution compares the first similarity between the first in-cabinet PD data and the second in-cabinet PD data, and the second similarity between the first in-cabinet PD data and the second out-cabinet PD data. This enables identification of environmental interference and location of in-cabinet PD faults using ultrasonic PD probes inside and outside the cabinet. Furthermore, machine-based anomaly detection saves human resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A schematic diagram of a dual ultrasonic partial discharge sensor deployment according to an embodiment of the present invention; Figure 2 This is a flowchart of a method for identifying partial discharge anomalies in a substation switch cabinet provided by an embodiment of the present invention; Figure 3 This is a structural block diagram of a device for identifying partial discharge anomalies in a substation switch cabinet provided by an embodiment of the present invention; Figure 4 A schematic diagram of a computing device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0023] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present disclosure, unless otherwise specified, "multiple" means two or more. In addition, the use of "based on" or "according to" means openness and inclusiveness, because the process, steps, calculations or other actions "based on" or "according to" one or more of the conditions or values may be based on additional conditions or values beyond the stated in practice.

[0024] According to one embodiment of the present invention, in order to monitor the substation, the present invention sets up the cloud side, edge side and terminal side in a cloud-edge collaborative manner.

[0025] The cloud-edge collaborative computing platform is an intelligent computing architecture developed to address the new challenges brought about by the development of the Internet of Things (IoT). By managing digital cloud computing, the platform shifts computing power from the centralized cloud to edge nodes close to data sources, optimizing data processing capabilities through a distributed computing model. This cloud-edge collaborative computing model not only extends cloud-native capabilities but also, through the deployment of edge nodes, enables data processing, business applications, and artificial intelligence (AI) models to be executed at the edge, close to data sources. This addresses challenges faced by IoT implementation, such as real-time response, data privacy, and ease of maintenance.

[0026] The cloud-edge collaborative computing platform has significantly improved data processing efficiency and system response speed in practical applications across multiple industries, enabling diverse intelligent applications and facilitating digital transformation across various sectors. The cloud can be used for advanced analysis and global optimization.

[0027] The cloud, edge and end sides together constitute the substation monitoring system.

[0028] Traditional switch cabinets face problems such as long partial discharge detection cycles, heavy workload, and high false alarm rates for equipment status detection. Currently, grassroots business personnel lack effective means and are unable to detect abnormal equipment status in a timely and accurate manner. The present invention deploys dual ultrasonic partial discharge sensors in substation switch cabinets and develops a substation switch cabinet partial discharge cabinet internal and external identification technology based on dual ultrasonic partial discharge sensors. This can effectively solve the problem of false alarms in traditional switch cabinet partial discharge monitoring, help operation and maintenance personnel quickly and accurately check whether the partial discharge fault is inside the cabinet, greatly improve the ability to accurately locate partial discharges, eliminate environmental interference outside the cabinet, reduce the equipment operation and maintenance workload of grassroots business personnel, improve operation and maintenance efficiency, and ensure power supply service reliability. The present invention also realizes autonomous control of the software and hardware platform through sensors such as the autonomously deployed dual ultrasonic partial discharge sensors, thereby improving the safety of substation operation.

[0029] The terminal side may include dual ultrasonic partial discharge sensors for substation monitoring. According to one embodiment of the present invention, the dual ultrasonic partial discharge sensors can be deployed on the panel of a switchgear, such as a switchgear panel in a 10kV switch room. The switchgear can be implemented as a high-voltage switchgear. The present invention does not limit the specific type of switchgear.

[0030] Specifically, the present invention can deploy dual ultrasonic partial discharge sensors on the wall of the switch room and on the cabinet surface of the switch cabinet.

[0031] Figure 1 A schematic diagram of a dual ultrasonic partial discharge sensor deployment provided by an embodiment of the present invention. Figure 1 As shown, the dual ultrasonic PD sensor includes two PD probes: an inner ultrasonic PD probe and an outer ultrasonic PD probe. The inner ultrasonic PD probe collects ultrasonic PD data from inside the cabinet, known as the in-cabinet PD data, while the outer ultrasonic PD probe collects ultrasonic PD data from outside the cabinet, known as the out-cabinet PD data. The inner portion of the dual ultrasonic PD sensor can be deployed actively, while the outer portion, such as the outer ultrasonic PD probe, can be magnetically deployed in the gaps between the cable compartment and the circuit breaker compartment outside the cabinet to identify PD inside and outside the cabinet.

[0032] Exemplarily, the parameters of the dual ultrasonic partial discharge sensor include: ultrasonic monitoring range 20-100kHz, center frequency 40Khz, ultrasonic sensitivity 0dBμV; wireless communication RF transmission power 15-17dBm, receiving sensitivity -109dBm, time synchronization error +-20μS.

[0033] The computing device in the present invention can be specifically implemented as a digital node device. The present invention's method for identifying partial discharges inside and outside a substation switchgear can be specifically executed by a digital node device deployed in the substation building. The digital node device connects northbound to the distribution network cloud master station and southbound to one or more end-side devices, such as one or more dual ultrasonic partial discharge sensors. The digital node device can be deployed in a 10KV switch room. The present invention does not limit the specific deployment method or location of the digital node device.

[0034] The deployment methods of digital node devices include fixed digital node devices and portable digital node devices.

[0035] Fixed digital node equipment can be deployed in a cabinet in the switch room and powered by a fixed power supply. Fixed digital node equipment is suitable for long-term monitoring of substations.

[0036] Portable digital node equipment is removable and reusable. It is suitable for short-term monitoring of substations and can be powered by a mobile power supply.

[0037] According to one embodiment of the present invention, the present invention can be divided into power-guarantee substations and non-power-guarantee substations according to the different operation and maintenance levels of the substations. The operation and maintenance level of the power-guarantee substation is higher than that of the non-power-guarantee substation. The present invention can set substations with high load importance, which need to be focused on, or substations that undertake some active guarantee tasks as power-guarantee substations, and the remaining substations as non-power-guarantee substations. The present invention can deploy fixed digital node equipment in power-guarantee substations and deploy portable digital node equipment in non-power-guarantee substations.

[0038] Digital node devices collect partial discharge sensor data using the Internet of Things (IoT) protocol for power transmission and transformation equipment and synchronize sensor time using the DRX scheduler. Partial discharge sensor data includes data collected by dual ultrasonic partial discharge sensors, including both in-cabinet and out-cabinet partial discharge data.

[0039] According to one embodiment of the present invention, the present invention does not limit the number of dual ultrasonic partial discharge sensors that can be connected to the digital node device, for example, it can be no less than 36. The digital node device supports data acquisition from various types of wired and wireless sensors, such as IEC61850, MQTT, Modbμs, wireless networking protocol for IoT node devices of power transmission and transformation equipment, and micro-power wireless network communication protocol for IoT of power transmission and transformation equipment; supports Ethernet port, RS-485 / RS-232, wireless 4G+APN, and LoRa wireless hardware communication, and supports wireless access capabilities such as 4G and 5G, as well as wired access capabilities such as optical fiber and Ethernet. The digital node device can access the distribution network cloud master station via the IEC104 protocol. The present invention does not limit the specific connection method between the digital node device and the distribution network cloud master station and the dual ultrasonic partial discharge sensor.

[0040] The cloud side includes a distribution network cloud master station. This station can be implemented as one or more computing devices. The present invention does not limit the specific components of the distribution network cloud master station. The distribution network cloud master station can be deployed within or outside the substation, communicating with digital node devices deployed in the substation building. The distribution network cloud master station can display data collected by dual ultrasonic partial discharge sensors in the switchroom, generate partial discharge alarms within the switchgear, and generate external interference signal alarms.

[0041] Figure 2 The present invention provides a flowchart of the steps of a method for identifying partial discharge anomalies in a substation switch cabinet.

[0042] This method is applied to the above-mentioned digital node equipment. Figure 2 As shown, the method includes the following steps: Step 101: Obtain in-cabinet partial discharge data of a first switch cabinet collected at the current moment by an inner ultrasonic partial discharge probe of a dual ultrasonic partial discharge sensor; the dual ultrasonic partial discharge sensor includes an inner ultrasonic partial discharge probe and an outer ultrasonic partial discharge probe, the inner ultrasonic partial discharge probe is arranged in the switch cabinet of the substation, and the outer ultrasonic partial discharge probe is arranged outside the switch cabinet.

[0043] Multiple dual ultrasonic PD sensors are deployed in substations. These sensors consist of an inner ultrasonic PD probe and an outer ultrasonic PD probe. The inner ultrasonic PD probe is located inside the substation's switchgear and collects PD data from inside the switchgear. The outer ultrasonic PD probe is located outside the switchgear and collects PD data from outside the switchgear.

[0044] A partial discharge signal is a signal generated by partial discharge in electrical equipment. Partial discharge occurs when the electric field strength in a local area of high-voltage equipment exceeds the breakdown strength of the dielectric due to factors such as electric field concentration or insulation material defects, resulting in instantaneous discharge in that area. This discharge is usually weak but can potentially affect the insulation condition of the equipment.

[0045] PD data includes unstructured data, such as PD pulse signals.

[0046] Step 102: When the signal strength of the partial discharge data in the cabinet is greater than or equal to a preset strength threshold, determine that the partial discharge of the first switch cabinet is abnormal.

[0047] Because switchgear is deployed in an environment that includes not only the switchgear but also other electrical appliances, such as lights and cameras, abnormalities in these components outside the switchgear, such as light flickering, can also generate partial discharges. These partial discharge signals are also detected by the partial discharge sensors monitoring the switchgear.

[0048] When components of other electrical appliances outside the switch cabinet produce abnormalities, the partial discharge they generate is superimposed on the normal partial discharge of the switch cabinet, thereby enhancing the signal strength of the partial discharge data inside the cabinet and the partial discharge data outside the cabinet.

[0049] Since the PD data outside the cabinet is easily affected by slight disturbances other than electrical appliances, in order to improve accuracy, the intensity of the PD data inside the cabinet is used to determine whether there is a PD anomaly.

[0050] When determining whether a switchgear exhibits abnormal PD based on the PD data within the cabinet, a PD reading of 8dBμV or higher and 20dBμV or lower indicates a concern level; a PD reading of 20dBμV or higher and 30dBμV or lower indicates a critical level; and a PD reading of 30dBμV or higher indicates a maintenance-required level. When the digital node equipment detects PD data within these ranges, it generates general, critical, and crisis alerts, respectively, to alert maintenance personnel. Therefore, a PD reading of 8dBμV or higher indicates an abnormal PD level.

[0051] Step 103: Obtain first in-cabinet partial discharge data and first out-cabinet partial discharge data of the first switch cabinet within a target time period after the current moment, and obtain second out-cabinet partial discharge data of at least one second switch cabinet located in the switch room of the first switch cabinet within the target time period.

[0052] Multiple dual ultrasonic PD sensors are deployed in a substation. Each dual ultrasonic PD sensor may have an independent start time and frequency for data collection. Therefore, the PD data collected by these dual ultrasonic PD sensors may not be collected simultaneously. For example, the PD data for the first switchgear may be collected at one time, and the PD data for the second switchgear may be collected at a second time. Or, the PD data inside the switchgear may be collected at one time, while the PD data outside the switchgear may be collected at a second time. If the PD data for the second switchgear is abnormal at the second time, it will be impossible to compare the data at different times to determine whether the abnormal PD is caused by interference such as external noise.

[0053] To this end, the present invention generates a time synchronization instruction from a digital node device, causing the inner and outer probes of all dual ultrasonic partial discharge sensors to begin collecting data at the same acquisition time and frequency specified by the next synchronized acquisition time. This allows the digital node device to obtain the first internal and first external partial discharge data for the first switchgear cabinet within the target time period.

[0054] A switch room is equipped with multiple switch cabinets. In addition to a first switch cabinet, there are multiple second switch cabinets. At least one second switch cabinet is selected. Specifically, one or multiple second switch cabinets can be selected. External partial discharge data of the second switch cabinet within a target time period is obtained to obtain the second external partial discharge data.

[0055] Step 104: Determine a first similarity between the first in-cabinet partial discharge data and the first out-cabinet partial discharge data, and a second similarity between the first in-cabinet partial discharge data and the second out-cabinet partial discharge data.

[0056] By comparing the first similarity between the partial discharge data inside the first cabinet and the partial discharge data outside the first cabinet, and comparing the second similarity between the partial discharge data inside the first cabinet and the partial discharge data outside the second cabinet, it can be determined whether the partial discharge abnormality is caused by external interference outside the cabinet or by a fault inside the switch cabinet.

[0057] Specifically, the in-cabinet feature vector can be determined based on the in-cabinet partial discharge spectrum of the in-cabinet partial discharge data, and the out-cabinet feature vector can be determined based on the out-cabinet partial discharge spectrum of the out-cabinet partial discharge data, and then the first similarity and the second similarity can be calculated based on the in-cabinet feature vector and the out-cabinet feature vector.

[0058] As an optional embodiment, determining a first similarity between the first in-cabinet partial discharge data and the first out-cabinet partial discharge data includes: Step 301: Obtain a first partial discharge spectrum of partial discharge data in the first cabinet, and extract a first eigenvector based on the first partial discharge spectrum; Step 302: Obtain a second partial discharge spectrum of the partial discharge data in the second cabinet, and extract a second eigenvector based on the second partial discharge spectrum; Step 303: respectively determine the similarity of each identical feature in the first feature vector and the second feature vector to obtain a plurality of sub-similarity; Step 304: Perform weighted averaging on the multiple sub-similarity scores based on preset weights of the features, and use the weighted averaging result as the first similarity score.

[0059] In steps 301-304, the PD data includes unstructured data, such as PD pulse signals. A PD map of the switchgear PD data can be drawn based on the PD pulse signals. PD maps of the PD data include PD phase distribution (PRPD) and PD pulse signal (PRPS). The present invention does not limit the specific type of PD map drawn from the PD data.

[0060] A first partial discharge spectrum of partial discharge data in a first cabinet is obtained, and a plurality of first characteristic vectors are extracted according to the first partial discharge spectrum; a second partial discharge spectrum of partial discharge data in a second cabinet is obtained, and a plurality of second characteristic vectors are extracted according to the second partial discharge spectrum.

[0061] As an optional embodiment, the first eigenvector is one or more of the first peak frequency, the first valley frequency, the first periodicity, and the first slope of the partial discharge spectrum; the second eigenvector is one or more of the second peak frequency, the second valley frequency, the second periodicity, and the second slope of the partial discharge spectrum.

[0062] The similarities of the same features in the first feature vector and the second feature vector are respectively determined to obtain a plurality of sub-similarity values; the plurality of sub-similarity values are weighted averaged according to preset weights of the features, and the weighted average result is used as the first similarity value.

[0063] As an optional embodiment, step 303 includes: Step 401: Determine the similarity between the first peak frequency and the second peak frequency to obtain a first sub-similarity; Step 402: Determine the similarity between the first valley frequency and the second valley frequency to obtain a second sub-similarity; Step 403: Determine the similarity between the first periodicity and the second periodicity to obtain a third sub-similarity; Step 404: Determine the similarity between the first slope and the second slope to obtain a fourth sub-similarity.

[0064] In steps 401-404, the peak frequencies corresponding to one or more peaks can be determined based on the partial discharge spectrum. Each peak corresponds to a peak in the waveform of the partial discharge spectrum. Each peak frequency is then used as a feature vector. It is understood that the number of peaks in the partial discharge spectrum can also be used as a feature vector.

[0065] The valley frequencies corresponding to one or more valleys can be determined based on the partial discharge spectrum. Each valley corresponds to a trough in the waveform of the partial discharge spectrum. Each valley frequency is then used as a feature vector. It is understood that the number of valleys present can also be used as a feature vector.

[0066] A characteristic vector may be determined according to whether the partial discharge spectrum has periodicity, and a characteristic vector may be determined according to the number of periods the partial discharge spectrum has.

[0067] Determining the characteristic vector according to the slope includes calculating the slope according to a group of adjacent peaks and troughs in the partial discharge spectrum to obtain a plurality of characteristic vectors.

[0068] According to an embodiment of the present invention, when calculating the first similarity based on the first eigenvector and the second eigenvector, the similarity corresponding to each eigenvector may be calculated.

[0069] For example, the peak frequencies in the first partial discharge spectrum include 50 Hz and 100 Hz, and the following first eigenvector can be determined: 50 Hz, 100 Hz, and 2 peaks; the peak frequencies in the second partial discharge spectrum include 48 Hz and 102 Hz, and the following second eigenvector can be determined: 48 Hz, 102 Hz, and 2 peaks.

[0070] The similarity can be calculated based on the peak frequencies and the number of peaks that appear in sequence, such as calculating the similarity based on 50 Hz and 48 Hz, calculating the sub-similarity based on 100 Hz and 102 Hz, and calculating the sub-similarity based on "2 peaks" and "2 peaks".

[0071] After calculating the sub-similarity of all types of eigenvectors, a weighted average of the multiple sub-similarity scores can be performed to obtain a first similarity. Specifically, weights can be set for peak frequency type eigenvectors, valley frequency type eigenvectors, periodic type eigenvectors, and slope type eigenvectors. The similarity of the partial discharge data is then determined based on the calculated similarity values and weights for each type of eigenvector. The present invention does not limit the specific method for setting the weights for each type of eigenvector and the specific values they take.

[0072] According to one embodiment of the present invention, when calculating the similarity of partial discharge data, an image similarity calculation model can also be pre-trained. The image similarity calculation model can automatically extract the features of each partial discharge pattern and perform comparisons. The present invention does not limit the specific training method or model type of the image similarity calculation model. For example, a deep neural network model can be used to train the image similarity calculation model. Next, the first partial discharge pattern of the partial discharge data in the first cabinet and the second partial discharge pattern of the partial discharge data in the second cabinet are input into the image similarity calculation model, and the first similarity is determined based on the output of the image similarity calculation model.

[0073] As an optional embodiment, step 401 includes: A cosine similarity between the first peak frequency and the second peak frequency is determined to obtain a first sub-similarity.

[0074] The present invention does not limit the specific algorithm used to calculate the similarity. For example, cosine similarity can be used for calculation. Cosine similarity is a commonly used method for measuring the similarity between vectors. It can be used to calculate the cosine value of the angle between two vectors.

[0075] Step 105: Determine a source of the abnormal partial discharge of the first switchgear according to the first similarity and / or the second similarity.

[0076] If the external PD data is small or absent, an abnormality may be occurring inside the switchgear. If the waveforms of the external and internal PD data are similar, the internal PD abnormality may be caused by external interference. Based on the above principles, the source of the abnormal PD in the first switchgear is determined.

[0077] As an optional embodiment, step 105 includes: Step 201: If at least one of the first similarity and the second similarity is greater than or equal to a preset similarity threshold, it is determined that the abnormal partial discharge of the first switchgear is caused by external interference; Step 202: If both the first similarity and the second similarity are smaller than the similarity threshold, it is determined that the abnormality of the first switch cabinet originates from within the cabinet.

[0078] In steps 201 and 202, a higher first similarity indicates that the inner and outer ultrasonic PD probes collected more similar PD data, suggesting that the PD anomaly may be caused by an external interference source. A higher second similarity indicates that the first and second switchgear collected more similar PD data outside the switchgear, suggesting that the PD anomaly may be caused by an external interference source.

[0079] Therefore, if at least one of the first similarity and the second similarity is greater than or equal to a preset similarity threshold, it is determined that the abnormal partial discharge of the first switch cabinet is caused by external interference.

[0080] If only the PD probe inside the first switch cabinet detects a PD, causing the PD data outside the first cabinet to be dissimilar to the PD data outside the first cabinet, or if the PD data outside the first cabinet is affected by the PD inside the first switch cabinet and is dissimilar to the PD data outside the second cabinet, then it can be determined that the PD abnormality in the switch cabinet is caused by the abnormality inside the switch cabinet.

[0081] Therefore, if the first similarity or the second similarity is less than the preset similarity threshold, it means that the abnormality may only occur inside the first switch cabinet, that is, the abnormality of the first switch cabinet originates from inside the first switch cabinet.

[0082] According to an embodiment of the present invention, the similarity threshold may be set to 50%. The present invention does not impose any limitation on the specific value of the similarity threshold.

[0083] As an optional embodiment, after step 102, the method further includes: Step 501: Obtaining first in-cabinet partial discharge data of a first switchgear cabinet within a target time period after a current moment, and obtaining second in-cabinet partial discharge data of at least one second switchgear cabinet located in a switch room of the first switchgear cabinet within the target time period; Step 502: Determine a third similarity between the first cabinet partial discharge data and the second cabinet partial discharge data; Step 503: If the third similarity is greater than or equal to the similarity threshold, determine that the source of the abnormal partial discharge of the first switch cabinet is external interference.

[0084] In steps 501 to 503 , a third similarity is determined based on the partial discharge data in the first switch cabinet where the partial discharge is abnormal and collected by the dual ultrasonic sensors and the partial discharge data in the second switch cabinet.

[0085] If the third similarity is greater than or equal to the similarity threshold, it is determined that the abnormal partial discharge of the first switch cabinet is caused by external interference. If the third similarity is less than the similarity threshold, it is determined that the abnormality is inside the first switch cabinet. This solution can also assist in determining whether the abnormal partial discharge of the first switch cabinet is caused by the switch cabinet by comparing the partial discharge data in different switch cabinets.

[0086] As an optional embodiment, after step 102, the method further includes: Sending a DRX (Discontinuous Reception) configuration instruction to the partial discharge probes of the plurality of dual ultrasonic partial discharge sensors, wherein the DRX configuration instruction includes a DRX cycle duration and a synchronization frame sequence number; the DRX configuration instruction is used to instruct the partial discharge probes to start collecting partial discharge data during the DRX cycle corresponding to the synchronization frame sequence number; Step 103 includes: Obtain the first cabinet internal partial discharge data and the first cabinet external partial discharge data collected by the inner ultrasonic partial discharge probe set in the first switch cabinet and the outer ultrasonic partial discharge probe set outside the first switch cabinet within the target time period after the start of the DRX cycle corresponding to the synchronization frame sequence number.

[0087] In an embodiment of the present invention, a digital node device generates a time synchronization instruction. The time synchronization instruction is used to cause all probes of multiple dual ultrasonic partial discharge sensors to begin collecting data at the same acquisition time specified by the next synchronized acquisition time, and to collect data at the same acquisition frequency specified by the synchronized acquisition frequency. The time synchronization instruction may specifically include the next synchronized acquisition time and the synchronized acquisition frequency. The digital node device then generates a DRX configuration instruction based on the time synchronization instruction and sends the DRX configuration instruction to each partial discharge sensor.

[0088] DRX configuration instructions include the DRX cycle duration, which specifies the duration of a DRX cycle, and the synchronization frame number, which specifies the DRX cycle in which synchronization data collection begins. The DRX cycle duration is determined by the synchronization acquisition frequency, and the synchronization frame number is determined by the next synchronization acquisition time. According to one embodiment of the present invention, the synchronization acquisition frequency and the corresponding DRX cycle duration can be set to 20μs, ensuring that the sensor's synchronization acquisition accuracy meets the computational requirements for identifying partial discharges inside and outside the cabinet, locating interference sources outside the cabinet, and locating partial discharges inside the cabinet. The synchronization acquisition accuracy of the collected data is ±20μs.

[0089] After time synchronization of multiple dual ultrasonic PD sensors, the sensors collect PD data at the same synchronization acquisition time and frequency and send the data to the digital node device. The PD data received by the digital node device from the multiple dual ultrasonic PD sensors is synchronously collected.

[0090] When comparing the PD data inside the cabinet with the PD data outside the cabinet, this solution ensures the simultaneity of the PD data through high-precision synchronous acquisition technology.

[0091] The present invention can be applied in the field of partial discharge monitoring of substation switch cabinets and can be used by operation and maintenance personnel to provide accurate identification of partial discharge anomalies inside and outside the cabinet when the switch cabinet partial discharge is abnormal, as well as the function of troubleshooting interference sources outside the cabinet. When it is determined that the partial discharge anomaly of the switch cabinet is caused by external interference, an external interference source alarm is issued; when it is determined that the partial discharge anomaly of the switch cabinet is caused by an internal abnormality of the switch cabinet, an internal partial discharge anomaly alarm of the switch cabinet is issued. Alarm information of the external interference source alarm or the internal partial discharge anomaly alarm is sent to the distribution network cloud master station so that the partial discharge anomaly can be handled; and corresponding data can also be sent to the distribution network cloud master station so that the specific data source of the abnormality is displayed at the distribution network cloud master station, including partial discharge maps, etc., to assist operation and maintenance personnel in further analysis.

[0092] In summary, in an embodiment of the present invention, the in-cabinet partial discharge data of the first switch cabinet collected by the inner ultrasonic partial discharge probe of the dual ultrasonic partial discharge sensor at the current moment is obtained; the dual ultrasonic partial discharge sensor includes an inner ultrasonic partial discharge probe and an outer ultrasonic partial discharge probe, the inner ultrasonic partial discharge probe is arranged in the switch cabinet of the substation, and the outer ultrasonic partial discharge probe is arranged outside the switch cabinet; when the signal strength of the in-cabinet partial discharge data is greater than or equal to a preset intensity threshold, the partial discharge abnormality of the first switch cabinet is determined; the first in-cabinet partial discharge data and the first outside-cabinet partial discharge data of the first switch cabinet within the target time period after the current moment are obtained, and the second outside-cabinet partial discharge data of at least one second switch cabinet located in the switch room of the first switch cabinet within the target time period are obtained; a first similarity between the first in-cabinet partial discharge data and the first outside-cabinet partial discharge data, and a second similarity between the first in-cabinet partial discharge data and the second outside-cabinet partial discharge data are determined; based on the first similarity and / or the second similarity, the source of the partial discharge abnormality of the first switch cabinet is determined. This solution compares the first similarity between the first in-cabinet PD data and the second in-cabinet PD data, and the second similarity between the first in-cabinet PD data and the second out-cabinet PD data. This enables identification of environmental interference and location of in-cabinet PD faults using ultrasonic PD probes inside and outside the cabinet. Furthermore, machine-based anomaly detection saves human resources.

[0093] Figure 3 This is a structural block diagram of a device for identifying abnormal partial discharge in a substation switch cabinet provided by an embodiment of the present invention. Figure 3 As shown, the apparatus 600 includes: A first data acquisition module 601 is configured to acquire in-cabinet partial discharge data of a first switch cabinet collected at the current moment by an inner ultrasonic partial discharge probe of a dual ultrasonic partial discharge sensor; the dual ultrasonic partial discharge sensor includes an inner ultrasonic partial discharge probe and an outer ultrasonic partial discharge probe, the inner ultrasonic partial discharge probe being disposed in the switch cabinet of the substation, and the outer ultrasonic partial discharge probe being disposed outside the switch cabinet; An abnormality determination module 602 is configured to determine that the first switch cabinet has an abnormal partial discharge when the signal strength of the partial discharge data in the cabinet is greater than or equal to a preset strength threshold; A second data acquisition module 603 is configured to acquire first in-cabinet partial discharge data and first out-cabinet partial discharge data of the first switchgear within a target time period after the current moment, and acquire second out-cabinet partial discharge data of at least one second switchgear located in the switch room of the first switchgear within the target time period; A similarity determination module 604 is configured to determine a first similarity between the first in-cabinet partial discharge data and the first out-cabinet partial discharge data, and a second similarity between the first in-cabinet partial discharge data and the second out-cabinet partial discharge data; The source determination module 605 is configured to determine a source of the first switchgear partial discharge anomaly based on the first similarity and / or the second similarity.

[0094] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0095] Figure 4 A schematic diagram of a computing device provided in an embodiment of the present invention. Figure 4 Schematic diagram of a computing device 700 according to an exemplary embodiment of the present invention is shown. Figure 4 As shown, the computing device 700 may include a central processing unit 710, a memory 720, an input / output interface 730, a communication interface 740, and a bus 750. The central processing unit 710, the memory 720, the input / output interface 730, and the communication interface 740 are connected to each other via the bus 750 within the computing device.

[0096] The central processing unit 710 can be implemented using a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0097] The memory 720 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage devices, dynamic storage devices, etc. The memory 720 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 720 and is called and executed by the central processing unit 710.

[0098] The input / output interface 730 is used to connect to input / output modules to enable information input and output. The communication interface 740 is used to enable communication between the computing device and other devices. The bus 750 comprises a pathway for transmitting information between the various components of the computing device (e.g., the CPU 710, memory 720, the input / output interface 730, and the communication interface 740).

[0099] It should be noted that although the computing device shown above only includes a central processing unit 710, a memory 720, an input / output interface 730, a communication interface 740, and a bus 750, in a specific implementation, the computing device may also include other components necessary for normal operation. In addition, those skilled in the art will understand that the computing device may only include the components necessary to implement the embodiments of this specification, and does not necessarily include all the components shown in the figure.

[0100] An embodiment of the present invention also provides a non-transitory readable storage medium storing instructions for causing the computing device to execute a method according to an embodiment of the present invention. The readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. The information can be a computer-readable instruction, a data structure, a program module, or other data. Examples of readable storage media include, but are not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape disk storage or other magnetic storage devices, or any other non-transitory readable storage medium.

Claims

1. A method for identifying abnormal partial discharge in a substation switch cabinet, characterized in that: The method comprises: Obtaining in-cabinet partial discharge data of a first switch cabinet collected at a current moment by an inner ultrasonic partial discharge probe of a dual ultrasonic partial discharge sensor, wherein the dual ultrasonic partial discharge sensor includes an inner ultrasonic partial discharge probe and an outer ultrasonic partial discharge probe, wherein the inner ultrasonic partial discharge probe is disposed in the switch cabinet of the substation, and the outer ultrasonic partial discharge probe is disposed outside the switch cabinet; When the signal strength of the partial discharge data in the cabinet is greater than or equal to a preset strength threshold, determining that the partial discharge of the first switch cabinet is abnormal; Obtaining first in-cabinet partial discharge data and first out-cabinet partial discharge data of the first switchgear within a target time period after the current moment, and obtaining second out-cabinet partial discharge data of at least one second switchgear located in the switch room of the first switchgear within the target time period; Determining a first similarity between the first in-cabinet partial discharge data and the first out-cabinet partial discharge data, and a second similarity between the first in-cabinet partial discharge data and the second out-cabinet partial discharge data; A source of the first switchgear partial discharge anomaly is determined according to the first similarity and / or the second similarity.

2. The method according to claim 1, characterized in that The determining, based on the first similarity and / or the second similarity, a source of the first switchgear partial discharge anomaly includes: If at least one of the first similarity and the second similarity is greater than or equal to a preset similarity threshold, it is determined that the abnormal partial discharge of the first switchgear is caused by external interference; If both the first similarity and the second similarity are smaller than the similarity threshold, it is determined that the abnormality of the first switch cabinet originates from within the cabinet.

3. The method according to claim 1, characterized in that The determining of a first similarity between the first in-cabinet partial discharge data and the first out-cabinet partial discharge data includes: Obtaining a first partial discharge spectrum of partial discharge data in the first cabinet, and extracting a first eigenvector based on the first partial discharge spectrum; Obtaining a second partial discharge spectrum of partial discharge data in the second cabinet, and extracting a second eigenvector based on the second partial discharge spectrum; respectively determining similarities of identical features in the first feature vector and the second feature vector to obtain a plurality of sub-similarity scores; According to the preset weights of the various features, a weighted average is performed on the multiple sub-similarity levels, and a result of the weighted average is used as the first similarity level.

4. The method according to claim 3, characterized in that The first eigenvector is one or more of a first peak frequency, a first valley frequency, a first periodicity, and a first slope of the partial discharge spectrum; the second eigenvector is one or more of a second peak frequency, a second valley frequency, a second periodicity, and a second slope of the partial discharge spectrum, and the similarity of each identical feature in the first eigenvector and the second eigenvector is determined separately to obtain multiple sub-similarity, including: determining a similarity between the first peak frequency and the second peak frequency to obtain a first sub-similarity; Determine a similarity between the first valley frequency and the second valley frequency to obtain a second sub-similarity; determining a similarity between the first periodicity and the second periodicity to obtain a third sub-similarity; A similarity between the first slope and the second slope is determined to obtain a fourth sub-similarity.

5. The method according to claim 4, characterized in that The determining the similarity between the first peak frequency and the second peak frequency to obtain a first sub-similarity includes: Determine a cosine similarity between the first peak frequency and the second peak frequency to obtain a first sub-similarity.

6. The method according to claim 1, characterized in that After determining that the partial discharge of the first switch cabinet is abnormal, the method further includes: Obtaining first in-cabinet partial discharge data of the first switchgear within a target time period after the current moment, and obtaining second in-cabinet partial discharge data of at least one second switchgear located in a switch room of the first switchgear within the target time period; Determining a third similarity between the first in-cabinet partial discharge data and the second in-cabinet partial discharge data; If the third similarity is greater than or equal to the similarity threshold, it is determined that the source of the abnormal partial discharge of the first switch cabinet is external interference.

7. The method according to claim 1, characterized in that After determining that the partial discharge of the first switch cabinet is abnormal, the method further includes: Sending a DRX configuration instruction to the partial discharge probes of the plurality of dual ultrasonic partial discharge sensors, wherein the DRX configuration instruction includes a DRX cycle duration and a synchronization frame sequence number; the DRX configuration instruction is used to instruct the partial discharge probes to start collecting partial discharge data in the DRX cycle corresponding to the synchronization frame sequence number; The obtaining of first in-cabinet partial discharge data and first out-cabinet partial discharge data of the first switch cabinet within a target time period after the current moment includes: Obtain the first cabinet internal partial discharge data and the first cabinet external partial discharge data collected by the inner ultrasonic partial discharge probe set in the first switch cabinet and the outer ultrasonic partial discharge probe set outside the first switch cabinet within the target time period after the start of the DRX cycle corresponding to the synchronization frame sequence number.

8. A device for identifying abnormal partial discharge in a substation switch cabinet, characterized in that: The device comprises: a first data acquisition module, configured to acquire in-cabinet partial discharge data of a first switch cabinet collected at a current moment by an inner ultrasonic partial discharge probe of a dual ultrasonic partial discharge sensor; the dual ultrasonic partial discharge sensor comprising an inner ultrasonic partial discharge probe and an outer ultrasonic partial discharge probe, the inner ultrasonic partial discharge probe being disposed in the switch cabinet of the substation, and the outer ultrasonic partial discharge probe being disposed outside the switch cabinet; an abnormality determination module, configured to determine that the first switch cabinet has an abnormal partial discharge when the signal strength of the partial discharge data in the cabinet is greater than or equal to a preset strength threshold; a second data acquisition module, configured to acquire first in-cabinet partial discharge data and first out-cabinet partial discharge data of the first switchgear within a target time period after the current moment, and acquire second out-cabinet partial discharge data of at least one second switchgear located in the switch room of the first switchgear within the target time period; a similarity determination module, configured to determine a first similarity between the first in-cabinet partial discharge data and the first out-cabinet partial discharge data, and a second similarity between the first in-cabinet partial discharge data and the second out-cabinet partial discharge data; A source determination module is configured to determine a source of the first switch cabinet partial discharge anomaly based on the first similarity and / or the second similarity.

9. The device according to claim 8, characterized in that The source determination module is specifically used for: If at least one of the first similarity and the second similarity is greater than or equal to a preset similarity threshold, it is determined that the abnormal partial discharge of the first switchgear is caused by external interference; If both the first similarity and the second similarity are smaller than the similarity threshold, it is determined that the abnormality of the first switch cabinet originates from within the cabinet.

10. The device according to claim 8, characterized in that The similarity determination module is specifically used to: Obtaining a first partial discharge spectrum of partial discharge data in the first cabinet, and extracting a first eigenvector based on the first partial discharge spectrum; Obtaining a second partial discharge spectrum of partial discharge data in the second cabinet, and extracting a second eigenvector based on the second partial discharge spectrum; respectively determining similarities of identical features in the first feature vector and the second feature vector to obtain a plurality of sub-similarity scores; According to the preset weights of the various features, a weighted average is performed on the multiple sub-similarity levels, and a result of the weighted average is used as the first similarity level.

11. The device according to claim 10, characterized in that The similarity determination module is specifically used to: determining a similarity between the first peak frequency and the second peak frequency to obtain a first sub-similarity; Determine a similarity between the first valley frequency and the second valley frequency to obtain a second sub-similarity; determining a similarity between the first periodicity and the second periodicity to obtain a third sub-similarity; A similarity between the first slope and the second slope is determined to obtain a fourth sub-similarity.

12. The device according to claim 11, characterized in that The similarity determination module is specifically used to: Determine a cosine similarity between the first peak frequency and the second peak frequency to obtain a first sub-similarity.

13. The device according to claim 8, characterized in that The device further comprises: a partial discharge data acquisition module, configured to acquire first in-cabinet partial discharge data of the first switchgear within a target time period after the current moment, and acquire second in-cabinet partial discharge data of at least one second switchgear located in the switch room of the first switchgear within the target time period; a third similarity determination module, configured to determine a third similarity between the first in-cabinet partial discharge data and the second in-cabinet partial discharge data; An interference determination module is configured to determine that the source of the partial discharge anomaly of the first switch cabinet is external interference if the third similarity is greater than or equal to the similarity threshold.

14. The device according to claim 8, characterized in that The device further comprises: A configuration instruction sending module is used to send a DRX configuration instruction to the partial discharge probes of the plurality of dual ultrasonic partial discharge sensors, wherein the DRX configuration instruction includes a DRX cycle duration and a synchronization frame sequence number; the DRX configuration instruction is used to instruct the partial discharge probe to start collecting partial discharge data in the DRX cycle corresponding to the synchronization frame sequence number; The second data acquisition module is specifically used to: Obtain the first cabinet internal partial discharge data and the first cabinet external partial discharge data collected by the inner ultrasonic partial discharge probe set in the first switch cabinet and the outer ultrasonic partial discharge probe set outside the first switch cabinet within the target time period after the start of the DRX cycle corresponding to the synchronization frame sequence number.

15. A computing device, characterized in that include: one or more processors; Memory; as well as One or more devices comprising instructions for performing the method according to any one of claims 1-7.

16. A computer-readable storage medium storing one or more programs, characterized in that: The one or more programs include instructions that, when executed by a computing device, cause the computing device to perform the method according to any one of claims 1-7.

Images