Distribution network line fault detection method and system based on data analysis

Abstract

The present invention relates to the field of distribution network detection technology, and specifically to a distribution network line fault detection method and system based on data analysis. A superposition model and a time-varying capacitance model are established based on the water film capacitance effect and wind vibration interference to determine the discharge peak amplitude and exponential decay constant. The degree of distortion of the pure discharge signal caused by the water film coupling and the degree of interference of the conductor-ground coupling on the pure discharge signal are measured based on the water film tailing coefficient and the wind vibration suppression coefficient to determine the time domain characteristics; a sparse reconstruction algorithm is used to separate and calculate the discharge source; thermal stress crack noise caused by ambient temperature changes and discharge sound confusion are combined to obtain thermal stress suppression acoustic energy; a pulse sliding window is dynamically adjusted based on the influence of water film thickness changes, and the pulse occurrence rate and phase window ratio are determined; a feature vector is established through the time domain characteristics, discharge source, thermal stress suppression acoustic energy, pulse occurrence rate and phase window ratio to realize the detection of cable arc faults or surface partial discharges.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network detection, and in particular to a distribution network line fault detection method and system based on data analysis. Background Art

[0002] In the real-world operating environment of distribution networks, two complex and overlapping climatic and environmental factors are often encountered. On the one hand, strong winds (such as those frequently seen in Gansu) can cause conductors to bend and sway, generating mechanical vibrations and sudden changes in conductor-to-ground capacitance. On the other hand, occasional rainfall, frost, or rapid changes in relative temperature at night, especially seasonal rainfall in the same region, can cause a thin layer of water film to form on the surfaces of distribution line insulation components (such as suspended porcelain insulators and constantan connectors) over layers of contaminants or salt spray deposits. These two environmental factors can cause different types of interference to the discharge signals measured by high-frequency current transformers (HFCTs), such as:

[0003] 1) Currently, most distribution network line fault detection systems use HFCT sensors or UHF antennas to monitor sudden discharges. Both arc faults and surface partial discharges (PDs) generate high-frequency transient pulses. Traditional methods often focus only on the signal's time-domain rise / decay characteristics or frequency-domain energy distribution, but fail to fully consider the combined effects of contaminants or salt spray deposits. In reality, when a water film forms on the contaminant layer, partial discharges or line arcs on insulator surfaces or constantan joints absorb a portion of the discharge energy, continuously changing the capacitance value as it evaporates.

[0004] 2) The salt content and pollutant ions in the water film are highly conductive, causing them to conduct electricity briefly near the voltage peak, then quickly evaporate, causing the capacitance to recover. As a result, the current waveform monitored by HFCT will be superimposed with a slow tail wave of capacitive coupling in addition to the pure exponential decay, usually manifesting as a double peak or tail in the signal.

[0005] 3) The time-domain distortion of water film tailing can cause inaccurate algorithms that rely on peak amplitude and exponential decay constants to distinguish weak arcs from PDs. For example, when the water film tailing component exceeds a certain amplitude, a rapidly decaying arc waveform may be misinterpreted as multiple high-speed PDs, or the PD signal's decay curve may be fitted with an abnormally large decay constant value, resulting in false alarms or missed alarms.

[0006] Along with water film interference, when wind speeds exceed a threshold, the elasticity and tension of the conductors cause them to oscillate and vibrate at high frequencies. This oscillation generates mechanical noise in the insulators (which is picked up by ultrasonic and acoustic channels). More critically, the instantaneous deviation of the conductor relative to the ground or nearby metal objects can cause a sudden change in the conductor-to-ground capacitance.

[0007] It can be seen that it is currently difficult to combine the two types of complex interference, namely water film capacitance tailing and ground capacitance coupling interference in strong wind environments, to achieve fault identification and detection of distribution network. Summary of the Invention

[0008] In response to the above-mentioned shortcomings of the existing technology, the present invention provides a distribution network line fault detection method and system based on data analysis, which can effectively solve the problem of the existing technology not considering the influence of two types of complex interference, namely water film capacitance tailing in air environment and conductor-to-ground capacitance coupling interference, on fault judgment of distribution network.

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0010] The present invention provides a distribution network line fault detection method and system based on data analysis, which at least includes:

[0011] Based on the water film capacitance effect and wind vibration interference, a superposition model and a time-varying capacitance model are established to determine the discharge peak amplitude and exponential decay constant. The degree of distortion of the pure discharge signal caused by water film coupling and the degree of interference of the conductor-ground coupling on the pure discharge signal are measured based on the water film tailing coefficient and the wind vibration suppression coefficient, respectively, to establish the time domain characteristics.

[0012] The non-uniform state of the contamination layer is introduced, and the sparse reconstruction algorithm is used to separate and calculate the discharge source;

[0013] Combining the confusion between thermal stress crack noise and discharge sound caused by ambient temperature changes, the ambient temperature is used to suppress thermal stress interference and obtain thermal stress suppression acoustic energy;

[0014] Dynamically adjust the pulse sliding window according to the influence of water film thickness changes, and determine the pulse occurrence rate and phase window ratio;

[0015] A feature vector is established through the time domain characteristics, discharge source number, thermal stress suppression acoustic energy, pulse occurrence rate and phase window ratio, and the detection and judgment of arc fault or surface partial discharge of distribution network cable is realized based on the logistic regression model.

[0016] By simultaneously fitting the exponential decay discharge, water film capacitance tail, and wind-induced vibration coupling spike within a single pulse time domain window, the pure discharge signal characteristics can be accurately extracted in complex environments without causing misjudgment due to the failure of a single interference model.

[0017] In an environment with both high salt fog and strong winds, if there is a rainy day and the water film thickness is the largest just after it, partial discharge is often more likely to occur. Traditionally, the superposition effect is often ignored. However, the present invention determines whether the water film tailing is dominant by using the water film tailing coefficient, and then removes this component. The remaining waveform is then removed from the wind vibration coupling determined by the wind vibration suppression coefficient. Ultimately, the waveform that remains is the closest to pure arc discharge or pure PD, making subsequent multi-source OMP positioning and feature fusion judgment more accurate.

[0018] In terms of judgment logic, the water film trailing coefficient is combined with the wind vibration suppression coefficient to avoid wasting the calculation amount of subsequent wind vibration fitting, saving online calculation amount and improving detection efficiency;

[0019] The introduction of the water film trailing coefficient and the wind vibration suppression coefficient allows us to directly skip the complete joint fitting when the interference of most pulse signals is relatively light, and only extract the original time domain exponential attenuation term. Only when the interference is judged to be significant, is the nonlinear least squares fitting performed after removing the wind vibration or water film. This avoids the situation where all pulses have to be deeply optimized using time-consuming double exponential or three-term superposition fitting, and in practice can shorten the average fitting time for each pulse.

[0020] The method to build the superposition model is:

[0021] Define the HFCT trigger threshold , if present:

[0022] When the current signal amplitude If the interval between the last trigger and the last trigger is greater than or equal to the minimum sampling point interval, a pulse is recorded and the trigger sampling number is ;

[0023] Extract the pulse time domain window:

[0024]

[0025] 、 Respectively represent the number of sampling points on both sides of the window width, Indicates the The intercepted signal within the time domain window where the trigger event is located, represents the pulse current time domain sequence of the original HFCT sampling;

[0026] Build a superposition model:

[0027]

[0028] Indicates the Sub-pulse pure discharge peak amplitude, represents the exponential decay constant, represents the transient reference voltage on the insulator surface, Indicates the first The sampling points correspond to time, Indicates the The instantaneous equivalent capacitance corresponding to the sampling point time is: Indicates the The time corresponding to the sampling point The instantaneous capacitance between the lower conductor and ground, Indicates the The line voltage value at each sampling point.

[0029] Furthermore, based on the water film capacitance effect and wind vibration interference, the water film trailing coefficient and wind-induced vibration suppression coefficient Determine surface partial discharge or line arc fault.

[0030] Furthermore, the method for obtaining the thermal stress suppression acoustic energy is:

[0031] Temperature gradient calculate:

[0032]

[0033] Indicates the time interval for environmental quantity sampling, Indicates the The ambient temperature corresponding to the moment of secondary pulse triggering;

[0034] Calculate the Ultrasonic energy during the sub-pulse :

[0035]

[0036] Indicates the The acoustic signal value at each sampling point;

[0037] Defining the thermal stress acoustic suppression coefficient

[0038] represents the reference temperature, represents the temperature sensitivity coefficient;

[0039] If satisfied 、 If the values ​​are greater than the corresponding thresholds, it is determined that the ultrasonic energy comes from the thermal stress crack noise, and the thermal stress suppression acoustic energy is obtained. .

[0040] Furthermore, the pulse rate is determined as follows:

[0041] Count the number of triggers in a time period

[0042] Indicates the index of the triggering event, Indicates the The discrete sampling point number corresponding to the pulse trigger, Indicates the high-frequency current transformer sampling rate, Indicates the The length of the dynamic sliding window corresponding to the trigger, Counting operations that represent the number of elements in a collection;

[0043] The pulse occurrence rate is calculated based on the ratio of the number of statistical triggers to the length of the dynamic sliding window.

[0044] Furthermore, the method for detecting cable arc fault or surface partial discharge using the logistic regression model is as follows:

[0045] Construct a logistic regression model and input the feature vector to output the arc discharge probability:

[0046] Construct cross entropy loss function;

[0047] Calculate the arc discharge probability based on the characteristic vector obtained each time;

[0048] If it is greater than the probability threshold, it is determined to be a cable arc fault, otherwise it is a surface partial discharge.

[0049] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0050] By simultaneously fitting exponential decay discharge, water film capacitance tail and wind-vibration coupling spikes in a single pulse time domain window, the pure discharge signal characteristics can be accurately extracted in complex environments without misjudgment due to the failure of a single interference model. The water film tail coefficient is used to determine whether the water film tail is dominant, and then this component is eliminated. The remaining waveform is then stripped of the wind-vibration coupling determined by the wind-vibration suppression coefficient. What remains is the waveform closest to pure arc discharge or pure PD, making subsequent multi-source OMP positioning and feature fusion judgment more accurate.

[0051] Based on the water film capacitance tailing and wind vibration coupling model, the clean pure discharge peak and attenuation characteristics are extracted. The HFCT signal after interference removal is used to perform multi-source discharge positioning on a sparse array. The filter bandwidth is adaptively adjusted in combination with the EIS aging index to ensure the capture of long-tail low-frequency and high-frequency signals. Thermal stress crack noise is suppressed based on the temperature gradient, the ultrasonic characteristics are optimized, and the statistical window is dynamically adjusted to ensure the reliability of pulse frequency and phase statistics. Through the synergistic effect of various steps and feature fusion, the feature fusion method can still accurately distinguish between arc faults and surface discharges under humid, windy and heavily polluted conditions, thereby improving detection accuracy and real-time performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0053] Figure 1 Schematic diagram of the overall method of the present invention. DETAILED DESCRIPTION

[0054] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0055] Traditional distribution network line fault detection, especially high-frequency discharge detection based on HFCT sensors, focuses on distinguishing cable arc discharge (Arc) from surface partial discharge (PD) in transient pulse signals. Early research and engineering practices were often based on the following scenarios:

[0056] When there is no obvious water film on the insulator surface and the contamination layer is dry, the discharge signal mainly appears as an exponentially decaying high-frequency pulse. In this case, traditional methods generally only need to perform simple time-domain exponential fitting to accurately extract the peak amplitude and exponential decay constant. This can then be combined with phase statistics, vibration signals, or fiber temperature monitoring to distinguish between Arc and PD.

[0057] In windy conditions, conductor swaying causes interference from mechanical vibration and electromagnetic coupling signals. Many solutions use additional wind speed sensors or vibration sensors to decouple vibration noise. However, most methods only address acoustic interference or mechanical vibration, and fail to fully consider how changes in conductor-ground capacitance can directly introduce high-frequency spikes into the HFCT signal. Consequently, wind-induced electromagnetic coupling is often overlooked or misinterpreted as weak arc discharges.

[0058] Distribution lines in some areas are subject to high pollution year-round. When pollutants (such as dirt, chemical fumes, and salt spray) deposit on the surfaces of conductor insulators or constantan connectors, they absorb atmospheric water vapor, forming a water film, particularly in humid weather. This water film generates surface partial discharge (PD) at the moment of discharge, accompanied by a sudden change in capacitance. This can manifest as tailing or "double-peak" distortion in the HFCT signal. Some studies have attempted to describe this using surface discharge models, but most methods rely solely on empirical threshold screening at the back end, making it difficult to directly incorporate transient changes in the water film coupling capacitance into the model.

[0059] In real-world scenarios, wind vibration and water film often occur simultaneously. For example, in industrial areas, dense fog and strong winds can cause both water film capacitance tailing and high-frequency spikes from wind vibration in the HFCT signal. If these signals are processed separately, even after removing the water film tailing, misjudgment due to the wind vibration spikes may still occur. Alternatively, even after removing the wind vibration, the residual water film tailing may affect the exponential fitting.

[0060] In addition, distribution network fault detection has strict real-time requirements. Once an Arc occurs on the line, which is a typical fault signal, an alarm must be issued and the protection strategy must be activated within milliseconds or less. Frequent large-scale data fitting to remove interference is not only time-consuming but also prone to delays or misjudgments.

[0061] In summary, current technical solutions are unable to simultaneously and quantitatively address the complex interference caused by water film capacitance tailing in air environments and ground capacitance coupling interference in windy environments. They can often only perform local optimization for a single environment, resulting in the following problems:

[0062] In windy and dry environments, water film trailing is ignored and PD is missed;

[0063] In the wet and windless environment after seasonal rain or frost, ignoring wind-vibration coupling can lead to misjudgment;

[0064] When both exist at the same time, a single model often cannot make the HFCT fitting converge to a reasonable solution, and may even lead to misjudgment of the secondary model, and small Arcs are treated as common PDs, thus delaying fault isolation.

[0065] Based on the above analysis, the present invention is proposed, and the present invention is further described below in conjunction with embodiments.

[0066] Example 1 (see Figure 1 ): A distribution network line fault detection method based on data analysis, including:

[0067] According to the environmental status of the distribution line, the current operating status of the distribution network can be judged as follows:

[0068] On rainy days or after rain, in an air environment with a high humidity, the water film on the contamination layer (dirt or salt spray attached to the surface of distribution line insulators, such as suspension insulators or constantan joints) generates instantaneous conductivity and evaporation-induced capacitive coupling, which can distort the discharge pulse waveform (causing double peaks or tailing distortion). This distortion seriously affects the extraction of signal features, especially the accuracy of the peak amplitude and exponential decay constant of the discharge signal, misleading the judgment of line arc discharge and surface partial discharge (PD).

[0069] Secondly, during strong winds, transient changes in the capacitance between the conductor and the ground can cause high-frequency spikes, manifesting as broadband interference. This is similar to weak line arc discharge (Arc) or multi-source surface partial discharge (PD) signals, leading to misjudgment. Especially when the conductor swings, it generates large transient coupling currents, which disturb the HFCT signal and make it difficult to distinguish between pulses caused by arc discharge and wind vibration. These interference signals may cause misjudgment, such as misinterpreting wind vibration signals as line arc discharge or misidentifying PD due to the influence of water film capacitance. Therefore, the calculation of the water film capacitance tailing coefficient and the wind vibration suppression coefficient is designed to quantify the interference effects and more accurately extract the pure arc discharge signal. The specific steps are as follows:

[0070] Define the trigger threshold of the high frequency current transformer (HFCT, installed at the bottom of the insulator or on the conductor busbar) , in a continuous HFCT data stream, if there is:

[0071] When The current signal amplitude of the sampling point And the interval with the previous trigger (minimum sampling point interval), it is considered that a discharge pulse event occurs at this moment, a pulse is recorded, and the trigger sampling sequence number is ;

[0072] Extract the pulse time domain window:

[0073]

[0074] 、 Respectively represent the number of sampling points on both sides of the window width, Indicates the The trigger event (i.e. The intercepted signal in the time domain window where the secondary discharge pulse is located, Indicates the The sampling point number when the pulse is triggered, Represents the time-domain series of pulse currents obtained by sampling the original HFCT (high-frequency current transformer). This includes the actual discharge current (arc current), transient currents caused by air breakdown or insulation damage, the time-varying capacitance effect formed by the water film on the contamination layer, which is essentially a displacement current and a capacitive current, and the coupling current caused by the change in conductor-to-ground capacitance due to wind vibration of the conductor, which is also a type of capacitive current.

[0075] Build a superposition model:

[0076]

[0077] represents the pure discharge signal term, Indicates the Sub-pulse pure discharge peak amplitude, represents the exponential decay constant, corresponding to the discharge energy decay, represents the water film capacitance tail term, represents the transient reference voltage on the insulator surface, Indicates the first Sampling points correspond to time; Indicates the The instantaneous equivalent capacitance corresponding to the sampling point time is: represents the wind-vibration coupling term;

[0078] , Indicates the The time corresponding to the sampling point The instantaneous capacitance between the lower conductor and ground, Indicates the static ground capacitance in the absence of wind or after recovery from wind vibration balance. Indicates that when the relative position or posture of the conductor changes instantaneously due to wind vibration, the relative The additional ground capacitance The exponential rate constant representing the return of the capacitance to its static value, , Indicates the Sampling points corresponding to time The line voltage value under Indicates the fundamental frequency of the power grid, Indicates the line peak voltage;

[0079] Creating a time-varying capacitor Model:

[0080] , express The water film capacitance at the moment, It represents the reference capacitance when the pollution layer is not saturated. Represents the additional capacitance after the water film is formed, represents the water film evaporation rate constant, represents a real number variable (unit: seconds), describing the moment when the capacitance is checked. Indicates the The trigger moment is used here to describe the The triggering time of the secondary discharge pulse helps fit the HFCT signal, especially the tail component that appears in the signal, by simulating the capacitance change caused by the water film during the arc discharge process. By modeling the water film capacitance and voltage change rate, this model can accurately extract the time domain characteristics of the discharge signal, such as peak amplitude and exponential decay constant, thereby improving the accuracy of the arc discharge signal.

[0081] Estimation by least squares fitting or nonlinear least squares method 、 、 as well as (I will not elaborate on this here);

[0082] Calculate the water film tailing coefficient , which measures the degree of distortion of the pure discharge signal caused by water film coupling:

[0083]

[0084] Therefore, if If it is greater than the first preset threshold, it means that the water film tailing effect is serious and tends to be surface partial discharge (PD);

[0085] Calculate wind vibration suppression coefficient , which measures the degree of interference of the conductive-ground coupling on the pure discharge signal:

[0086]

[0087] like If the value is greater than the second preset threshold, it is determined that the wind vibration interference is significant, and the corresponding interference should be eliminated before extracting the time domain features. Therefore:

[0088] like is greater than a first preset threshold, If the value is less than or equal to the second preset threshold, it is judged that the water film tailing is serious, the wind vibration interference can be ignored, and the pulse is judged to be a surface PD. The time domain characteristics are output. 、 、 、 ;

[0089] like is less than or equal to a first preset threshold, If the value is greater than the second preset threshold, it is judged that the wind vibration interference is serious and the water film tail is light. The wind vibration term is first removed and then fitted. The corrected time domain characteristics are obtained by fitting according to the least squares method. 、 、 、 , I will not go into details here.

[0090] In summary, by introducing the water film capacitance effect and wind vibration interference correction, the distortion problem of HFCT signals caused by the water film effect and wind vibration electromagnetic interference in strong wind environments is solved, providing a more accurate method for distinguishing line arc discharge and surface partial discharge (PD) signals. By jointly modeling interference terms and quantifying the interference impact, this solution can improve the accuracy of time domain feature extraction, avoid misjudgment in traditional methods, and enhance the overall performance of the system, especially its adaptability in complex environments.

[0091] 2) When the contamination layer is non-uniform, multiple discharge sources (surface partial discharge (PD) or line arc fault (Arc)) may be located on the same insulator. Single-channel or dual-channel positioning may experience aliasing (multiple surface partial discharge (PD) sources will generate superimposed signals, resulting in multiple solutions or deviations in one-way positioning). A sparse reconstruction algorithm is used to separate and locate multiple discharge sources and distinguish between multiple PDs and single-point Arcs, including:

[0092] Array observation matrix construction:

[0093] Set the total number of array channels , mixed observation vector ;

[0094] Discrete the suspected discharge position of the insulator surface and the conductor into candidate points ;

[0095] Constructing the observation matrix , No. Column indicates if at position When a discharge occurs, the delay and attenuation characteristics of the discharge signal in each channel;

[0096] OMP sparse reconstruction solution:

[0097] For the first The observation value vector within the sub-pulse window , represents transpose, Indicates the The sensor channels are in sequence number The sample value at Assume there is a sparse source vector ,satisfy:

[0098] , represents the residual tolerance, Indicates the The residual vector of the trigger event, represents the observation matrix;

[0099] Solved using the orthogonal matching pursuit algorithm , Indicates that the constraints listed below are met and a sparse coefficient vector is obtained , calculate the discharge source ;

[0100] Among them, if Greater than 1 and all ( represents the index set) all fall in the high area, then the event tends to be multi-source surface PD;

[0101] If it is equal to 1 and the corresponding candidate point position Near the middle of the conductor ( The elements in the candidate point set represent the candidate discharge locations, describing theoretically possible discharge locations, It indicates the actual candidate point position corresponding to the non-zero component obtained after sparse reconstruction. It points to a specific candidate point position, corresponding to the most likely discharge source position in the non-zero sparse coefficient), that is, a non-insulator surface, which tends to be a single point Arc.

[0102] 3) Drastic changes in ambient temperature can cause thermal stress crack noise, generating ultrasonic pulses similar to PD / Arc. Specifically, the temperature difference between day and night or the thermal expansion difference between the metal constantan joint and the ceramic insulator will produce microcrack friction or thermal stress acoustic pulses. The spectrum covers the ultrasonic and low-frequency acoustic regions, which is confused with the PD acoustic signal. The ultrasonic channel cannot distinguish between crack friction sound and PD sound, resulting in distorted acoustic characteristics. Therefore, these interferences are suppressed by using temperature information. The specific steps are as follows:

[0103] Execute temperature gradient calculate:

[0104] ,like Greater than the thermal stress threshold , then there may be a cracking sound at this moment, Indicates the time interval for sampling environmental quantities (temperature, humidity, etc.), Indicates the The ambient temperature corresponding to the moment of secondary pulse triggering, Indicates Second pulse triggering time Before Ambient temperature at

[0105] Calculate the Ultrasonic energy during the sub-pulse (i.e., the energy integral of the acoustic signal collected in the ultrasonic sensing channel):

[0106] , Indicates the The acoustic signal value collected by the ultrasonic sensor at each sampling point;

[0107] Defining the thermal stress acoustic suppression coefficient , represents the reference temperature, represents the temperature sensitivity coefficient;

[0108] If satisfied , represents the suppression threshold, it is believed that the ultrasonic energy mainly comes from the thermal stress crack noise, and the thermal stress suppression acoustic energy is obtained by suppressing it. :

[0109] This correction ensures that the acoustic energy in subsequent eigenvectors is not interfered with by thermal stress noise. Therefore, the comparison between the comprehensive temperature gradient and the ultrasonic energy itself can effectively distinguish between thermal stress crack friction sound and discharge sound when the temperature changes sharply. Secondly, when there is a large temperature difference between day and night or when there is a difference in thermal expansion between the constantan connector and the ceramic insulator, the outbreak of thermal crack sound can be predicted in advance, allowing the system to actively suppress the ultrasonic characteristics during this period to prevent them from being mistakenly called for PD / Arc judgment.

[0110] 4) Furthermore, after rainfall or dew, the rapid change in water film thickness can cause the pulse triggering timing to drift, destroying the original pulse frequency and phase concentration indicators. Pulse density and phase statistics based on a fixed window width are prone to pseudo-random distribution and misjudgment as PD. Therefore, the statistical window is dynamically adjusted according to the change rate of the water film thickness to ensure the reliability of the statistical sample. The specific steps are as follows:

[0111] Collect water film thickness change rate (When calculating its value, the sampling time interval is ), calculate the dynamic sliding window length , Indicates the initial sliding window, such as 1s, Represents the water film change sensitivity coefficient, if If the value is large, increase the window length to ensure that there are still enough trigger events to count after nanosecond drift.

[0112] Pulse rate statistics:

[0113] Count the number of triggers in a time period , Indicates the index of the triggering event, Indicates the The discrete sampling point number corresponding to the pulse trigger, Indicates the high-frequency current transformer sampling rate, Indicates the The length of the dynamic sliding window corresponding to the trigger, The counting operation representing the number of elements in a set is equivalent to the cardinality of the set;

[0114] Calculate the pulse rate ;

[0115] Use grid phase calculation Phase window ratio at the second trigger , grid fundamental frequency , Indicates the The second trigger moment, describing the phase window ratio in the The moment of the second trigger (in this solution, Both represent trigger moments and are only used to describe the time point of each event or measurement in the system). Indicates the reference initial phase angle of the grid phase, and then the water film change rate Automatically widen the window when it is severe ,ensure a sufficient number of trigger samples so that pulse rate statistics and phase concentration are performed on a more balanced time base, avoiding pseudo-random distribution caused by microsecond drift;

[0116] Secondly, combining the pulse occurrence rate and phase window ratio of the adaptive window can more reliably distinguish between continuous multiple PD triggers and occasional Arc triggers.

[0117] 5) Therefore, the time domain features calculated above ( 、 、 、 )、Discharge source number( ), thermal stress suppression of acoustic energy ( ), pulse rate ( ) and the phase window ratio ( ) Fusion to create feature vector , by building a logistic regression model, the arc discharge probability is output :

[0118] , represents the weight vector of the logistic regression model, Represents the bias term of the logistic regression model;

[0119] Constructing the cross entropy loss function , , Indicates the sample label (Arc is 1, PD is 0), which is solved by gradient descent or Adam optimizer , I will not elaborate on this here, and then, based on each obtained calculate If it is greater than the probability threshold, it is determined to be a cable arc fault Arc, otherwise it is a surface partial discharge PD. Therefore, the judgment is performed:

[0120] If a line arc fault occurs, a fault alarm is generated and reported to the control center, triggering the isolation and positioning module (based on travel wave positioning or cross-correlation positioning) to quickly remove the faulty section, triggering the network self-healing (reconstruction) strategy, and restoring power supply to unaffected users;

[0121] If partial discharge occurs on the surface, the number of PD events of the insulator is accumulated. If the number of PD events exceeds the threshold within a certain period of time, a PD alarm is generated, prompting the operation and maintenance to perform on-site cleaning or take anti-pollution measures.

[0122] In summary, the present invention extracts clean pure discharge peak and attenuation characteristics based on the water film capacitance tail term and wind-induced vibration coupling model, uses the interference-removed HFCT signal to perform multi-source discharge positioning on a sparse array, and adaptively adjusts the filter bandwidth in combination with the EIS aging index to ensure the capture of long-tail low-frequency and high-frequency signals. It suppresses thermal stress crack noise based on the temperature gradient, optimizes ultrasonic features, and dynamically adjusts the statistical window to ensure reliable pulse frequency and phase statistics. Through the synergistic effect of each step and feature fusion, the present invention can accurately distinguish between arc faults and surface discharges under strong winds and heavy pollution conditions based on feature fusion, thereby improving detection accuracy and real-time performance.

[0123] Finally, the present invention also provides:

[0124] A distribution network line fault detection system is implemented according to the distribution network line fault detection method based on data analysis;

[0125] A computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method when executing the computer program;

[0126] A computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, the steps of the method described are implemented; no further details are given here, and reference can be made to the above method.

[0127] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A distribution network line fault detection method based on data analysis, characterized in that: The steps include: Monitor the current signal amplitude and determine whether to extract the pulse time domain window to obtain the intercepted signal within the time domain window where the trigger event is located; In response to the input intercepted signal, a pure discharge signal term is constructed based on the pure discharge peak amplitude and exponential decay constant. A water film capacitance tail term is constructed based on the instantaneous equivalent capacitance, reference voltage, and capacitance change rate. A wind-induced vibration coupling term is constructed based on the instantaneous capacitance between the conductor and the ground and the grid voltage. Based on this, a superposition model is established to solve the pure discharge peak amplitude and exponential decay constant. The maximum deviation of the water film capacitance tail term and the maximum amplitude of the wind vibration coupling term are calculated, and the water film tail coefficient and wind vibration suppression coefficient are determined in turn by combining them with the pure discharge peak amplitude. Construct time domain features; If the contamination layer is non-uniform, the sparse reconstruction algorithm is used to separate and calculate the discharge source; Combining the confusion between thermal stress crack noise and discharge sound caused by ambient temperature changes, the ambient temperature is used to suppress thermal stress interference and obtain thermal stress suppression acoustic energy; Collect the initial sliding window, and dynamically adjust the pulse sliding window length based on the water film thickness change rate, and determine the pulse occurrence rate based on the number of triggering power grid fault events; In response to the input time domain characteristics, discharge source number, thermal stress suppression acoustic energy, and pulse occurrence rate, a feature vector is established in combination with the phase window ratio, and the fault status of the distribution network is detected based on the discriminant model.

2. The method for detecting line faults in a distribution network based on data analysis according to claim 1, characterized in that: When the current signal amplitude If the value is greater than the trigger threshold and the interval between the last trigger and the last trigger is greater than or equal to the minimum sampling point interval, a pulse is recorded and the trigger sampling sequence number is ; Extract the pulse time domain window: ； 、 Respectively represent the number of sampling points on both sides of the window width, Indicates the The intercepted signal within the time domain window where the trigger event is located, represents the pulse current time domain sequence of the original HFCT sampling; Build a superposition model: ； Indicates the Sub-pulse pure discharge peak amplitude, represents the exponential decay constant, represents the transient reference voltage on the insulator surface, Indicates the first The sampling points correspond to time, Indicates the The instantaneous equivalent capacitance corresponding to the sampling point time is: Indicates the The time corresponding to the sampling point The instantaneous capacitance between the lower conductor and ground, Indicates the The line voltage value at each sampling point.

3. The method for detecting line faults in a distribution network based on data analysis according to claim 1, characterized in that: A time-varying capacitance model is introduced to compensate for the tailing effect of the water film dynamic process on the capacitance current, so as to extract the pure discharge peak amplitude and exponential decay constant. The specific time-varying capacitance model is: ； express The water film capacitance at the moment, It represents the reference capacitance when the pollution layer is not saturated. Represents the additional capacitance after the water film is formed, represents the water film evaporation rate constant, represents a real variable, Indicates the The trigger moment.

4. The method for detecting line faults in a distribution network based on data analysis according to claim 2, wherein: Water film trailing coefficient , which measures the degree of distortion of the pure discharge signal caused by water film coupling: ； Wind vibration suppression coefficient , which measures the degree of interference of the conductive-ground coupling on the pure discharge signal: ； According to the water film trailing coefficient and wind-induced vibration suppression coefficient Determine surface partial discharge or line arc fault.

5. The method for detecting line faults in a distribution network based on data analysis according to claim 1, characterized in that: The method for obtaining thermal stress suppression acoustic energy is: Temperature gradient calculate: ； Indicates the time interval for environmental quantity sampling, Indicates the The ambient temperature corresponding to the moment of secondary pulse triggering; Calculate the Ultrasonic energy during the sub-pulse : ； Indicates the The acoustic signal value at each sampling point; Defining the thermal stress acoustic suppression coefficient ; represents the reference temperature, represents the temperature sensitivity coefficient; If satisfied 、 If the values ​​are greater than the corresponding thresholds, it is determined that the ultrasonic energy comes from the thermal stress crack noise, and the thermal stress suppression acoustic energy is obtained. : , represents the thermal stress threshold, Indicates the suppression threshold.

6. The method for detecting line faults in a distribution network based on data analysis according to claim 2, characterized in that: The method for determining the pulse occurrence rate is: Count the number of triggers in a time period ; Indicates the index of the triggering event, Indicates the The discrete sampling point number corresponding to the pulse trigger, Indicates the high-frequency current transformer sampling rate, Indicates the The length of the dynamic sliding window corresponding to the trigger, Counting operations that represent the number of elements in a collection; The pulse occurrence rate is calculated based on the ratio of the number of statistical triggers to the length of the dynamic sliding window.

7. The method for detecting line faults in a distribution network based on data analysis according to claim 1, characterized in that: The method for detecting the fault state of the distribution network by the discriminant model is: Construct a logistic regression model and input the feature vector to output the arc discharge probability: Construct a cross entropy loss function; Calculate the arc discharge probability based on the characteristic vector obtained each time; If it is greater than the probability threshold, it is determined to be a cable arc fault, otherwise it is a surface partial discharge.

8. A distribution network line fault detection system, characterized in that: The method for detecting line faults in a distribution network based on data analysis is implemented according to any one of claims 1 to 7.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images