10kv switch cabinet partial discharge remote detection system based on sensor technology

Abstract

The application belongs to the technical field of switch cabinet discharge detection, in particular to a 10kV switch cabinet partial discharge remote detection system based on sensor technology, which determines the high partial discharge area through three-dimensional electrostatic field simulation and historical fault data clustering analysis, determines the optimal ultraviolet detector layout scheme in combination with the performance indicators of coverage completeness rate and average positioning error, captures the ultraviolet photon signals in the switch cabinet in real time, processes the ultraviolet photon signals through environmental interference suppression and shielding signal compensation, identifies the partial discharge type based on the phase characteristics, amplitude characteristics and discharge frequency of the processed ultraviolet photon signals, locates the discharge position according to the time difference of multiple signals, establishes the mapping relationship between the ultraviolet photon intensity and the discharge energy, determines the severity level according to the discharge duration and the mapped discharge energy, and realizes the accurate detection and efficient early warning of the switch cabinet partial discharge risk.

Description

Technical Field

[0001] This invention belongs to the field of switchgear discharge detection technology, specifically a 10kV switchgear partial discharge remote detection system based on sensor technology. Background Technology

[0002] As the core switching equipment of the power distribution network, the insulation condition of 10kV switchgear is directly related to the safe and stable operation of the power system. Partial discharge is an important early sign of insulation degradation, so partial discharge detection of switchgear is of great significance.

[0003] Currently, partial discharge detection methods based on sensor technology have been widely used, including pulsed current method, ultrasonic method and ultraviolet detection method. Gallium nitride-based ultraviolet detectors, with their wide bandgap characteristics, have a high selective response capability in the 200-365nm deep ultraviolet band, which can effectively avoid the interference of visible light and infrared light. They also have the characteristics of fast response speed, high quantum efficiency, high temperature resistance and excellent electromagnetic radiation resistance, showing significant application potential in the ultraviolet detection of partial discharge.

[0004] However, existing technologies for partial discharge detection in 10kV switchgear based on gallium nitride-based ultraviolet detectors still have many limitations, specifically: 1. Existing technologies rely heavily on experience to deploy sensors without fully considering the internal electric field distribution and historical fault patterns of the switchgear for scientific planning, resulting in incomplete coverage of high-incidence areas of partial discharge or waste of sensor resources. At the same time, the sensors may experience signal attenuation due to obstruction by the internal structure or other components of the cabinet. Existing technologies lack effective signal attenuation compensation mechanisms, which affect the reliability and accuracy of detection.

[0005] 2. Existing methods rely solely on a single signal feature to identify discharge types, resulting in slow identification and limited ability to distinguish types. Furthermore, since the discharge type is not combined with the dynamic state of the equipment, the severity assessment only uses a fixed threshold or a simple linear model, leading to large assessment bias, delayed early warning, and a lack of effective exploration and utilization of energy gain potential. Summary of the Invention

[0006] To overcome the shortcomings in the background art, embodiments of the present invention provide a 10kV switchgear partial discharge remote detection system based on sensor technology, which can effectively solve the problems involved in the background art.

[0007] The objective of this invention can be achieved through the following technical solution: a 10kV switchgear partial discharge remote detection system based on sensor technology, comprising: a detection point deployment module, a signal acquisition and processing module, a discharge feature recognition module, and a detection and early warning module.

[0008] The detection point deployment module is connected to the signal acquisition and processing module, the signal acquisition and processing module is connected to the discharge feature recognition module, and the discharge feature recognition module is connected to the detection and early warning module.

[0009] Detection point deployment module: Based on the structural simulation analysis of the switchgear and the statistical results of historical faults, the deployment points and installation density of gallium nitride-based ultraviolet detectors are planned in areas with high incidence of partial discharge.

[0010] Signal acquisition and processing module: Based on the ultraviolet photon signals captured in real time by the ultraviolet detector in the switch cabinet, the ultraviolet photon signals are processed through environmental interference suppression and signal obstruction compensation.

[0011] Discharge feature identification module: Based on the phase characteristics, amplitude characteristics and discharge frequency of the processed ultraviolet photon signal, it identifies the type of partial discharge, which includes surface discharge, corona discharge or internal discharge, and locates the discharge position based on the time difference of multi-source signals.

[0012] Detection and early warning module: Establishes a mapping relationship between ultraviolet photon intensity and discharge energy, determines the severity level of the discharge location based on the discharge duration and the mapped discharge energy, and generates early warning information for feedback.

[0013] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) The present invention accurately defines the high-incidence area of ​​partial discharge by combining the electrostatic field simulation of the three-dimensional model of the switch cabinet with the spatial clustering analysis of historical fault data, and selects the optimal ultraviolet detector deployment scheme by combining the performance indicators of coverage completeness rate and average positioning error, thus solving the problem of regional missed detection and resource waste caused by the traditional method relying on experience deployment.

[0014] (2) The present invention effectively distinguishes the discharge type by combining multiple features, and determines the occlusion state by comparing the theoretical and actual signal strength ratios, and performs signal strength gain compensation on the occluded detector, which effectively compensates for signal attenuation and ensures the speed and accuracy of signal acquisition.

[0015] (3) Based on the discharge type, the present invention constructs a quantitative mapping function of normalized ultraviolet radiation intensity and apparent discharge quantity, calculates the energy of a single discharge, combines the cumulative discharge energy, and establishes a dynamic evaluation benchmark based on historical monitoring data. The severity of the discharge is evaluated by the ratio of the real-time monitoring value to the benchmark, thereby achieving accurate assessment and efficient early warning of partial discharge risk of switchgear. Attached Figure Description

[0016] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the module connection of the present invention.

[0018] Figure 2 This is a logic flowchart showing the deployment locations and installation density of the ultraviolet detector of the present invention.

[0019] Figure 3 This is a logic diagram for determining whether the ultraviolet detector of the present invention is blocked. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Reference Figure 1 As shown, the present invention provides a remote detection system for partial discharge of 10kV switchgear based on sensor technology, including: a detection point deployment module, a signal acquisition and processing module, a discharge feature recognition module, and a detection and early warning module.

[0022] The detection point deployment module is connected to the signal acquisition and processing module, the signal acquisition and processing module is connected to the discharge feature recognition module, and the discharge feature recognition module is connected to the detection and early warning module.

[0023] Detection point deployment module: Based on the structural simulation analysis of the switchgear and the statistical results of historical faults, the deployment points and installation density of gallium nitride-based ultraviolet detectors are planned in areas with high incidence of partial discharge.

[0024] Reference Figure 2 As shown, in a preferred embodiment of the present invention, the layout points and installation density are specifically determined by: electrostatic field simulation analysis based on the three-dimensional model of the switch cabinet to determine the high electric field intensity area.

[0025] Based on historical fault statistics, spatial clustering analysis is performed on fault points in a 3D model to identify historical fault hotspot areas.

[0026] The high electric field intensity region and the historical fault hotspot region are spatially superimposed, and their union is used to define the region with high incidence of partial discharge.

[0027] In this embodiment, considering that the generation of partial discharge is closely related to the electric field strength, electrostatic field simulation is carried out by using a three-dimensional model of the switch cabinet. The electric field distribution in different areas of the cabinet can be accurately calculated. Combined with spatial clustering analysis of historical fault data, historical fault hotspots with frequent faults and concentrated discharge events can be identified. The two are spatially superimposed and the union is taken, so that neither potential high electric field discharge areas are missed nor historical fault areas are ignored. Finally, the high-incidence areas of partial discharge are accurately defined.

[0028] Within the designated high-incidence area of ​​partial discharge, multiple sets of random initial distribution schemes for ultraviolet detectors are generated under the constraints of coverage and allowable overlap. The performance indicators of each distribution scheme for simulated discharge signals are calculated and compared. The performance indicators include at least coverage completeness and average positioning error. A multi-stage screening method is used to select the optimal distribution scheme for ultraviolet detectors. The scheme includes installation density and coordinates of each deployment point.

[0029] It should be noted that, in order to avoid missed detections, waste of ultraviolet detector resources, and signal redundancy interference, two hard constraints are set: coverage and permissible overlap. The coverage requirement is that the solution must cover more than 90% of the potential discharge points in the high-incidence area, and the permissible overlap requirement is that the overlapping area of ​​the monitoring range of different ultraviolet detectors shall not exceed 30%. Implementers can dynamically adjust the coverage and permissible overlap according to equipment operation and maintenance and sensor performance.

[0030] Based on the above division and constraints, multiple sets of random initial deployment schemes are generated by computer algorithms. The specific number can be adjusted according to computing resources and accuracy requirements. Each scheme must specify: the number of ultraviolet detectors to be installed, the monitoring angle and effective monitoring radius of each ultraviolet detector, and the specific deployment point coordinates of each detector in the high-incidence area. The deployment point coordinates are derived based on the spatial coordinate system of the three-dimensional model.

[0031] For each initial scheme, a simulation test is performed. A preset simulated discharge signal is input. The simulated discharge signal is set based on the ultraviolet photon characteristics of common discharge types in switchgear. Two sets of core performance indicators are calculated. The calculation of coverage completeness rate is specifically: the coverage ratio of the monitoring range of all ultraviolet detectors in the scheme to the simulated discharge signal points. The coverage ratio is the ratio of the number of covered simulated discharge points to the total number of simulated discharge points.

[0032] The calculation of the average positioning error is specifically as follows: based on the actual coordinates of the simulated discharge signal and the positioning coordinates calculated by the detector in the scheme using the time difference positioning method, the average positioning error of each scheme is calculated. It should also be noted that other performance indicators, such as signal-to-noise ratio and response delay, can be added as needed to further optimize the scheme evaluation dimensions.

[0033] Based on the above description, we first eliminate the initial schemes that do not meet the above conditions in terms of coverage and allowable overlap. Then, we sort the schemes after the initial screening according to the priority of coverage completeness from high to low and average positioning error from low to high.

[0034] If there is a solution with the highest coverage completion rate and the smallest average positioning error, then it should be selected as the optimal solution.

[0035] If multiple solutions have similar performance, then the number of sensors and signal redundancy of the solutions should be further compared, and the solution with fewer sensors and lower redundancy should be selected first to reduce costs and interference.

[0036] Finally, an optimal distribution scheme for ultraviolet detectors, including installation density and coordinates of each deployment point, was determined.

[0037] This invention accurately defines high-incidence areas of partial discharge by combining electrostatic field simulation of a 3D model of a switchgear with spatial cluster analysis of historical fault data. It also selects the optimal deployment scheme for ultraviolet detectors by combining performance indicators such as coverage completeness and average positioning error, thus solving the problem of missed detections and resource waste caused by the reliance on experience-based deployment in traditional methods.

[0038] Signal acquisition and processing module: Based on the ultraviolet photon signals captured in real time by the ultraviolet detector in the switch cabinet, the ultraviolet photon signals are processed through environmental interference suppression and signal obstruction compensation.

[0039] In a preferred embodiment of the present invention, the environmental interference suppression specifically involves: extracting the pulse characteristics and spectral characteristics of the ultraviolet photon signal as signal characteristics through joint time-frequency domain analysis.

[0040] The extracted signal features are matched and compared with a preset ambient light source ultraviolet interference feature library.

[0041] If the similarity between the signal features and the light source features in the ambient light source ultraviolet interference feature library exceeds a preset permissible threshold, then the interference components are identified and filtered out in a targeted manner.

[0042] In this embodiment, the preset ambient light source ultraviolet interference feature library covers typical natural and artificial light sources in the 10kV switchgear operating environment, such as midday sunlight, LED lighting, and arc radiation from adjacent equipment. Based on the above light sources, the following measures are implemented: First, in an environment without partial discharge, various light source signals are collected using the same type of ultraviolet detector, covering different working conditions such as distance, angle, and intensity. Then, the sampled signals are subjected to joint time-frequency domain analysis to extract pulse, spectrum, and photon count features. Finally, the extracted features are classified and stored according to the structure of light source type—scene parameters—feature set to form an ambient light source ultraviolet interference feature library.

[0043] It should be noted that the similarity calculation logic is as follows: the features of the signal to be identified and the features of the interfering light sources in the ambient light source ultraviolet interference feature library are both converted into feature vectors with the same dimension, and the similarity is calculated using the cosine similarity method. It should be noted that the cosine similarity method is an existing technical means, so it will not be described in detail here.

[0044] It should also be noted that the preset permission threshold is obtained through experimental calibration. Specifically, in a laboratory environment, a mixed test dataset containing standard partial discharge signals and environmental interference signals is constructed. By calculating the similarity between each signal and the interference feature library, the maximum similarity value of the effective discharge signal and the minimum similarity value of the interference signal are determined. The initial threshold is set between the above two values. Then, the initial threshold is put into actual operation environment for verification, and the threshold is iteratively adjusted based on the field monitoring data: if a discharge signal is misjudged as interference, the threshold is increased; if an interference signal is misjudged as discharge, the threshold is decreased. After continuous optimization, until no misjudgment occurs in the field operation for a continuous period of time, the threshold is the final preset permission threshold.

[0045] Reference Figure 3 As shown, in a preferred embodiment of the present invention, the occlusion signal compensation specifically involves: acquiring the response signal intensity of each ultraviolet detector and arranging them in descending order, selecting the one with the highest signal intensity as the reference ultraviolet detector, and assuming that it is in an unoccupied state.

[0046] Based on the coordinates of each ultraviolet detector deployment point, the fixed spatial distance between each ultraviolet detector is obtained, and the theoretical signal strength ratio between each ultraviolet detector and the reference ultraviolet detector is calculated according to the inverse square law.

[0047] Calculate the ratio of the actual signal strength of the detector being evaluated to that of the reference detector.

[0048] Understandably, based on the coordinates of each ultraviolet detector deployment point, the fixed spatial distance between any two ultraviolet detectors can be directly calculated using the spatial distance formula. Then, the target ultraviolet signal source is regarded as a point radiation source. According to the propagation characteristics of radiation energy in physics, the radiant energy that can be intercepted per unit area is inversely proportional to the square of the propagation distance, as shown by the spherical area formula. Therefore, the theoretical signal strength ratio is equal to the square of the distance between the detector being evaluated and the signal source, divided by the square of the distance between the reference detector and the signal source.

[0049] If the actual signal strength ratio is consistently and significantly lower than the theoretical signal strength ratio, the detector being evaluated is determined to be blocked.

[0050] If the actual ratios of all evaluated detectors match the theoretical ratios, then the reference detector is not obstructed.

[0051] If the actual signal strength ratio of all evaluated detectors is consistently and significantly higher than the theoretical signal strength ratio, then the inverse reference detector is blocked.

[0052] For detectors that are determined to be blocked, the response signal strength is increased based on the degree to which their inference fails to meet the standard.

[0053] It should be noted that the degree of non-compliance refers to the difference between the theoretical signal strength and the actual signal strength. The gain processing uses this difference as a compensation amount and restores it to the theoretical level by superimposing the compensation amount. The compensation amount is the difference between the theoretical signal strength and the actual signal strength.

[0054] It should also be noted that the gain processing can also be achieved by dividing the compensation amount by the actual signal strength to obtain a compensation coefficient and then performing a multiplication operation.

[0055] Discharge feature identification module: Based on the phase characteristics, amplitude characteristics and discharge frequency of the processed ultraviolet photon signal, it identifies the type of partial discharge, which includes surface discharge, corona discharge or internal discharge, and locates the discharge position based on the time difference of multi-source signals.

[0056] In a preferred embodiment of the present invention, the identification of partial discharge type specifically means: if the discharge frequency is continuously and stably in the high range, it is preferentially determined to be corona discharge.

[0057] Conversely, the phase and amplitude characteristics of the processed ultraviolet photon signal are evaluated to see if they are dispersed. If both phase and amplitude characteristics are dispersed, it is preferentially determined to be internal discharge.

[0058] If neither of the above two conditions is met, then surface discharge is the preferred diagnosis.

[0059] The above logic for identifying partial discharge types is based on the consideration that partial discharges differ in location, discharge channel morphology, and energy release patterns. Therefore, rapid discharge type identification can be achieved by analyzing the significant and stable differences in discharge frequency, ultraviolet photon signal phase characteristics, and amplitude characteristics detected by sensors.

[0060] Specifically, because the corona discharge occurs in the gas surrounding the conductor, the discharge process is stable and has a high repetition rate, so its discharge frequency is continuously and stably in the high range.

[0061] Because the internal discharge occurs due to defects inside the insulation, its pulses have no fixed concentrated phase within the power frequency cycle and exhibit a uniformly dispersed scattering characteristic. At the same time, since the discharge channel develops randomly inside the insulation, the energy of each discharge varies greatly, resulting in a significant dispersion in its amplitude distribution. Therefore, dual dispersion of phase and amplitude is its typical characteristic.

[0062] The surface discharge characteristics are between corona discharge and internal discharge. They usually occur along the insulating surface and appear in clusters within a fixed half-wave of the power frequency voltage. Therefore, their phase distribution and amplitude distribution are between corona discharge and internal discharge.

[0063] It should also be noted that the discharge frequency refers to the number of ultraviolet photon signal pulses occurring per unit time.

[0064] In a preferred embodiment of the present invention, the evaluation process for the dispersion of the phase characteristics and amplitude characteristics is as follows: calculate the statistical entropy value of the discharge pulse distribution on the power frequency phase; if the statistical entropy value approaches the theoretical maximum value under uniform distribution and spans multiple phase regions, it is determined that the phase distribution is a scattered distribution.

[0065] Calculate the coefficient of variation of the amplitude of all discharge pulses within a fixed time window. If the coefficient of variation is significantly greater than 1, the pulse amplitude distribution is determined to be dispersed.

[0066] In one embodiment of the present invention, the evaluation process for the dispersion of the phase characteristics and amplitude characteristics is as follows: calculate the statistical entropy value of the discharge pulse distribution on the power frequency phase, and when the statistical entropy value approaches the theoretical maximum value and the discharge crosses multiple phase regions, it is determined that the phase distribution is dispersed.

[0067] The coefficient of variation refers to the ratio of the standard deviation to the mean of all partial discharge pulse amplitude data within a fixed time window. It is used to quantify the dispersion of pulse amplitude data. When the coefficient of variation is significantly greater than 1, it indicates that the deviation of the discharge pulse amplitude exceeds the average level, the amplitude difference between different pulses is extremely large, the dispersion is significant, and it conforms to the characteristics of a dispersed amplitude distribution.

[0068] This invention effectively distinguishes discharge types by combining multiple features, determines the occlusion state by comparing the ratio of theoretical to actual signal strength, and compensates for signal strength gain of the occluded detector, effectively making up for signal attenuation and ensuring fast and accurate signal acquisition.

[0069] In a preferred embodiment of the present invention, the specific method for locating the discharge position based on the time difference of multi-source signals is as follows: based on each ultraviolet detector synchronized with the same time, the arrival time of the ultraviolet pulse signal of the same partial discharge event is recorded.

[0070] Using the ultraviolet detector where the signal arrives first as the time reference point, the time difference between the other ultraviolet detectors and this reference point is calculated, and a time difference sequence is constructed.

[0071] Based on the known installation coordinates of each ultraviolet detector and the time difference sequence, and according to the principle of signal propagation at the speed of light, the spatial distance difference between the partial discharge power source and each ultraviolet detector is derived from the signal arrival time difference. Through spatial iterative search, a spatial point is obtained such that the distance difference between it and each ultraviolet detector best matches the measured distance difference. The coordinates of this point are then determined as the location of the partial discharge power source.

[0072] It should be noted that the spatial iterative search specifically involves: First, based on the ultraviolet signal propagating at a preset speed of light, the theoretical distance difference between the power supply and each detector can be derived according to the signal arrival time difference. The theoretical distance difference is the product of the speed of light and the time difference.

[0073] Next, using the coordinates of each detector as the center of the sphere and the theoretical distance difference as a reference, a spatial region covering all possible solutions is determined as the initial search range. This range can be set according to the actual physical size of the switch cabinet to balance search efficiency and completeness.

[0074] Finally, in each iteration, several candidate points are selected within the current spatial range, and the calculated distance difference between each candidate point and all detectors is calculated. The candidate point with the smallest error is used as the center to narrow the search range. The above calculation, comparison, and screening steps are repeated until the error is less than the preset threshold. The candidate point at this time is the optimal matching point, and the coordinates of this point are determined as the location of the local discharge power source.

[0075] It should also be noted that the preset threshold is set according to the operation and maintenance requirements. For example, if it is necessary to locate a specific insulator or terminal, the preset threshold is set to the centimeter level. If it is only necessary to locate the cabinet area, the preset threshold can be set to the decimeter level. The narrowing of the search range can be reduced according to the actual situation by a certain ratio. For example, the initial ratio is 1 / 10 of the search range, and the ratio is reduced to 1 / 100 as the range is narrowed.

[0076] In a preferred embodiment of the present invention, the mapping relationship between the ultraviolet photon intensity and the discharge energy is specifically as follows: the processed ultraviolet photon signal intensity and the distance parameters between the discharge source and the ultraviolet detector are obtained, and based on the transmission attenuation characteristics of ultraviolet light in the atmosphere, the signal intensity actually received by the ultraviolet detector is uniformly corrected to the ultraviolet radiation intensity at a standard observation distance.

[0077] Understandably, based on the transmission attenuation characteristics of ultraviolet light in the atmosphere, the corresponding atmospheric transmission attenuation formula, such as the Beer-Lambert law, can be used to calculate the relationship between the actual received signal strength of the detector and the theoretical radiation intensity of the discharge source at the distance from the detector.

[0078] Based on a unified standard observation distance set for the monitoring scenario, and combined with the above attenuation formula, we deduce in reverse: if the power source is at the standard observation distance, what should its radiation intensity be? Then, we correct the actual received signal intensity of each detector according to this derivation result, so that the signal intensity of all detectors corresponds to the ultraviolet radiation intensity at the standard observation distance, thus eliminating the signal deviation caused by distance differences.

[0079] Based on the physical relationship between the number of ultraviolet photons and the discharge energy, and combined with the discharge type identification results, a quantitative mapping function is constructed with the normalized ultraviolet radiation intensity as input and the local discharge quantity as output.

[0080] In a preferred embodiment of the present invention, the quantitative mapping function is specifically a functional relationship established by fitting laboratory calibration and field data, with normalized ultraviolet radiation intensity as the independent variable and apparent discharge quantity as the dependent variable. The form and parameters of the functional relationship are adaptively selected according to the identified discharge type. Specifically, when corona discharge is identified, a linear function is used.

[0081] When surface discharge is identified, a power function is used.

[0082] When internal discharge is identified, a non-analytical mapping relationship based on piecewise linear or lookup table methods is adopted.

[0083] The above-mentioned quantitative mapping function determination logic takes into account the different physical correlation characteristics between ultraviolet radiation intensity and apparent discharge quantity for different discharge types. Corona discharge often occurs in strong electric field regions such as conductor tips and burrs. The discharge process is a continuous and stable process of localized weak ionization of air, and its ultraviolet radiation intensity has a linear correlation with the apparent discharge quantity. Therefore, the normalized ultraviolet radiation intensity increases proportionally with the increase of apparent discharge quantity, and the data points of the two can be accurately fitted by a linear equation, which can be specifically: .

[0084] in, Apparent discharge quantity characterizes the intensity and scale of partial discharge. To normalize the ultraviolet radiation intensity, reflecting the ultraviolet photon flux generated by the discharge, This is a proportionality coefficient, representing the increase in discharge quantity corresponding to a unit of radiation intensity. This is a constant term that reflects background noise or the discharge initiation threshold.

[0085] In the above calculation formula, The main part constituting the mapping relationship reflects the proportionality between radiation intensity and discharge quantity, and is a constant. The baseline for the relationship was determined, and the normalized ultraviolet radiation intensity was used. Apparent discharge quantity as the independent variable As dependent variables, the two are quantitatively correlated through a linear equation, when radiation intensity When increased, apparent discharge amount It grows linearly, with a proportionality coefficient It determines the slope of growth.

[0086] It should be noted that the aforementioned proportionality coefficient and constant terms It was determined in a controlled laboratory environment using the standard pulse current method and linear regression analysis.

[0087] It is known that surface discharge occurs on the surface of insulating materials. The discharge process is affected by uneven surface electric field distribution and fluctuations in insulation properties caused by dirt / humidity. The relationship between ultraviolet radiation intensity and apparent discharge quantity is non-linear and power-law related. Therefore, the normalized ultraviolet radiation intensity with increasing apparent discharge quantity can be represented by a power function, which can be specifically: .

[0088] in, and Consistent with the above explanation, This is a proportionality coefficient, related to the initial conditions of the discharge and the characteristics of the surface insulating material, and determines the benchmark scale of the functional relationship. The power exponent is a key morphological parameter, and its value is usually greater than 1. It determines the curvature of the nonlinear growth and directly reflects the nonlinear strength of the surface discharge.

[0089] In the above calculation formula, radiation intensity The growth rate will be faster than the apparent discharge. linear growth, power exponent It is the mathematical marker that distinguishes surface discharge from linear corona discharge. Its value directly reflects the severity of the nonlinear enhancement effect of surface discharge, and the proportionality coefficient. This is related to the specific insulation material, contaminant composition, and environmental conditions, and sets the baseline position for the entire relationship curve.

[0090] The proportionality coefficient With power exponent The calibration and the above proportional coefficient and constant terms The calibration is consistent.

[0091] Internal discharge occurs at defects such as air gaps and impurities within the insulating material. The discharge process is affected by the randomness of defect distribution and the uncertainty of the internal breakdown path of the insulation. Therefore, there is no unified, continuous, analytical law governing the relationship between ultraviolet radiation intensity and apparent discharge quantity. The discharge energy release varies greatly among different defects, and the same apparent discharge quantity may correspond to different ultraviolet radiation intensities. According to experimental data fitting analysis, the two only exhibit a locally linear relationship within a specific discharge quantity range; no continuous function covers the entire range. Therefore, it is necessary to establish a non-analytical mapping relationship through piecewise linearity or table lookup methods to ensure the quantitative accuracy of each range. The piecewise linearity method obtains the locally linear relationship within different discharge quantity ranges through calibration, while the table lookup method establishes the radiation intensity through calibration. and apparent discharge quantity The lookup table for data pairs allows for the calculation of apparent discharge using interpolation during actual measurements.

[0092] In a preferred embodiment of the present invention, the single discharge energy of partial discharge is calculated in real time based on the mapping function, the cumulative discharge energy is calculated in combination with the discharge duration, and the maximum value of single discharge energy is retrieved simultaneously.

[0093] A dynamic evaluation benchmark is established for the maximum single discharge energy and cumulative discharge energy, which is obtained through statistical analysis of historical monitoring data.

[0094] The specific steps for statistical analysis of historical monitoring data are as follows: S1, collect historical partial discharge monitoring data of the target equipment during normal operation cycle, which needs to cover complete operating conditions such as different loads, ambient temperature and humidity, while filtering abnormal interference data such as false discharge signals caused by equipment operation and electromagnetic noise, to ensure that the data only contains real and valid discharge energy records.

[0095] S2, statistically analyze the instantaneous energy values ​​of all single discharge events in the historical data to form a single energy dataset, and statistically analyze the cumulative discharge energy within a unit period in the historical data according to a fixed time period to form a periodic cumulative energy dataset.

[0096] S3. Calculate the statistical characteristic values ​​of the two datasets mentioned above, such as the average value. Based on the normal distribution or the device-specific discharge energy distribution law, determine the normal operation range and use the upper limit of the corresponding range as the dynamic evaluation benchmark.

[0097] It should be noted that in other embodiments, the implementer may also use standard deviation or other statistical characteristic values ​​instead of statistical characteristic values, and the normal operating range is an interval that covers more than 95% of normal operating condition data.

[0098] S4. Regularly incorporate new historical monitoring data, repeat the above screening and statistical steps, update the statistical characteristic values ​​and benchmark range, and ensure that the benchmark can adapt to the changes in discharge energy characteristics caused by factors such as equipment aging and changes in operating conditions, so as to avoid evaluation failure due to benchmark solidification.

[0099] Based on the maximum single discharge energy and cumulative discharge energy obtained from real-time monitoring, their ratios to the corresponding dynamic evaluation benchmarks are calculated.

[0100] Based on the different intervals in which the ratio of the real-time monitoring value to the dynamic benchmark falls, the severity levels are progressively divided into normal level, attention level, warning level, and danger level. The division of these intervals is set according to established operating procedures and grading standards.

[0101] It should be noted that the logic for classifying the severity levels is as follows: when the ratio of the maximum single discharge energy and the cumulative discharge energy to the corresponding dynamic assessment benchmark both fall within the normal range, it means that the real-time discharge energy has not exceeded the normal upper limit of the dynamic benchmark, and is judged as normal. If the above ratio exceeds the normal upper limit but does not exceed twice it, it means that the real-time discharge energy has begun to deviate from the benchmark, and is judged as alert. When the above ratio further increases to 2 to 3 times the normal range, it means that the real-time discharge energy significantly exceeds the normal benchmark, and is judged as warning. When the above ratio exceeds 3 times the normal range, it means that the real-time discharge energy far exceeds the normal benchmark, and is judged as dangerous.

[0102] It should also be noted that the above range values ​​are only examples. The actual ranges should be set strictly according to the specific equipment's operating procedures, industry classification standards, and equipment operation and maintenance experience to ensure the rationality and safety of the classification.

[0103] This invention constructs a quantitative mapping function of normalized ultraviolet radiation intensity and apparent discharge quantity based on discharge type, calculates the energy of a single discharge, combines it with the cumulative discharge energy, and establishes a dynamic evaluation benchmark based on historical monitoring data. The severity of the discharge is evaluated by the ratio of the real-time monitoring value to the benchmark, thereby achieving accurate assessment and efficient early warning of partial discharge risk in switchgear.

[0104] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0105] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0106] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0107] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0108] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0109] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A remote detection system for partial discharge of 10kV switchgear based on sensor technology, characterized in that, include: Detection point deployment module: Based on the structural simulation analysis of the switchgear and the statistical results of historical faults, the deployment points and installation density of gallium nitride-based ultraviolet detectors are planned in areas with high incidence of partial discharge; Signal acquisition and processing module: Based on the ultraviolet photon signals captured in real time by the ultraviolet detector in the switch cabinet, the ultraviolet photon signals are processed through environmental interference suppression and signal obstruction compensation; Discharge feature identification module: Based on the phase characteristics, amplitude characteristics and discharge frequency of the processed ultraviolet photon signal, it identifies the type of partial discharge, which includes surface discharge, corona discharge or internal discharge, and locates the discharge position based on the time difference of multi-source signals; Detection and early warning module: Establishes a mapping relationship between ultraviolet photon intensity and discharge energy, determines the severity level of the discharge location based on the discharge duration and the mapped discharge energy, and generates early warning information for feedback; The specific locations and installation density are as follows: Electrostatic field simulation analysis based on a 3D model of the switchgear was used to determine the region with high electric field intensity. Based on historical fault statistics, spatial clustering analysis is performed on fault points in a 3D model to identify historical fault hotspot areas. The high electric field intensity region and the historical fault hotspot region are spatially superimposed, and their union is taken to define the high-incidence region of partial discharge. Within the designated high-incidence area of ​​partial discharge, multiple sets of random initial distribution schemes for ultraviolet detectors are generated under the constraints of coverage and allowable overlap. The performance indicators of each distribution scheme for simulated discharge signals are calculated and compared. The performance indicators include at least coverage completeness and average positioning error. A multi-stage screening method is used to select the optimal distribution scheme for ultraviolet detectors. The scheme includes installation density and coordinates of each deployment point. The environmental interference suppression specifically refers to: By combining time-frequency domain analysis, the pulse characteristics and spectral characteristics of the ultraviolet photon signal are extracted as signal features; The extracted signal features are matched and compared with a preset ambient light source ultraviolet interference feature library; If the similarity between the signal features and the light source features in the ambient light source ultraviolet interference feature library exceeds a preset permissible threshold, then the interference components are identified and filtered out in a targeted manner. The occlusion signal compensation specifically includes: The response signal intensity of each ultraviolet detector is obtained and sorted in descending order. The detector with the highest signal intensity is selected as the reference ultraviolet detector, and it is assumed to be in an unobstructed state. Based on the coordinates of each ultraviolet detector deployment point, the fixed spatial distance between each ultraviolet detector is obtained, and the theoretical signal strength ratio between each ultraviolet detector and the reference ultraviolet detector is calculated according to the inverse square law. Calculate the ratio of the actual signal strength of the detector being evaluated to that of the reference detector; If the actual signal strength ratio is consistently and significantly lower than the theoretical signal strength ratio, the detector being evaluated is determined to be blocked. If the actual ratios of all evaluated detectors match the theoretical ratios, then the reference detector is not obstructed. If the actual signal strength ratio of all evaluated detectors is consistently and significantly higher than the theoretical signal strength ratio, then the reference detector is likely blocked. For detectors that are determined to be blocked, the response signal strength is increased based on the degree to which their inference fails to meet the standard.

2. The 10kV switchgear partial discharge remote detection system based on sensor technology according to claim 1, characterized in that: The specific method for identifying the type of partial discharge is as follows: If the discharge frequency remains consistently high, it is preliminarily identified as corona discharge. Conversely, the phase and amplitude characteristics of the processed ultraviolet photon signal are evaluated to determine whether they exhibit dispersion. If both phase and amplitude characteristics show dispersion, it is preferentially determined to be internal discharge. If neither of the above two conditions is met, then surface discharge is the preferred diagnosis.

3. The 10kV switchgear partial discharge remote detection system based on sensor technology according to claim 2, characterized in that: The evaluation process for the dispersion of the phase and amplitude characteristics is as follows: Calculate the statistical entropy value of the discharge pulse distribution on the power frequency phase. If the statistical entropy value approaches the theoretical maximum value under uniform distribution and spans multiple phase regions, it is determined that the phase distribution is a scattered distribution. Calculate the coefficient of variation of the amplitude of all discharge pulses within a fixed time window. If the coefficient of variation is significantly greater than 1, the pulse amplitude distribution is determined to be dispersed.

4. The 10kV switchgear partial discharge remote detection system based on sensor technology according to claim 1, characterized in that: The specific method for locating the discharge position based on the time difference of multi-source signals is as follows: Based on the unified time synchronization of each ultraviolet detector, the arrival time of the ultraviolet pulse signal of the same partial discharge event is recorded. Using the ultraviolet detector where the signal arrives first as the time reference point, calculate the signal arrival time difference between the other ultraviolet detectors and this reference point, and construct a time difference sequence; Based on the known installation coordinates of each ultraviolet detector and the time difference sequence, and according to the principle of signal propagation at the speed of light, the spatial distance difference between the partial discharge power source and each ultraviolet detector is derived from the signal arrival time difference. Through spatial iterative search, a spatial point is obtained such that the distance difference between it and each ultraviolet detector best matches the measured distance difference. The coordinates of this point are then determined as the location of the partial discharge power source.

5. The 10kV switchgear partial discharge remote detection system based on sensor technology according to claim 1, characterized in that: The mapping relationship between ultraviolet photon intensity and discharge energy is as follows: The processed ultraviolet photon signal intensity and the distance parameters between the discharge source and the ultraviolet detector are obtained. Based on the transmission attenuation characteristics of ultraviolet light in the atmosphere, the signal intensity actually received by the ultraviolet detector is uniformly corrected to the ultraviolet radiation intensity at a standard observation distance. Based on the physical relationship between the number of ultraviolet photons and the discharge energy, and combined with the discharge type identification results, a quantitative mapping function is constructed with the normalized ultraviolet radiation intensity as input and the local discharge quantity as output.

6. The 10kV switchgear partial discharge remote detection system based on sensor technology according to claim 5, characterized in that: The quantitative mapping function is specifically: A functional relationship was established by fitting laboratory calibration data and field data, with normalized ultraviolet radiation intensity as the independent variable and apparent discharge quantity as the dependent variable. The form and parameters of this functional relationship are adaptively selected based on the identified discharge type. Specifically: When corona discharge is identified, a linear function is used. When surface discharge is identified, a power function is used. When internal discharge is identified, a non-analytical mapping relationship based on piecewise linear or lookup table methods is adopted.

7. The 10kV switchgear partial discharge remote detection system based on sensor technology according to claim 5, characterized in that: The method for classifying the severity levels is as follows: The single discharge energy of partial discharge is calculated in real time based on the mapping function, and the cumulative discharge energy is calculated in combination with the discharge duration. The maximum value of single discharge energy is retrieved simultaneously. A dynamic evaluation benchmark is established for the maximum single discharge energy and cumulative discharge energy, which is obtained through statistical analysis of historical monitoring data; Based on the maximum single discharge energy and cumulative discharge energy obtained from real-time monitoring, their ratios to the corresponding dynamic evaluation benchmarks are calculated respectively. Based on the different intervals in which the ratio of the real-time monitoring value to the dynamic benchmark falls, the severity levels are progressively divided into normal level, attention level, warning level, and danger level. The division of these intervals is set according to established operating procedures and grading standards.

Images