Partial discharge monitoring method, system, medium and equipment

Abstract

The invention relates to a partial discharge monitoring method and system, a medium and equipment. The method comprises the following steps: I, capturing spatial distribution characteristics of multi-channel partial discharge signals by using a radar sensor through an array scanning technology; step II, carrying out amplification, filtering and anti-interference conditioning on the partial discharge signal; and step III, carrying out time difference positioning and visualization on the conditioned partial discharge signal. According to the partial discharge monitoring method, the radar array scanning technology is adopted, and high-sensitivity monitoring of partial discharge of the power equipment can be achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power equipment state monitoring; in particular, the present application relates to a partial discharge monitoring method, system, medium and equipment. BACKGROUND

[0002] Partial discharge is a common precursor to failure in the insulation system of power equipment, which is usually caused by the decline of partial insulation strength due to insulation material aging, mechanical damage or environmental factors (such as humidity, temperature, etc.). The long-term existence of partial discharge can lead to further deterioration of insulation performance, and eventually may cause equipment failure or even major accidents of power systems. Therefore, the monitoring and diagnosis of partial discharge is of great significance to ensure the safe and stable operation of power systems.

[0003] As an important tool for power equipment state monitoring, the partial discharge monitoring intelligent terminal can monitor the partial discharge of power equipment in real time, discover potential faults in time and provide diagnostic basis. At present, the partial discharge monitoring technology has been widely applied in switchgear, cable, transformer and other power equipment, and has become an important part of power system state monitoring and fault warning.

[0004] The existing partial discharge monitoring technology is mainly based on ultrasonic principle, transient earth voltage (TEV) principle or ultra-high frequency (UHF) principle, although certain achievements have been made in practical application, but there are still various problems. SUMMARY

[0005] Therefore, the present application provides a partial discharge monitoring method, system, medium and equipment, thereby solving or at least alleviating one or more of the above problems and other aspects in the prior art.

[0006] In order to achieve the foregoing purpose, the first aspect of the present application provides a partial discharge monitoring method, wherein the partial discharge monitoring method comprises: Step I, using a radar sensor to capture the spatial distribution characteristics of multi-channel partial discharge signals by array scanning technology; Step II, amplifying, filtering and anti-interference conditioning the partial discharge signals; Step III, time difference positioning and visualization of the conditioned partial discharge signals.

[0007] In the partial discharge monitoring method as described above, optionally, in the step I, the radar sensor is a radar array sensor composed of MxN antenna units, and its working frequency band is 300MHz-3GHz, which is used to receive electromagnetic wave signals generated by partial discharge.

[0008] In the partial discharge monitoring method as described above, optionally, in the step II, the conditioning step comprises adaptive filtering, pulse alignment and feature band extraction, wherein the adaptive filtering adopts a wavelet transform and Kalman filtering fusion algorithm, the noise printing ratio is ≥ 40 dB, and the implementation steps of the wavelet transform and Kalman filtering fusion algorithm comprise: performing wavelet transform on the partial discharge signal to decompose the partial discharge signal into approximation and detail coefficients of different frequency bands; in each scale, estimating the distribution characteristics of noise by a statistical method; according to the estimated noise level, dynamically adjusting the threshold of the wavelet coefficient, and performing hard thresholding or soft thresholding processing to remove the noise part; applying Kalman filtering to the signal after wavelet transform, establishing a state space model of the signal, and estimating the true value of the signal by a recursive algorithm to further optimize the signal quality; and reconstructing the wavelet coefficient after denoising to obtain the signal after denoising.

[0009] In the partial discharge monitoring method as described above, optionally, in the step III, a three-dimensional imaging of the discharge source of the partial discharge signal is performed by a back projection algorithm to generate a heat map or a point cloud model, and the spatial resolution is ≤ 1 cm.

[0010] In the partial discharge monitoring method as described above, optionally, the back projection algorithm comprises: meshing the monitoring space of the radar sensor to discretize it into voxels, each voxel representing a potential discharge source position; determining the geometric parameters of the array to construct a propagation model of the partial discharge signal; extracting data features related to back projection in the multi-channel partial discharge signal; performing back projection calculation based on the data features; performing three-dimensional imaging optimization and threshold processing on the data of the back projection calculation; and generating a three-dimensional visual image based on the optimized data and the processed threshold.

[0011] In the partial discharge monitoring method as described above, optionally, the step of meshing the monitoring space of the radar sensor comprises: defining a three-dimensional coordinate system according to the coverage capability of the array, taking the center of the array as the origin, and the x, y and z axes corresponding to the length, width and height respectively to determine the spatial boundary of the monitoring space, wherein , discretizing the monitoring space into N×P×Q voxels according to the resolution requirement, and setting the edge length Δ of each voxel to 0.5-1 cm, and the voxel coordinates are wherein, i = 1..N, j = 1..P, k = 1..Q; The step of determining the geometric parameters of the array comprises: Measuring the three-dimensional coordinates of the MxN antenna units in the array wherein, m = 1..M, n = 1..N, ensuring that the error is ≤0.1 cm; The step of constructing the propagation model of the partial discharge signal comprises: Calculating the straight-line distance from each voxel to the antenna is , Based on the signal strength attenuation law, combined with the dielectric loss coefficient correction, record the attenuation model as G(d), The propagation time of the voxel to the antenna is The step of extracting data features related to back projection comprises: For each antenna channel (m, n), extract the peak amplitude of the partial discharge pulse from the denoised signal , and remove the environmental noise base, Determine the accurate time when each antenna receives the discharge signal through the pulse front edge detection algorithm ; The step of the back projection technology comprises: For each antenna (m, n), traverse all voxels , calculate the theoretical signal strength that the antenna (m, n) should receive if the voxel is a discharge source , and compare it with the actual received strength , convert the matching degree into the energy contribution of the voxel, wherein, The theoretical strength calculation formula is wherein K is the inherent strength of the discharge source, which is assumed to be a constant, The energy contribution calculation formula is: if wherein Δt is the time error threshold, then the voxel obtains energy , otherwise the contribution is 0, The total energy of each voxel is calculated by the sum of the back projection contributions of all antennas , The higher the total energy , the greater the probability that the voxel is a real discharge source.

[0012] To achieve the foregoing purposes, a second aspect of the present application provides a partial discharge monitoring system for executing the partial discharge monitoring method according to any one of the preceding first aspects, wherein the partial discharge monitoring system comprises: a radar sensor module, which adopts a radar array to differentially detect a partial discharge signal and capture spatial distribution characteristics of the partial discharge signal; a signal conditioning module, which is configured to amplify, filter and / or remove noise interference from the signal collected by the radar sensor; and a data processing module, which is configured to analyze the signal and realize positioning and visualization of the partial discharge source.

[0013] In the partial discharge monitoring system as described above, optionally, the partial discharge monitoring system further comprises: a communication module, which is configured to realize remote transmission of monitoring data through wired communication or wireless communication; and a power supply module, which is configured to supply power to the system and ensure normal operation of the device in the case of power failure, and the visualization step of the data processing module comprises: gridding an imaging space; constructing a radar array geometry parameter and a signal propagation model; extracting data characteristics related to back projection from the multi-channel signal; performing back projection calculation based on the data characteristics; performing three-dimensional imaging optimization and threshold processing on the data of the back projection calculation; and generating a three-dimensional visualization image based on the optimized data and the processed threshold.

[0014] To achieve the foregoing object, a third aspect of the present application provides a computer readable medium having a computer program stored thereon, wherein the program, when executed by a processing device, implements the steps of the method according to any one of the preceding first aspect.

[0015] To achieve the foregoing object, a fourth aspect of the present application provides an electronic device, wherein the electronic device comprises: a storage device having a computer program stored thereon; a processing device configured to execute the computer program in the storage device to implement the steps of the method according to any one of the preceding first aspect.

[0016] The partial discharge monitoring method of the present application adopts a radar array scanning technology and can realize high-sensitivity monitoring of partial discharge of power equipment such as switch cabinets and cables. In the optional technical solution, the method can have the advantages of high sensitivity, high precision, strong anti-interference capability and real-time performance.

[0017] The present application further provides a partial discharge monitoring system for executing the partial discharge monitoring method of the present application, and therefore the system also has the advantages described above.

[0018] The present application further provides a computer readable medium and an electronic device for performing the partial discharge monitoring method of the present application, thus the medium and device also have the above advantages. BRIEF DESCRIPTION OF DRAWINGS

[0019] The disclosure of the present application will become more apparent from the following description in conjunction with the accompanying drawings. It is to be understood that the drawings are only for the purpose of illustration and are not to be construed as limiting the scope of the present application. In the drawings: Figure 1 is a schematic block diagram of an embodiment of the partial discharge monitoring method according to the present application; and Figure 2 is a schematic block diagram of an embodiment of the partial discharge monitoring system according to the present application. DETAILED DESCRIPTION

[0020] The structure, composition, features and advantages of the partial discharge monitoring method, system, medium and device of the present application will be described below in an exemplary manner with reference to the accompanying drawings and specific embodiments, however, all the descriptions are not to be used to form any limitation on the present application.

[0021] In addition, for any single technical feature described or implied in the embodiments mentioned herein, or any single technical feature shown or implied in the drawings, the present application still allows any combination or deletion to be continued between these technical features (or their equivalents) without any technical obstacles, thus it should be considered that more embodiments according to the present application are also within the scope of the description herein.

[0022] It should also be noted that the terms "rear end" and the like indicate the orientation or positional relationship based on the orientation or positional relationship of the components of the partial discharge monitoring system shown in the drawings, which is only for the convenience of describing the present disclosure and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present disclosure.

[0023] Figure 1 is a schematic block diagram of an embodiment of the partial discharge monitoring method according to the present application.

[0024] As shown in the figure, in this embodiment, the partial discharge monitoring method includes the following steps: the radar array captures the discharge signal; signal conditioning; and time difference positioning and visualization.

[0025] In the aforementioned step I, the spatial distribution characteristics of the multi-channel partial discharge signals can be captured by the radar sensor through array scanning technology. This step is a signal acquisition step. Specifically, the radar sensor captures the spatial distribution characteristics of the multi-channel partial discharge signals through array scanning technology, and in some cases, the ambient noise signals can be acquired simultaneously. The radar sensor can cover the entire surface of the power equipment such as switchgear or cable, avoiding monitoring blind areas. Specifically, a radar array sensor composed of MxN antenna units can be used, which can have a working frequency band of 300MHz-3GHz to receive the electromagnetic wave signals generated by partial discharge. The high sensitivity and wide frequency band characteristics of the radar array can effectively capture the high-frequency pulse signals generated during partial discharge.

[0026] In the aforementioned step II, the partial discharge signals can be amplified, filtered, and anti-interference conditioned. Specifically, in this step, the signal conditioning module performs amplification, filtering, anti-interference, and other preprocessing on the acquired multi-channel signals, and outputs clear signals to the data processing module. The preprocessing can include but is not limited to adaptive filtering, pulse alignment, and feature band extraction. Among them, the adaptive filtering can use a wavelet transform and Kalman filter fusion algorithm, which can achieve a noise suppression ratio ≥40dB in this embodiment.

[0027] The implementation steps of the wavelet transform and Kalman filter fusion algorithm include signal decomposition, noise estimation, adaptive thresholding, Kalman filtering, and signal reconstruction.

[0028] In the signal decomposition step, the original partial discharge signal can be subjected to wavelet transform to decompose it into approximation and detail coefficients of different frequency bands. In the noise estimation step, the noise distribution characteristics can be estimated in each scale by statistical methods. In the adaptive thresholding step, the wavelet coefficient threshold can be dynamically adjusted according to the estimated noise level, and hard thresholding or soft thresholding processing can be performed to remove noise components. In the Kalman filtering step, the wavelet-transformed signal can be subjected to Kalman filtering to establish a signal state space model, dynamically track the time-varying characteristics of the partial discharge signal, suppress residual noise (especially time-varying noise), and optimize the signal estimate value in real time, further optimizing the signal quality and providing a more stable and accurate signal basis for subsequent data processing. In the signal reconstruction step, the denoised wavelet coefficients can be reconstructed to obtain the denoised signal.

[0029] Specifically, in the wavelet transform, the signal is recursively decomposed into approximation coefficients and detail coefficients through a series of low-pass / high-pass filters and down-sampling. Among them, Approximation coefficients are the abstract representation of low frequency part of the signal, which preserves the main profile and macroscopic form of the signal. It is the result of processing the signal by low-pass filter and down-sampling. It can be considered as the abstract of the smooth and trend part of the signal, which represents the low frequency information of the signal at each decomposition level. In partial discharge, approximation coefficients contain the slowly changing part of the partial discharge process or the low frequency interference of the background environment, such as the power frequency signal of the power grid. The role of approximation coefficients is to maintain the low frequency structure of the signal, assist in judging the basic form and long-term trend of the signal, and at the same time avoid mistaking the low frequency background as an abnormal signal, which helps to ensure the accuracy of signal reconstruction from a macroscopic point of view.

[0030] Detail coefficients are the abstract representation of high frequency part of the signal, which emphasizes the edges, jumps, noise and detail information of the signal. It is the result of processing the signal by high-pass filter and down-sampling. It captures the high frequency details or instantaneous changes in the signal, representing the high frequency information of this level. Partial discharge usually produces high frequency short pulse signals, which are concentrated in the detail coefficients. The role of detail coefficients is to highlight the high frequency characteristics of partial discharge, which helps to distinguish real discharge pulses from noise and interference signals. The effective extraction of detail coefficients is the key to partial discharge recognition and positioning.

[0031] Through wavelet transform, the useful part and interference part hidden in the signal can be more clearly distinguished, because noise usually exists in the high frequency detail part. Through decomposition, detail coefficients can be analyzed separately, which facilitates threshold processing of noise components to suppress high frequency noise while retaining partial discharge pulse signals. Then the approximation coefficients and processed detail coefficients can be recombined through inverse wavelet transform to obtain a clear signal.

[0032] Specifically, in the noise estimation, the noise level can be extracted from the detail coefficients obtained by wavelet transform decomposition, and the steps include: after the collected partial discharge signal is subjected to wavelet transform, the detail coefficients of one or several high frequency detail levels are taken; the median absolute deviation (Median Absolute Deviation) statistics are applied to these coefficients to dynamically calculate the noise level at the current time or in the current sampling window, and the noise standard deviation is estimated as where MAD is the original median absolute deviation of the detail coefficients, and the constant 0.6745 is the reciprocal of the median absolute deviation of the standard normal distribution; the noise level is updated in real time as the monitoring environment conditions change, realizing adaptive adjustment of the subsequent threshold, and ensuring that the denoising process can adapt to complex and variable working environments.

[0033] Specifically, in the adaptive thresholding, the threshold of the wavelet coefficient can adopt the Donoho threshold for soft thresholding processing. Through the calculation of the aforementioned noise level, the Donoho threshold can be calculated as where N is the data length. In some embodiments, the threshold λ can be adjusted according to different decomposition levels, noise estimates and actual observations, and can also be weighted and corrected according to an adaptive strategy of the environment. Since the partial discharge signal is usually collected in a strong noise environment, the noise level may fluctuate with the environment, electromagnetic interference, equipment state and the like, and dynamic adjustment of the threshold can avoid the problem of false deletion of discharge pulse signals caused by too high threshold or noise residue caused by too low threshold. Adaptive thresholding dynamically adjusts the threshold size according to the noise level of the current signal, and balances the balance between "filtering noise" and "keeping effective signals". It can remove random noise and small irrelevant signals that affect the accuracy of the subsequent positioning and imaging system, while retaining the short-time high-frequency pulse characteristics generated by partial discharge, preventing key signals from being lost due to false noise removal, thereby improving the signal-to-noise ratio (SNR) of the signal, providing a more accurate and clear signal basis for subsequent time difference of arrival (TDOA) positioning and back-projection three-dimensional imaging, and thus reducing false positives caused by noise misjudgment, ensuring the reliability of the monitoring system.

[0034] Specifically, in the Kalman filtering, the recursive algorithm and filtering operation can reduce the interference of noise on the detection of signal arrival time, such as pulse front jitter caused by noise, and by dynamically tracking the signal, ensure that the signal pulses of different antenna channels are more accurately aligned, reduce the time difference error between channels, and finally control the time difference error to the nanosecond level to support the millimeter-level positioning of the discharge source in the subsequent steps. Moreover, the signal-to-noise ratio of the Kalman filtered signal is higher, the strength characteristics of the real discharge signal are more prominent, and the time domain continuity is better, and the consistency of the signals received by different antennas is stronger, avoiding or reducing the "junk points" (false discharge sources) or "blurring" (unclear boundaries of real discharge sources) phenomenon caused by residual noise or artifacts in the signal in subsequent three-dimensional imaging.

[0035] In the foregoing step III, the radar array scanning technology can be used to realize millimeter-level positioning and visualization of the partial discharge source by multi-sensor cooperation, high-precision time difference positioning and three-dimensional imaging algorithm, solving the problem of weak anti-interference and low precision of traditional methods. This step is a data processing step.

[0036] Specifically, the time difference positioning is based on the time difference of arrival (TDOA) algorithm, using the multi-antenna nodes of the radar array to calculate the arrival time difference of the partial discharge signals received by multiple antennas, thereby determining the initial position coordinates of the discharge source.

[0037] Specifically, the three-dimensional imaging can employ a back-projection algorithm to perform three-dimensional imaging on the discharge source to generate a high-resolution thermal map or point cloud model, and in this embodiment, the spatial resolution is ≤1 cm. The core is to reconstruct the three-dimensional position and energy distribution of the discharge source in reverse through the spatial distribution characteristics of the multi-channel signals of the radar array, combined with the electromagnetic wave propagation characteristics (such as spherical wave diffusion, attenuation law) of the partial discharge signal and the geometric parameters of the radar array, to realize the visualization of millimeter-level precision.

[0038] The specific steps of the back-projection algorithm can include imaging space gridding, radar array geometric parameter and signal propagation model construction, multi-channel data feature extraction, back-projection calculation, three-dimensional imaging optimization and threshold processing, and three-dimensional visualization.

[0039] In the step of imaging space gridding, a high-resolution foundation can be laid. The monitoring area (such as the inside of a switch cabinet, the surface coverage of a cable) is discretized into three-dimensional pixels, i.e. voxels. Each voxel represents a potential discharge source position, and the size of the voxel directly determines the spatial resolution, for example, in this embodiment it can be ≤1 cm. When determining the monitoring space range, for example, according to the coverage capability of the radar array, such as monitoring a space range of a switch cabinet with dimensions of 1 m x 0.5 m x 0.5 m, a three-dimensional coordinate system is defined with the array center as the origin, and the x, y, z axes correspond to the length, width and height respectively, to determine the spatial boundary, i.e. When performing voxel division, the three-dimensional space is discretized into N x P x Q voxels according to the resolution requirement (such as resolution ≤1 cm), and the edge length Δ of each voxel is set to 0.5-1 cm to ensure that the resolution is ≤1 cm, and the voxel coordinates are denoted as where i = 1..N, j = 1..P, k = 1..Q. In this example, a space of 1 m x 0.5 m x 0.5 m is divided according to a resolution of 1 cm, and 100 x 50 x 50 = 250,000 voxels can be obtained.

[0040] In the step of radar array geometric parameter and signal propagation model construction, the accuracy of back-projection can be determined, because the accuracy of back-projection depends on the spatial position calibration of the radar array and the modeling of the electromagnetic wave propagation law. In the antenna element coordinate calibration, the three-dimensional coordinates of the M x N antenna elements in the array are measured where m = 1..M, n = 1..N, to ensure that the error is ≤0.1 cm, avoiding imaging distortion caused by antenna position deviation. Regarding the electromagnetic wave propagation model, a propagation model of the electromagnetic wave generated by the partial discharge signal is constructed, and the signal strength decays with distance and there is a propagation time delay when the electromagnetic wave propagates in the medium, where the medium includes air, insulating materials, etc. Specifically, the construction of the electromagnetic wave propagation model requires distance calculation, attenuation model and propagation time, including calculating the distance between each voxel To the antenna is a straight line distance ; based on the signal strength attenuation law (such as spherical wave attenuation, intensity is proportional to ), combined with the dielectric loss coefficient (such as the reflection loss of the metal components in the switch cabinet), the correction is recorded as the attenuation factor function G(d); according to the electromagnetic wave propagation speed c, the propagation time of the voxel to the antenna is , wherein, in the air, .

[0041] In the step of multi-channel data feature extraction, key data features related to back projection can be extracted from the preprocessed multi-channel signal, such as signal strength, arrival time, etc. Specifically, for signal strength extraction, the data feature of signal strength can be extracted by extracting the peak amplitude of the partial discharge pulse from the denoised signal to represent the signal energy received by the antenna, and the environmental noise floor is removed by baseline calibration; for arrival time extraction, the data feature of arrival time can be determined by the pulse front detection algorithm to determine the exact time when the discharge signal is received by each antenna, which is used to verify the consistency of the propagation time. The threshold trigger method can be used in the pulse front detection algorithm.

[0042] In the step of back projection calculation, the core can be to project the signal energy of each antenna back to all possible discharge source positions, i.e. voxels, and accumulate the energy, and the voxel with the highest energy is the discharge source position. Specifically, for single antenna back projection, for each antenna (m, n), all voxels are traversed , and the theoretical signal strength that the antenna (m, n) should receive if the voxel is a discharge source is calculated , and compared with the actual received intensity , and the matching degree is converted into the energy contribution of the voxel. Specifically, the theoretical intensity calculation formula is , wherein K is the inherent intensity of the discharge source, which is assumed to be a constant. Specifically, the energy contribution calculation formula is: if the signal propagation model matches the propagation time obtained by multi-channel data feature extraction, i.e. , wherein Δt is the time error threshold, which can be set to 1 ns, then the voxel obtains energy to back-propagate the discharge source intensity; otherwise, the contribution is 0, and the invalid position of the propagation time mismatch is excluded. For multi-antenna energy accumulation, the total energy of each voxel is calculated by the sum of the contributions of all antenna back projections . Because the signal contribution of the real discharge source to all antennas should be consistent, while the contribution of noise is randomly distributed, the higher the total energy , the greater the probability that the voxel is the real discharge source.

[0043] In the step of three-dimensional imaging optimization and threshold processing, in order to improve the imaging clarity, the voxel energy distribution can be optimized to remove noise interference. First, combined with the initial positioning optimization of TDOA, the initial coordinates of the discharge source obtained by the time difference positioning , its surrounding voxels can be given higher weights to accelerate convergence to the true position, wherein, as an example, the surrounding voxels can be taken within 5cm, and the higher weight can be taken as multiplied by a factor of 1.2. Secondly, threshold filtering is performed, which sets the energy threshold , the voxel energy is set to 0 to remove noise-induced low-energy noise and retain high-energy areas, wherein, as an example, the energy threshold can be taken as the 90th percentile of all voxel energies.

[0044] In the step of three-dimensional visualization, a heat map and a point cloud model can be generated. The processed voxel energy distribution is converted into an intuitive three-dimensional image, such as a heat map, a point cloud model, etc. Specifically, the heat map generation includes: smoothing the voxel energy distribution by a three-dimensional interpolation algorithm, representing the energy intensity with a color gradient, generating a three-dimensional heat map of the discharge source, and visually displaying the spatial distribution, wherein, as an example, the three-dimensional interpolation algorithm can be linear interpolation. The point cloud model generation includes: extracting the voxel coordinates with energy , assigning the size / color of the point according to the energy intensity, the higher the energy, the larger / bright the point, generating a point cloud model, supporting interactive operations such as rotation and scaling, and facilitating observation of the three-dimensional morphology of the discharge source.

[0045] In the partial discharge monitoring method of the present application, in order to meet the spatial resolution requirement, the voxel size, antenna array density and time accuracy parameters in the above steps need to be controlled. In this embodiment, the spatial resolution is required to be ≤1cm; in order to achieve a spatial resolution of ≤1cm, in the embodiment of the present application, the following can be set: regarding the voxel size, the voxel edge length is ≤1cm, ensuring that the minimum resolvable unit meets the requirements; regarding the antenna array density, the number of MxN antennas is sufficient, which can be a 16x16 array, providing multi-directional projection data to avoid blurring caused by insufficient spatial sampling; regarding the time accuracy, the time of arrival measurement error is ≤1ns, corresponding to a spatial error of ≤3mm, ensuring the accuracy of the propagation time matching verification. In some embodiments, optionally, the spatial resolution of imaging can be ensured to be ≤1cm by optimizing the parameters of the matched filter and convolution operation.

[0046] In this embodiment, further, the partial discharge monitoring method can further comprise data transmission. The processed monitoring data can be transmitted to a monitoring terminal (e.g. a ground monitoring center) through wired or wireless communication to position the results and send alarm information to the staff for remote monitoring and diagnosis. When an abnormality is detected, an alarm signal is triggered and alarm information is sent, thereby achieving real-time monitoring and early warning to ensure timely maintenance measures are taken.

[0047] In this embodiment, further, the partial discharge monitoring method can further comprise alarm and display. The partial discharge monitoring system is equipped with an audible and visual alarm device that issues a warning prompt when an abnormal partial discharge is detected. The data processing results can be viewed through a local display screen or a remote terminal.

[0048] Figure 2 A schematic block diagram of one embodiment of the partial discharge monitoring system according to the present application.

[0049] As shown in the figure, in this embodiment, the partial discharge monitoring system can include a radar sensor module, a signal conditioning module, and a data processing module, which can be used to perform the partial discharge monitoring method as described in any of the preceding embodiments. In some embodiments, the radar sensor module can use a radar array to differentially detect partial discharge signals and capture the spatial distribution characteristics of the partial discharge signals; the signal conditioning module can be used to amplify, filter, and / or remove noise interference from the signals collected by the radar sensor; and the data processing module can be used to analyze the signals and achieve localization and visualization of the partial discharge source. The partial discharge monitoring system can be a monitoring terminal device that can ensure that weak partial discharge signals can be collected.

[0050] In further embodiments, the partial discharge monitoring system can include a communication module for remote transmission of monitoring data through wired or wireless communication, and a power supply module for powering the system to ensure normal operation of the device in the event of a power outage. The visualization step of the data processing module includes: gridding the imaging space; constructing a radar array geometry parameter and signal propagation model; extracting data features related to back projection from the multi-channel signals; performing back projection calculation based on the data features; performing three-dimensional imaging optimization and threshold processing on the data of the back projection calculation; and generating a three-dimensional visualization image based on the optimized data and processed threshold. For detailed descriptions of these steps, please refer to the descriptions above in connection with the partial discharge monitoring method, which will not be repeated here.

[0051] In further embodiments, the partial discharge monitoring system can have a central controller that can be used to control the radar sensor module, the signal conditioning module, and the data processing module to perform the aforementioned partial discharge monitoring method.

[0052] In specific embodiments, the radar sensor module can be selected as AE503D differential sensor, which adopts radar array scanning technology and can capture the spatial distribution characteristics of partial discharge signals. The radar sensor module supports high-sensitivity signal detection and can support the data processing module to distinguish partial discharge signals from environmental noise.

[0053] In specific embodiments, the signal conditioning module can include amplification circuit, filtering circuit and anti-interference circuit. By amplifying and filtering the signals collected by the radar sensor, noise interference can be removed to ensure the accuracy of the signals. In this embodiment, the amplification circuit can use a common local operational amplifier circuit, and the filtering circuit can use a common RC low-pass filter circuit.

[0054] In specific embodiments, the data processing module can include an ARM processor and / or an FPGA. In this embodiment, the data processing module can be built-in a high-performance embedded microcontroller STM32 series based on the ARM Cortex-M series kernel, such as STM32F103C8T6, which is used for feature extraction and analysis of collected signals. The data processing module supports time domain, frequency domain and spatial domain signal processing algorithms, and can extract parameters such as amplitude and spectral characteristics of partial discharge signals through spectral analysis. In some embodiments, the data processing module can integrate intelligent pattern recognition algorithms, which can automatically diagnose the type and severity of partial discharge according to the characteristic map of partial discharge.

[0055] In this embodiment, further, the partial discharge monitoring system can further include a housing, a communication module, and a power module, etc.

[0056] According to different embodiments, the housing can be designed with waterproof, dustproof and shockproof to adapt to complex power equipment operating environment. It can also have a magnetic or bundled mounting structure for quick deployment and disassembly. For example, the radar sensor can be attached or tied to the rear end of the housing. In some embodiments, the housing can also include an LCD display and control buttons for output display and input control.

[0057] According to different embodiments, the communication module can support wired communication (such as RS485, Ethernet) and wireless communication (such as Wi-Fi, LoRa, 5G) to realize remote transmission of monitoring data for real-time viewing and analysis by the ground monitoring center. According to specific needs, data transmission can use encryption protocols to ensure communication security.

[0058] According to different embodiments, the power module can be designed with low power consumption and support multiple power supply modes (such as lithium batteries, external power supply, etc.). In addition, the system can also be equipped with a backup power supply to ensure that the device can still work normally in case of power failure.

[0059] Further, the present application also provides a computer readable medium, which can have a computer program stored thereon, the program being executed by a processing device to implement the steps of the partial discharge monitoring method according to any one of the preceding embodiments.

[0060] Further, the present application also provides an electronic device, which can include a storage device and a processing device, the storage device can have a computer program stored thereon, and the processing device can be configured to execute the computer program in the storage device to implement the steps of the partial discharge monitoring method according to any one of the preceding embodiments.

[0061] In particular, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, embodiments of the present application can include a computer program product comprising a computer program carried on a non-transitory computer readable medium, the computer program comprising program code for implementing the methods illustrated by the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device. When the computer program is executed by a processing device, the above-mentioned functions defined in the methods of the embodiments of the present application are implemented.

[0062] It should be noted that the computer readable medium of the present application can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may, for example, be, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or apparatus. In the present application, the computer readable signal medium can include a data signal carried in a baseband or as a part of a carrier wave, which carries computer readable program code. Such a propagated data signal can take many forms, including but not limited to an electromagnetic signal, an optical signal or any suitable combination of the above. The computer readable signal medium can also be any computer readable medium other than the computer readable storage medium, which can send, propagate or transmit a program for use by or in conjunction with an instruction execution system, device or apparatus. The program code contained in the computer readable medium can be transmitted by any suitable medium, including but not limited to a wire, a cable, an RF (radio frequency) or the like, or any suitable combination of the above.

[0063] The computer readable medium described above can be contained in the electronic device described above, or can exist separately and not be assembled into the electronic device.

[0064] The computer readable medium described above carries one or more programs, which, when executed by the electronic device, cause the electronic device to: capture the spatial distribution characteristics of the multi-channel partial discharge signal by array scanning technology using the radar sensor; amplify, filter and anti-interference condition the partial discharge signal; and perform time difference positioning and visualization on the conditioned partial discharge signal, which uses radar array scanning technology to achieve high sensitivity monitoring of the partial discharge of the power equipment.

[0065] The modules described in the embodiments of the present application can be implemented in software or hardware. In some cases, the name of the module does not constitute a limitation on the module itself.

[0066] Some embodiments of the partial discharge monitoring method of the present application can have one or more of the following advantages compared with the prior art: (1) high sensitivity: receiving partial discharge signals through a radar array sensor can detect weak discharge signals; (2) high precision: three-dimensional imaging based on the back projection algorithm realizes a spatial resolution of ≤1 cm, and the positioning accuracy is significantly improved; (3) strong anti-interference ability: the adaptive filtering algorithm realizes a noise suppression ratio ≥40 dB, ensuring signal quality; (4) real-time: the positioning results and alarm information are transmitted in real time through a wireless communication module, meeting the actual application requirements.

[0067] In particular, the partial discharge monitoring method of the present application uses radar array scanning technology, which, compared with the existing technology based on ultra-high frequency (UHF) monitoring, although both involve detection of electromagnetic signals, differs in frequency range, purpose, technical principle and application scenario. The radar sensor in the present application focuses more on active emission and echo analysis to realize target detection and distance measurement, and has a wider working frequency and technology; while ultra-high frequency detection is more directed at receiving and analyzing radio wave signals in a specific frequency band, and may be applied to the fields of radio monitoring and communication, and the working frequency is usually slightly lower and passive detection is the main. Radar array scanning technology is active detection, while ultra-high frequency is passive detection; radar array scanning technology can detect partial discharge, mainly based on the fact that electromagnetic wave signals are generated during partial discharge, and the radar system, especially the array scanning radar, has high sensitivity, high resolution and spatial positioning capability, and can capture these signals and locate the discharge source.

[0068] The technical scope of the present application is not limited to the content in the above description, and those skilled in the art can make various modifications and changes to the above embodiments without departing from the technical idea of the present application, and these modifications and changes should all belong to the scope of the present application.

Claims

1. A method for monitoring partial discharge, characterized in that, The method includes: Step 1: Use radar sensors to capture the spatial distribution characteristics of multi-channel partial discharge signals through array scanning technology; Step II involves amplifying, filtering, and anti-interference conditioning the partial discharge signal. Step III: Perform time-difference localization and visualization on the conditioned partial discharge signal.

2. The method as described in claim 1, characterized in that, In step I, the radar sensor is a radar array sensor composed of M×N antenna elements, with an operating frequency band of 300MHz-3GHz, used to receive electromagnetic wave signals generated by partial discharge.

3. The method as described in claim 1, characterized in that, In step II, the conditioning steps include adaptive filtering, pulse alignment, and feature band extraction. The adaptive filtering employs a wavelet transform and Kalman filter fusion algorithm with a noise suppression ratio ≥40dB. The implementation steps of the wavelet transform and Kalman filter fusion algorithm include: The partial discharge signal is decomposed into approximate and detail coefficients in different frequency bands by performing wavelet transform. At each scale, the distribution characteristics of the noise are estimated using statistical methods; Based on the estimated noise level, the threshold of the wavelet coefficients is dynamically adjusted to perform hard or soft thresholding to remove noise components. Kalman filtering is applied to the wavelet-transformed signal to establish a state-space model. The true value of the signal is estimated using a recursive algorithm, further optimizing the signal quality. The denoised wavelet coefficients are reconstructed to obtain the denoised signal.

4. The method as described in claim 1, characterized in that, In step III, the discharge source of the partial discharge signal is three-dimensionally imaged using a back-projection algorithm to generate a heat map or point cloud model with a spatial resolution ≤1cm.

5. The method as described in claim 4, characterized in that, In step III, the back-projection algorithm includes: The monitoring space of the radar sensor is gridded and discretized into voxels, each voxel representing a potential discharge source location; Determine the geometric parameters of the array and construct a propagation model for the partial discharge signal; Extract the back-projection related data features from the multi-channel partial discharge signal; Perform back projection calculation based on the data features; Perform 3D imaging optimization and thresholding on the back-projection calculation data; and 3D visualization images are generated based on optimized data and processed thresholds.

6. The method as described in claim 5, characterized in that, The step of gridding the monitoring space of the radar sensor includes: A three-dimensional coordinate system is defined based on the coverage capability of the array, with the center of the array as the origin, and the x, y, and z axes corresponding to length, width, and height, respectively, to determine the spatial boundary of the monitoring space. , The monitoring space is discretized into N×P×Q voxels according to the resolution requirements, with the side length Δ of each voxel set to 0.5-1cm, and the voxel coordinates are as follows: Where i=1..N, j=1..P, k=1..Q; The steps for determining the geometric parameters of the array include: Measure the three-dimensional coordinates of the M×N antenna elements in the array Where m = 1..M, n = 1..N, ensuring an error ≤ 0.1cm. The steps for constructing the propagation model of the partial discharge signal include: Calculate each voxel to antenna The straight-line distance is: ， Based on the signal strength attenuation law and combined with the correction of the dielectric loss coefficient, the attenuation model is denoted as follows: ; With the speed of electromagnetic wave propagation The propagation time from the voxel to the antenna is then... ; The steps for extracting data features related to back projection include: For each antenna channel Extract the peak amplitude of the partial discharge pulse from the denoised signal. And remove the environmental noise base. The precise moment when each antenna receives the discharge signal is determined using a pulse leading edge detection algorithm. ; The steps of the reverse projection technique include: For each antenna traverse all voxels Calculate the antenna if the voxel is the discharge source. Theoretical signal strength to be received and the actual received strength The comparison converts the matching degree into the energy contribution of the voxel, whereby... The theoretical strength calculation formula is as follows Where K is the inherent strength of the discharge source, assumed to be constant. The formula for calculating energy contribution is: If Where Δt is the time error threshold, then the voxel gains energy. Otherwise, the contribution is 0. The total energy of each voxel is calculated by summing the back-projection contributions of all antennas: Total Energy The higher the value, the greater the probability that the voxel is a real discharge source.

7. A partial discharge monitoring system for performing the partial discharge monitoring method as described in any one of claims 1 to 6, characterized in that, The system includes: The radar sensor module uses a radar array to perform differential detection on the partial discharge signal and capture the spatial distribution characteristics of the partial discharge signal. The signal conditioning module is used to amplify, filter, and / or remove noise interference from the signals acquired by the radar sensor; and The data processing module is used to analyze the signal and realize the location and visualization of the partial discharge power source.

8. The system as described in claim 7, characterized in that, The system further includes: A communication module, wherein the communication module enables remote transmission of monitoring data via wired or wireless communication; and The power module supplies power to the system, ensuring the equipment functions normally during power outages. Furthermore, the visualization steps of the data processing module include: Gridded imaging space; Construct a model of radar array geometric parameters and signal propagation; Extract back-projection related data features from multi-channel signals; Perform back projection calculation based on the data features; Perform 3D imaging optimization and thresholding on the back-projection calculation data; and 3D visualization images are generated based on optimized data and processed thresholds.

9. A computer-readable medium having a computer program stored thereon, characterized in that, When the program is executed by the processing device, it performs the steps of the method as described in any one of claims 1 to 6.

10. An electronic device, characterized in that, The electronic device includes: A storage device on which computer programs are stored; A processing device for executing the computer program in the storage device to implement the steps of the method according to any one of claims 1 to 6.

Images