A method, device, medium, and product for detecting partial discharge.

Abstract

This invention relates to the field of power technology and discloses a method, device, medium, and product for detecting partial discharge. The method includes: acquiring the initial sound source location; adjusting the detection direction of a microphone array based on the initial sound source location; acquiring the pulse density based on the partial discharge signal; acquiring the current sampling frequency based on the pulse density; sampling discharge pulses based on the current sampling frequency to acquire each target discharge pulse and its corresponding power frequency phase angle; generating a phase-analyzed partial discharge map based on each target discharge pulse and its corresponding power frequency phase angle; and acquiring the partial discharge type based on the phase-analyzed partial discharge map using a pre-trained feature extraction model and a partial discharge type recognition model. This method can efficiently and accurately detect and locate partial discharges, achieve automatic identification of partial discharge types, and ensure detection accuracy and reliability in complex environments.

Description

Technical Field

[0001] This invention relates to the field of power technology, and in particular to a method, device, medium, and product for detecting partial discharge. Background Technology

[0002] In heavy-haul railways, the detection and treatment of partial discharge in traction substations is of great significance for improving the safety and reliability of railway operation.

[0003] Currently, existing methods for detecting partial discharge mainly rely on periodic inspections or manual testing. This requires staff to carry testing equipment and check each point individually, which not only consumes a lot of time and manpower but also makes it difficult to achieve continuous, comprehensive, and real-time testing. Furthermore, it cannot automatically identify the type of partial discharge and has low accuracy and reliability in complex environments. Summary of the Invention

[0004] This invention provides a method, device, medium, and product for detecting partial discharge, which can efficiently and accurately detect and locate partial discharge, automatically identify the type of partial discharge, and ensure detection accuracy and reliability in complex environments.

[0005] According to one aspect of the present invention, a method for detecting partial discharge is provided, comprising:

[0006] During partial discharge inspection, the initial sound source position is obtained, and after adjusting the detection direction of the microphone array according to the initial sound source position, the partial discharge signal is obtained through the microphone array.

[0007] Based on the partial discharge signal, the pulse density is obtained, and based on the pulse density and the mapping relationship between the preset pulse density range and the sampling frequency, the current sampling frequency is obtained.

[0008] Discharge pulses are sampled according to the current sampling frequency to obtain each target discharge pulse and its corresponding power frequency phase angle, and a phase analysis partial discharge map is generated based on each target discharge pulse and its corresponding power frequency phase angle.

[0009] The partial discharge type is obtained based on the phase-analyzed partial discharge map by using a pre-trained feature extraction model and a partial discharge type identification model.

[0010] According to another aspect of the present invention, a partial discharge detection device is provided, comprising:

[0011] The partial discharge signal acquisition module is used to acquire the initial sound source position during partial discharge inspection, and to acquire the partial discharge signal through the microphone array after adjusting the detection direction of the microphone array according to the initial sound source position.

[0012] The sampling frequency acquisition module is used to acquire the pulse density based on the partial discharge signal, and to acquire the current sampling frequency based on the pulse density and the mapping relationship between the preset pulse density range and the sampling frequency.

[0013] The phase-analyzed partial discharge map generation module is used to sample discharge pulses according to the current sampling frequency, obtain each target discharge pulse and its corresponding power frequency phase angle, and generate a phase-analyzed partial discharge map based on each target discharge pulse and its corresponding power frequency phase angle.

[0014] The partial discharge type acquisition module is used to acquire the partial discharge type based on the phase-analyzed partial discharge map by using a pre-trained feature extraction model and a partial discharge type identification model.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] At least one processor; and

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the partial discharge detection method according to any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program for causing a processor to execute and implement the partial discharge detection method according to any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the partial discharge detection method described in any embodiment of the present invention.

[0021] The technical solution of this invention, during partial discharge inspection, obtains the initial sound source position, and after adjusting the detection direction of the microphone array according to the initial sound source position, acquires the partial discharge signal through the microphone array; based on the partial discharge signal, obtains the pulse density, and based on the pulse density and the mapping relationship between the preset pulse density range and the sampling frequency, obtains the current sampling frequency; samples the discharge pulses according to the current sampling frequency, obtains each target discharge pulse and its corresponding power frequency phase angle, and generates a phase-analyzed partial discharge map based on each target discharge pulse and its corresponding power frequency phase angle; through a pre-trained feature extraction model and a partial discharge type recognition model, the partial discharge type is obtained based on the phase-analyzed partial discharge map; by automatically adjusting the detection direction of the microphone array according to the sound source position, setting a dynamic sampling frequency according to the pulse density, and combining the feature extraction model and the partial discharge type recognition model, the partial discharge type is identified based on the phase-analyzed partial discharge map, which can efficiently and accurately detect and locate partial discharge, achieve automatic identification of partial discharge types, and ensure detection accuracy and reliability in complex environments.

[0022] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart of a partial discharge detection method provided according to Embodiment 1 of the present invention;

[0025] Figure 2 This is a schematic diagram of the PRPD spectrum of suspended discharge according to Embodiment 1 of the present invention;

[0026] Figure 3 This is a flowchart of a partial discharge detection method provided according to Embodiment 2 of the present invention;

[0027] Figure 4 This is a schematic diagram of the structure of a partial discharge detection device according to Embodiment 3 of the present invention;

[0028] Figure 5 This is a schematic diagram of the structure of an electronic device that implements the partial discharge detection method of this invention. Detailed Implementation

[0029] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "first," "second," "modification," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0031] Example 1

[0032] Figure 1 This is a flowchart of a partial discharge detection method provided in Embodiment 1 of the present invention. This embodiment is applicable to the detection of partial discharge in heavy-haul railway traction substations. The method can be executed by a partial discharge detection device, which can be implemented in hardware and / or software. Typically, the partial discharge detection device can be configured in electronic equipment, such as computer equipment, servers, etc. Figure 1 As shown, the method includes:

[0033] S110. During partial discharge inspection, the initial sound source position is obtained, and after adjusting the detection direction of the microphone array according to the initial sound source position, the partial discharge signal is obtained through the microphone array.

[0034] In this embodiment, the partial discharge detection system can consist of a spiral microphone array, a visible light camera, a pan-tilt unit, and a processor. The spiral microphone array and the visible light camera can be deployed on the pan-tilt unit, and the detection area can be adjusted by the pan-tilt unit. The system can perform partial discharge inspections on traction power supply equipment within a specified radius, thereby reducing the number of detection systems used for each traction unit. When a partial discharge occurs, ultrasonic waves are emitted. These ultrasonic signals are captured by the spiral microphone array and converted into a visual image, which can intuitively display the specific location of the partial discharge. Furthermore, the captured ultrasonic signals can be fused with the image captured by the visible light camera to generate a visual image containing the location of the partial discharge, allowing inspectors to intuitively see the location of the partial discharge.

[0035] In a specific example, when performing partial discharge inspection, the system is first started, and the microphone array and camera are initialized. Then, based on the preset detection point coordinates, the microphone array and camera are rotated via a pan-tilt unit to inspect devices in different directions. The preset detection points can be predefined locations of the equipment to be inspected and key points. Key points can be specific parts of the equipment, such as connectors or insulators in switchgear, or specific areas surrounding the equipment. The coordinates of the detection points can be based on their relative positions according to the equipment layout diagram or their absolute positions based on the actual installation environment. Coordinate values ​​can include parameters such as horizontal angle, vertical angle, and distance.

[0036] For pan-tilt control, before the inspection begins, the pan-tilt is initialized to its starting position, ensuring the microphone array and camera are in their initial state. Then, based on preset detection point coordinates, the required rotation angles of the microphone array and camera are calculated, and the horizontal and vertical movements of the pan-tilt are driven by motors to align the microphone array and camera with the target detection points. During the inspection, under the control of the pan-tilt, the microphone array and camera align with each detection point and begin data acquisition. Specifically, the microphone array acquires ultrasonic signals, and the camera acquires visible light images. Further, the edge computing core board (processor) processes the ultrasonic signals to extract the initial sound source location. Simultaneously, the visible light images are processed to identify key parts of the equipment. For example, the coordinates of locations with strong sound waves (amplitude greater than a preset threshold) can be used as the sound source location.

[0037] After determining the initial sound source location, the microphone array can be fine-tuned using a pan-tilt unit to ensure precise alignment with the initial sound source. For example, the angular deviation between the current detection direction (array center direction) and the initial sound source location can be calculated, and the microphone array can be controlled to move horizontally and vertically based on this angular deviation, so that the microphone array is aligned with the initial sound source location. After adjustment, the ultrasonic signal generated by partial discharge is re-acquired through the microphone array as the partial discharge signal.

[0038] Optionally, multiple detection points and corresponding detection sequences can be pre-set. After the inspection is started, the system controls the pan-tilt unit to rotate sequentially according to this detection sequence, so that the microphone array is aimed at each detection point for data acquisition. At each detection point, the system can control the dwell time to ensure that sufficient data is collected. To improve inspection efficiency, the system can also use a path optimization algorithm to calculate the optimal inspection path based on the distribution of detection points and the motion characteristics of the pan-tilt unit. The optimized path can reduce the number of pan-tilt unit rotations and the rotation angle, thereby improving inspection speed and efficiency. This embodiment does not specifically limit the path optimization algorithm.

[0039] Secondly, during the inspection process, the edge computing core board can feed back the processed data to the back-end control center in real time. Based on this feedback, the back-end control center monitors the inspection process in real time and adjusts the movement of the gimbal. If an abnormal signal or unclear image is detected at a certain detection point during the inspection, the back-end control center can dynamically adjust the gimbal's movement parameters and re-align it with that detection point for data acquisition.

[0040] S120. Based on the partial discharge signal, obtain the pulse density, and based on the pulse density and the mapping relationship between the preset pulse density range and the sampling frequency, obtain the current sampling frequency.

[0041] In this embodiment, the number of pulses NP within a unit time T can be obtained first based on the partial discharge signal. Then, the pulse density D can be calculated using the formula D=NP / T. For example, the unit time T can be 20 milliseconds. Next, the pulse density range where the current pulse density falls can be found according to the preset mapping relationship between the pulse density range and the sampling frequency, and the sampling frequency corresponding to this pulse density range can be determined as the current sampling frequency.

[0042] The mapping relationship between pulse density range and sampling frequency can be preset. For example, the mapping relationship can be shown in Table 1. The pulse density is divided into three ranges: less than 10, greater than or equal to 10 and less than or equal to 50, and greater than 50. Each range corresponds to a sampling frequency of 100, 50, and 20 kilohertz (kHz), respectively, and each range has its own applicable scenarios. This embodiment can balance detail capture and redundancy control by dynamically adjusting the sampling frequency based on pulse density.

[0043] Table 1. Mapping Relationship between Pulse Density Range and Sampling Frequency

[0044]

[0045] S130. Sampling of discharge pulses is performed according to the current sampling frequency to obtain each target discharge pulse and its corresponding power frequency phase angle, and a phase analysis partial discharge map is generated based on each target discharge pulse and its corresponding power frequency phase angle.

[0046] In this embodiment, either an external synchronization method or an internal synchronization method can be used to synchronize the sampling with the 50 Hz power frequency to ensure the accuracy of the electrical pulse phase angle. Then, discharge pulses can be acquired based on the current sampling frequency to obtain each discharge pulse and its corresponding pulse time. Based on this pulse time, the power frequency phase angle corresponding to each discharge pulse can be calculated. Finally, a phase-resolved partial discharge (PRPD) map can be plotted based on the power frequency phase angle and amplitude corresponding to each discharge pulse. The PRPD map uses phase as the horizontal axis, pulse amplitude as the vertical axis, and pulse density as the color level (the darker the color, the higher the density). For example, the PRPD map of a levitating discharge can be shown as follows: Figure 2 As shown, each point represents a discharge pulse.

[0047] S140. Using a pre-trained feature extraction model and a partial discharge type identification model, the partial discharge type is obtained based on the phase-analyzed partial discharge map.

[0048] In this embodiment, deep learning algorithms can be used to extract features and classify the generated PRPD map to determine whether partial discharge phenomena exist, and, if partial discharge phenomena are determined to exist, to identify the specific type of partial discharge. The feature extraction model can be built based on a Convolutional Neural Network (CNN), with the PRPD map (resolution 3600) as input. (1000 pixels), the output is a 64-dimensional feature vector. The partial discharge type identification model can be built based on a Long Short-Term Memory (LSTM) network. The input is the 64-dimensional feature vector output by the feature extraction model, and the output is the partial discharge type.

[0049] Specifically, the PRPD map can be preprocessed first, then input into the feature extraction model to obtain a 64-dimensional feature vector. Finally, the 64-dimensional feature vector and a time step of 100 (covering 100 power frequency cycles) are input into the partial discharge type identification model to obtain the partial discharge type. The partial discharge type can include suspended discharge, corona discharge, and surface discharge.

[0050] Preprocessing of the PRPD spectrum includes normalization and data augmentation. During normalization, the amplitude of the PRPD spectrum is normalized to [0,1], and the color levels (pulse density) are normalized to [0,255]. During data augmentation, phase shifting is performed... The training set was expanded by scaling the magnitude and amplitude by 0.8-1.2 times and adding noise (SNR=10dB) to improve the model's generalization ability.

[0051] In a specific example, the structure of the feature extraction model can be shown in Table 2. It includes an input layer, convolutional layer 1, pooling layer 1, convolutional layer 2, pooling layer 2, and a fully connected layer. The partial discharge type recognition model can include an input layer, an LSTM layer, and an output layer. The LSTM layer includes two hidden layers, each with 128 units. The output layer includes a Softmax activation function, achieving a classification accuracy greater than or equal to 95%.

[0052] Table 2 Feature Extraction Model Structure

[0053]

[0054] Optionally, after obtaining the partial discharge type, the following may also be included:

[0055] A processing suggestion corresponding to the partial discharge type is generated, and an operation and maintenance report is generated based on the partial discharge type and the processing suggestion. The operation and maintenance report is then sent to the target user through a preset method.

[0056] In this embodiment, if partial discharge is detected, an operation and maintenance report can be automatically generated. Specifically, firstly, based on the preset mapping relationship between partial discharge types and processing suggestions, the processing suggestion corresponding to the current partial discharge type is found; alternatively, the partial discharge type and sound source location are input into a pre-trained large language model, and the processing suggestion output by the large language model is obtained. Then, based on a preset format, the partial discharge type and processing suggestion can be combined to generate an operation and maintenance report. Finally, the operation and maintenance report can be automatically sent to a pre-specified target user (such as operation and maintenance personnel) using a preset method, such as SMS or email.

[0057] Optionally, the maintenance report can also include information such as the location, phase, and amplitude of the sound source. For example, the maintenance report could be: "Connector #1 switchgear, corona discharge, maintenance recommended within 24 hours." While sending the maintenance report to the target user, it can also be visualized on the system interface. Maintenance personnel can perform on-site inspections and handle issues based on the report, and upload the results and feedback information. The system can persistently store the results and feedback information.

[0058] The advantage of the above settings is that they enable a direct display of partial discharge detection results, thereby improving operation and maintenance efficiency.

[0059] The technical solution of this invention, during partial discharge inspection, obtains the initial sound source position, and after adjusting the detection direction of the microphone array according to the initial sound source position, acquires the partial discharge signal through the microphone array; based on the partial discharge signal, obtains the pulse density, and based on the pulse density and the mapping relationship between the preset pulse density range and the sampling frequency, obtains the current sampling frequency; samples the discharge pulses according to the current sampling frequency, obtains each target discharge pulse and its corresponding power frequency phase angle, and generates a phase-analyzed partial discharge map based on each target discharge pulse and its corresponding power frequency phase angle; through a pre-trained feature extraction model and a partial discharge type recognition model, the partial discharge type is obtained based on the phase-analyzed partial discharge map; by automatically adjusting the detection direction of the microphone array according to the sound source position, setting a dynamic sampling frequency according to the pulse density, and combining the feature extraction model and the partial discharge type recognition model, the partial discharge type is identified based on the phase-analyzed partial discharge map, which can efficiently and accurately detect and locate partial discharge, achieve automatic identification of partial discharge types, and ensure detection accuracy and reliability in complex environments.

[0060] Example 2

[0061] Figure 3This is a flowchart of a partial discharge detection method provided in Embodiment 2 of the present invention. This embodiment is a further refinement of the above technical solution, and the technical solution in this embodiment can be combined with one or more of the above implementation methods. Figure 3 As shown, the method includes:

[0062] S210. During partial discharge inspection, obtain the initial sound source location.

[0063] In this embodiment, during partial discharge inspection, initialization and guided coarse localization can be performed first, followed by guided fine localization. During initialization and guided coarse localization, the microphone array is divided into eight microphone subarrays (each subarray contains 16 microphones arranged in a spiral pattern). During initialization, all subarrays are activated, and a full-area coarse scan is performed according to a preset range, step size, and duration. Then, based on preset localization accuracy and calculation speed, the suspected sound source region is quickly located. If a suspected sound source region is successfully detected—for example, if the signal amplitude exceeds a preset threshold and is confirmed by two consecutive scans—the guided localization mode is triggered, and the initial sound source location is extracted. If no suspected sound source region is detected, the inspection proceeds along a preset path.

[0064] S220. Obtain the risk level corresponding to the initial sound source location. If the risk level is determined to be high risk, control the microphone array to perform an initial detection direction adjustment based on the initial sound source location and the first preset speed. After the adjustment is completed, activate the microphone subarray corresponding to the initial sound source location.

[0065] In this embodiment, detection points can be divided into three risk levels based on historical fault data, and coordinate accuracy and inspection priority can be set differently. For example, the detection point classification can be as shown in Table 3, where the risk levels include high risk, medium risk and low risk, and each risk level has a corresponding coverage equipment type, coordinate accuracy and inspection priority.

[0066] Table 3. Classification of Testing Points

[0067]

[0068] Specifically, the device type corresponding to the initial sound source location can be obtained, and based on this device type, the risk level corresponding to the initial sound source location can be determined by consulting a detection point classification table. Then, if the risk level is high, precise localization can be directly guided based on the initial sound source location. However, if the risk level is medium or low, it is necessary to wait for higher-priority detection points to complete their inspections according to a priority order of 1-3 before performing precise localization based on the initial sound source location.

[0069] During guided precise positioning, the microphone array can be rotated to the center of the suspected area corresponding to the initial sound source position at a first preset speed. After rotation, the microphone subarray corresponding to the initial sound source position is activated. For example, if the initial sound source position is on the right, then microphone subarrays 5-8 on the right are activated. The first preset speed can be a relatively large preset speed value, such as 5 degrees / second horizontally and 3 degrees / second vertically.

[0070] S230. Acquire the sampling signal of the microphone subarray, and obtain the updated sound source position based on the sampling signal.

[0071] Furthermore, the current partial discharge signal is reacquired using a microphone subarray, and a sound source localization algorithm is employed to calculate precise coordinates based on this signal, which are then used to update the sound source location. For example, the accuracy of the updated sound source location can be 0.02 meters. This embodiment does not specifically limit the sound source localization algorithm.

[0072] S240. Based on the updated sound source position and the second preset speed, control the microphone array to perform a re-detection direction adjustment so that the center of the microphone array is aligned with the sound source.

[0073] The second preset speed is less than the first preset speed. The second preset speed can be a small preset speed value, for example, 0.5 degrees / second. In this embodiment, the current rotation angle can be calculated based on the updated sound source position and the current detection direction, and the microphone array can be controlled to adjust the rotation angle according to the second preset speed to ensure that the center of the microphone array is aligned with the sound source.

[0074] Optionally, based on the updated sound source position and the second preset speed, controlling the microphone array to re-adjust the detection direction so that the center of the microphone array is aligned with the sound source may include:

[0075] Calculate the positional deviation between the updated sound source position and the center of the microphone array;

[0076] If the detected positional deviation meets the preset adjustment conditions, then based on the updated sound source position and the second preset speed, the microphone array is controlled to perform a re-detection direction adjustment so that the center of the microphone array is aligned with the sound source.

[0077] In this embodiment, feedback closed-loop logic can be used to achieve guided positioning. Specifically, a sound source localization algorithm is used to calculate and update the positional deviation between the sound source position and the center of the microphone array. x, The system then calculates the position deviation (y) and determines whether it meets the preset adjustment conditions. If it does, the guided positioning is considered complete. If it does not, the system returns to collecting partial discharge signals via the microphone subarray and recalculates the position deviation and checks if it meets the preset adjustment conditions, until the current position deviation meets the preset adjustment conditions. This forms a closed loop of "acquisition-calculation-adjustment".

[0078] The preset adjustment conditions can be preset conditions that the position deviation must meet when the adjustment is completed, for example, they could be... Rice, and rice.

[0079] The advantage of the above settings is that they can improve the accuracy of microphone array adjustment and ensure that the center of the microphone array is aligned with the sound source.

[0080] Optionally, in this embodiment, in addition to guided positioning, temporary path insertion, high-risk point repetitive inspection, and path shortening are also supported. Temporary path insertion means that if a non-preset suspected point is detected, it is designated as a temporary high-risk point and inserted into the current inspection path for priority inspection. High-risk point repetitive inspection means that after inspecting a high-risk point, it is inspected again at 5-minute intervals (a total of 3 times) to avoid sudden partial discharges. Path shortening refers to optimizing the path based on experimental data using a minimum path algorithm, shortening the time by 20%-30% compared to a fixed path. Furthermore, in guided positioning mode, high-risk point data can be stored as high-priority data (3 backups, stored for 1 year), and low-risk point data can be stored as normal-priority data (1 backup, stored for 6 months). When partial discharge is detected, an alarm can be generated and pushed to the backend control center.

[0081] S250. Acquire a partial discharge signal through the microphone array, obtain a pulse density based on the partial discharge signal, and obtain the current sampling frequency based on the pulse density and the mapping relationship between the preset pulse density range and the sampling frequency.

[0082] S260. Based on the partial discharge signal, obtain the background noise amplitude, and based on the background noise amplitude, obtain the current threshold.

[0083] In this embodiment, a background noise component can be extracted from the partial discharge signal, and the background noise amplitude can be obtained based on this component. Then, the background noise amplitude can be multiplied by a preset ratio to obtain the product as the current threshold. The preset ratio can be set based on historical data.

[0084] S270. When the amplitude of the partial discharge signal is detected to be greater than the current threshold, discharge pulse sampling is triggered, and the power frequency voltage signal is acquired.

[0085] Specifically, when the amplitude of the detected partial discharge signal (ultrasonic signal) is greater than the current threshold, it indicates that a partial discharge phenomenon may occur, and at this time, discharge pulse sampling is triggered.

[0086] S280. Using a preset phase-locked loop, a synchronization clock is obtained based on the power frequency voltage signal, and discharge pulses are sampled based on the current sampling frequency and the synchronization clock to obtain each target discharge pulse and its corresponding power frequency phase angle.

[0087] In this embodiment, to ensure the accuracy of the power frequency phase angle, power frequency phase synchronization calibration can be performed. Specifically, an external synchronization method is adopted. First, a 50 Hz power frequency voltage signal is obtained from the traction substation through voltage transformer division. Then, an ADF4159 phase-locked loop module is used to convert the power frequency voltage signal into a synchronization clock, which is then connected to the processor as a synchronization source. Finally, discharge pulses are sampled according to the current sampling frequency to obtain each target discharge pulse and its corresponding power frequency phase angle. The phase of the synchronization clock is consistent with the power frequency voltage signal, with a phase resolution of 0.1 degrees.

[0088] Optionally, sampling the discharge pulses according to the current sampling frequency and the synchronization clock to obtain each target discharge pulse and its corresponding power frequency phase angle may include:

[0089] Discharge pulses are sampled according to the current sampling frequency and the synchronization clock to obtain each initial discharge pulse and the corresponding power frequency phase angle.

[0090] Based on the power frequency phase angle corresponding to each initial discharge pulse and the mapping relationship between the preset phase layer and the power frequency phase angle range, the phase layer corresponding to each initial discharge pulse is obtained.

[0091] Based on the phase layer corresponding to each initial discharge pulse, the standard deviation of the pulse amplitude corresponding to each phase layer is obtained, and based on the standard deviation of the pulse amplitude, the density adjustment coefficient corresponding to each phase layer is obtained.

[0092] Based on the density adjustment coefficient, pulse interpolation is performed on each phase layer to obtain each target discharge pulse and the corresponding power frequency phase angle.

[0093] In this embodiment, phase layering can be performed, dividing the 360-degree power frequency phase into 12 intervals (30 degrees each), with each interval constituting one phase layer. A mapping relationship between the phase layers and the power frequency phase angle range is generated. Therefore, when generating a target discharge pulse, firstly, actual discharge pulses are sampled based on the current sampling frequency and synchronization clock to obtain each initial discharge pulse and its corresponding power frequency phase angle. Then, based on the preset mapping relationship between phase layers and the power frequency phase angle range, the phase layer corresponding to each initial discharge pulse is determined by finding the power frequency phase angle range corresponding to the initial discharge pulse. Next, based on the amplitude of each initial discharge pulse in each phase layer, the standard deviation of the pulse amplitude corresponding to each phase layer is calculated. Then, based on the current standard deviation of the pulse amplitude and the preset mapping relationship between the standard deviation range and the density adjustment coefficient, the density adjustment coefficient corresponding to each phase layer is obtained. Finally, the number of initial discharge pulses in each phase layer is multiplied by the corresponding density adjustment coefficient to calculate the number of pulse interpolations. Discharge pulses with the corresponding number of pulse interpolations are added to each phase layer, thereby obtaining all target discharge pulses and their corresponding power frequency phase angles.

[0094] The advantage of the above settings is that by adjusting the sampling point density according to phase layering, the feature resolution of the PRPD spectrum can be improved, which can improve the accuracy of partial discharge type identification.

[0095] Optionally, obtaining the density adjustment coefficient corresponding to each phase layer based on the standard deviation of the pulse amplitude may include:

[0096] Obtain the maximum pulse amplitude corresponding to each phase layer, and obtain the density adjustment threshold corresponding to each phase layer based on the maximum pulse amplitude;

[0097] The density adjustment coefficients corresponding to each phase layer are obtained based on the pulse amplitude standard deviation and the density adjustment threshold.

[0098] In an optional example, based on the amplitude of each initial discharge pulse in each phase layer, the maximum pulse amplitude corresponding to each phase layer is obtained, and the maximum pulse amplitude is multiplied by a preset value to obtain the product as the corresponding density adjustment threshold. Next, in each phase layer, it is determined whether the standard deviation of the pulse amplitude is greater than the density adjustment threshold. If so, the corresponding density adjustment coefficient is determined to be a preset non-zero value (increasing the sampling point density); if the standard deviation of the pulse amplitude is less than or equal to the density adjustment threshold, the density adjustment coefficient is determined to be zero.

[0099] The advantage of the above settings is that they can improve the accuracy of the density adjustment coefficient and achieve effective adjustment of the sampling point density.

[0100] S290. Based on each target discharge pulse and its corresponding power frequency phase angle, a phase-analyzed partial discharge map is generated. The partial discharge type is obtained based on the phase-analyzed partial discharge map using a pre-trained feature extraction model and a partial discharge type identification model.

[0101] The technical solution of this invention involves obtaining the risk level corresponding to the initial sound source location. If the risk level is determined to be high, the microphone array is controlled to perform an initial detection direction adjustment based on the initial sound source location and a first preset speed. After the adjustment is completed, the microphone subarray corresponding to the initial sound source location is activated. The acquisition signal of the microphone subarray is acquired, and the updated sound source location is obtained based on the acquisition signal. Based on the updated sound source location and a second preset speed, the microphone array is controlled to perform a second detection direction adjustment so that the center of the microphone array is aligned with the sound source. By adopting a microphone array adjustment method that combines guided coarse positioning with fine positioning, the accuracy of microphone array adjustment can be improved, and the accuracy of partial discharge detection can be improved. Based on the partial discharge signal, the background noise amplitude is obtained, and the current threshold is obtained based on the background noise amplitude. When the amplitude of the detected partial discharge signal is greater than the current threshold, discharge pulse sampling is triggered, and the power frequency voltage signal is obtained. Through a preset phase-locked loop, a synchronization clock is obtained based on the power frequency voltage signal, and discharge pulse sampling is performed based on the current sampling frequency and the synchronization clock to obtain each target discharge pulse and the corresponding power frequency phase angle. By setting a dynamic threshold based on the background noise amplitude, the probability of false sampling can be reduced. By achieving sampling and power frequency synchronization based on the power frequency voltage signal, the accuracy of the power frequency phase angle can be improved, which can improve the accuracy of phase resolution partial discharge spectrum.

[0102] In a specific example, the partial discharge detection system of this invention was deployed in a heavy-haul railway traction substation. It collects ultrasonic signals of partial discharge and identifies them using deep learning algorithms. The system automatically plans inspection paths, monitors and locates partial discharge positions in real time, and has successfully identified and issued warnings for multiple partial discharge events, effectively improving the safety and reliability of equipment operation.

[0103] In this embodiment, by fusing acoustic imaging and partial discharge signals, the location of partial discharge can be accurately pinpointed with an accuracy of 0.02 meters, significantly improving the accuracy and reliability of detection. The introduction of deep learning algorithms enables partial discharge type identification accuracy to exceed 95%, effectively distinguishing different types of partial discharge and providing a more accurate basis for equipment maintenance and repair. The combination of the automatic partial discharge inspection system and intelligent partial discharge methods reduces the workload of manual inspections, improves inspection efficiency and coverage, and lowers operation and maintenance costs. The system's intelligent analysis and prediction functions can detect potential faults in advance, reducing equipment downtime and improving equipment operational safety and reliability. The system and deep learning algorithms are highly adaptable to complex environments, maintaining high detection accuracy and identification accuracy under complex conditions such as high noise and strong electromagnetic interference, ensuring the reliability of detection results. Through the system's high coverage and automated inspection system, the number of devices required to be deployed in each traction substation is reduced, lowering equipment costs and maintenance workload.

[0104] Example 3

[0105] Figure 4 This is a schematic diagram of a partial discharge detection device provided in Embodiment 3 of the present invention. Figure 4 As shown, the device includes: a partial discharge signal acquisition module 310, a sampling frequency acquisition module 320, a phase-analyzed partial discharge map generation module 330, and a partial discharge type acquisition module 340; wherein,

[0106] The partial discharge signal acquisition module 310 is used to acquire the initial sound source position during partial discharge inspection, and to acquire the partial discharge signal through the microphone array after adjusting the detection direction of the microphone array according to the initial sound source position.

[0107] The sampling frequency acquisition module 320 is used to acquire the pulse density based on the partial discharge signal, and to acquire the current sampling frequency based on the pulse density and the mapping relationship between the preset pulse density range and the sampling frequency.

[0108] The phase-analyzed partial discharge map generation module 330 is used to sample discharge pulses according to the current sampling frequency, obtain each target discharge pulse and the corresponding power frequency phase angle, and generate a phase-analyzed partial discharge map according to each target discharge pulse and the corresponding power frequency phase angle.

[0109] The partial discharge type acquisition module 340 is used to acquire the partial discharge type based on the phase-analyzed partial discharge map by using a pre-trained feature extraction model and a partial discharge type identification model.

[0110] The technical solution of this invention, during partial discharge inspection, obtains the initial sound source position, and after adjusting the detection direction of the microphone array according to the initial sound source position, acquires the partial discharge signal through the microphone array; based on the partial discharge signal, obtains the pulse density, and based on the pulse density and the mapping relationship between the preset pulse density range and the sampling frequency, obtains the current sampling frequency; samples the discharge pulses according to the current sampling frequency, obtains each target discharge pulse and its corresponding power frequency phase angle, and generates a phase-analyzed partial discharge map based on each target discharge pulse and its corresponding power frequency phase angle; through a pre-trained feature extraction model and a partial discharge type recognition model, the partial discharge type is obtained based on the phase-analyzed partial discharge map; by automatically adjusting the detection direction of the microphone array according to the sound source position, setting a dynamic sampling frequency according to the pulse density, and combining the feature extraction model and the partial discharge type recognition model, the partial discharge type is identified based on the phase-analyzed partial discharge map, which can efficiently and accurately detect and locate partial discharge, achieve automatic identification of partial discharge types, and ensure detection accuracy and reliability in complex environments.

[0111] Optionally, the phase analysis partial discharge pattern generation module 330 includes:

[0112] The threshold acquisition unit is used to acquire the background noise amplitude based on the partial discharge signal, and to acquire the current threshold based on the background noise amplitude.

[0113] The power frequency voltage signal acquisition unit is used to trigger discharge pulse sampling and acquire the power frequency voltage signal when the amplitude of the detected partial discharge signal is greater than the current threshold.

[0114] The discharge pulse sampling unit is used to obtain a synchronization clock based on the power frequency voltage signal through a preset phase-locked loop, and to sample discharge pulses based on the current sampling frequency and the synchronization clock to obtain each target discharge pulse and the corresponding power frequency phase angle.

[0115] Optionally, the discharge pulse sampling unit is specifically used to sample the discharge pulses according to the current sampling frequency and the synchronization clock, and to obtain each initial discharge pulse and the corresponding power frequency phase angle;

[0116] Based on the power frequency phase angle corresponding to each initial discharge pulse and the mapping relationship between the preset phase layer and the power frequency phase angle range, the phase layer corresponding to each initial discharge pulse is obtained.

[0117] Based on the phase layer corresponding to each initial discharge pulse, the standard deviation of the pulse amplitude corresponding to each phase layer is obtained, and based on the standard deviation of the pulse amplitude, the density adjustment coefficient corresponding to each phase layer is obtained.

[0118] Based on the density adjustment coefficient, pulse interpolation is performed on each phase layer to obtain each target discharge pulse and the corresponding power frequency phase angle.

[0119] Optionally, the discharge pulse sampling unit is further configured to obtain the maximum pulse amplitude corresponding to each of the phase layers, and obtain the density adjustment threshold corresponding to each of the phase layers based on the maximum pulse amplitude;

[0120] The density adjustment coefficients corresponding to each phase layer are obtained based on the pulse amplitude standard deviation and the density adjustment threshold.

[0121] Optionally, the partial discharge signal acquisition module 310 is specifically used to acquire the risk level corresponding to the initial sound source position. If the risk level is determined to be high risk, the microphone array is controlled to perform an initial detection direction adjustment based on the initial sound source position and the first preset speed. After the adjustment is completed, the microphone subarray corresponding to the initial sound source position is activated.

[0122] Acquire the sampling signals from the microphone subarray, and obtain and update the sound source position based on the sampling signals;

[0123] Based on the updated sound source location and the second preset speed, the microphone array is controlled to re-adjust its detection direction so that the center of the microphone array is aligned with the sound source.

[0124] The second preset speed is less than the first preset speed.

[0125] Optionally, the partial discharge signal acquisition module 310 is also used to calculate the positional deviation between the updated sound source position and the center of the microphone array;

[0126] If the detected positional deviation meets the preset adjustment conditions, then based on the updated sound source position and the second preset speed, the microphone array is controlled to perform a re-detection direction adjustment so that the center of the microphone array is aligned with the sound source.

[0127] Optionally, the partial discharge detection device also includes:

[0128] The operation and maintenance report generation module is used to generate processing suggestions corresponding to the partial discharge type, generate an operation and maintenance report based on the partial discharge type and the processing suggestions, and send the operation and maintenance report to the target user through a preset method.

[0129] The partial discharge detection device provided in this embodiment of the invention can execute the partial discharge detection method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0130] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0131] Example 4

[0132] Figure 5 A schematic diagram of an electronic device 40 that can be used to implement embodiments of the present invention is shown. The electronic device 40 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device 40 can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0133] like Figure 5 As shown, the electronic device 40 includes at least one processor 41 and a memory, such as a read-only memory (ROM) 42 or a random access memory (RAM) 43, communicatively connected to the at least one processor 41. The memory stores computer programs executable by the at least one processor. The processor 41 can perform various appropriate actions and processes based on the computer program stored in the read-only memory 42 or loaded from the storage unit 48 into the random access memory 43. The RAM 43 can also store various programs and data required for the operation of the electronic device 40. The processor 41, ROM 42, and RAM 43 are interconnected via a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0134] Multiple components in electronic device 40 are connected to I / O interface 45, including: input unit 46, such as keyboard, mouse, etc.; output unit 47, such as various types of monitors, speakers, etc.; storage unit 48, such as disk, optical disk, etc.; and communication unit 49, such as network card, modem, wireless transceiver, etc. Communication unit 49 allows electronic device 40 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0135] Processor 41 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, central processing units, graphics processing units, various special-purpose artificial intelligence computing chips, various processors running machine learning model algorithms, digital signal processors, and any suitable processor, controller, microcontroller, etc. Processor 41 performs the various methods and processes described above, such as methods for detecting partial discharge.

[0136] In some embodiments, the partial discharge detection method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 48. In some embodiments, part or all of the computer program may be loaded and / or mounted on electronic device 40 via ROM 42 and / or communication unit 49. When the computer program is loaded into RAM 43 and executed by processor 41, one or more steps of the partial discharge detection method described above may be performed. Alternatively, in other embodiments, processor 41 may be configured to perform the partial discharge detection method by any other suitable means (e.g., by means of firmware).

[0137] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays, application-specific integrated circuits (ASICs), application-specific standard products (ASICs), system-on-a-chip (SoCs), complex programmable logic devices, computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0138] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0139] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, optical fibers, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0140] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device 40, which includes: a display device (e.g., a cathode ray tube or liquid crystal display) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device 40. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0141] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0142] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact via a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server.

[0143] This embodiment may also include a computer program product, which includes a computer program that, when executed by a processor, implements the partial discharge detection method provided in any embodiment of the present invention.

[0144] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0145] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for detecting partial discharge, characterized in that, include: During partial discharge inspection, the initial sound source position is obtained, and after adjusting the detection direction of the microphone array according to the initial sound source position, the partial discharge signal is obtained through the microphone array. Based on the partial discharge signal, the pulse density is obtained, and based on the pulse density and the mapping relationship between the preset pulse density range and the sampling frequency, the current sampling frequency is obtained. Based on the partial discharge signal, the background noise amplitude is obtained, and based on the background noise amplitude, the current threshold is obtained; When the amplitude of the partial discharge signal is detected to be greater than the current threshold, discharge pulse sampling is triggered, and the power frequency voltage signal is acquired. By using a preset phase-locked loop, a synchronous clock is obtained based on the power frequency voltage signal, and discharge pulses are sampled based on the current sampling frequency and the synchronous clock to obtain each initial discharge pulse and the corresponding power frequency phase angle. Based on the power frequency phase angle corresponding to each initial discharge pulse and the mapping relationship between the preset phase layer and the power frequency phase angle range, the phase layer corresponding to each initial discharge pulse is obtained. Based on the phase layer corresponding to each initial discharge pulse, the standard deviation of the pulse amplitude corresponding to each phase layer is obtained, and based on the standard deviation of the pulse amplitude, the density adjustment coefficient corresponding to each phase layer is obtained. Pulse interpolation is performed on each phase layer according to the density adjustment coefficient to obtain each target discharge pulse and the corresponding power frequency phase angle, and a phase analysis partial discharge map is generated according to each target discharge pulse and the corresponding power frequency phase angle. The partial discharge type is obtained based on the phase-analyzed partial discharge map by using a pre-trained feature extraction model and a partial discharge type identification model.

2. The method according to claim 1, characterized in that, Based on the standard deviation of the pulse amplitude, the density adjustment coefficient corresponding to each phase layer is obtained, including: Obtain the maximum pulse amplitude corresponding to each phase layer, and obtain the density adjustment threshold corresponding to each phase layer based on the maximum pulse amplitude; The density adjustment coefficients corresponding to each phase layer are obtained based on the pulse amplitude standard deviation and the density adjustment threshold.

3. The method according to claim 1, characterized in that, The microphone array is adjusted for detection direction based on the initial sound source location, including: Obtain the risk level corresponding to the initial sound source location. If the risk level is determined to be high risk, then control the microphone array to perform an initial detection direction adjustment based on the initial sound source location and the first preset speed. After the adjustment is completed, activate the microphone subarray corresponding to the initial sound source location. Acquire the sampling signals from the microphone subarray, and obtain and update the sound source position based on the sampling signals; Based on the updated sound source location and the second preset speed, the microphone array is controlled to re-adjust its detection direction so that the center of the microphone array is aligned with the sound source. The second preset speed is less than the first preset speed.

4. The method according to claim 3, characterized in that, Based on the updated sound source location and the second preset speed, the microphone array is controlled to perform a re-detection direction adjustment so that the center of the microphone array is aligned with the sound source, including: Calculate the positional deviation between the updated sound source position and the center of the microphone array; If the detected positional deviation meets the preset adjustment conditions, then based on the updated sound source position and the second preset speed, the microphone array is controlled to perform a re-detection direction adjustment so that the center of the microphone array is aligned with the sound source.

5. The method according to claim 1, characterized in that, After obtaining the partial discharge type, the following is also included: A processing suggestion corresponding to the partial discharge type is generated, and an operation and maintenance report is generated based on the partial discharge type and the processing suggestion. The operation and maintenance report is then sent to the target user through a preset method.

6. An electronic device, characterized in that, The electronic device includes: At least one processor, and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the partial discharge detection method according to any one of claims 1-5.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the partial discharge detection method according to any one of claims 1-5.

8. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the partial discharge detection method according to any one of claims 1-5.

Images