Direct-current power equipment partial discharge map drawing method and related device

Abstract

The invention belongs to the technical field of high-voltage direct-current power transmission, and relates to a direct-current power equipment partial discharge map drawing method and a related device. Acquiring a section of continuous partial discharge pulse signals to obtain a pulse time sequence queue; determining a spectrum amplitude threshold value, and dividing the pulse time sequence queue into a main sequence formed by high pulses and a plurality of subsequences formed by continuous low pulses; taking the X axis as a time sequence axis, constructing an XOY rectangular coordinate system, dividing a pulse region and a background region, and delimiting a threshold value band; the pulse time sequence queue is divided into a short-time queue and a long-time queue; when the pulse time sequence queue is a short-time queue, the discharge map comprises a short-time direct-current partial discharge map in a three-dimensional form and a short-time direct-current partial discharge map in a two-dimensional form; and when the pulse time sequence queue is a long-time queue, the discharge map is a two-dimensional long-time DC partial discharge map. The problem that an ultrahigh frequency method combined with a pulse sequence phase distribution map cannot be suitable for direct-current power equipment is solved.

Description

Technical Field

[0001] This invention belongs to the field of high voltage direct current transmission technology, specifically relating to a method and related apparatus for drawing partial discharge patterns of DC power equipment. Background Technology

[0002] Various defects generated in high-voltage power equipment throughout its entire lifecycle—manufacturing, transportation, installation, and operation—can trigger partial discharge signals of varying intensities and forms. These partial discharges further accelerate the degradation of insulation performance. A scientifically sound and clear spectral characterization method facilitates the reliable implementation of feature extraction and pattern recognition functions, thereby enabling accurate diagnosis, timely early warning, and effective prevention of sudden power system failures.

[0003] Currently, the ultra-high frequency (UHF) method combined with pulse sequence phase distribution (PRPS) mapping utilizes the fact that the voltage of alternating current (AC) varies periodically with the phase (e.g., a sine wave, cycling from 0° to 360°). The probability of partial discharge and its signal strength are strongly correlated with the phase of the AC current (e.g., discharges from certain defects are more likely to occur near the voltage peak). PRPS mapping leverages this characteristic: it arranges the detected discharge pulses along two dimensions—time (period axis) and AC phase (phase axis)—and uses pulse amplitude to represent signal strength. This two-dimensional mapping clearly shows the discharge pattern. Therefore, the UHF method combined with PRPS mapping, with its high sensitivity, real-time performance, and data identifiability, has become the dominant technology for partial discharge detection in AC power equipment.

[0004] High-voltage direct current (HVDC) transmission has advantages such as large capacity, long distance, and flexible dispatch and control, and has developed rapidly in power systems in recent years. However, partial discharge under DC voltage is independent of the power frequency phase, and time and amplitude are only two basic parameters that can be obtained. The richness of its features is far less than that of AC, which undoubtedly places more stringent requirements on waveform description and spectrum construction.

[0005] Unfolding one-dimensional data into two dimensions can increase information density and adapt to numerous image recognition-based machine learning algorithms. However, its arrangement, especially the timing of starting a new line, needs to be supported by physical meaning, thus rendering the periodic and phase axes of the PRPS graph no longer applicable.

[0006] Spectrum analysis requires high-specification instruments that can meet the requirements of high-bandwidth raw sampling, high-frequency computing and processing, and large-capacity communication storage. These instruments are expensive and require highly skilled operators. Existing AC hardware is not suitable for migration or reuse, and equipping a complete station with such equipment is not feasible.

[0007] The lack of high-quality spectra also hinders the statistical analysis and pattern summarization of typical defect discharge cases. Currently, the identification and diagnosis technology for DC partial discharge, whether in theoretical or experimental research, lags significantly behind AC partial discharge, falling far short of practical needs and has not yet achieved formal engineering application. Summary of the Invention

[0008] The purpose of this invention is to provide a method and related apparatus for drawing partial discharge patterns of DC power equipment, which solves the problem that the existing ultra-high frequency method combined with pulse sequence phase distribution patterns cannot be applied to DC power equipment.

[0009] This invention is achieved through the following technical solution: This invention discloses a method for drawing partial discharge patterns of DC power equipment, comprising the following steps: A continuous partial discharge pulse signal is acquired to obtain the pulse timing queue; The amplitude threshold of the spectrum is determined based on the pulse time sequence queue, and the pulse time sequence queue is divided into a main sequence consisting of high pulses and several subsequences consisting of continuous low pulses. Using the X-axis as the time axis, an XOY rectangular coordinate system is constructed to divide the pulse region and background region, and a threshold band is defined based on the spectrum amplitude threshold. Pulse timing queues are divided into short-time queues and long-time queues; When the pulse timing queue is a short-time queue, the discharge spectrum includes a three-dimensional short-time DC partial discharge spectrum and a two-dimensional short-time DC partial discharge spectrum; the specific drawing process is as follows: Each high pulse in the main sequence has a line segment length determined by its amplitude and is arranged sequentially on one side of the pulse region. A Z-axis perpendicular to the XOY plane is introduced, and each subsequence contains several columns representing low pulses parallel to the Z-axis direction. The column height is determined by its amplitude and they are arranged sequentially in a columnar queue along the Y-axis on one side of the background region. The elements on both sides are interspersed and staggered to obtain a three-dimensional short-time DC partial discharge spectrum. A top view of a three-dimensional short-time DC partial discharge pattern is obtained by vertically flipping it along the X-axis, reducing each low-pulse column to pixels, thus obtaining a two-dimensional short-time DC partial discharge pattern. When the pulse timing queue is a long-time queue, the discharge pattern is a two-dimensional long-time DC partial discharge pattern, and the specific drawing process is as follows: Each high pulse in the main sequence has a segment length determined by its amplitude and is arranged sequentially on one side of the pulse region; each subsequence is determined by the Log number of low pulses it contains. 10 The length is determined by logarithm, and the data are arranged sequentially on one side of the background region to obtain a two-dimensional long-term DC partial discharge pattern.

[0010] Furthermore, the step of determining the spectrum amplitude threshold based on the pulse time sequence queue involves dividing the pulse time sequence queue into a main sequence consisting of high pulses and several subsequences consisting of consecutive low pulses; specifically: Determine the upper limit of the pulse capacity that can be displayed based on the pixel width of the drawing interface; the upper limit of the pulse capacity should not exceed half of the horizontal pixel count of the screen. Using the maximum pulse amplitude of the pulse timing queue as a temporary threshold, start traversing and count the total number of high pulses. If it is less than the upper limit of pulse capacity, gradually reduce the temporary threshold until the total number of high pulses is about to exceed the upper limit of pulse capacity. Set the temporary threshold at this time as the spectrum amplitude threshold. Calculate the difference X between the upper limit of pulse capacity and the total number of high pulses corresponding to the spectrum amplitude threshold; Pulses with amplitude values ​​above the spectrum threshold are all counted as high pulses, and those with amplitude values ​​below the spectrum threshold are counted as low pulses. For the remaining pulses whose amplitude equals the spectral amplitude threshold, calculate their respective neighborhood relative heights: Take the X neighborhoods with the largest relative heights and include them in the high pulses, which will fill the pulse capacity limit. The remaining neighborhoods with smaller relative heights are all low pulses. High pulses constitute the main sequence, and each consecutive low pulse constitutes a subsequence.

[0011] Furthermore, the expression for calculating the relative height of the neighborhood is: ΔV N = 3V N - 0.5V N-2 - V N-1 - V N+1 - 0.5V N+2 ; Where, ΔV N V represents the relative height of the neighborhood of the Nth pulse; N V represents the amplitude of the Nth pulse; N-2 V represents the amplitude of the (N-2)th pulse; N-1 V represents the amplitude of the (N-1)th pulse; N+1 V represents the amplitude of the (N+1)th pulse; N+2 This represents the amplitude of the (N+2)th pulse.

[0012] Furthermore, the pulse region is divided as follows: the planar region formed by the positive half-axis of the Y-axis and the X-axis is used to draw a high-pulse histogram, and each line segment represents an independent pulse.

[0013] Furthermore, the background area is divided as follows: the plane area formed by the negative half-axis of the Y-axis and the X-axis is used as the background area to draw the subsequence histogram, and each queue or line segment represents a continuous low pulse.

[0014] Furthermore, the threshold band is determined as follows: within the pulse region, a ray is emitted vertically from the scale of the spectrum amplitude threshold located on the Y-axis, and the region between this ray and the X-axis is taken as the threshold band.

[0015] This invention also discloses a system for plotting partial discharge patterns of DC power equipment, comprising: The data acquisition module is used to acquire a continuous partial discharge pulse signal; The sequence segmentation module is used to determine the spectrum amplitude threshold based on the pulse time sequence queue, and to segment the pulse time sequence queue into a main sequence consisting of high pulses and several subsequences consisting of consecutive low pulses. The coordinate system construction module is used to construct an XOY rectangular coordinate system with the X-axis as the time axis, divide the pulse region and background region, and define the threshold band according to the spectrum amplitude threshold. The pulse timing queue is divided into a short-time queue and a long-time queue. When the pulse timing queue is a short-time queue, the discharge spectrum includes a three-dimensional short-time DC partial discharge spectrum and a two-dimensional short-time DC partial discharge spectrum. When the pulse timing queue is a long-time queue, the discharge spectrum is a two-dimensional long-time DC partial discharge spectrum. The 3D short-time partial discharge map construction module is used to draw short-time DC partial discharge maps in three-dimensional form. The specific drawing process is as follows: Each high pulse in the main sequence has a line segment length determined by its amplitude and is arranged sequentially on one side of the pulse region. A Z-axis perpendicular to the XOY plane is introduced, and each subsequence contains several columns representing low pulses parallel to the Z-axis direction. The column height is determined by its amplitude and they are arranged sequentially in a columnar queue along the Y-axis on one side of the background region. The elements on both sides are interspersed and staggered to obtain a three-dimensional short-time DC partial discharge spectrum. The two-dimensional short-time map construction module is used to draw a two-dimensional short-time DC partial discharge map. The specific process is as follows: take the top view of the two-dimensional short-time DC partial discharge map, flip it vertically based on the X-axis, and reduce each low pulse column to pixels to obtain the two-dimensional short-time DC partial discharge map. The two-dimensional long-time partial discharge pattern construction module is used to draw two-dimensional long-time DC partial discharge patterns; the specific drawing process is as follows: Each high pulse in the main sequence has a segment length determined by its amplitude and is arranged sequentially on one side of the pulse region; each subsequence is determined by the Log number of low pulses it contains. 10 The length is determined by logarithm, and the data are arranged sequentially on one side of the background region to obtain a two-dimensional long-term DC partial discharge pattern.

[0016] The present invention also discloses a computer device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the method for drawing partial discharge patterns of DC power equipment.

[0017] The present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method for drawing partial discharge patterns of DC power equipment.

[0018] Compared with the prior art, the present invention has the following beneficial technical effects: This invention discloses a method for drawing partial discharge maps of DC power equipment. By adopting a map structure that replaces the traditional phase axis with a time axis (X-axis) and innovatively defining the main sequence and sub-sequence, this invention fundamentally solves the core problem that existing PRPS-type maps cannot be applied to the analysis of partial discharge of DC power equipment without a fixed period due to their reliance on periodic phase.

[0019] By dynamically determining the spectrum amplitude threshold based on the upper limit of pulse capacity and using "neighborhood relative height" for precise pulse screening, this scheme can adaptively optimize and segment the original pulse sequence, ensuring that, under limited display resources, significant discharge pulses (main sequence) are highlighted while effectively preserving continuous low pulse (subsequence) information in the background noise.

[0020] By designing different two-dimensional and three-dimensional plotting rules for short-term and long-term queues, this invention can provide both high temporal resolution (approximately 0.2 milliseconds) for detailed observation of discharge details (short-term plot) and effective display of long-term trends through logarithmic compression (long-term plot), thus meeting the needs of different analysis scenarios.

[0021] Compared to AC partial discharge PRPS spectra, this invention alters the arrangement rules and descriptive weights, replacing the physically meaningless periodic and phase axes with main and sub-sequences. The hardware can still directly utilize the AC version of the UHF monitoring device, requiring no changes to the communication protocol and data format, and no additional manpower or instrumentation investment, offering a cost advantage and meeting the needs of whole-site deployment, unattended operation, and long-term data retention.

[0022] This invention generates a spectrum that highlights impulses and segments background noise while preserving temporal information. It is a reversible, lossless transformation that can restore the original waveform of the acquired signal, facilitating the removal of noise interference. The spectrum can be loaded in real-time, continuously, and in a rolling manner on the software side, similar to PRPS, making it more user-friendly for manual observation and interpretation. The spectrum has high dimensionality and density, providing a rich space for feature mining, and inherently possesses the conditions to achieve ideal recognition accuracy regardless of whether traditional neural networks or deep learning algorithms are used. Attached Figure Description

[0023] Figure 1 This is a flowchart of a method for drawing partial discharge patterns of DC power equipment according to the present invention; Figure 2 This is a two-dimensional schematic diagram of a short-term partial discharge pattern; Figure 3 This is a case study of a three-dimensional application of short-time real-time signal acquisition. Figure 4 This is a two-dimensional application case of short-time real-time signal acquisition; Figure 5 This is a two-dimensional schematic diagram of a long-term partial discharge pattern; Figure 6 This is a case study of a two-dimensional application of long-term real-time signal acquisition. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the present invention clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention; that is, the described embodiments are only a part of the embodiments of the present invention, and not all of them.

[0025] The components described and illustrated in the accompanying drawings and embodiments of this invention can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the invention provided in the following drawings is not intended to limit the scope of the claimed invention, but merely to illustrate one selected embodiment of the invention. All other embodiments obtained by those skilled in the art based on the accompanying drawings and embodiments of this invention without inventive effort are within the scope of protection of this invention.

[0026] The following is an explanation of the relevant terms: Partial discharge: An electrical discharge in which the insulation between conductors is only partially bridged. This discharge may or may not occur near the conductor. It can occur inside both AC and DC power equipment.

[0027] Ultra-high frequency (UHF): Radio waves with wavelengths ranging from 1 m to 1 dm and frequencies from 300 to 3000 MHz, commonly used in mobile communications and broadcasting. Partial discharge phenomena can also generate UHF signals, typically located in the 300–1500 MHz band.

[0028] DC power systems are power networks that utilize direct current (DC) for power transmission, distribution, and use. They complement traditional alternating current (AC) power systems and offer unique advantages in specific application scenarios. Key features include: current characteristics (DC current has a constant direction and no periodic changes, avoiding the frequency variation problems of AC current); high-efficiency transmission (no inductive or capacitive reactance losses, suitable for long-distance or high-loss environments); and compatibility (directly matching with DC power sources such as solar cells and batteries, reducing AC-DC conversion steps and improving efficiency).

[0029] Power frequency phase: In an AC power system, the angular position (e.g., 0°~360°) of the voltage waveform at a given moment, with a sine wave as the reference. Many defects or fault indicators within AC power equipment are related to voltage amplitude or rate of change, thus exhibiting periodic variation characteristics. DC power systems do not have this concept.

[0030] PRPS: Phase Resolved Pulse Sequence. The x-axis represents the phase, with values ​​depending on the fraction of a power frequency cycle (0.02 seconds); the y-axis represents 50 cycles; and the z-axis represents the discharge amplitude. The PRPS graph is considered to contain all the information of a 1-second partial discharge signal (it can be presented as a 3D bar chart or a 2D top view (the z-axis is represented by color or grayscale instead of bar height)).

[0031] Sampling rate: The number of samples extracted from a continuous signal and used to assemble a discrete signal per unit time. For AC systems, if only the peak value, pulse density, phase distribution, and other characteristics of the signal are being analyzed, a 100MHz-level sampling module is sufficient. For DC systems, if frequency domain analysis is required, the sampling rate should be at least twice that of the original signal, which will drastically increase the cost of equipment and supporting hardware.

[0032] This invention discloses a method for drawing partial discharge patterns of DC power equipment, comprising the following steps: A continuous partial discharge pulse signal is acquired to obtain the pulse timing queue; The amplitude threshold of the spectrum is determined based on the pulse time sequence queue, and the pulse time sequence queue is divided into a main sequence consisting of high pulses and several subsequences consisting of continuous low pulses. Using the X-axis as the time axis, an XOY rectangular coordinate system is constructed to divide the pulse region and background region, and a threshold band is defined based on the spectrum amplitude threshold. Pulse timing queues are divided into short-time queues and long-time queues; When the pulse timing queue is a short-time queue, the discharge spectrum includes a three-dimensional short-time DC partial discharge spectrum and a two-dimensional short-time DC partial discharge spectrum; the specific drawing process is as follows: Each high pulse in the main sequence has a line segment length determined by its amplitude and is arranged sequentially on one side of the pulse region. A Z-axis perpendicular to the XOY plane is introduced, and each subsequence contains several columns representing low pulses parallel to the Z-axis direction. The column height is determined by its amplitude and they are arranged sequentially in a columnar queue along the Y-axis on one side of the background region. The elements on both sides are interspersed and staggered to obtain a three-dimensional short-time DC partial discharge spectrum. A top view of a three-dimensional short-time DC partial discharge pattern is obtained by vertically flipping it along the X-axis, reducing each low-pulse column to pixels, thus obtaining a two-dimensional short-time DC partial discharge pattern. When the pulse timing queue is a long-time queue, the discharge pattern is a two-dimensional long-time DC partial discharge pattern, and the specific drawing process is as follows: Each high pulse in the main sequence has a segment length determined by its amplitude and is arranged sequentially on one side of the pulse region; each subsequence is determined by the Log number of low pulses it contains. 10 The length is determined by logarithm, and the data are arranged sequentially on one side of the background region to obtain a two-dimensional long-term DC partial discharge pattern.

[0033] To overcome the difficulties in describing and identifying DC partial discharge waveforms caused by limited sampling rate and lack of phase information, this invention proposes a method for constructing main and sub-sequences and presenting spectra with different dimensions, based on threshold segmentation and time sequence preservation. Specific features include: High-pulse data are plotted as lines, and low-pulse data as bars. The main sequence represents the overall trend, while the subsequence describes the discharge interval and noise floor fluctuations. The two sets of data are displayed on either side of the time-series axis. The grayscale and width of the threshold band intuitively reflect the overall level and fluctuation of the signal. Short-time and long-time spectra are used to cover different time scales, respectively replacing the PRPS and PRPD spectra commonly used in AC partial discharge analysis.

[0034] The features and performance of the present invention will be further described in detail below with reference to embodiments.

[0035] Example 1 like Figure 1 As shown, this invention discloses a method for drawing partial discharge patterns of DC power equipment, and the specific implementation steps are as follows: 1. A short-time queue of 5000 sampling points was obtained by continuously collecting partial discharge pulse signals for 1 second and taking the peak value every 0.2 milliseconds.

[0036] 2. Determine the upper limit of the pulse capacity that can be displayed based on the pixel width of the drawing interface; Using the maximum pulse amplitude of the short-time queue as a temporary threshold, we begin to traverse downwards. When the total number of high pulses that meet the conditions just exceeds the pulse capacity limit, we determine the temporary threshold at this point as the spectrum amplitude threshold. Pulses exceeding the spectrum amplitude threshold and a subset of pulses equal to the spectrum amplitude threshold are called high pulses, and the rest are low pulses. Using each high pulse as a dividing point, the short-time queue is divided into a main sequence consisting of high pulses and several subsequences consisting of consecutive low pulses.

[0037] In step 2, the method for determining the spectrum amplitude threshold, high pulse, and low pulse is as follows: 2.1 To ensure visual quality, when determining the maximum pulse capacity that can be displayed, the maximum pulse capacity should not exceed half the number of horizontal pixels on the screen. 2.2. Using the maximum pulse amplitude of the short-time queue as a temporary threshold, start traversing and count the total number of high pulses. If it is less than the pulse capacity limit, gradually reduce the temporary threshold until the total number of high pulses is about to exceed the pulse capacity limit. Set the temporary threshold at this time as the spectrum amplitude threshold. Pulses with amplitude values ​​above the spectrum threshold are all counted as high pulses, and pulses with amplitude values ​​below the spectrum threshold are counted as low pulses. Calculate the difference X between the pulse capacity upper limit and the current total number of high pulses. For the remaining pulses whose amplitude is exactly equal to the spectrum amplitude threshold, calculate their respective neighborhood relative heights: ΔV N = 3V N - 0.5V N-2 - V N-1 - V N+1 - 0.5V N+2 Where, ΔV N V represents the relative height of the neighborhood of the Nth pulse; N V represents the amplitude of the Nth pulse; N-2 V represents the amplitude of the (N-2)th pulse; N-1 V represents the amplitude of the (N-1)th pulse; N+1 V represents the amplitude of the (N+1)th pulse; N+2 This represents the amplitude of the (N+2)th pulse; The X neighborhoods with the largest relative heights are selected and included in the high pulses, thus filling the pulse capacity limit. The remaining neighborhoods with smaller relative heights are all low pulses.

[0038] like Figure 2 As shown, high pulses constitute the main sequence, and each consecutive low pulse constitutes a subsequence.

[0039] For example, if the pulse capacity limit is 200 and the temporary threshold is 2000mV, the total number of high pulses exceeding 2000mV at the start time is 0. If the temporary threshold is lowered step by step, the total number of high pulses is 70 when it is reduced to 1600mV, 120 when it is reduced to 1000mV, 190 when it is reduced to 900mV, and 230 when it is reduced to 850mV. However, the minimum resolution scale of the instrument is 50mV.

[0040] Therefore, all 190 pulses with amplitudes greater than 900mV were counted as high pulses. There were still 10 empty slots before filling the maximum pulse capacity, but 40 pulses had an amplitude equal to 900mV. The relative height ΔV of each pulse's neighborhood was then calculated. N Only the 10 results with the largest values ​​are included in the high pulse.

[0041] 2.3 Members in the pulse timing queue that are not included in the high pulse are counted as low pulses.

[0042] 3. Construct an XOY plane rectangular coordinate system with the X-axis as the time axis (with the starting acquisition time as the origin, and only the positive half axis), divide the background area and pulse area, and define the threshold band according to the spectrum amplitude threshold.

[0043] The pulse region refers to the planar region formed by the positive half-axis of the Y-axis and the X-axis, which is used to draw a high-pulse histogram. Each line segment represents an independent pulse, so the dimension is the electrical signal amplitude, such as millivolt, picosecond, or decibel.

[0044] The background region refers to the planar region formed by the negative half-axis of the Y-axis and the X-axis, which is used to draw the subsequence histogram. Each queue (three-dimensional) or line segment (two-dimensional) represents a continuous low pulse, so the dimension is time or time sequence, such as milliseconds or seconds. The threshold band refers to the area between the X-axis and the ray perpendicularly emitted from the Y-axis scale of the spectrum amplitude threshold within the pulse region. The purpose is to uniformly color the parts below the threshold of each line segment in the pulse region, thereby highlighting the degree to which the high pulse exceeds the threshold as a whole and the differences between them.

[0045] 4. To integrate the partial discharge spectrum into the online monitoring system and observe the complete waveform of cyclic rolling loading in real time, a Z-axis perpendicular to the XOY plane was introduced to plot a three-dimensional partial discharge spectrum suitable for short-time pulse signals on the order of 1 second. The specific process is as follows: like Figure 3As shown, each high pulse in the main sequence is presented as a high pulse line segment, forming a histogram, and arranged sequentially on one side of the pulse region located near the human eye's observation point; each subsequence sandwiched between adjacent high pulses contains several low pulse bars parallel to the Z-axis, which are arranged sequentially along the Y-axis to form a columnar queue, arranged on one side of the background region; the length of each high pulse line segment in the pulse region is determined according to the pulse amplitude, the length of each queue in the background region is determined according to the number of low pulses contained therein, and the height of each low pulse bar in the queue is determined by the pulse amplitude; the elements on both sides are interspersed and staggered, with equal line width and spacing.

[0046] The high-pulse line segments and low-pulse bars are displayed in different shades of gray or colors to distinguish them from each other. Taking gray as an example, the specific method is as follows: 4.1 Linearly map the range of pulse amplitude values ​​to grayscale [0, 255]; 4.2. For each high pulse segment in the pulse region, the portion located in the threshold band will uniformly display the grayscale corresponding to the amplitude threshold of the spectrum, while the portion exceeding the threshold will display the grayscale corresponding to its actual amplitude. 4.3. In each subsequence of the background area, all low pulse bars display the grayscale corresponding to their own amplitude.

[0047] 5. For example Figure 4 As shown, in order to apply short-time maps to pattern recognition, it is necessary to eliminate factors such as near-far distortion and occlusion to draw a two-dimensional short-time map. Therefore, the Z-axis is removed, reducing each low-pulse bar in the background region to a planar pixel; the Y-axis is vertically flipped so that the positive half-axis is at the top. This map also retains all temporal information, and the data can be restored from the image without loss, making it suitable for machine algorithm reading and cross-program transfer.

[0048] The high-pulse line segments and low-pulse pixels are displayed as different gray levels or colors to distinguish them from each other, and the specific method is the same as in step 4.

[0049] Example 2 This invention discloses a method for drawing partial discharge patterns of DC power equipment, and the specific implementation steps are as follows: 1. A long-term queue of 100,000 sampling points was obtained by continuously collecting partial discharge pulse signals for 20 seconds and taking the peak value every 0.2 milliseconds.

[0050] 2. Determine the upper limit of the pulse capacity that can be displayed based on the pixel width of the drawing interface; Using the maximum pulse amplitude of the short-time queue as a temporary threshold, we begin to traverse downwards. When the total number of high pulses that meet the conditions just exceeds the pulse capacity limit, we determine the temporary threshold at this point as the spectrum amplitude threshold. Pulses exceeding the spectrum amplitude threshold and a subset of pulses equal to the spectrum amplitude threshold are called high pulses, and the rest are low pulses. Using each high pulse as a dividing point, the short-time queue is divided into a main sequence consisting of high pulses and several subsequences consisting of consecutive low pulses.

[0051] The method for determining the amplitude threshold, high pulse, and low pulse of the spectrum is the same as in Example 1.

[0052] 3. Construct an XOY plane rectangular coordinate system with the X-axis as the time axis (with the starting acquisition time as the origin, and only the positive half axis), divide the background area and pulse area, and define the threshold band according to the spectrum amplitude threshold.

[0053] 4. To extend the coverage time of a single spectrum and reveal waveform characteristics and distribution patterns over a larger timescale, it is necessary to increase the spectrum amplitude threshold to compress the number of high pulses and sacrifice the timing information of low pulses, as shown in the diagram. Figure 5 The diagram shown is a two-dimensional partial discharge pattern applicable to long-duration pulse signals of 20 seconds. The specific process is as follows: like Figure 6 As shown, each high pulse in the main sequence is presented as a high pulse line segment, forming a histogram, and arranged sequentially on one side of the pulse area in the upper half of the image; each subsequence sandwiched between adjacent high pulses no longer specifically distinguishes its internal low pulses, but is simplified into a whole subsequence line segment, based on the Log of the number of low pulses contained. 10 The lengths of the line segments are determined logarithmically and arranged on one side of the background area in the lower half of the image; the elements on both sides are interspersed and staggered, with equal line widths and intervals. This graph preserves the main timing information of the signal, and both sides of the Y-axis are in histogram form.

[0054] The high-pulse segment and sub-sequence segment are displayed in different shades of gray or colors to distinguish them from each other. Taking gray as an example, the specific method is as follows: 4.1 Linearly map the range of pulse amplitude values ​​to grayscale [0, 255]; 4.2. For each line segment in the pulse region, the portion within the threshold band will uniformly display the gray level corresponding to the threshold, while the portion exceeding the threshold will display the gray level corresponding to its actual amplitude. 4.3. Each subsequence line segment in the background area is displayed in two shades of gray. The gray level at the far end of the X-axis corresponds to the maximum pulse amplitude of the subsequence, while the gray level at the near end corresponds to the average pulse amplitude of the subsequence.

[0055] The DC partial discharge pattern drawing method provided by this invention can use existing AC version online monitoring devices and communication protocols to achieve accurate, reasonable and lossless graphical description of DC partial discharge signals. It can provide a useful reference for pattern recognition of early fault symptoms, thereby providing a reliable basis for the evaluation, diagnosis, early warning and maintenance of DC power equipment.

[0056] This invention also discloses a system for plotting partial discharge patterns of DC power equipment, comprising: The data acquisition module is used to acquire a continuous partial discharge pulse signal; The sequence segmentation module is used to determine the spectrum amplitude threshold based on the pulse time sequence queue, and to segment the pulse time sequence queue into a main sequence consisting of high pulses and several subsequences consisting of consecutive low pulses. The coordinate system construction module is used to construct an XOY rectangular coordinate system with the X-axis as the time axis, divide the pulse region and background region, and define the threshold band according to the spectrum amplitude threshold. The pulse timing queue is divided into a short-time queue and a long-time queue. When the pulse timing queue is a short-time queue, the discharge spectrum includes a three-dimensional short-time DC partial discharge spectrum and a two-dimensional short-time DC partial discharge spectrum. When the pulse timing queue is a long-time queue, the discharge spectrum is a two-dimensional long-time DC partial discharge spectrum. The 3D short-time partial discharge map construction module is used to draw short-time DC partial discharge maps in three-dimensional form. The specific drawing process is as follows: Each high pulse in the main sequence has a line segment length determined by its amplitude and is arranged sequentially on one side of the pulse region. A Z-axis perpendicular to the XOY plane is introduced, and each subsequence contains several columns representing low pulses parallel to the Z-axis direction. The column height is determined by its amplitude and they are arranged sequentially in a columnar queue along the Y-axis on one side of the background region. The elements on both sides are interspersed and staggered to obtain a three-dimensional short-time DC partial discharge spectrum. The two-dimensional short-time map construction module is used to draw a two-dimensional short-time DC partial discharge map. The specific process is as follows: take the top view of the two-dimensional short-time DC partial discharge map, flip it vertically based on the X-axis, and reduce each low pulse column to pixels to obtain the two-dimensional short-time DC partial discharge map. The two-dimensional long-time partial discharge pattern construction module is used to draw two-dimensional long-time DC partial discharge patterns; the specific drawing process is as follows: Each high pulse in the main sequence has a segment length determined by its amplitude and is arranged sequentially on one side of the pulse region; each subsequence is determined by the Log number of low pulses it contains. 10 The length is determined by logarithm, and the data are arranged sequentially on one side of the background region to obtain a two-dimensional long-term DC partial discharge pattern.

[0057] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for drawing partial discharge patterns of DC power equipment, characterized in that, Includes the following processes: A continuous partial discharge pulse signal is acquired to obtain the pulse timing queue; The amplitude threshold of the spectrum is determined based on the pulse time sequence queue, and the pulse time sequence queue is divided into a main sequence consisting of high pulses and several subsequences consisting of continuous low pulses. Using the X-axis as the time axis, an XOY rectangular coordinate system is constructed to divide the pulse region and background region, and a threshold band is defined based on the spectrum amplitude threshold. Pulse timing queues are divided into short-time queues and long-time queues; When the pulse timing queue is a short-time queue, the discharge spectrum includes a three-dimensional short-time DC partial discharge spectrum and a two-dimensional short-time DC partial discharge spectrum; the specific drawing process is as follows: Each high pulse in the main sequence has a line segment length determined by its amplitude and is arranged sequentially on one side of the pulse region. A Z-axis perpendicular to the XOY plane is introduced, and each subsequence contains several columns representing low pulses parallel to the Z-axis direction. The column height is determined by its amplitude and they are arranged sequentially in a columnar queue along the Y-axis on one side of the background region. The elements on both sides are interspersed and staggered to obtain a three-dimensional short-time DC partial discharge spectrum. A top view of a three-dimensional short-time DC partial discharge pattern is obtained by vertically flipping it along the X-axis, reducing each low-pulse column to pixels, thus obtaining a two-dimensional short-time DC partial discharge pattern. When the pulse timing queue is a long-time queue, the discharge pattern is a two-dimensional long-time DC partial discharge pattern, and the specific drawing process is as follows: Each high pulse in the main sequence has a segment length determined by its amplitude and is arranged sequentially on one side of the pulse region; each subsequence is determined by the Log number of low pulses it contains. 10 The length is determined by logarithm, and the data are arranged sequentially on one side of the background region to obtain a two-dimensional long-term DC partial discharge pattern.

2. The method for drawing partial discharge patterns of DC power equipment according to claim 1, characterized in that, The method for determining the spectrum amplitude threshold based on the pulse time sequence queue involves dividing the pulse time sequence queue into a main sequence consisting of high pulses and several sub-sequences consisting of consecutive low pulses; specifically: Determine the upper limit of the pulse capacity that can be displayed based on the pixel width of the drawing interface; the upper limit of the pulse capacity should not exceed half of the horizontal pixel count of the screen. Using the maximum pulse amplitude of the pulse timing queue as a temporary threshold, start traversing and count the total number of high pulses. If it is less than the upper limit of pulse capacity, gradually reduce the temporary threshold until the total number of high pulses is about to exceed the upper limit of pulse capacity. Set the temporary threshold at this time as the spectrum amplitude threshold. Calculate the difference X between the upper limit of pulse capacity and the total number of high pulses corresponding to the spectrum amplitude threshold; Pulses with amplitude values ​​above the spectrum threshold are all counted as high pulses, and those with amplitude values ​​below the spectrum threshold are counted as low pulses. For the remaining pulses whose amplitude equals the spectral amplitude threshold, calculate their respective neighborhood relative heights: Take the X neighborhoods with the largest relative heights and include them in the high pulses, which will fill the pulse capacity limit. The remaining neighborhoods with smaller relative heights are all low pulses. High pulses constitute the main sequence, and each consecutive low pulse constitutes a subsequence.

3. The method for drawing partial discharge patterns of DC power equipment according to claim 2, characterized in that, The expression for calculating the relative height of the neighborhood is: ΔV N = 3V N - 0.5V N-2 - V N-1 - V N+1 - 0.5V N+2 ; Where, ΔV N V represents the relative height of the neighborhood of the Nth pulse; N V represents the amplitude of the Nth pulse; N-2 V represents the amplitude of the (N-2)th pulse; N-1 V represents the amplitude of the (N-1)th pulse; N+1 V represents the amplitude of the (N+1)th pulse; N+2 This represents the amplitude of the (N+2)th pulse.

4. The method for drawing partial discharge patterns of DC power equipment according to claim 1, characterized in that, The pulse region is divided as follows: the plane region formed by the positive half-axis of the Y-axis and the X-axis is used to draw a high-pulse histogram, and each line segment represents an independent pulse.

5. The method for drawing partial discharge patterns of DC power equipment according to claim 1, characterized in that, The background area is divided as follows: the plane area formed by the negative half-axis of the Y-axis and the X-axis is used as the background area to draw the subsequence histogram. Each queue or line segment represents a continuous low pulse.

6. The method for drawing partial discharge patterns of DC power equipment according to claim 1, characterized in that, The threshold band is determined as follows: within the pulse region, a ray is emitted vertically from the scale of the spectrum amplitude threshold located on the Y-axis, and the area between this ray and the X-axis is taken as the threshold band.

7. A system for plotting partial discharge patterns of DC power equipment, characterized in that, include: The data acquisition module is used to acquire a continuous partial discharge pulse signal; The sequence segmentation module is used to determine the spectrum amplitude threshold based on the pulse time sequence queue, and to segment the pulse time sequence queue into a main sequence consisting of high pulses and several subsequences consisting of consecutive low pulses. The coordinate system construction module is used to construct an XOY rectangular coordinate system with the X-axis as the time axis, divide the pulse region and background region, and define the threshold band according to the spectrum amplitude threshold. The pulse timing queue is divided into a short-time queue and a long-time queue. When the pulse timing queue is a short-time queue, the discharge spectrum includes a three-dimensional short-time DC partial discharge spectrum and a two-dimensional short-time DC partial discharge spectrum. When the pulse timing queue is a long-time queue, the discharge spectrum is a two-dimensional long-time DC partial discharge spectrum. The 3D short-time partial discharge map construction module is used to draw short-time DC partial discharge maps in three-dimensional form. The specific drawing process is as follows: Each high pulse in the main sequence has a line segment length determined by its amplitude and is arranged sequentially on one side of the pulse region. A Z-axis perpendicular to the XOY plane is introduced, and each subsequence contains several columns representing low pulses parallel to the Z-axis direction. The column height is determined by its amplitude and they are arranged sequentially in a columnar queue along the Y-axis on one side of the background region. The elements on both sides are interspersed and staggered to obtain a three-dimensional short-time DC partial discharge spectrum. The two-dimensional short-time map construction module is used to draw a two-dimensional short-time DC partial discharge map. The specific process is as follows: take the top view of the two-dimensional short-time DC partial discharge map, flip it vertically based on the X-axis, and reduce each low pulse column to pixels to obtain the two-dimensional short-time DC partial discharge map. The two-dimensional long-time partial discharge pattern construction module is used to draw two-dimensional long-time DC partial discharge patterns; the specific drawing process is as follows: Each high pulse in the main sequence has a segment length determined by its amplitude and is arranged sequentially on one side of the pulse region; each subsequence is determined by the Log number of low pulses it contains. 10 The length is determined by logarithm, and the data are arranged sequentially on one side of the background region to obtain a two-dimensional long-term DC partial discharge pattern.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method for drawing partial discharge patterns of DC power equipment as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for drawing partial discharge patterns of DC power equipment as described in any one of claims 1 to 6.

Images