Methods, apparatus, equipment, and storage media for determining internal discharge and external noise.

Abstract

This application discloses a method, apparatus, equipment, and storage medium for determining in-cabinet discharge and external noise, relating to the field of partial discharge detection. The method includes: acquiring a standard waveform and a real-time waveform of the in-cabinet current signal; determining a standard time difference between a standard original pulse peak and a standard reflected pulse peak based on the standard waveform; determining a real-time time difference between a real-time original pulse peak and a real-time reflected pulse peak based on the real-time waveform; obtaining a standard distance based on the standard time difference and the propagation speed of the current signal in the cable; obtaining a real-time distance based on the real-time time difference and the propagation speed of the current signal in the cable; and determining the signal source based on a comparison between the real-time distance and the standard distance. This method can improve the accuracy of signal source determination.

Description

Technical Field

[0001] This application relates to the field of partial discharge detection, and in particular to a method, apparatus, equipment and storage medium for determining in-cabinet discharge and external noise. Background Technology

[0002] In power systems, ring main units and switchgear are key nodes in power distribution and control, and their operational reliability directly affects the overall safety and stability of the power grid. During long-term operation, these devices are prone to partial discharge due to factors such as insulation material aging, uneven electric field distribution, and changes in ambient temperature and humidity. If partial discharge is not identified and addressed in a timely manner, it may escalate into serious faults such as insulation breakdown and equipment damage, causing widespread power outages and severely impacting industrial production and residential electricity consumption.

[0003] Currently, the pulse current method based on built-in bushing sensors is commonly used for on-site detection of partial discharge in ring main units / switchgear. This method requires no structural modifications to the equipment and has the advantages of low cost and convenient operation. In addition, the industry often supplements this method with ultra-high frequency detection and ultrasonic detection to improve the integrity of the detection. These methods mainly achieve discharge identification by collecting and analyzing the amplitude, phase, frequency and other characteristics of the signal, and have accumulated certain technical expertise in practical applications.

[0004] However, existing detection methods are inaccurate in determining the source of signals. Summary of the Invention

[0005] This application provides a method, apparatus, equipment, and storage medium for determining in-cabinet discharge and external noise, which can improve the accuracy of signal source determination.

[0006] To achieve the above objectives, this application adopts the following technical solution:

[0007] Firstly, this application provides a method for determining the difference between in-cabinet discharge and external noise, including:

[0008] Acquire the standard waveform and the real-time waveform of the cabinet current signal;

[0009] Based on the standard waveform, determine the standard time difference between the standard original pulse peak and the standard reflected pulse peak;

[0010] Based on the real-time waveform, determine the real-time time difference between the real-time original pulse peak and the real-time reflected pulse peak;

[0011] The standard distance is obtained based on the standard time difference and the propagation speed of the current signal in the cable;

[0012] The real-time distance is obtained based on the real-time time difference and the propagation speed of the current signal in the cable;

[0013] The source of the signal is determined based on the comparison between the real-time distance and the standard distance.

[0014] Optionally, determining the signal source based on the comparison result between the real-time distance and the standard distance includes:

[0015] The standard attenuation coefficient is obtained based on the standard original pulse peak and the standard reflected pulse peak;

[0016] The real-time attenuation coefficient is obtained based on the real-time original pulse peak and the real-time reflected pulse peak.

[0017] The source of the signal is determined based on the comparison between the standard attenuation coefficient and the real-time attenuation coefficient, and the comparison between the real-time distance and the standard distance.

[0018] Optionally, determining the signal source based on the comparison between the standard attenuation coefficient and the real-time attenuation coefficient, and the comparison between the real-time distance and the standard distance, includes:

[0019] Extract the pulse rising edge characteristic parameters of the standard waveform to obtain the standard rising edge characteristics;

[0020] Extract the pulse rising edge characteristic parameters of the real-time waveform to obtain the real-time rising edge characteristics;

[0021] The source of the signal is determined by comparing the standard attenuation coefficient with the real-time attenuation coefficient, the real-time distance with the standard distance, and the standard rising edge characteristics with the real-time rising edge characteristics.

[0022] Optionally, determining the signal source based on the comparison results of the standard attenuation coefficient and the real-time attenuation coefficient, the comparison results of the real-time distance and the standard distance, and the comparison results of the standard rising edge characteristics and the real-time rising edge characteristics includes:

[0023] Obtain the distance difference between the real-time distance and the standard distance, as well as the attenuation difference between the real-time attenuation coefficient and the standard attenuation coefficient;

[0024] Calculate the rate of change of the real-time rising edge feature relative to the standard rising edge feature;

[0025] A comprehensive score is obtained based on the distance difference, attenuation difference, and rising edge change rate.

[0026] The source of the signal is determined based on the comprehensive score and the score threshold.

[0027] Optionally, determining the signal source based on the comprehensive score and the score threshold includes:

[0028] Acquire historical noise level data under the detection environment;

[0029] The scoring threshold is dynamically adjusted based on the historical noise level data.

[0030] If the overall score is less than or equal to the adjusted score threshold, the signal source is determined to be discharge inside the cabinet;

[0031] If the overall score is greater than the adjusted score threshold, the signal source is determined to be external noise.

[0032] Optionally, the method further includes:

[0033] Extract the kurtosis and skewness of the standard waveform within the effective window to form a standard feature vector;

[0034] Extract the kurtosis and skewness of the real-time waveform within the same window to form a real-time feature vector;

[0035] Calculate the cosine similarity between the standard feature vector and the real-time feature vector;

[0036] If the cosine similarity is less than the similarity threshold, the signal source is determined to be external noise.

[0037] If the cosine similarity is greater than or equal to the similarity threshold, the signal source is determined to be discharge inside the cabinet.

[0038] Optionally, the method further includes:

[0039] Obtain the real-time temperature and relative humidity of the detection environment;

[0040] The propagation speed is corrected based on real-time temperature and relative humidity.

[0041] Secondly, this application provides a device for determining internal discharge and external noise, comprising:

[0042] The acquisition module is used to acquire the standard waveform and the real-time waveform of the current signal inside the cabinet.

[0043] The processing module is used to determine the standard time difference between the standard original pulse peak and the standard reflected pulse peak based on the standard waveform; determine the real-time time difference between the real-time original pulse peak and the real-time reflected pulse peak based on the real-time waveform; obtain the standard distance according to the standard time difference and the propagation speed of the current signal in the cable; and obtain the real-time distance according to the real-time time difference and the propagation speed of the current signal in the cable.

[0044] The determination module is used to determine the source of the signal based on the comparison result between the real-time distance and the standard distance.

[0045] Thirdly, this application provides a computing device, including a memory and a processor;

[0046] The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in any one of the first aspects.

[0047] Fourthly, this application provides a computer-readable storage medium for storing a computer program for performing the method as described in any one of the first aspects.

[0048] As can be seen from the above technical solution, this application has at least the following beneficial effects:

[0049] First, the solution breaks through the limitations of traditional single-feature detection. It constructs a multi-dimensional judgment system based on distance features and supplemented by attenuation coefficient and rising edge features. Combined with cosine similarity verification of waveform statistical features such as kurtosis and skewness, it captures the essential difference between discharge and noise from multiple dimensions such as signal propagation law, energy attenuation characteristics and waveform morphology differences. This greatly reduces the risk of misjudgment based on single features and improves the accuracy of signal source determination under complex working conditions.

[0050] Secondly, by introducing a dynamic threshold adjustment strategy that correlates a comprehensive scoring mechanism with historical noise levels, the judgment criteria can adapt to the noise interference intensity of different detection environments. This effectively avoids judgment bias in high-noise or low-noise scenarios caused by fixed thresholds, enhancing the method's environmental adaptability and robustness. Simultaneously, by combining real-time correction of signal propagation speed based on ambient temperature and humidity, the influence of environmental factors on distance calculation is further offset, ensuring the reliability of the feature parameters.

[0051] Finally, the solution relies entirely on the built-in bushing sensor of the equipment to collect signals, without the need for additional auxiliary sensors or structural modifications to the equipment. While inheriting the advantages of low cost and convenient operation of the traditional pulse current method, it achieves a leapfrog improvement in detection accuracy through algorithm optimization. It can timely and accurately identify partial discharge in the cabinet, avoiding missed fault detection or over-maintenance due to signal misjudgment, providing strong protection for the safe and stable operation of ring main unit / switch cabinet, thereby maintaining the continuity and reliability of power grid supply and reducing economic losses caused by fault outages.

[0052] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0053] Figure 1 A flowchart illustrating a method for determining in-cabinet discharge and external noise, provided as an embodiment of this application;

[0054] Figure 2 A schematic diagram of a device for determining internal discharge and external noise provided in an embodiment of this application;

[0055] Figure 3 This is a schematic diagram of a computing device provided in an embodiment of this application. Detailed Implementation

[0056] The terms "first," "second," and "third," etc., used in this application specification and accompanying drawings are used to distinguish different objects, not to limit a specific order.

[0057] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0058] To ensure clarity and conciseness in the description of the following embodiments, a brief introduction to the related technologies is given first:

[0059] Ring main units / switchgear are devices for power distribution and control in power systems. They are widely used in distribution network lines, substations and other scenarios, and undertake the key functions of line connection and disconnection, fault isolation and power distribution. Their operating status directly determines the stability and security of power grid supply.

[0060] Partial discharge is a localized, non-penetrating discharge phenomenon that occurs when the insulation material inside a ring main unit / switch cabinet develops local defects due to factors such as aging, electric field distortion, and environmental corrosion. The electric field strength at the defect exceeds the insulation tolerance threshold, which is an early warning signal of equipment insulation deterioration.

[0061] External noise originates from external cables connected to the ring main unit / switch cabinet, such as cable insulation defects, external electromagnetic interference, and local electric arcs in the cable. The interference signals propagate into the cabinet through the cables, and their amplitude, phase, and other characteristics are easily overlapped with the partial discharge signals inside the cabinet.

[0062] Ring main units / switchgear, as key nodes in power grid distribution and control, directly affect the overall safety and continuity of power supply of the grid due to their insulation condition. During long-term operation, due to factors such as natural aging of insulation materials, uneven distribution of the internal electric field, and alternating changes in ambient temperature and humidity, partial discharge can easily occur inside the equipment. If this discharge signal is not identified and processed in a timely and accurate manner, the discharge will continue to erode the insulation material, eventually leading to serious faults such as insulation breakdown and equipment damage, which can then cause large-scale power outages, significantly impacting industrial production and normal residential power consumption.

[0063] Currently, on-site testing commonly uses the pulse current method based on built-in sleeve sensors, supplemented by ultra-high frequency and ultrasonic testing methods. However, the main bottleneck is the inaccurate determination of the signal source, making it difficult to effectively distinguish between partial discharge inside the cabinet and noise outside the station.

[0064] The main causes include: First, the noise outside the station and the discharge signal inside the cabinet are very similar in terms of basic characteristics such as amplitude and phase. The reflection and attenuation phenomena when the signal propagates in long-distance cables, as well as the influence of ambient temperature and humidity on the signal propagation speed, further confuse the characteristics of the two.

[0065] Second, existing methods mostly rely on single basic features and fixed judgment thresholds, failing to fully explore deeper information such as signal propagation patterns, energy loss, and morphological characteristics. They cannot adapt to changes in noise levels in different scenarios, leading to a high risk of misjudgment and missed judgment under complex operating conditions. This makes it difficult to provide timely warnings of real discharge faults and may also increase unnecessary operation and maintenance costs due to misjudgments.

[0066] In view of this, embodiments of this application provide a method for determining in-cabinet discharge and external noise, which can be executed by a processing device. This processing device can be a terminal or a server. Terminals include, but are not limited to, smartphones, tablets, laptops, personal digital assistants, or smart wearable devices. Servers can be cloud servers, such as central servers in a central cloud computing cluster or edge servers in an edge cloud computing cluster. Of course, servers can also be servers in a local data center. A local data center refers to a data center directly controlled by the user.

[0067] To address the major problem of accurately distinguishing between internal discharge and external noise in the partial discharge detection of existing ring main units / switchgear, this application firstly utilizes the time-domain reflectometry method to capture the distance characteristics of signal propagation, using the difference between the standard distance and the real-time distance as the distinguishing criterion; secondly, it mines the signal energy attenuation characteristics (attenuation coefficient) and waveform morphology characteristics (rising edge characteristics, kurtosis, skewness), and strengthens the identification of the essential differences between discharge and noise through multi-dimensional feature complementarity verification; finally, it introduces a dynamic threshold adjustment mechanism that corrects environmental parameters and correlates historical noise to offset the impact of environmental interference and scene differences on the judgment results, ultimately achieving accurate and stable determination of signal sources under complex operating conditions, while maintaining compatibility with existing equipment, allowing for application without additional modifications.

[0068] To make the technical solution of this application clearer and easier to understand, the method for determining in-cabinet discharge and external noise provided by an embodiment of this application will be described below with reference to the accompanying drawings. Figure 1 As shown, this figure is a flowchart of a method for determining in-cabinet discharge and external noise according to an embodiment of this application. The method includes:

[0069] S201. The processing equipment acquires the standard waveform and the real-time waveform of the cabinet current signal.

[0070] The internal current signal refers to the pulse current signal generated in the internal components (such as busbars, switches, bushings) or connected external cables during the operation of the ring main unit / switch cabinet. It includes discharge signals caused by partial discharge inside the cabinet, and may also include interference signals transmitted from outside the station through the cables.

[0071] The standard waveform is the pure internal current signal waveform collected by the processing equipment under the normal operation of the ring main unit / switch cabinet and without external noise interference. This waveform can truly reflect the propagation path, energy attenuation and morphological characteristics of the signal inside the cabinet, and serves as a benchmark reference for subsequent comparison with the real-time waveform.

[0072] Real-time waveform refers to the waveform of the current signal inside the cabinet that is collected in real time by the processing equipment during the normal operation of the ring main unit / switch cabinet. This waveform may contain partial discharge signals inside the cabinet or may be superimposed with external noise signals. It is an analytical object that needs to be compared with standard waveforms to determine the source of the signal.

[0073] The detailed acquisition process is as follows:

[0074] Preparations before data acquisition: First, connect the processing equipment to the three-phase connection port of the live indicator in the ring main unit / switch cabinet using a compatible interface cable (such as a banana plug-BNC interface cable) to ensure a reliable connection between the equipment and the inherent bushing sensor inside the cabinet. At the same time, complete the communication pairing between the processing equipment and the supporting terminal (such as a mobile phone), such as WIFI connection, to ensure that waveform data can be transmitted and stored in real time. After the test personnel have taken safety precautions, confirm that the equipment has sufficient power and normal self-test, and enter the data acquisition ready state.

[0075] The standard waveform acquisition process is as follows: Under normal operating conditions and with no external noise interference detected, a time-domain reflectometry (TDR) pulse signal of preset amplitude is injected into the cabinet through the processing equipment. The pulse signal propagates through the components inside the cabinet to the external cable terminal and is reflected back into the cabinet. The processing equipment captures the current signal of this complete propagation process in real time through the built-in bushing sensor. After signal amplification and filtering within the equipment, the complete waveform containing the standard original pulse peak and the standard reflected pulse peak is extracted, which is the standard waveform. This waveform is simultaneously stored in the processing equipment or the matching terminal as a reference for subsequent comparison.

[0076] The steps for acquiring the real-time waveform are as follows: keeping the connection method between the processing equipment and the live indicator unchanged and the acquisition parameters consistent, under the normal load operation of the ring main unit / switch cabinet, the processing equipment continuously acquires the current signal inside the cabinet through the bushing sensor. At this time, the acquired signal may contain the pulse signal generated by partial discharge inside the cabinet, or it may be superimposed with the noise signal transmitted from the external cable. The processing equipment also performs amplification, filtering and other preprocessing on the acquired raw signal, removes invalid noise, and extracts the complete waveform containing the real-time raw pulse peak and the real-time reflected pulse peak, which is the real-time waveform, and stores it to the equipment or terminal in real time.

[0077] Through the detailed acquisition process described above, standard waveforms and real-time waveforms are obtained, providing data support for subsequent calculation of the pulse peak time difference between the two types of waveforms, derivation of the corresponding distances, and determination of the signal source through multi-dimensional comparison.

[0078] S202. The processing equipment determines the standard time difference between the standard original pulse peak and the standard reflected pulse peak based on the standard waveform; and determines the real-time time difference between the real-time original pulse peak and the real-time reflected pulse peak based on the real-time waveform.

[0079] The standard raw pulse peak is the first peak amplitude of the time-domain reflection (TDR) pulse signal injected into the cabinet by the processing equipment in the standard waveform. It is a characteristic point when the pulse signal is emitted from the equipment and initially propagates, and intuitively reflects the initial energy intensity of the pulse.

[0080] The standard reflected pulse peak is the peak amplitude of the initial pulse signal in the standard waveform, which is reflected back into the cabinet and captured by the processing equipment after the initial pulse signal propagates through the cabinet components and external cables to the cable terminal. The amplitude is smaller than the standard original pulse peak, and it is the characteristic point of the pulse signal completing the closed loop of emission, propagation and reflection.

[0081] The standard time difference is the difference between the time point corresponding to the standard original pulse peak and the time point corresponding to the standard reflected pulse peak. It reflects the complete propagation time of the pulse signal from inside the cabinet to being reflected back into the cabinet via the cable terminal.

[0082] The real-time raw pulse peak is the first peak amplitude of the pulse signal (which may be a discharge signal or a noise signal) actually generated inside the cabinet in the real-time waveform. It is a characteristic point of the initial propagation of the real-time signal.

[0083] The real-time reflected pulse peak is the peak amplitude of the signal corresponding to the original pulse peak in the real-time waveform. After the signal propagates to the reflection point (cable terminal or noise source), it is reflected back into the cabinet and captured. It is a characteristic point of the real-time signal propagation closed loop.

[0084] The real-time time difference is the difference between the time point corresponding to the real-time original pulse peak and the time point corresponding to the real-time reflected pulse peak, reflecting the actual propagation time of the real-time pulse signal in the cabinet and cable.

[0085] The specific steps are as follows:

[0086] The processing device first performs noise reduction optimization on the stored standard waveform and real-time waveform respectively. It uses a built-in filtering algorithm to remove invalid noise in the waveform, such as random interference with small amplitude, to ensure that the pulse peak characteristics in the two types of waveforms are clearly distinguishable and to avoid noise affecting the peak recognition accuracy.

[0087] For standard waveforms, the processing equipment automatically identifies and marks the time coordinates corresponding to the standard original pulse peak through a peak detection algorithm, that is, the initial time point when the pulse signal is emitted from the equipment. Then, it locates the time coordinates corresponding to the standard reflected pulse peak, that is, the time point when the reflected signal returns to the cabinet and is captured. Subsequently, the standard time difference between the standard original pulse peak and the standard reflected pulse peak is calculated through the built-in calculation module of the equipment. This time difference directly corresponds to the complete propagation time of the pulse signal in the cabinet, cable terminal, and the path in the cabinet.

[0088] The expression for calculating the standard time difference is:

[0089]

[0090] in, Indicates the standard time difference. This represents the time coordinates corresponding to the standard original pulse peak. This represents the time coordinate corresponding to the standard reflected pulse peak.

[0091] For real-time waveforms, the same peak detection algorithm and judgment threshold are used as for standard waveforms. The processing device performs feature recognition on the real-time waveform, marks the time coordinates corresponding to the original real-time pulse peaks (i.e., the time point when the real-time signal is initially generated) and the time coordinates corresponding to the real-time reflected pulse peaks (i.e., the time point when the real-time signal returns after reflection). Then, through the same calculation logic, the real-time time difference is obtained, which reflects the total time consumed by the real-time signal in the actual propagation path.

[0092] The expression for calculating the real-time time difference is:

[0093]

[0094] in, Indicates the real-time time difference. This represents the time coordinate corresponding to the real-time raw pulse peak. This represents the time coordinate corresponding to the real-time reflected pulse peak.

[0095] This step serves as a bridge between waveform acquisition and distance calculation. By extracting the time coordinates of key characteristic peaks in the two types of waveforms, the morphological characteristics of the waveforms are transformed into quantifiable time parameters, providing an accurate quantitative basis for subsequently deriving the actual propagation distance by combining the signal propagation speed and distinguishing the signal source.

[0096] S203. The processing equipment obtains the standard distance based on the standard time difference and the propagation speed of the current signal in the cable; and obtains the real-time distance based on the real-time time difference and the propagation speed of the current signal in the cable.

[0097] The propagation speed of a current signal in a cable refers to the rate at which a pulse current signal is transmitted in the external cable connecting a ring main unit / switch cabinet. Its magnitude is determined by the cable material (such as copper core or aluminum core), the type of insulation medium, and the temperature and humidity of the detection environment. It is a physical parameter that can be determined through theoretical calculation or actual calibration, and can be dynamically corrected in conjunction with environmental parameters.

[0098] The standard distance is a distance parameter calculated based on the standard time difference and the signal propagation speed. It corresponds to the complete propagation path length of the pulse signal in the standard waveform from the inside of the cabinet to the reflection back into the cabinet via the cable terminal. Because the internal structure of the cabinet is compact, the internal propagation distance can be ignored. Essentially, it is twice the total length of the external cable and serves as the distance benchmark for determining the discharge signal inside the cabinet.

[0099] Real-time distance is a distance parameter calculated based on the real-time time difference and the signal propagation speed. It corresponds to the actual propagation path length of the pulse signal (discharge or noise) in the real-time waveform from the source to the reflection point back into the cabinet. It is a quantitative indicator for distinguishing the source of the signal.

[0100] Before calculating the standard distance and real-time distance, the method also includes a correction for the propagation speed, as follows:

[0101] First, the processing device acquires the real-time temperature and relative humidity of the detection environment; then, it corrects the propagation speed based on the real-time temperature and relative humidity.

[0102] The processing device has a built-in temperature and humidity sensor that captures the temperature and relative humidity data of the detection site in real time during the synchronous period of waveform acquisition and time difference calculation. The acquisition frequency is consistent with the signal sampling frequency, for example, once every millisecond, to ensure that the acquired environmental parameters can truly reflect the on-site conditions when the signal propagates. After the acquisition is completed, the device automatically filters abnormal data, such as values ​​with instantaneous fluctuations that exceed the reasonable range, and retains stable and valid temperature and humidity data for subsequent calculations.

[0103] The processing equipment first retrieves the standard propagation speed corresponding to the current cable type from the built-in database. This standard value is a reference parameter determined by theoretical calculation or previous actual measurement under the reference environmental conditions (temperature 25℃, relative humidity 60%).

[0104] The processing equipment adjusts the standard propagation speed based on a preset environmental adaptation correction formula and real-time temperature and humidity data. The correction formula for temperature effects is as follows:

[0105]

[0106] in, This indicates the temperature-corrected propagation speed. Indicates the standard propagation speed, This represents the temperature correction factor, which is determined by the cable material. This indicates the real-time temperature of the monitored environment.

[0107] The correction formula for the effect of humidity is:

[0108]

[0109] in, Indicates the final propagation speed. This represents the humidity correction factor, which is determined based on the cable insulation type. This indicates the relative humidity of the environment being tested.

[0110] The processing device uses the established standard time difference, combined with the corrected propagation speed, to calculate the standard distance using the following expression:

[0111]

[0112] in, Indicates standard distance.

[0113] Since the propagation path of the pulse signal corresponding to the standard waveform is: emitted from inside the cabinet → reflected by the cable terminal → returned to the cabinet, and the cabinet structure is compact, the internal propagation distance is negligible. Essentially twice the total length of the external cable, it serves as a distance benchmark for measuring signals originating inside the cabinet.

[0114] Maintain the speed of transmission To ensure consistency in the baseline for both distance calculations and avoid additional errors, the processing device uses the pre-determined real-time time difference to calculate the real-time distance using the same expression:

[0115]

[0116] in, Indicates real-time distance.

[0117] This real-time distance value corresponds to the actual propagation path length of the real-time signal. If the real-time signal is an in-cabinet discharge, its propagation path is consistent with the standard waveform and will be very close to the standard distance. If the real-time signal is external noise (originating from a defect point inside the cable), its propagation path is only the round-trip distance from the noise generation point to the return cabinet. It will be significantly smaller .

[0118] This step serves as a bridge between time difference calculation and signal source determination. By using quantification expressions, abstract time characteristics are transformed into concrete distance indicators, making the differences in the propagation paths of discharge and noise intuitively apparent, thus laying the foundation for distinguishing signal sources through distance comparison in the future.

[0119] S204. The processing equipment determines the source of the signal based on the comparison between the real-time distance and the standard distance.

[0120] The comparison result refers to the quantitative judgment basis obtained by calculating the difference between the real-time distance and the standard distance through a preset algorithm and comparing it with a set threshold.

[0121] The signal source refers to the origin of the pulse signal collected inside the ring main unit / switch cabinet, which can be divided into internal discharge (discharge signal caused by internal insulation defects of the equipment) and external noise (interference signal from external cable defects, electromagnetic interference, etc.).

[0122] Other indicators for determining the source of a signal include the following:

[0123] The processing equipment obtains the standard attenuation coefficient based on the standard original pulse peak and the standard reflected pulse peak; and obtains the real-time attenuation coefficient based on the real-time original pulse peak and the real-time reflected pulse peak.

[0124] The standard attenuation coefficient is a quantitative index calculated based on the standard original pulse peak and the standard reflected pulse peak. It reflects the degree of energy attenuation of the pulse signal after it propagates through the cabinet, cable terminal and the complete path inside the cabinet under standard conditions. It is the benchmark for determining the energy characteristics of the discharge signal inside the cabinet.

[0125] The real-time attenuation coefficient is a quantitative index calculated based on the real-time original pulse peak and the real-time reflected pulse peak. It reflects the degree of energy attenuation of the real-time signal after it has traveled through the actual propagation path and is an energy characteristic parameter that helps distinguish the source of the signal.

[0126] The processing equipment first performs feature localization on the pre-processed (noise reduction, filtering) standard waveform and real-time waveform, and then uses a peak detection algorithm to identify and extract the amplitude of the standard original pulse peak, the amplitude of the standard reflected pulse peak, as well as the amplitude of the real-time original pulse peak and the amplitude of the real-time reflected pulse peak, to ensure that the extracted amplitude parameters can truly reflect the strength of the signal energy.

[0127] The processing equipment calculates the standard attenuation coefficient using the following expression to quantify the energy attenuation pattern under the standard path:

[0128]

[0129] in, Indicates the standard attenuation coefficient. This represents the amplitude of the standard reflected pulse peak. This indicates the amplitude of the standard original pulse peak.

[0130] This standard attenuation coefficient essentially reflects the proportion of energy attenuation of a pulse signal within the cabinet, cable terminals, and along its path within the cabinet due to factors such as cable insulation loss and dielectric absorption. It serves as a benchmark for the inherent energy characteristics of discharge signals within the cabinet. For example, the standard attenuation coefficient... This indicates that the standard signal attenuates by 70% after propagation.

[0131] Maintaining the same calculation logic as the standard attenuation factor, the processing device calculates the real-time attenuation factor using the same expression, quantifying the actual energy attenuation ratio of the real-time signal:

[0132]

[0133] in, Indicates the real-time attenuation coefficient. This indicates the amplitude of the real-time reflected pulse peak. This indicates the amplitude of the real-time raw pulse peak.

[0134] If the real-time signal is discharge inside the cabinet, its propagation path is consistent with the standard path, and its energy attenuation pattern is similar, resulting in a similar height. If the real-time signal is external noise (originating from inside the cable), the propagation path is shorter, and energy loss is less. It will be significantly larger than For example, the noise signal propagation path is only the middle section of the cable, resulting in less energy attenuation. .

[0135] This step is an important supplement to distance feature determination. By exploring the energy attenuation characteristics of the signal, it effectively avoids misjudgments that may occur due to a single distance feature. For example, the propagation distance of special noise signals is close to the standard distance, but the energy attenuation pattern is significantly different, which further improves the accuracy of signal source determination.

[0136] The processing equipment determines the source of the signal based on the comparison between the standard attenuation coefficient and the real-time attenuation coefficient, as well as the comparison between the real-time distance and the standard distance.

[0137] Before determining the source of the signal, it is necessary to obtain the comparison results between the standard rising edge characteristics and the real-time rising edge characteristics. The specific steps are as follows:

[0138] The processing equipment extracts the pulse rising edge characteristic parameters of the standard waveform to obtain the standard rising edge characteristics; it also extracts the pulse rising edge characteristic parameters of the real-time waveform to obtain the real-time rising edge characteristics.

[0139] The rising edge of a pulse refers to the process segment in which the pulse signal rises from 10% to 90% of its amplitude. It is one of the morphological characteristics of a pulse signal, and its variation pattern is determined by the physical characteristics of the signal source (such as discharge type and interference nature) and the propagation path. It is the basis for distinguishing the morphology of signals from different sources.

[0140] Rising edge characteristic parameters are quantitative indicators that describe the shape of the rising edge of a pulse. They include rising edge time (the time it takes for the signal to rise from 10% amplitude to 90% amplitude), rising edge slope (the ratio of amplitude change to rising edge time), and number of rising edge inflection points (the number of inflection points in amplitude change during the rising process). These parameters can characterize the steepness, smoothness, and other morphological differences of the rising edge.

[0141] The standard rising edge feature is a morphological benchmark constructed based on the pulse rising edge feature parameters of the standard waveform. It reflects the inherent morphological pattern of the rising edge of the discharge signal in the pure cabinet, such as the standard rising edge time and standard slope range. It serves as the morphological reference for subsequent comparisons.

[0142] Real-time rising edge characteristics are actual morphological data constructed based on the pulse rising edge characteristic parameters of real-time waveforms. They reflect the rising edge shape of real-time acquired signals (discharge or noise). By comparing with standard rising edge characteristics, the morphological differences of signal generation sources can be captured.

[0143] The processing device has built-in preset feature parameter extraction rules. Based on the detection scenario (such as cable type and equipment voltage level), it determines the rising edge feature parameters to be extracted. The rising edge feature parameters are selected from two parameters: rising edge time and rising edge slope, balancing discrimination and computational efficiency. Precision thresholds for parameter extraction are also set, for example, time precision is retained to 0.01μs, and slope precision is retained to... .

[0144] The processing equipment performs focused analysis on the pre-processed standard waveform (pure cabinet discharge signal waveform) to locate the pulse rising edge segment corresponding to the original standard pulse peak (from...). Amplitude point to Amplitude point); and calculate the characteristic parameters of this segment, standard rise time. (The time difference between two points) and the rising slope (the ratio of the magnitude change to time). The standard rising slope is calculated as follows:

[0145]

[0146] in, Indicates the standard rising slope. Indicates the standard rise time.

[0147] The calculated parameters are integrated into a standard rising edge characteristic. The morphological reference of the discharge signal inside the cabinet is stored in the device database.

[0148] Maintaining extraction rules, parameter types, and accuracy thresholds consistent with standard waveforms, the processing equipment synchronously analyzes the preprocessed real-time waveform to locate the pulse rising edge segment corresponding to the real-time original pulse peak (from...). Amplitude point to Amplitude point); the same algorithm is used to calculate the real-time rise time. And the real-time rising edge slope. The expression for calculating the real-time rising edge slope is:

[0149]

[0150] in, Indicates the real-time rising edge slope. Indicates the real-time rising edge time.

[0151] Integrate parameters into real-time rising edge characteristics This serves as morphological analysis data for real-time signals.

[0152] Due to the different physical characteristics of the signal sources, there are fundamental differences in the rising edge morphology of discharge inside the cabinet and noise outside the station. Discharge inside the cabinet is caused by insulation defects, and the electric field distribution is stable, so the rising edge is relatively gentle (longer and smaller). Noise outside the station is mostly caused by local electric arcs or electromagnetic interference in cables, and the energy release is violent, so the rising edge is steeper (shorter and larger).

[0153] This step extracts two types of morphological features and constructs a three-dimensional judgment system of distance, energy, and morphology, effectively avoiding the risk of misjudgment based on a single feature (such as some noise signals having distance and attenuation coefficients close to standard values, but significant differences in morphological features), providing key morphological basis for subsequent comprehensive judgment.

[0154] The processing equipment determines the source of the signal based on the comparison results of the standard attenuation coefficient and the real-time attenuation coefficient, the comparison results of the real-time distance and the standard distance, and the comparison results of the standard rising edge characteristics and the real-time rising edge characteristics.

[0155] Specifically, the processing device acquires the distance difference between the real-time distance and the standard distance, as well as the attenuation difference between the real-time attenuation coefficient and the standard attenuation coefficient; and calculates the rising edge change rate of the real-time rising edge feature relative to the standard rising edge feature.

[0156] Distance difference refers to the quantified difference between real-time distance and standard distance (denoted as ). This reflects the deviation between the real-time signal propagation path length and the standard discharge path length within the cabinet; it is a difference index in the distance dimension. The expression for the distance difference is:

[0157]

[0158] in, This represents the distance difference.

[0159] Attenuation difference refers to the quantitative difference between the real-time attenuation coefficient and the standard attenuation coefficient (denoted as ). This reflects the degree of deviation between the real-time signal energy attenuation pattern and the standard attenuation pattern of discharge within the cabinet; it is an indicator of the difference in the energy dimension. The expression for the attenuation difference is:

[0160]

[0161] in, This represents the attenuation difference.

[0162] Rising edge change rate refers to the proportion of change of the real-time rising edge feature relative to the standard rising edge feature (denoted as ). , The rise time and rise slope are calculated separately to reflect the deviation of the real-time signal pattern from the standard discharge pattern inside the cabinet, serving as a difference indicator in the morphological dimension. The expression for the rise time change rate is:

[0163]

[0164] in, This represents the rate of change over the rising edge time.

[0165] The expression for the rate of change of the rising slope is:

[0166]

[0167] in, This indicates the rate of change of the slope along the rising edge.

[0168] Through the above calculations, the original three-dimensional feature parameters, which differed in dimensions and magnitudes, were transformed into four standardized difference indicators: distance difference, attenuation difference, rise time change rate, and rise slope change rate. These indicators intuitively quantify the degree of deviation between the real-time signal and the standard discharge signal inside the cabinet in each dimension. The smaller the deviation, the closer the real-time signal is to the discharge characteristics inside the cabinet; the larger the deviation, the more likely it is external noise. This lays the foundation for subsequent weighted fusion calculations to obtain a comprehensive score and establish a unified judgment standard.

[0169] The processing equipment obtains a comprehensive score based on the distance difference, attenuation difference, and rise time change rate.

[0170] The comprehensive score is a quantitative score obtained by integrating three types of difference indicators—distance difference, attenuation difference, and rise time change rate (including rise time change rate and rise slope change rate)—through a preset weighted fusion algorithm. It is a comprehensive indicator that reflects the overall deviation between the real-time signal and the standard signal of discharge inside the cabinet. The higher the score, the greater the deviation, and the more likely it is to be external noise; the lower the score, the closer it is to the characteristics of discharge inside the cabinet.

[0171] Since the distance difference, attenuation difference, and rising edge rate of change have different dimensions and magnitudes, they need to be mapped to the same interval, such as [0, 10], through a normalization algorithm to eliminate the influence of the difference in dimensions on the fusion result. After normalization, the standardized distance difference, standardized attenuation difference, standardized rising edge time rate of change, and standardized rising edge slope rate of change are obtained respectively.

[0172] The processing device has a built-in fixed weight matrix trained based on a large amount of measured data (covering different cable types, equipment operating conditions, and noise scenarios), such as distance difference weights. (Differences in propagation paths are the basis for distinguishing signal sources and have the highest weight), and attenuation difference weight. (The energy decay law is an important supplement), weighting of the rate of change of the rising edge time. Weight of the rate of change of the rising slope (The two types of morphological indicators work together to enhance the differentiation effect).

[0173] The standardized indicators are multiplied by their corresponding weights and then summed to obtain the comprehensive score. The calculation expression is as follows:

[0174]

[0175] in, This indicates the overall score. This represents the standardized distance difference. This represents the standardized attenuation difference. This represents the standardized rate of change over the rising edge. This represents the rate of change of the rising slope of the standardized rising edge.

[0176] This process eliminates the magnitude interference of different indicators through standardization, reflects the differences in importance of each dimension feature to the judgment result through weight allocation, and finally transforms the scattered differences of multiple dimensions into a single, comparable comprehensive score.

[0177] The processing equipment determines the signal source based on a comprehensive score and a scoring threshold. The specific steps are as follows:

[0178] The processing equipment acquires historical noise level data of the detection environment; and dynamically adjusts the scoring threshold based on the historical noise level data.

[0179] Historical noise level data under the testing environment refers to a set of external noise characteristic data accumulated and stored through long-term testing, which is related to the current testing scenario (same model equipment, same cable type, same location). It includes the distance difference, attenuation difference, rise time change rate and corresponding comprehensive score corresponding to historical noise, and is also associated with scenario information such as testing time and ambient temperature and humidity. It is a historical statistical basis reflecting the noise interference intensity of the scenario.

[0180] The processing equipment accesses the database via its built-in storage module or network connection to retrieve historical noise level data that matches the current detection scenario. The filtering criteria include equipment model, cable type, detection area, and ambient temperature and humidity range (e.g., if the current temperature is 20~25℃ and the humidity is 50%~60%, then historical data within the same range will be matched) to ensure that the acquired data is strongly correlated with the current operating conditions. At the same time, abnormal data (such as extreme values ​​that exceed three times the normal noise intensity) are removed, and the valid sample set is retained.

[0181] The processing equipment performs statistical analysis on the filtered valid historical noise data and calculates characteristic indicators, including the maximum comprehensive score corresponding to the historical noise. ,average value and standard deviation The maximum value reflects the highest possible noise interference intensity in the scenario, while the average value and standard deviation reflect the distribution pattern of the noise intensity.

[0182] Based on historical noise statistical characteristics, the initial scoring threshold is adjusted using a preset algorithm as follows:

[0183] If the maximum value of the historical noise comprehensive score Approaching the initial scoring threshold Then the adjusted scoring threshold will be set as follows:

[0184]

[0185] in, This indicates the adjusted scoring threshold. Indicates the safety factor. Indicates the initial scoring threshold. It represents the standard deviation.

[0186] If the overall intensity of historical noise is low, i.e., the average value is low Then the adjusted scoring threshold will be set as follows:

[0187]

[0188] Appropriately lower the initial scoring threshold to improve the detection rate of weak discharge signals; if the historical data sample size is insufficient (e.g., less than 50 groups), keep the initial scoring threshold unchanged to ensure the reliability of the adjustment.

[0189] Through this process, the scoring threshold is no longer a fixed, universal value, but can be dynamically adapted according to the historical patterns of noise interference in the current scenario. This effectively solves the problem of misjudging discharge as noise in high-noise scenarios and misjudging noise as discharge in low-noise scenarios when the fixed threshold is used. This makes the judgment criteria more in line with the actual scenario and further improves the environmental adaptability and judgment accuracy of the technical solution.

[0190] If the overall score is less than or equal to the adjusted score threshold, the signal source is determined to be discharge inside the cabinet; if the overall score is greater than the adjusted score threshold, the signal source is determined to be noise outside the station.

[0191] The significance of the comprehensive score is to quantify the overall difference between the real-time signal and the standard signal of discharge inside the cabinet. If the real-time signal is discharge inside the cabinet, its propagation path (corresponding to the distance difference), energy attenuation law (corresponding to the attenuation difference), and waveform shape (corresponding to the rate of change of rising edge) are highly consistent with the standard signal, and the differences in each dimension are small, resulting in a lower comprehensive score. If the real-time signal is external noise, its propagation path is shorter, its energy attenuation is less, and its waveform shape is steeper, resulting in greater differences in each dimension, resulting in a higher comprehensive score.

[0192] when If the signal is within the allowable range, it indicates that the overall deviation between the real-time signal and the standard signal of discharge inside the cabinet is within the allowable range, and the characteristics of each dimension are consistent with the inherent laws of discharge inside the cabinet. Therefore, the source of the signal is determined to be discharge inside the cabinet.

[0193] when When the signal deviates from the standard discharge signal inside the cabinet, it indicates that the overall deviation of the real-time signal from the standard discharge signal inside the cabinet exceeds the threshold boundary. Its characteristics are more consistent with the difference pattern of external noise. Therefore, the source of the signal is determined to be external noise.

[0194] The method also includes:

[0195] The processing device extracts the kurtosis and skewness of the standard waveform within the effective window to form a standard feature vector; it also extracts the kurtosis and skewness of the real-time waveform within the same window to form a real-time feature vector.

[0196] The effective window refers to a fixed time interval set according to the duration of the pulse signal, such as the time range corresponding to 50 sampling points before and after the pulse peak. It is used to focus on the waveform segment of the pulse signal, eliminate interference from irrelevant noise before and after, and ensure the relevance and accuracy of kurtosis and skewness calculations.

[0197] Kurtosis is a statistic that describes the steepness of the probability distribution of a waveform. The waveform of the discharge signal inside the cabinet is relatively flat and has a low kurtosis value, while the amplitude distribution of the noise signal outside the station is more concentrated (and prone to spikes) and has a higher kurtosis value.

[0198] Skewness is a statistic that describes the degree of asymmetry in the probability distribution of a waveform. The waveform of the discharge signal inside the cabinet is approximately symmetrical, and the skewness value is close to 0. The waveform of the noise signal outside the station is mostly asymmetrical, and the skewness value deviates from 0 to a greater extent.

[0199] The standard eigenvector is a two-dimensional vector formed by integrating the kurtosis and skewness of the standard waveform within the effective window. It is a benchmark vector characterizing the statistical distribution characteristics of the discharge signal inside the cabinet.

[0200] The real-time feature vector is a two-dimensional vector formed by integrating the kurtosis and skewness of the real-time waveform within the same effective window. It is a comparison vector that characterizes the statistical distribution characteristics of the real-time signal (discharge or noise).

[0201] The processing device calculates the cosine similarity between the standard feature vector and the real-time feature vector. The expression for the cosine similarity between the standard feature vector and the real-time feature vector is:

[0202]

[0203] in, Represents cosine similarity. Represents the standard eigenvector. Represents the real-time feature vector. Indicates the standard waveform kurtosis. Indicates the real-time waveform kurtosis. Indicates the standard waveform skewness. This indicates the real-time waveform skewness.

[0204] If the cosine similarity is less than the similarity threshold, the signal source is determined to be external noise; if the cosine similarity is greater than or equal to the similarity threshold, the signal source is determined to be internal discharge.

[0205] like This indicates that the statistical distribution characteristics (kurtosis, skewness) of the real-time signal are highly similar to the standard signal of discharge inside the cabinet, and it is determined to be discharge inside the cabinet;

[0206] like This indicates that the statistical distribution characteristics of the real-time signal differ significantly from those of the standard signal, and are more consistent with the distribution pattern of external noise, thus it is determined to be external noise.

[0207] Based on the above description, this application has the following beneficial effects:

[0208] First, the solution breaks through the limitations of traditional single-feature detection. It constructs a multi-dimensional judgment system based on distance features and supplemented by attenuation coefficient and rising edge features. Combined with cosine similarity verification of waveform statistical features such as kurtosis and skewness, it captures the essential difference between discharge and noise from multiple dimensions such as signal propagation law, energy attenuation characteristics and waveform morphology differences. This greatly reduces the risk of misjudgment based on single features and improves the accuracy of signal source determination under complex working conditions.

[0209] Secondly, by introducing a dynamic threshold adjustment strategy that correlates a comprehensive scoring mechanism with historical noise levels, the judgment criteria can adapt to the noise interference intensity of different detection environments. This effectively avoids judgment bias in high-noise or low-noise scenarios caused by fixed thresholds, enhancing the method's environmental adaptability and robustness. Simultaneously, by combining real-time correction of signal propagation speed based on ambient temperature and humidity, the influence of environmental factors on distance calculation is further offset, ensuring the reliability of the feature parameters.

[0210] Finally, the solution relies entirely on the built-in bushing sensor of the equipment to collect signals, without the need for additional auxiliary sensors or structural modifications to the equipment. While inheriting the advantages of low cost and convenient operation of the traditional pulse current method, it achieves a leapfrog improvement in detection accuracy through algorithm optimization. It can timely and accurately identify partial discharge in the cabinet, avoiding missed fault detection or over-maintenance due to signal misjudgment, providing strong protection for the safe and stable operation of ring main unit / switch cabinet, thereby maintaining the continuity and reliability of power grid supply and reducing economic losses caused by fault outages.

[0211] The above text combined Figure 1 The method for determining in-cabinet discharge and external noise provided in the embodiments of this application has been described in detail. The apparatus and equipment provided in the embodiments of this application will be described below with reference to the accompanying drawings.

[0212] like Figure 2 As shown in the figure, this is a schematic diagram of a device for determining in-cabinet discharge and external noise according to an embodiment of this application. The device includes:

[0213] The acquisition module 301 is used to acquire the standard waveform and the real-time waveform of the current signal inside the cabinet.

[0214] Processing module 302 is used to determine the standard time difference between the standard original pulse peak and the standard reflected pulse peak based on the standard waveform; determine the real-time time difference between the real-time original pulse peak and the real-time reflected pulse peak based on the real-time waveform; obtain the standard distance according to the standard time difference and the propagation speed of the current signal in the cable; and obtain the real-time distance according to the real-time time difference and the propagation speed of the current signal in the cable.

[0215] The determination module 303 is used to determine the source of the signal based on the comparison result between the real-time distance and the standard distance.

[0216] Optionally, the processing module 302 is specifically used to obtain the standard attenuation coefficient based on the standard original pulse peak and the standard reflected pulse peak; and to obtain the real-time attenuation coefficient based on the real-time original pulse peak and the real-time reflected pulse peak.

[0217] The determination module 303 is specifically used to determine the source of the signal based on the comparison results of the standard attenuation coefficient and the real-time attenuation coefficient, and the comparison results of the real-time distance and the standard distance.

[0218] Optionally, the processing module 302 is specifically used to extract the pulse rising edge feature parameters of the standard waveform to obtain the standard rising edge feature; and to extract the pulse rising edge feature parameters of the real-time waveform to obtain the real-time rising edge feature.

[0219] The determination module 303 is specifically used to determine the source of the signal based on the comparison results of the standard attenuation coefficient and the real-time attenuation coefficient, the comparison results of the real-time distance and the standard distance, and the comparison results of the standard rising edge characteristics and the real-time rising edge characteristics.

[0220] Optionally, the acquisition module 301 is specifically used to acquire the distance difference between the real-time distance and the standard distance, as well as the attenuation difference between the real-time attenuation coefficient and the standard attenuation coefficient.

[0221] The processing module 302 is specifically used to calculate the rising edge change rate of the real-time rising edge feature relative to the standard rising edge feature; and to obtain a comprehensive score based on the distance difference, attenuation difference, and rising edge change rate.

[0222] The determination module 303 is specifically used to determine the source of the signal based on the comprehensive score and the score threshold.

[0223] Optionally, the acquisition module 301 is specifically used to acquire historical noise level data under the detection environment;

[0224] Processing module 302 is specifically used to dynamically adjust the scoring threshold based on the historical noise level data;

[0225] The determination module 303 is specifically used to determine the signal source as cabinet discharge if the comprehensive score is less than or equal to the adjusted score threshold; and to determine the signal source as external noise if the comprehensive score is greater than the adjusted score threshold.

[0226] Optionally, the processing module 302 is also used to extract the kurtosis and skewness of the standard waveform within the effective window to form a standard feature vector; extract the kurtosis and skewness of the real-time waveform within the same window to form a real-time feature vector; and calculate the cosine similarity between the standard feature vector and the real-time feature vector.

[0227] The determination module 303 is further configured to determine that the signal source is external noise if the cosine similarity is less than the similarity threshold, and to determine that the signal source is internal discharge if the cosine similarity is greater than or equal to the similarity threshold.

[0228] Optionally, the acquisition module 301 is also used to acquire the real-time temperature and relative humidity of the detection environment;

[0229] The processing module 302 is also used to correct the propagation speed based on real-time temperature and relative humidity.

[0230] The device for determining in-cabinet discharge and external noise according to the embodiments of this application can correspondingly execute the method described in the embodiments of this application, and the other operations and / or functions of each module / unit of the device for determining in-cabinet discharge and external noise are respectively for implementing Figure 1For the sake of brevity, the corresponding processes of each method in the illustrated embodiments will not be described in detail here.

[0231] This application also provides a computing device. For example... Figure 3 As shown in the figure, this is a schematic diagram of a computing device provided in an embodiment of this application. The computing device 700 includes a bus 701, a processor 702, a communication interface 703, and a memory 704. The processor 702, the memory 704, and the communication interface 703 communicate with each other via the bus 701.

[0232] The 701 bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0233] The processor 702 can be any one or more of the following processors: central processing unit (CPU), graphics processing unit (GPU), microprocessor (MP), or digital signal processor (DSP).

[0234] The communication interface 703 is used for communication with external devices.

[0235] Memory 704 may include volatile memory, such as random access memory (RAM). Memory 704 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0236] The memory 704 stores executable code, and the processor 702 executes the executable code to perform the aforementioned method for determining discharge inside the cabinet and noise outside the station.

[0237] Specifically, in achieving Figure 2 In the case of the illustrated embodiment, and Figure 2When the modules or units of the device for determining in-cabinet discharge and external noise described in the embodiment are implemented by software, the execution... Figure 2 The software or program code required for the functions of each module / unit can be partially or entirely stored in the memory 704. The processor 702 executes the program code corresponding to each unit stored in the memory 704 to perform the aforementioned method for determining internal discharge and external noise.

[0238] This application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that a computing device can store, or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive). The computer-readable storage medium includes instructions that instruct the computing device to perform the above-described method for determining in-cabinet discharge and external noise.

[0239] This application also provides a computer program product comprising one or more computer instructions. When the computer instructions are loaded and executed on a computing device, all or part of the processes or functions described in this application are generated.

[0240] The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0241] When the computer program product is executed by a computer, the computer performs any of the aforementioned methods for determining internal discharge and external noise. The computer program product can be a software installation package; when any of the aforementioned methods for determining internal discharge and external noise is required, the computer program product can be downloaded and executed on a computer.

[0242] The descriptions of the processes or structures corresponding to the above figures each have their own emphasis. For parts of a process or structure that are not described in detail, please refer to the relevant descriptions of other processes or structures.

[0243] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be covered within the scope of protection of this application.

Claims

1. A method for determining internal discharge and external noise, characterized in that, The method includes: The standard waveform and the real-time waveform of the cabinet current signal are obtained; the standard waveform is obtained by simulating cabinet discharge under noise-free laboratory conditions. Based on the standard waveform, determine the standard time difference between the standard original pulse peak and the standard reflected pulse peak; based on the real-time waveform, determine the real-time time difference between the real-time original pulse peak and the real-time reflected pulse peak. The standard distance is obtained based on the standard time difference and the propagation speed of the current signal in the cable; the real-time distance is obtained based on the real-time time difference and the propagation speed of the current signal in the cable. The source of the signal is determined based on the comparison between the real-time distance and the standard distance. The step of determining the signal source based on the comparison result between the real-time distance and the standard distance includes: The standard attenuation coefficient is obtained based on the standard original pulse peak and the standard reflected pulse peak; The real-time attenuation coefficient is obtained based on the real-time original pulse peak and the real-time reflected pulse peak. The source of the signal is determined based on the comparison between the standard attenuation coefficient and the real-time attenuation coefficient, and the comparison between the real-time distance and the standard distance. The step of determining the signal source based on the comparison results of the standard attenuation coefficient and the real-time attenuation coefficient, and the comparison results of the real-time distance and the standard distance, includes: Extract the pulse rising edge characteristic parameters of the standard waveform to obtain the standard rising edge characteristics; Extract the pulse rising edge characteristic parameters of the real-time waveform to obtain the real-time rising edge characteristics; The source of the signal is determined by comparing the standard attenuation coefficient with the real-time attenuation coefficient, the real-time distance with the standard distance, and the standard rising edge characteristics with the real-time rising edge characteristics. The determination of the signal source based on the comparison results of the standard attenuation coefficient and the real-time attenuation coefficient, the comparison results of the real-time distance and the standard distance, and the comparison results of the standard rising edge characteristics and the real-time rising edge characteristics includes: Obtain the distance difference between the real-time distance and the standard distance, and the attenuation difference between the real-time attenuation coefficient and the standard attenuation coefficient; the expression for the distance difference is: in, Indicates the distance difference. Indicates real-time distance. Indicates standard distance; The expression for the attenuation difference is: in, Indicates the attenuation difference. Indicates the real-time attenuation coefficient. Indicates the standard attenuation coefficient. Calculate the rising edge rate of change of the real-time rising edge feature relative to the standard rising edge feature; the rising edge rate of change refers to the proportion of change of the real-time rising edge feature relative to the standard rising edge feature, calculated separately for the rising edge slope and rising edge time. The expression for the rising edge time rate of change is: in, Indicates the rate of change over the rising edge. Indicates the standard rise time. Indicates the real-time rise time; The expression for the rate of change of the rising slope is: in, This indicates the rate of change of the slope along the rising edge. Indicates the standard rising slope. Indicates the real-time rising edge slope; A comprehensive score is obtained based on the distance difference, attenuation difference, and rising edge change rate. in, This indicates the overall score. This represents the standardized distance difference. This represents the standardized attenuation difference. This represents the standardized rate of change over the rising edge. This represents the standardized rate of change of the rising slope; where , , and They are respectively for , , and The result after normalization; Indicates the distance difference weight. Indicates the weight of the decay difference. Indicates the weight of the rate of change over the rising edge. Indicates the weight of the rate of change of the rising slope; The source of the signal is determined based on the comprehensive score and the score threshold.

2. The method according to claim 1, characterized in that, The determination of the signal source based on the comprehensive score and the score threshold includes: Acquire historical noise level data under the detection environment; The scoring threshold is dynamically adjusted based on the historical noise level data. If the overall score is less than or equal to the adjusted score threshold, the signal source is determined to be discharge inside the cabinet; If the overall score is greater than the adjusted score threshold, the signal source is determined to be external noise.

3. The method according to claim 1, characterized in that, The method further includes: Extract the kurtosis and skewness of the standard waveform within the effective window to form a standard feature vector; Extract the kurtosis and skewness of the real-time waveform within the same window to form a real-time feature vector; Calculate the cosine similarity between the standard feature vector and the real-time feature vector; If the cosine similarity is less than the similarity threshold, the signal source is determined to be external noise. If the cosine similarity is greater than or equal to the similarity threshold, the signal source is determined to be discharge inside the cabinet.

4. The method according to claim 1, characterized in that, The method further includes: Obtain the real-time temperature and relative humidity of the detection environment; The propagation speed is corrected based on real-time temperature and relative humidity.

5. A device for determining internal discharge and external noise, characterized in that, The device includes: The acquisition module is used to acquire the standard waveform and the real-time waveform of the current signal inside the cabinet. The processing module is used to determine the standard time difference between the standard original pulse peak and the standard reflected pulse peak based on the standard waveform; determine the real-time time difference between the real-time original pulse peak and the real-time reflected pulse peak based on the real-time waveform; obtain the standard distance according to the standard time difference and the propagation speed of the current signal in the cable; and obtain the real-time distance according to the real-time time difference and the propagation speed of the current signal in the cable. The determination module is used to determine the signal source based on the comparison result of the real-time distance and the standard distance. Determining the signal source based on the comparison result includes: obtaining a standard attenuation coefficient based on the standard original pulse peak and the standard reflected pulse peak; obtaining a real-time attenuation coefficient based on the real-time original pulse peak and the real-time reflected pulse peak; determining the signal source based on the comparison result of the standard attenuation coefficient and the real-time attenuation coefficient, and the comparison result of the real-time distance and the standard distance; determining the signal source based on the comparison result of the standard attenuation coefficient and the real-time attenuation coefficient, and the comparison result of the real-time distance and the standard distance, includes: extracting the pulse of the standard waveform. The rising edge characteristic parameters are used to obtain the standard rising edge characteristic; the pulse rising edge characteristic parameters of the real-time waveform are extracted to obtain the real-time rising edge characteristic; the signal source is determined based on the comparison results of the standard attenuation coefficient and the real-time attenuation coefficient, the comparison results of the real-time distance and the standard distance, and the comparison results of the standard rising edge characteristic and the real-time rising edge characteristic; the determination of the signal source based on the comparison results of the standard attenuation coefficient and the real-time attenuation coefficient, the comparison results of the real-time distance and the standard distance, and the comparison results of the standard rising edge characteristic and the real-time rising edge characteristic includes: obtaining the distance difference between the real-time distance and the standard distance, and the attenuation difference between the real-time attenuation coefficient and the standard attenuation coefficient; wherein, the expression for the distance difference is: in, Indicates the distance difference. Indicates real-time distance. Indicates standard distance; The expression for the attenuation difference is: in, Indicates the attenuation difference. Indicates the real-time attenuation coefficient. Indicates the standard attenuation coefficient. Calculate the rising edge rate of change of the real-time rising edge feature relative to the standard rising edge feature; the rising edge rate of change refers to the proportion of change of the real-time rising edge feature relative to the standard rising edge feature, calculated separately for the rising edge slope and rising edge time. The expression for the rising edge time rate of change is: in, Indicates the rate of change over the rising edge. Indicates the standard rise time. Indicates the real-time rise time; The expression for the rate of change of the rising slope is: in, This indicates the rate of change of the slope along the rising edge. Indicates the standard rising slope. Indicates the real-time rising edge slope; A comprehensive score is obtained based on the distance difference, attenuation difference, and rising edge change rate. in, This indicates the overall score. This represents the standardized distance difference. This represents the standardized attenuation difference. This represents the standardized rate of change over the rising edge. This represents the standardized rate of change of the rising slope; where , , and They are respectively for , , and The result after normalization; Indicates the distance difference weight. Indicates the weight of the decay difference. Indicates the weight of the rate of change over the rising edge. Indicates the weight of the rate of change of the rising slope; The source of the signal is determined based on the comprehensive score and the score threshold.

6. A computing device, characterized in that, Including memory and processor; The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program for performing the method as described in any one of claims 1 to 4.

Images