Partial discharge map generation method and device and storage medium

Abstract

The invention provides a partial discharge map generation method and device, and a storage medium, which can visually reflect the cycle characteristics of a discharge phenomenon in an AC voltage period, and improve the visualization effect of a partial discharge map. The method comprises the following steps: acquiring a partial discharge pulse signal sequence of power equipment, wherein the partial discharge pulse signal sequence comprises a plurality of partial discharge pulse signals; based on the power frequency period of the partial discharge pulse signal sequence, converting the arrival time of each partial discharge pulse signal into a corresponding phase angle, and normalizing each phase angle into an angle interval of a polar axis in a polar coordinate system; determining each phase angle as an angle coordinate in a polar coordinate system, determining the amplitude of the corresponding partial discharge pulse signal as a radial coordinate in the polar coordinate system, and obtaining a plurality of polar coordinate data points; a partial discharge map is generated based on the plurality of polar coordinate data points.

Description

Technical Field

[0001] This application relates to the field of partial discharge detection and analysis technology, and in particular to a method, apparatus and storage medium for generating partial discharge patterns. Background Technology

[0002] Partial discharge (PD) is a key indicator for assessing the insulation condition of high-voltage power equipment. By analyzing the discharge characteristics at different phases, the type and severity of insulation defects can be effectively identified, thereby enabling the detection and analysis of partial discharge.

[0003] Currently, the analysis of discharge characteristics at different phases mainly relies on phase-resolved partial discharge (PRPD) maps, such as... Figure 1 As shown, traditional PRPD maps are generated by statistically analyzing the distribution of discharge signals in a Cartesian coordinate system (horizontal axis represents voltage phase from 0° to 360°, and vertical axis represents the amplitude of the discharge signal), and can be used to identify the type of insulation defect. However, because traditional PRPD maps use a linearly distributed phase axis, they cannot intuitively reflect the cyclic characteristics of the discharge phenomenon within the AC voltage cycle. This makes it difficult for analysts to effectively observe the onset phase, extinction phase, and periodicity of the discharge, thus weakening the ability to characterize the periodic features of partial discharge.

[0004] Therefore, the visualization effect of the current partial discharge spectrum needs to be further improved. Summary of the Invention

[0005] This application provides a method, apparatus, and storage medium for generating partial discharge patterns, which can intuitively reflect the cyclic characteristics of discharge phenomena within an AC voltage cycle, thereby improving the visualization effect of partial discharge patterns.

[0006] Firstly, a method for generating partial discharge patterns is provided, including: Acquire the partial discharge pulse signal sequence of the power equipment, which includes multiple partial discharge pulse signals; Based on the power frequency period of the partial discharge pulse signal sequence, the arrival time of each partial discharge pulse signal is converted into the corresponding phase angle, and each phase angle is normalized to the angle range of the polar axis in the polar coordinate system. Each phase angle is defined as an angular coordinate in the polar coordinate system, and the amplitude of the corresponding partial discharge pulse signal is defined as a radial coordinate in the polar coordinate system, thus obtaining multiple polar coordinate data points. Partial discharge maps are generated based on multiple polar coordinate data points.

[0007] In one feasible design, a partial discharge spectrum is generated based on multiple polar coordinate data points, including: Density estimation is performed on each polar coordinate data point from multiple polar coordinate data points to obtain the probability density value corresponding to each polar coordinate data point; The corresponding visual attribute is determined based on the probability density value of each polar coordinate data point; A partial discharge map is plotted in the polar coordinate system based on the coordinates of each polar coordinate data point and its corresponding visual attributes.

[0008] In a feasible design, based on the power frequency period of the partial discharge pulse signal sequence, the arrival time of each partial discharge pulse signal is converted into a corresponding phase angle, including: Obtain the power frequency period of the partial discharge pulse signal sequence; The arrival time of each partial discharge pulse signal is converted into the corresponding phase angle using the following formula: ; Where i represents the number of the partial discharge pulse signal, This represents the arrival time of the i-th partial discharge pulse signal, where T represents the power frequency period. This represents the phase angle corresponding to the i-th partial discharge pulse signal.

[0009] In a feasible design, density estimation is performed on each of the multiple polar coordinate data points to obtain the probability density value corresponding to each polar coordinate data point, including: Based on the phase angle and amplitude corresponding to each polar coordinate data point, the Gaussian kernel density estimation method is used to estimate the density of the polar coordinate data points, thereby obtaining the probability density value corresponding to each polar coordinate data point.

[0010] In a feasible design, the Gaussian kernel density estimation method uses the following formula: ; Where N represents the number of polar coordinate data points. This represents the bandwidth parameter corresponding to the phase dimension. This represents the bandwidth parameter corresponding to the amplitude dimension. Represents the kernel density function. This represents the phase angle corresponding to the polar coordinate data points for which Gaussian kernel density estimation is to be performed. This represents the magnitude of the polar coordinate data point for which Gaussian kernel density estimation is to be performed. This represents the phase angle corresponding to the i-th partial discharge pulse signal. This represents the amplitude of the i-th partial discharge pulse signal.

[0011] In one feasible design, a partial discharge spectrum is generated based on multiple polar coordinate data points, including: The amplitude corresponding to each polar coordinate data point is normalized to obtain the normalized amplitude. A partial discharge spectrum is generated based on the coordinates of each polar coordinate data point and its corresponding normalized amplitude.

[0012] In a feasible design, the normalization process uses the following formula: ; Where i represents the number of the partial discharge pulse signal, This represents the initial amplitude of the i-th partial discharge pulse signal. This represents the minimum amplitude in the partial discharge pulse signal sequence. This represents the maximum amplitude in the partial discharge pulse signal sequence. This represents the normalized amplitude of the i-th partial discharge pulse signal.

[0013] Secondly, an apparatus for generating partial discharge patterns is provided, comprising: The data acquisition module is used to acquire the partial discharge pulse signal sequence of the power equipment. The partial discharge pulse signal sequence includes multiple partial discharge pulse signals. The phase angle determination module is used to convert the arrival time of each partial discharge pulse signal into the corresponding phase angle based on the power frequency period of the partial discharge pulse signal sequence, so that each phase angle is normalized to the angle range of the polar axis in the polar coordinate system. The data mapping module is used to determine each phase angle as an angular coordinate in the polar coordinate system and the amplitude of the corresponding partial discharge pulse signal as a radial coordinate in the polar coordinate system, thereby obtaining multiple polar coordinate data points. The map generation module is used to generate partial discharge maps based on multiple polar coordinate data points.

[0014] In one feasible design, a partial discharge spectrum is generated based on multiple polar coordinate data points, including: Density estimation is performed on each polar coordinate data point from multiple polar coordinate data points to obtain the probability density value corresponding to each polar coordinate data point; The corresponding visual attribute is determined based on the probability density value of each polar coordinate data point; A partial discharge map is plotted in the polar coordinate system based on the coordinates of each polar coordinate data point and its corresponding visual attributes.

[0015] Thirdly, a storage medium is provided that stores a computer program, which, when executed by a processor, implements the steps of the aforementioned method embodiment for generating partial discharge patterns.

[0016] This application is based on the core idea that "the discharge characteristics of partial discharge are closely related to the phase of the applied AC voltage," recognizing that the arrival time of the partial discharge pulse signal directly corresponds to the periodic variation of the power frequency voltage. Therefore, this application converts the arrival time of the partial discharge pulse signal into a normalized phase angle (the numerical range of the phase angle is the numerical range of the polar axis of the polar coordinate system), thereby transforming time-series data into phase information that directly reflects the voltage periodicity. Furthermore, utilizing the inherent cyclic characteristics of the polar coordinate system, the phase angle is mapped to angular coordinates, and the amplitude is mapped to radial coordinates, so that each partial discharge pulse signal forms a data point in the polar coordinate plane. This integrates the periodicity and amplitude distribution of the discharge within the framework of a visualized polar coordinate system, effectively overcoming the inherent defects of the traditional linear coordinate system in expressing cyclical patterns, and achieving a more intuitive and accurate characterization of the features of partial discharge. Attached Figure Description

[0017] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is an example of a conventional partial discharge pattern provided in an exemplary embodiment of this application; Figure 2 This is a schematic flowchart illustrating a method for generating a partial discharge pattern according to an exemplary embodiment of this application; Figure 3 This is an example of a conventional PRPD map and PRPS map provided in an exemplary embodiment of this application; Figure 4 This is an example of a conventional PRPD map and a polar coordinate PRPD map provided in an exemplary embodiment of this application; Figure 5 This is a schematic diagram of an exemplary apparatus for generating a partial discharge pattern provided in an exemplary embodiment of this application. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] like Figure 1In traditional partial discharge maps, the linearly distributed phase axes fail to visually represent the cyclic characteristics of the discharge phenomenon within the same voltage cycle, making it difficult to effectively analyze the discharge's initiation phase, extinction phase, and periodicity. To address this problem, such as... Figure 2 As shown, this application provides a method for generating partial discharge patterns, including: S110, acquire the partial discharge pulse signal sequence of the power equipment.

[0021] The partial discharge pulse signal sequence includes multiple partial discharge pulse signals, which can be obtained through a partial discharge detection system.

[0022] Specifically, after the partial discharge detection system performs online or offline detection on the tested power equipment, the partial discharge pulse signal sequence output by the partial discharge detection system is obtained, and the arrival time and amplitude of each partial discharge pulse signal are recorded, as shown in the following formula (1): , formula (1); Where i represents the number of the partial discharge pulse signal, and n represents the number of partial discharge pulse signals contained in the partial discharge pulse signal sequence. This indicates the arrival time of the i-th partial discharge pulse signal. This represents the amplitude of the i-th partial discharge pulse signal. Each This represents a partial discharge pulse signal. This represents the sequence of partial discharge pulse signals.

[0023] S120, based on the power frequency period of the partial discharge pulse signal sequence, converts the arrival time of each partial discharge pulse signal into the corresponding phase angle, and normalizes each phase angle to the angle range of the polar axis in the polar coordinate system.

[0024] In a feasible design, based on the power frequency period of the partial discharge pulse signal sequence, the arrival time of each partial discharge pulse signal is converted into a corresponding phase angle, including: Obtain the power frequency period of the partial discharge pulse signal sequence; The arrival time of each partial discharge pulse signal is converted into the corresponding phase angle using the following formula (2): , formula (2); Where i represents the number of the partial discharge pulse signal, This represents the arrival time of the i-th partial discharge pulse signal, where T represents the power frequency period. This represents the phase angle corresponding to the i-th partial discharge pulse signal.

[0025] The above formula is obtained through " "right Take the modulus to obtain The relative time within the current power frequency cycle (i.e.) Subtract the time of the most recent complete cycle (the result ranges from 0 to T), and then... calculate The proportion of the relative time within the current power frequency cycle to one cycle, and this proportion is compared with... Multiplication achieves the goal of This is converted into the phase angle within the angular interval of the polar axis in the polar coordinate system.

[0026] The above embodiments are based on the principle that "the occurrence of partial discharge is closely related to the phase of the applied AC current". The above formula (2) is used to convert the pulse arrival time into the phase angle within the corresponding AC voltage period (the phase angle range is [0°, 360°)). Essentially, it maps the discrete pulse in the time dimension to the periodic phase dimension to adapt to the closed periodic characteristics of polar coordinates, so as to facilitate the subsequent display of the complete periodic characteristics of the partial discharge spectrum in the polar coordinate system.

[0027] S130, each phase angle is determined as an angular coordinate in the polar coordinate system, and the amplitude of the corresponding partial discharge pulse signal is determined as a radial coordinate in the polar coordinate system, thus obtaining multiple polar coordinate data points.

[0028] Taking the i-th partial discharge pulse signal as an example, its corresponding phase angle Determine the angular coordinates in the polar coordinate system and set their magnitudes. Determined as radial coordinates in polar coordinate system The obtained polar coordinate data points for( , ).

[0029] Because the amplitude dynamic range of partial discharge pulse signals is large, 99% of the pulse amplitudes may be concentrated at a low level, with only 1% of the pulse amplitudes being high. This results in the radial coordinate being dominated by a very small number of high-amplitude pulses in the polar coordinate system, causing most polar coordinate data points in the partial discharge spectrum to be very close to the origin, making them visually indistinguishable. To address this issue, one feasible design method includes: The amplitude corresponding to each polar coordinate data point is normalized to obtain the normalized amplitude. A partial discharge spectrum is generated based on the coordinates of each polar coordinate data point and its corresponding normalized amplitude.

[0030] Specifically, the amplitude corresponding to each polar coordinate data point is normalized to obtain the normalized amplitude. Then, the radial coordinate of the polar coordinate data point is replaced with the normalized amplitude, and a partial discharge spectrum is generated based on each replaced polar coordinate data point.

[0031] For example, the normalization process uses the following formula (3): , formula (3); Where i represents the number of the partial discharge pulse signal, This represents the initial amplitude of the i-th partial discharge pulse signal. This represents the minimum amplitude in the partial discharge pulse signal sequence. This represents the maximum amplitude in the partial discharge pulse signal sequence. This represents the normalized amplitude of the i-th partial discharge pulse signal.

[0032] The above example enhances the display effect of small-amplitude pulses by normalizing the amplitude corresponding to each polar coordinate data point. Furthermore, by using the max-min normalization method of formula (3), the amplitude of all pulses can be accurately and effectively linearly compressed to a uniform scale range (such as [0,1]), making small-amplitude pulses that were originally difficult to distinguish due to large amplitude differences clearly visible on the spectrum, providing more accurate and intuitive data support for subsequent partial discharge mode identification and fault diagnosis.

[0033] S140 generates a partial discharge map based on multiple polar coordinate data points.

[0034] Specifically, based on the coordinate values ​​of each polar coordinate data point determined above, each polar coordinate data point is plotted in the polar coordinate system to generate a partial discharge spectrum.

[0035] Currently, PRPD maps are typically obtained from phase-resolved pulse sequence (PRPS) maps, such as... Figure 3 As shown, the PRPS map displays the phase and amplitude sequence of partial discharge pulses within each power frequency cycle in chronological order, preserving the pulse's time-series information and intuitively reflecting the dynamic changes of discharge pulses across different cycles. In obtaining the PRPD map from the PRPS map, key information of all partial discharge pulses within multiple power frequency cycles is first extracted from the PRPS map, including the phase angle and amplitude of each pulse. Then, the distribution of pulse amplitude at each phase point (e.g., all amplitudes appearing at a certain phase point) is statistically summarized from the extracted multi-power frequency cycle pulse data. Combined with... Figure 3 As can be seen, the PRPS spectrum provides raw data of multi-cycle pulse sequences, while the PRPD spectrum is the result of statistically summarizing these sequence data along the phase dimension. By discarding the time series information of individual cycles, it focuses on the overall phase-amplitude distribution of the discharge. Therefore, the PRPD spectrum also has the following limitations: The PRPD (Pulse-Range Discharge) map superimposes discharge pulses from all power frequency cycles onto a single phase-amplitude plane. When the data volume is massive, a huge number of discrete points will excessively cluster and overlap within specific phase intervals. This simple point superposition method results in an information density in the map far exceeding the resolution limit of human vision. The statistical significance of the PRPD map is not intuitive and cannot directly convey the statistical significance of the pulses in spatial distribution. The human eye cannot easily determine from the density of points which signals are frequently occurring and stable discharges (i.e., primary discharge modes) and which are occasional and secondary discharges (i.e., secondary discharge modes). The conversion efficiency from statistical information to visual information is extremely low.

[0036] Therefore, in order to improve the statistical significance of partial discharge maps, a feasible design can be implemented by generating partial discharge maps based on multiple polar coordinate data points in the following way: Density estimation is performed on each polar coordinate data point from multiple polar coordinate data points to obtain the probability density value corresponding to each polar coordinate data point; The corresponding visual attribute is determined based on the probability density value of each polar coordinate data point; A partial discharge map is plotted in the polar coordinate system based on the coordinates of each polar coordinate data point and its corresponding visual attributes.

[0037] For example, determining the corresponding visual attribute based on the probability density value of each polar coordinate data point includes: Based on the probability density value of each polar coordinate data point, its corresponding color is determined through a preset color mapping table; The color mapping table is configured such that: the lower the probability density value, the lighter the mapped color and the higher the brightness; the higher the probability density value, the darker the mapped color and the lower the brightness.

[0038] Alternatively, based on the probability density value of each polar coordinate data point, its corresponding gray level can be determined through a preset gray-level mapping table; The grayscale mapping table is configured such that the lower the probability density value, the smaller the mapped grayscale value; and the higher the probability density value, the larger the mapped grayscale value.

[0039] Since each polar coordinate data point corresponds to a partial discharge pulse signal, the above example estimates the probability density value of each polar coordinate data point to reflect the frequency of the corresponding partial discharge pulse signal. Furthermore, by mapping the probability density value to visual attributes (such as color depth or brightness), the quantified statistical significance is presented in a direct visual comparison form, such as... Figure 4As shown, high-density areas, endowed with strong visual attributes (such as dark colors), occupy the visual focus of the human eye, corresponding to stable and frequent primary discharge patterns. Low-density areas, with weaker visual attributes (such as light colors), do not occupy the visual focus of the human eye, corresponding to occasional and scattered secondary discharges. Therefore, the above example establishes a three-dimensional encoding scheme of phase (angle), amplitude (radial distance), and statistical significance (color gradient) within a polar coordinate framework. This scheme conveys rich diagnostic information through a single view, greatly improving the information entropy and interpretation efficiency of a unit image. This allows analysts to quickly and accurately identify statistically significant primary and secondary discharge patterns from the spectrum without complex interpretation.

[0040] In a feasible design, density estimation is performed on each of the multiple polar coordinate data points to obtain the probability density value corresponding to each polar coordinate data point. Based on the phase angle and amplitude corresponding to each polar coordinate data point, the Gaussian kernel density estimation method is used to estimate the density of the polar coordinate data points, thereby obtaining the probability density value corresponding to each polar coordinate data point.

[0041] For example, based on the normalized amplitude corresponding to each polar coordinate data point, the Gaussian kernel density is estimated using the following formula (4): , formula (4); Where N represents the number of polar coordinate data points. This represents the bandwidth parameter corresponding to the phase dimension. This represents the bandwidth parameter corresponding to the amplitude dimension. This represents the phase angle corresponding to the polar coordinate data points for which Gaussian kernel density estimation is to be performed. This represents the magnitude of the polar coordinate data point for which Gaussian kernel density estimation is to be performed. This represents the phase angle corresponding to the i-th partial discharge pulse signal. This represents the amplitude corresponding to the i-th partial discharge pulse signal (based on the normalized amplitude for each polar coordinate data point). (This represents the normalized amplitude of the i-th partial discharge pulse signal).

[0042] The kernel density function is represented as shown in the following formula (5): , formula (5); in, The parameter representing the kernel density function corresponds to the parameter in formula (4). ,or .

[0043] The bandwidth parameter in formula (4) and Based on the Silverman rule, which is an empirical rule used to automatically estimate the optimal bandwidth in kernel density estimation, the bandwidth in the j-th dimension is calculated using the following formula (6) for d-dimensional data. : , formula (6); Where j is less than or equal to d, in this application, the value of d is 2 (i.e., the 2D data includes phase dimension data and amplitude dimension data), and the value of j is 1 when corresponding to the phase dimension (i.e. When j is 2, it corresponds to the magnitude dimension (i.e. ). This represents the estimated standard deviation of the data in the j-th dimension.

[0044] The above embodiments process polar coordinate data points using Gaussian kernel density estimation. Leveraging the unique smoothing and decay characteristics of the Gaussian kernel function, discrete phase-amplitude data points are transformed into a continuous probability density distribution field. This processing not only effectively quantifies the pulse aggregation degree in the region where each data point is located, but also suppresses random outlier interference while preserving the main distribution trend through the smoothing effect of the kernel function. It also reveals the inherent aggregation characteristics of the discharge data, including deep features such as multi-peak structures and distribution boundaries. Visual attributes determined based on probability density values ​​allow statistically significant high-density regions to naturally stand out in the generated density field, making the main discharge mode and occasional secondary discharges significantly distinguishable in terms of probability density values.

[0045] This application is based on the core idea that "the discharge characteristics of partial discharge are closely related to the phase of the applied AC voltage," recognizing that the arrival time of the partial discharge pulse signal directly corresponds to the periodic variation of the power frequency voltage. Therefore, this application converts the arrival time of the partial discharge pulse signal into a normalized phase angle (the numerical range of the phase angle is the numerical range of the polar axis of the polar coordinate system), thereby transforming time-series data into phase information that directly reflects the voltage periodicity. Furthermore, utilizing the inherent cyclic characteristics of the polar coordinate system, the phase angle is mapped to angular coordinates, and the amplitude is mapped to radial coordinates, so that each partial discharge pulse signal forms a data point in the polar coordinate plane. This integrates the periodicity and amplitude distribution of the discharge within the framework of a visualized polar coordinate system, effectively overcoming the inherent defects of the traditional linear coordinate system in expressing cyclical patterns, and achieving a more intuitive and accurate characterization of the features of partial discharge.

[0046] The following is based on Figure 4 For example, this application demonstrates the effect of projecting a partial discharge pulse signal sequence onto a polar coordinate system. Figure 4The traditional PRPD map and the polar coordinate PRPD map in the image correspond to the same partial discharge pulse signal sequence: like Figure 4 In the traditional PRPD graph shown, the sine curve represents the power frequency voltage signal, the data points represent the partial discharge pulse signal, the horizontal axis represents the phase (0° to 360°), and the vertical axis represents the amplitude. From Figure 4 It can be observed that the partial discharge pulse signals are mainly concentrated in the following three phase intervals: (0°~80°), (155°~265°), and (335°~355°). However, when performing phase difference analysis on these three clusters of signals, it was found that the phase difference between the clusters of signals does not satisfy the 180° symmetry. Therefore, according to traditional analysis methods, the insulation defect type reflected by this partial discharge pulse signal sequence is easily misjudged as corona discharge or noise interference.

[0047] In the polar coordinate PRPD map generated according to the scheme of this application, the partial discharge pulse signal sequence is mapped to polar coordinate space, which allows for a more intuitive observation of its distribution characteristics. After polar coordinate representation, it can be clearly found that the first cluster signal (0°~80°) and the third cluster signal (335°~355°) actually belong to the partial discharge activity of the same cluster signal (the first cluster signal and the third cluster signal constitute the fourth cluster signal). The fourth cluster signal and the second cluster signal (155°~265°) are located at the diagonal position of the polar coordinate map, with a phase difference of approximately 180°, which is consistent with typical floating discharge characteristics.

[0048] Furthermore, in the partial discharge spectrum of this application, the polar coordinate angle axis is a closed loop from 0° to 360°, and the direction of angle increase is consistent with the time flow direction of the power frequency voltage phase (i.e., rotating counterclockwise from 0° to 360°, corresponding to the continuous advancement of the voltage cycle). Determining which of the two boundary phases of a signal cluster is the starting phase and which is the extinguishing phase requires consideration of the time sequence. If the boundary phases of a cluster of signals are respectively and ( < (and did not cross 0°): This is the initial phase (the discharge occurs first in time). This is the extinction phase (discharge ends later in time). For example, the boundary phases of the second cluster signal are respectively... (155°) (265°), with an initial phase of 155° and an extinction phase of 265°.

[0049] If a cluster of signals crosses the 0° closure point, the boundary phases are respectively and ( < And crossing 0°), then As the initial phase, The extinction phase. For example, the boundary phases of the fourth cluster signal are respectively... (80°) (335°), due to the cyclic angle axis, the time progression from 335° to 360° (0°) to 80° is continuous, so its starting phase is 335° and its extinguishing phase is 80°.

[0050] However Figure 4 The traditional PRPD diagram shown uses a Cartesian coordinate system, with its phase axis distributed linearly from 0° to 360°. This cannot intuitively reflect the cyclic characteristics of the discharge phenomenon within the same voltage cycle, making it difficult to effectively analyze the start-up phase and extinction phase of the discharge.

[0051] Furthermore, traditional PRPD maps do not provide a clear visual indication of whether each signal appears once or multiple times within multiple power frequency cycles. However, the partial discharge map of this application can further express the density of partial discharge pulses through the color intensity of the dots, i.e., the periodic patterns such as the phase and amplitude of frequently occurring signals mainly concentrated in which range, providing richer and more intuitive information.

[0052] In practical applications, the partial discharge pattern of this application also has the following effects: (1) Improve feature separability and optimize model input quality

[0053] Traditional partial discharge pulse (PRPD) images, displayed in Cartesian coordinates, often exhibit overlapping of different PRPD pulse signals and blurred cluster boundaries, resulting in low feature discriminativeness and posing challenges for feature extraction in machine learning models. This invention, through polar coordinate mapping, transforms the phase distribution of PRPD pulse signals into a circular periodic layout. This enables PRPD signals from different sources to form clearly separated and structurally distinct clusters in polar coordinate space, significantly improving the separability and interpretability of image features and providing high-quality, structured input data for machine learning models.

[0054] (2) Enhance the visibility and completeness of key identification information

[0055] This application's polar coordinate-based partial discharge map not only clearly displays the phase distribution of partial discharges, but also enhances the expression of key features such as discharge amplitude and density through radial dimension and color coding, making discrimination information such as phase difference and aggregation regions more intuitive. This improvement effectively assists the model in capturing detailed features that are easily overlooked in traditional maps, thereby enhancing the sufficiency of classification criteria.

[0056] (3) Improve model training efficiency and classification accuracy

[0057] Because the partial discharge spectrum features provided in this application are more significant and distinct, machine learning models (such as convolutional neural networks) can learn effective feature representations faster and more accurately, significantly accelerating model convergence and reducing training difficulty. In practical applications, this directly translates to higher automatic identification accuracy, stronger generalization ability, and more stable system performance, providing a reliable technical foundation for intelligent monitoring and diagnosis of the insulation status of power equipment.

[0058] like Figure 5 As shown, this application also provides an apparatus for generating partial discharge patterns, comprising: The data acquisition module is used to acquire the partial discharge pulse signal sequence of the power equipment. The partial discharge pulse signal sequence includes multiple partial discharge pulse signals. The phase angle determination module is used to convert the arrival time of each partial discharge pulse signal into the corresponding phase angle based on the power frequency period of the partial discharge pulse signal sequence, so that each phase angle is normalized to the angle range of the polar axis in the polar coordinate system. The data mapping module is used to determine each phase angle as an angular coordinate in the polar coordinate system and the amplitude of the corresponding partial discharge pulse signal as a radial coordinate in the polar coordinate system, thereby obtaining multiple polar coordinate data points. The map generation module is used to generate partial discharge maps based on multiple polar coordinate data points.

[0059] In a feasible design, the spectrum generation module generates a partial discharge spectrum based on multiple polar coordinate data points in the following way: Density estimation is performed on each polar coordinate data point from multiple polar coordinate data points to obtain the probability density value corresponding to each polar coordinate data point; The corresponding visual attribute is determined based on the probability density value of each polar coordinate data point; A partial discharge map is plotted in the polar coordinate system based on the coordinates of each polar coordinate data point and its corresponding visual attributes.

[0060] In a feasible design, the phase angle determination module realizes the power frequency cycle based on the partial discharge pulse signal sequence by converting the arrival time of each partial discharge pulse signal into the corresponding phase angle in the following way: Obtain the power frequency period of the partial discharge pulse signal sequence; The arrival time of each partial discharge pulse signal is converted into the corresponding phase angle using the following formula: ; Where i represents the number of the partial discharge pulse signal, This represents the arrival time of the i-th partial discharge pulse signal, where T represents the power frequency period. This represents the phase angle corresponding to the i-th partial discharge pulse signal.

[0061] In a feasible design, the map generation module performs density estimation for each of multiple polar coordinate data points in the following manner to obtain the probability density value corresponding to each polar coordinate data point: Based on the phase angle and amplitude corresponding to each polar coordinate data point, the Gaussian kernel density estimation method is used to estimate the density of the polar coordinate data points, thereby obtaining the probability density value corresponding to each polar coordinate data point.

[0062] In a feasible design, the Gaussian kernel density estimation method uses the following formula: ; Where N represents the number of polar coordinate data points. This represents the bandwidth parameter corresponding to the phase dimension. This represents the bandwidth parameter corresponding to the amplitude dimension. Represents the kernel density function. This represents the phase angle corresponding to the polar coordinate data points for which Gaussian kernel density estimation is to be performed. This represents the magnitude of the polar coordinate data point for which Gaussian kernel density estimation is to be performed. This represents the phase angle corresponding to the i-th partial discharge pulse signal. This represents the amplitude of the i-th partial discharge pulse signal.

[0063] In one feasible design, a partial discharge spectrum is generated based on multiple polar coordinate data points, including: The amplitude corresponding to each polar coordinate data point is normalized to obtain the normalized amplitude. A partial discharge spectrum is generated based on the coordinates of each polar coordinate data point and its corresponding normalized amplitude.

[0064] In a feasible design, the normalization process uses the following formula: ; Where i represents the number of the partial discharge pulse signal, This represents the initial amplitude of the i-th partial discharge pulse signal. This represents the minimum amplitude in the partial discharge pulse signal sequence. This represents the maximum amplitude in the partial discharge pulse signal sequence. This represents the normalized amplitude of the i-th partial discharge pulse signal.

[0065] Other implementation methods and effects of the above-described device can be found in the description of the partial discharge pattern generation method embodiment, and will not be repeated here.

[0066] This application also provides a storage medium storing a computer program that, when executed by a processor, implements the steps of the aforementioned method embodiment for generating partial discharge patterns.

[0067] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.

[0068] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0069] The block diagrams of devices, apparatuses, devices, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0070] It should also be noted that in the apparatus, equipment, and methods of this application, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of this application.

[0071] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this application. Therefore, this application is not intended to be limited to the aspects shown herein, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0072] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this application to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A method of partial discharge pattern generation, characterized by, The method comprises the following steps: obtaining a partial discharge pulse signal sequence of a power equipment, the partial discharge pulse signal sequence comprising a plurality of partial discharge pulse signals; transforming an arrival time of each partial discharge pulse signal into a corresponding phase angle based on a power frequency cycle of the partial discharge pulse signal sequence, and normalizing each phase angle into an angle interval of a polar axis in a polar coordinate system; determining each phase angle as an angle coordinate in the polar coordinate system, and determining an amplitude of a corresponding partial discharge pulse signal as a radial coordinate in the polar coordinate system, to obtain a plurality of polar coordinate data points; generating a partial discharge map based on the plurality of polar coordinate data points.

2. The method of claim 1, wherein, The generating of the partial discharge map based on the plurality of polar coordinate data points comprises: performing density estimation on each polar coordinate data point of the plurality of polar coordinate data points to obtain a probability density value corresponding to each polar coordinate data point; determining a corresponding visual attribute based on the probability density value of each polar coordinate data point; drawing the partial discharge map in the polar coordinate system according to the coordinates of each polar coordinate data point and the corresponding visual attribute.

3. The method according to claim 1 or 2, characterized in that, The transforming of the arrival time of each partial discharge pulse signal into the corresponding phase angle based on the power frequency cycle of the partial discharge pulse signal sequence comprises: obtaining the power frequency cycle of the partial discharge pulse signal sequence; transforming the arrival time of each partial discharge pulse signal into the corresponding phase angle by using the following formula: ； wherein i represents the number of partial discharge pulse signals, denotes the arrival time of the i-th partial discharge pulse signal, T denotes the power frequency period, denotes the phase angle corresponding to the i-th partial discharge pulse signal.

4. The method of claim 2, wherein, The performing of the density estimation on each polar coordinate data point of the plurality of polar coordinate data points to obtain the probability density value corresponding to each polar coordinate data point comprises: performing density estimation on the polar coordinate data point by using a Gaussian kernel density estimation method according to the phase angle and the amplitude corresponding to each polar coordinate data point, to obtain the probability density value corresponding to each polar coordinate data point.

5. The method of claim 4, wherein, The Gaussian kernel density estimation method uses the following formula: ； wherein N represents a number of the plurality of polar coordinate data points, represents a bandwidth parameter corresponding to the phase dimension, represents a bandwidth parameter corresponding to the amplitude dimension, represents a kernel density function, represents a phase angle corresponding to a polar coordinate data point to be subjected to the Gaussian kernel density estimation, represents an amplitude corresponding to a polar coordinate data point to be subjected to the Gaussian kernel density estimation, represents a phase angle corresponding to the i-th partial discharge pulse signal, represents an amplitude corresponding to the i-th partial discharge pulse signal.

6. The method of claim 1, wherein, The generating of the partial discharge map based on the plurality of polar coordinate data points comprises: performing normalization processing on the amplitude corresponding to each polar coordinate data point to obtain a normalized amplitude; generating the partial discharge map according to the coordinates of each polar coordinate data point and the corresponding normalized amplitude.

7. The method of claim 6, wherein, The formula used in the normalization processing is: ； wherein i denotes the number of the partial discharge pulse signal, denotes the initial amplitude of the i-th partial discharge pulse signal, denotes the minimum amplitude of the sequence of partial discharge pulse signals, denotes the maximum amplitude of the sequence of partial discharge pulse signals, denotes the normalized amplitude of the i-th partial discharge pulse signal.

8. An apparatus for generating a partial discharge pattern, characterized by The method comprises the following steps: a data acquisition module is configured to obtain a partial discharge pulse signal sequence of a power equipment, the partial discharge pulse signal sequence comprising a plurality of partial discharge pulse signals; a phase angle determination module is configured to transform an arrival time of each partial discharge pulse signal into a corresponding phase angle based on a power frequency cycle of the partial discharge pulse signal sequence, and normalize each phase angle into an angle interval of a polar axis in a polar coordinate system; a data mapping module is configured to determine each phase angle as an angle coordinate in the polar coordinate system, and determine an amplitude of a corresponding partial discharge pulse signal as a radial coordinate in the polar coordinate system, to obtain a plurality of polar coordinate data points; a map generation module is configured to generate a partial discharge map based on the plurality of polar coordinate data points.

9. The apparatus of claim 8, wherein, The generating of the partial discharge map based on the plurality of polar coordinate data points comprises: performing density estimation on each polar coordinate data point of the plurality of polar coordinate data points to obtain a probability density value corresponding to each polar coordinate data point; determining a corresponding visual attribute based on a probability density value of each polar coordinate data point; drawing a partial discharge map in the polar coordinate system according to the coordinates of each polar coordinate data point and the corresponding visual attribute.

10. A storage medium storing a computer program, characterized by The computer program, when executed by a processor, implements the steps of the method for generating a partial discharge map according to any one of claims 1 to 7. The computer program, when executed by a processor, implements the steps of the method for generating a partial discharge map according to any one of claims 1 to 7.

Images