Low-voltage electric leakage fault positioning method and system, medium and product

Abstract

The invention discloses a low-voltage leakage fault positioning method and system, a medium and a product, and relates to the field of power detection. The method comprises the following steps: acquiring a zero-sequence current signal at a neutral line grounding down lead at the low-voltage side of a distribution transformer of a target power distribution network through a flexible zero-sequence current detection unit; analyzing the zero-sequence current signal through an analysis control module to obtain a signal analysis result; when the signal analysis result represents that a potential electric leakage area exists, electric leakage pre-judgment information is generated; controlling a signal injection module to inject a feature detection signal into the target power distribution network, and controlling a high-altitude signal sensing unit and an auxiliary environment sensing unit to cooperatively perform signal detection in a potential electric leakage area to obtain electric leakage detection data; and analyzing the electric leakage detection data through an analysis control module, and determining the position of an electric leakage fault point of the target power distribution network. According to the invention, the low-voltage leakage positioning efficiency and reliability can be improved.

Description

Technical Field

[0001] This application relates to the field of power detection technology, and in particular to a method, system, medium and product for locating low-voltage leakage faults. Background Technology

[0002] Low-voltage distribution networks, especially 0.4kV overhead lines, are widely distributed in urban and rural areas. Their operating environment is complex, and leakage faults occur frequently. Traditional fault diagnosis mainly relies on manual pole climbing inspection or segmented power outage testing. This method is inefficient, affects user power supply, and has high operational safety risks, making it difficult to achieve uninterrupted, fast, and safe leakage location. Existing technologies using the signal injection method for live-line detection typically include a host (10) capable of generating and injecting a specific frequency signal, and one or more portable receiving devices for detecting the signal. During operation, the characteristic signal is applied to the distribution system, and the testing personnel carry the receiving device and move along the line. By listening to or observing the changes in signal strength indicated by the receiving device, they can determine the approximate direction of the fault point. This method avoids a complete power outage and improves the flexibility of detection to some extent. However, since the entire detection process lacks a preliminary judgment of the fault range, the testing personnel usually need to start from the signal injection point and perform a comprehensive signal strength scan of the entire possible line path to find the location of signal abrupt change. This segment-by-segment scanning method "from the start point to the end point" is time-consuming in complex multi-branch lines. Meanwhile, stray signal interference in the environment may also affect the accurate identification of valid feature signals, thus posing a challenge to the accuracy and success rate of positioning. It is difficult to effectively narrow the initial detection range during live-line positioning of leakage current in low-voltage distribution networks, thereby improving positioning efficiency and reliability. Summary of the Invention

[0003] This application provides a method, system, medium, and product for locating low-voltage leakage faults, which addresses the technical problem of improving the efficiency and reliability of low-voltage leakage fault location.

[0004] In a first aspect, embodiments of this application provide a low-voltage leakage fault location method, applied to a low-voltage leakage fault location device. The low-voltage leakage fault location device includes a main unit 10, a flexible zero-sequence current detection unit 20, a high-altitude signal sensing unit 30, and an auxiliary environmental sensing unit 40. The main unit 10 includes a signal injection module and an analysis and control module. The signal output terminal 21 of the flexible zero-sequence current detection unit 20 is detachably electrically connected to the signal input terminal 11 of the main unit 10 via an adapter cable. The high-altitude signal sensing unit 30 and the auxiliary environmental sensing unit 40 are wirelessly connected to the main unit 10, respectively. The method includes: The zero-sequence current signal at the grounding lead of the neutral line on the low-voltage side of the distribution transformer of the target distribution network is obtained through the flexible zero-sequence current detection unit 20. The zero-sequence current signal is analyzed by the analysis and control module to obtain the signal analysis results; When the signal analysis results indicate the existence of a potential leakage area, leakage prediction information is generated. Based on the leakage current prediction information, the analysis and control module controls the signal injection module to inject a characteristic detection signal into the target power distribution network, and controls the high-altitude signal sensing unit 30 and the auxiliary environmental sensing unit 40 to work together to perform signal detection in the potential leakage current area to obtain leakage current detection data. The analysis and control module analyzes the leakage detection data to determine the location of the leakage fault point in the target power distribution network.

[0005] Optionally, the step of analyzing the zero-sequence current signal through the analysis and control module to obtain signal analysis results includes: filtering and extracting feature parameters from the zero-sequence current signal through the analysis and control module to obtain a set of feature parameters characterizing the signal characteristics; acquiring network topology data of the target distribution network, the network topology data including electrical parameters and section identifiers of each line section; and analyzing and processing the set of feature parameters and the network topology data through a preset evaluation model to generate the signal analysis results, the signal analysis results being used to identify the potential leakage probability of each line section in the target distribution network.

[0006] Optionally, the step of filtering and extracting feature parameters from the zero-sequence current signal through the analysis and control module to obtain a set of feature parameters characterizing the signal characteristics includes: performing power frequency filtering and bandpass filtering on the zero-sequence current signal through the analysis and control module to obtain a power frequency component and at least one preset non-power frequency component; calculating the effective current value of the power frequency component and calculating the frequency band energy value with the highest signal energy among the non-power frequency components of the at least one preset non-power frequency band; calculating the waveform distortion rate and dynamic fluctuation coefficient of the zero-sequence current signal based on the waveform sampling data of the zero-sequence current signal, wherein the dynamic fluctuation coefficient characterizes the ratio of the average difference between the signal peak and the signal valley of the zero-sequence current signal within a preset time period to the effective signal value; and obtaining the set of feature parameters based on the effective current value, the frequency band energy value, the waveform distortion rate, and the dynamic fluctuation coefficient.

[0007] Optionally, the step of analyzing and processing the set of feature parameters and the network topology data using a preset evaluation model to generate the signal analysis result includes: inputting the set of feature parameters into the preset evaluation model, wherein the preset evaluation model is an evaluation model trained based on historical fault data, the preset evaluation model includes a preset feature region association database, the preset feature region association database stores historical association weights of different combinations of feature parameters and each of the line segments; based on the historical association weights and the electrical parameters of each of the line segments in the network topology data, obtaining the potential leakage probability output by the preset evaluation model corresponding to each of the line segments; and taking the line segments with the potential leakage probability greater than a preset probability threshold as the potential leakage region to obtain the signal analysis result.

[0008] Optionally, controlling the high-altitude signal sensing unit 30 and the auxiliary environmental sensing unit 40 to collaboratively perform signal detection within the potential leakage area to obtain leakage detection data includes: controlling the high-altitude signal sensing unit 30 to perform non-contact signal sensing on the overhead lines within the potential leakage area to obtain first characteristic signal strength data for each line segment; determining the target segment to be detected based on the first characteristic signal strength data; controlling the auxiliary environmental sensing unit 40 to perform near-field signal sensing on ground environmental objects associated with the target segment to obtain second characteristic signal strength data for each ground environmental object; and obtaining the leakage detection data based on the first characteristic signal strength data and the second characteristic signal strength data.

[0009] Optionally, the step of analyzing the leakage current detection data through the analysis and control module to determine the location of the leakage current fault point in the target distribution network includes: sorting the first characteristic signal strength data corresponding to each detection point in the leakage current detection data based on the current flow direction of the line in the potential leakage current area; when the signal strength attenuation rate between adjacent detection points is greater than a preset attenuation threshold, determining the area between the two adjacent detection points corresponding to the signal strength attenuation rate as the fault location interval; and determining the location of the ground environment object corresponding to the maximum strength value among all the second characteristic signal strength data in the fault location interval as the location of the leakage current fault point.

[0010] Optionally, before controlling the signal injection module to inject the feature detection signal into the target power distribution network, the method further includes: controlling the high-altitude signal sensing unit 30 or the auxiliary environmental sensing unit 40 to perform background electromagnetic spectrum scanning in the area to be detected, and obtaining the background noise intensity of each frequency point within a preset frequency range; calculating a candidate score for each frequency point based on the background noise intensity and the signal propagation efficiency coefficient of each frequency point, wherein the candidate score is negatively correlated with the background noise intensity and positively correlated with the signal propagation efficiency coefficient; and configuring the frequency point corresponding to the maximum candidate score as the operating frequency of the feature detection signal.

[0011] Secondly, embodiments of this application provide a low-voltage leakage fault location system, which includes: one or more processors and a memory; the memory is coupled to one or more processors, and the memory is used to store computer program code, the computer program code including computer instructions, and the one or more processors call the computer instructions to cause the low-voltage leakage fault location system to perform the method described in the first aspect and any possible implementation thereof.

[0012] Thirdly, embodiments of this application provide a computer program product containing instructions that, when the computer program product is run on a low-voltage leakage fault location system, cause the low-voltage leakage fault location system to perform the method described in the first aspect and any possible implementation thereof.

[0013] Fourthly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a low-voltage leakage fault location system, cause the low-voltage leakage fault location system to perform the method described in the first aspect and any possible implementation thereof.

[0014] In summary, one or more technical solutions provided in this application have at least the following technical effects or advantages: 1. By applying a low-voltage leakage current locator that integrates a main unit, a flexible zero-sequence current detection unit, and high-altitude and auxiliary environmental sensing units, and employing a method of "passive pre-judgment followed by directional and coordinated precision measurement," it is possible to achieve rapid live-line location of leakage faults in low-voltage distribution networks without affecting normal power supply. This effectively overcomes the inherent defects of traditional methods, which require power outages and pole climbing for inspection, resulting in high operational risks, a large impact range, and low efficiency, significantly improving the efficiency and reliability of leakage fault location.

[0015] 2. By extracting multi-dimensional features from the zero-sequence current signal and using an evaluation model based on historical data for analysis to generate predictive results that identify potential leakage areas, early intelligent diagnosis of leakage faults is achieved. This enables in-depth mining of fault features from complex signals, accurately narrowing down the scope of subsequent precise detection, thereby fundamentally improving the efficiency and accuracy of the entire positioning process.

[0016] 3. By adopting a collaborative detection mechanism of high-altitude rapid scanning for initial screening and ground-based near-field precise measurement for confirmation, and combining the dual criteria of signal mutation analysis and near-field peak positioning, the system achieves accurate location of fault points in complex field environments. This balances detection efficiency and positioning accuracy, reduces reliance on operator experience, and improves the reliability and first-time success rate of the positioning results.

[0017] 4. By automatically scanning background noise and adaptively selecting the optimal operating frequency before signal injection, the system's adaptability in complex electromagnetic environments is significantly enhanced, effectively improving the signal-to-noise ratio of the detection signal, ensuring the sensitivity and reliability of subsequent detection, and enabling the positioning system to work stably under various interference environments. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the low-voltage leakage current locator provided in the embodiments of this application; Figure 2 This is a flowchart illustrating the low-voltage leakage fault location method provided in the embodiments of this application; Figure 3 This is another flowchart illustrating the low-voltage leakage fault location method provided in the embodiments of this application; Figure 4 This is a schematic diagram of the low-voltage leakage fault location system provided in the embodiments of this application.

[0019] Explanation of reference numerals in the attached figures: 10, Main unit; 11, Signal input terminal; 20, Flexible zero-sequence current detection unit; 21, Signal output terminal; 30, High-altitude signal sensing unit; 40, Auxiliary environmental sensing unit; 601, Central processing unit; 602, Read-only memory; 603, Random access memory; 604, Bus; 605, Input / output interface; 606, Input section; 607, Output section; 608, Storage section; 609, Communication section; 610, Driver; 611, Removable media. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] In the description of the embodiments of this application, words such as "illustrative," "for example," or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "illustrative," "for example," or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Rather, the use of words such as "illustrative," "for example," or "for example" is intended to present the relevant concepts in a specific manner.

[0022] In the description of the embodiments of this application, the terms "first, second, third" are used only to distinguish similar objects and do not represent a specific order of objects. It is understood that "first, second, third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0023] In the embodiments of this application, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0024] In the implementation of this application, the collection and processing of relevant data should strictly comply with the requirements of relevant national laws and regulations, obtain the informed consent or separate consent of the personal information subject, and carry out subsequent data use and processing within the scope of laws and regulations and the authorization of the personal information subject.

[0025] Unless otherwise defined, all technical and scientific terms used in the embodiments of this application have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in the embodiments of this application is for the purpose of describing the embodiments of this application only and is not intended to limit this application.

[0026] This application provides a method, system, medium, and product for locating low-voltage leakage faults, which can improve the efficiency and reliability of low-voltage leakage fault location.

[0027] Figure 1 This is a schematic diagram of the low-voltage leakage current locator provided in the embodiments of this application.

[0028] This application discloses a low-voltage leakage current locator, such as... Figure 1 As shown, the system includes a host 10, a flexible zero-sequence current detection unit 20, a high-altitude signal sensing unit 30, and an auxiliary environmental sensing unit 40. The host 10 includes a signal injection module and an analysis and control module. The signal output terminal 21 of the flexible zero-sequence current detection unit 20 is detachably electrically connected to the signal input terminal 11 of the host 10 via an adapter cable. The high-altitude signal sensing unit 30 and the auxiliary environmental sensing unit 40 are wirelessly connected to the host 10, respectively.

[0029] The host unit 10 represents the core control and processing equipment of the positioning device, integrating functional circuits for signal generation, data processing, logic control, and communication coordination. The signal injection module is a circuit unit integrated within the host unit, used to generate a characteristic current signal of a specific frequency and amplitude, and to inject this signal into the power distribution system via coupling. The analysis and control module is a core processing unit integrated within the host unit, containing a microprocessor and related peripheral circuits, used to execute signal processing algorithms, fault analysis logic, and control the coordinated operation of all other units. The flexible zero-sequence current detection unit 20 represents a current measuring device that uses a non-rigid, bendable Rogowski coil as the sensing element, specifically designed to measure current by wrapping around and snapping it around the conductor without cutting the wire, particularly suitable for measuring zero-sequence currents containing rich high-frequency components. The signal output terminal 21 is the physical interface on the flexible zero-sequence current detection unit 20 used to convert the sensed current signal into an electrical signal and output it. The signal input terminal 11 is the physical interface on the host unit 10 used to receive signals from the flexible zero-sequence current detection unit 20 or other wired sensors. A matching cable refers to a wire with connectors at both ends that match the signal output terminal 21 and the signal input terminal 11, used to achieve a stable electrical connection and signal transmission between the two. The high-altitude signal sensing unit 30 refers to a portable detection device specifically designed for non-contact detection of specific frequency signal strength in overhead lines, typically employing a clamp-on current transformer or a high-sensitivity magnetic field induction antenna. The auxiliary environmental sensing unit 40 refers to a portable detection device used for contact or near-field sensing of surfaces or surrounding objects such as distribution towers, grounding devices, guy wires, or nearby buildings in a ground environment to detect leakage signal strength, typically equipped with a probe and a strength indicator.

[0030] Figure 2 This is a flowchart illustrating the low-voltage leakage fault location method provided in the embodiments of this application.

[0031] This application discloses a method for locating low-voltage leakage faults, applicable to... Figure 1 The low-voltage leakage current locator shown is, for example Figure 2 As shown, the steps include the following.

[0032] S101. The zero-sequence current signal at the grounding lead of the neutral line on the low-voltage side of the distribution transformer of the target distribution network is obtained through the flexible zero-sequence current detection unit 20.

[0033] Specifically, the flexible zero-sequence current detection unit 20 is physically deployed and connected to the neutral grounding lead of the low-voltage side of the distribution transformer in the target distribution network. The flexible zero-sequence current detection unit 20 continuously or on demand senses and measures the current flowing through the lead through its built-in sensing mechanism. Instead of measuring the total current, it specifically senses, extracts and quantifies the zero-sequence current signal component contained therein. This process is electrically isolated and does not require power interruption. The acquired raw zero-sequence current signal (usually an analog quantity that varies with time or a data stream that has undergone preliminary digitization) is then transmitted to the subsequent processing unit through the signal output terminal 21 of the unit via the adapter cable, providing the initial and most critical electrical state input data for the entire fault location process.

[0034] The target distribution network refers to the specific low-voltage (usually 0.4kV) power distribution network for which leakage fault investigation is to be conducted. Its scope covers the lines and equipment from the secondary side of the distribution transformer to the end-user's meter. The low-voltage side of the distribution transformer refers to the winding and its outgoing terminals on the side with the lower voltage level in the distribution transformer; in a 0.4kV distribution network, this typically refers to the side with a voltage of 400V / 230V. The neutral grounding down conductor is a dedicated conductor that leads from the neutral point (or neutral busbar) on the low-voltage side of the distribution transformer and ultimately connects to the earth's grounding grid. This conductor is the main path for the zero-sequence current to converge and flow into the earth. The zero-sequence current signal refers to the sum of the instantaneous values ​​of the three-phase currents (I0) in a three-phase four-wire system. A +I B +I C The current component generated by the fault is theoretically zero under normal symmetrical and fault-free conditions; however, when an asymmetrical fault such as single-phase grounding or leakage occurs, this component will appear significantly, and its amplitude, waveform, and spectrum characteristics carry fault information.

[0035] S102. The zero-sequence current signal is analyzed by the analysis control module to obtain the signal analysis results.

[0036] Specifically, the analysis and control module receives or retrieves the raw, continuously changing zero-sequence current signal from the data acquisition unit in real time. Since the raw signal inevitably contains power frequency fundamental components, high-frequency noise, and other normal disturbances from system operation, the analysis and control module immediately initiates a series of digital signal processing algorithms. These algorithms include, but are not limited to, digital filtering (such as band-stop filters to remove 50Hz power frequency and low-pass filters to remove irrelevant high-frequency noise) and signal transformation. Based on signal purification and standardization, the module further mines the fault information contained within the signal from both time and frequency dimensions. For example, in the time domain, it calculates the transient peak value, initial polarity, and pulse width; in the frequency domain, it analyzes the signal's spectral structure using methods such as Fast Fourier Transform (FFT) or Wavelet Transform, extracting the energy distribution and amplitude of high-frequency components in specific frequency bands caused by leakage faults. Finally, these quantified and extracted key indicators that most effectively characterize the signal state are integrated into a structured dataset—the signal analysis results.

[0037] The signal analysis results refer to the output information generated by the analysis and control module after processing the zero-sequence current signal, which is used to guide subsequent positioning operations. It is not the raw waveform data, but rather refined and interpreted conclusive, structured data, such as a dataset that identifies potential fault areas and their corresponding probabilities or confidence levels.

[0038] Based on the above embodiments, as an optional embodiment, for Figure 2 The step S102 shown can be implemented through steps S201-S203, which will be explained in detail below.

[0039] S201. The zero-sequence current signal is filtered and its characteristic parameters are extracted by the analysis and control module to obtain a set of characteristic parameters that characterize the signal.

[0040] Specifically, the raw waveform data of the monitored zero-sequence current signal is acquired. This raw data is typically mixed with various noises, high-order harmonics, and other interference components. To eliminate the negative impact of this irrelevant information on subsequent analysis, the signal needs to be filtered first. Filtering can employ digital filters (such as Butterworth low-pass filters and Kalman filters) or frequency domain analysis methods (such as Fast Fourier Transform, FFT). The core objective is to filter out high-frequency noise and frequency components irrelevant to the specific application scenario, while retaining or enhancing key frequency components that reflect the true state of the system (such as power frequency components). After obtaining a clean signal waveform, a series of mathematical algorithms and signal processing techniques are applied to quantify and analyze the filtered signal from multiple dimensions, including the time and frequency domains. For example, in the time domain, the peak value, root mean square value, and waveform steepness (current change rate d) of the signal are calculated. i / d t In the frequency domain, the fundamental amplitude, the content of specific harmonics (such as the 3rd and 5th harmonics), and the spectral energy distribution are analyzed. Finally, these calculated values, which describe the core characteristics of the signal from different perspectives, are integrated to form a structured dataset, namely, "a set of feature parameters." This set of parameters constitutes a concise, accurate, and quantitative mathematical description of the original complex signal.

[0041] Filtering is a signal processing technique used to separate specific frequency components of interest from mixed signals or to filter out unwanted frequency components. For example, low-pass filtering is used to remove high-frequency noise, while band-pass filtering is used to extract signals within a specific frequency range. Signal characteristics represent the unique properties of a signal in different dimensions, such as signal amplitude, rate of change, frequency composition, phase relationship, and energy distribution. A set of feature parameters represents a collection or vector consisting of multiple (two or more) feature parameters. This set can describe the state of the signal more comprehensively and three-dimensionally than a single parameter. For example, this set of feature parameters can be represented as [peak value, root mean square value, third harmonic content, signal energy].

[0042] Based on the above embodiments, as an optional embodiment, step S201 can be implemented through steps S2011-S2014, which will be explained in detail below.

[0043] S2011. The zero-sequence current signal is subjected to power frequency filtering and bandpass filtering by the analysis and control module to obtain the power frequency component and at least one preset non-power frequency component.

[0044] Specifically, the analysis and control module applies two different digital filtering operations to the input zero-sequence current signal digital sequence in parallel or serially. The first is power frequency filtering, which accurately separates the signal components closely related to the fundamental frequency of the power grid (e.g., 50Hz) to obtain the power frequency component. This type of filtering is usually designed to be highly selective, aiming to retain power frequency information to the maximum extent while effectively suppressing harmonics and noise near the power frequency, thereby obtaining a relatively pure fundamental frequency signal. The second type is bandpass filtering. Its design goal is to allow frequency components within one or more specific, pre-defined non-power frequency bands to pass through, based on pre-established knowledge or experience, thereby obtaining one or more non-power frequency components. These pre-defined frequency bands are determined based on historical fault data, electrical fault theory, or experimental analysis, and are considered to potentially contain characteristic frequency signals generated by typical leakage current or early fault phenomena such as arcing, corona discharge, and partial insulation discharge. Through this step, the analysis and control module successfully decouples a raw signal containing broadband information into two (or more) signal components focused in the frequency domain, each representing a different physical phenomenon: the power frequency component mainly reflects the steady-state, large-amplitude ground leakage current; while the non-power frequency component may carry characteristic fingerprint information reflecting the nature and development stage of the fault. This separation lays a direct signal foundation for subsequent calculation of the characteristic parameters of different components (such as the power frequency RMS value and non-power frequency energy).

[0045] Power frequency filtering refers to a specially designed digital or analog filtering process whose passband center frequency is set to the rated operating frequency of the power system (such as 50Hz or 60Hz). Its purpose is to efficiently and faithfully extract or isolate the fundamental frequency component from complex signals. Bandpass filtering refers to a filtering operation that allows signal components within its passband frequency range (i.e., between the lower and upper cutoff frequencies) to pass through with minimal attenuation, while significantly attenuating frequency components outside the passband. Power frequency components refer to the signal portion obtained after power frequency filtering, with the fundamental frequency of the power grid as the main energy concentration area. These represent the low-frequency main component of zero-sequence current directly related to normal system operation or steady-state leakage. Preset non-power frequency bands refer to one or more specific frequency ranges predefined based on technical knowledge, equipment characteristics, or application scenarios before implementing filtering. These ranges are located outside the power frequency range (e.g., a sub-band from several hundred hertz to several thousand hertz) and are set as the characteristic frequency region that may be related to the target detection phenomenon (such as a specific type of electrical fault). Non-power frequency components refer to signal components located within the corresponding frequency band after undergoing one or more bandpass filtering processes targeting a preset non-power frequency band. They represent specific frequency components in the zero-sequence current signal, besides the fundamental power frequency, that are considered to potentially contain diagnostically valuable information.

[0046] S2012. Calculate the effective value of the current of the power frequency component, and calculate the frequency energy value of the highest signal energy in the non-power frequency component of at least one preset non-power frequency band.

[0047] Specifically, the analysis and control module performs calculations on the power frequency component. Since the power frequency component is a relatively "pure" periodic or near-periodic signal dominated by the power grid fundamental frequency (e.g., 50Hz), the module uses the Root Mean Square (RMS) algorithm to calculate it. This calculation typically involves averaging the squares of the discrete sequence of the power frequency component over a continuous sampling period and then taking the square root. The result is a scalar, which physically represents the DC current value that produces the equivalent thermal effect of the power frequency current within one cycle. This RMS current value is a key indicator for quantifying the magnitude of steady-state leakage current. Simultaneously, the analysis and control module needs to perform energy analysis on the non-power frequency component in at least one preset non-power frequency band. Since there may be more than one preset non-power frequency band (e.g., multiple narrowband filters are set to cover different suspected characteristic frequency bands), it is necessary to iterate and calculate the signal energy of the non-power frequency component corresponding to each preset frequency band. The signal energy calculation typically involves summing the squares of the amplitudes at each sampling point of the discrete sequence of the component in that frequency band (or integrating for continuous signals). This value directly reflects the strength of signal activity within that specific frequency range. After calculation, the energy values ​​calculated from all preset non-power frequency bands are compared to identify the frequency band with the highest signal energy. Finally, this highest energy value is recorded as the frequency band energy value, indicating which frequency band has the most significant activity among the preset multiple frequency bands that may contain fault characteristics, thus providing the most prominent clue for judging the possible fault type or activity level.

[0048] The effective value of current refers to the square root of the average of the squared values ​​of an alternating current over a complete cycle. It is a commonly used physical quantity for measuring the magnitude of alternating current (such as heating effect or work capacity), and its unit is ampere (A). Signal energy refers to the measure of a signal's intensity in the time or frequency domain. For discrete signals, it is often defined as the sum of the squares of the amplitudes at each point in the signal sequence; for a specific frequency band, it can also be obtained by integrating the power spectral density of that band, reflecting the total activity level of the signal within the analyzed object (time period or frequency band). The frequency band energy value refers to the specific signal energy value calculated for the identified "frequency band with the highest signal energy."

[0049] S2013. Based on the waveform sampling data of the zero-sequence current signal, calculate the waveform distortion rate and dynamic fluctuation coefficient of the zero-sequence current signal. The dynamic fluctuation coefficient represents the ratio of the average difference between the peak value and the valley value of the zero-sequence current signal within a preset time period to the effective value of the signal.

[0050] Specifically, the analysis and control module directly calls the stored waveform sampling data of the zero-sequence current signal (i.e., the original or appropriately preprocessed discrete-time sequence), performs spectral analysis (such as Fast Fourier Transform) on the sampling data, identifies the fundamental component (usually the power frequency) and each harmonic component, and then calculates it according to the definition of total harmonic distortion (THD). THD is the ratio of the square root of the sum of the squares of the effective values ​​of all harmonic components except the fundamental component to the effective value of the fundamental component. The result is a dimensionless percentage value that quantitatively describes the degree of deviation of the signal waveform from the ideal sine wave and reflects the total level of harmonic pollution in the signal. Secondly, parallel calculation of the dynamic fluctuation coefficient requires defining a preset duration as the analysis time window. Within this window (which may contain multiple power grid cycles), the local peaks (maximums) and local valleys (minimums) of the signal are traversed and identified. For multiple consecutive time windows (or a long window is segmented), the difference between the signal peak and the signal valley (i.e., the peak-to-peak fluctuation range) within each window is calculated. The arithmetic mean of all these differences is calculated to obtain an index characterizing the typical peak-to-peak fluctuation amplitude of the signal. Finally, this average is divided by the effective value of the zero-sequence current signal over the entire analysis period to obtain the dynamic fluctuation coefficient. This coefficient is also a dimensionless ratio that standardizes the fluctuation amplitude and correlates it with the overall strength (effective value) of the signal, thus more fairly measuring the relative instability or the severity of dynamic changes of the signal.

[0051] The waveform sampling data of the zero-sequence current signal refers to a series of discrete amplitude data points arranged in chronological order, obtained by periodically measuring the zero-sequence current analog signal through an analog-to-digital converter. Waveform distortion rate is a parameter used to quantify the degree to which a periodic alternating current waveform deviates from its standard sine wave; it typically refers specifically to the total harmonic distortion rate, which is the ratio (expressed as a percentage) of the sum of the squares of the effective values ​​of all harmonic components to the effective value of the fundamental component. Dynamic fluctuation coefficient is a dimensionless parameter used to characterize the relative drasticness or stability of the amplitude variation of an alternating current signal over a short time scale. Preset duration refers to the length of the time period used for peak-valley statistical analysis before calculating the dynamic fluctuation coefficient. Signal peak value refers to the amplitude corresponding to the local maximum value point on the signal waveform within the preset duration. Signal valley value refers to the amplitude corresponding to the local minimum value point on the signal waveform within the preset duration. Effective value (RMS) refers to the root mean square value calculated for the alternating current signal over the entire analysis period (or a complete cycle); it is a standard physical quantity for measuring the overall strength of the signal.

[0052] S2014. Based on the effective value of current, frequency band energy value, waveform distortion rate and dynamic fluctuation coefficient, a set of characteristic parameters are obtained.

[0053] Specifically, the analysis and control module gathers the four key values ​​calculated in the preceding steps: the effective current value characterizing steady-state leakage current intensity, the frequency band energy value reflecting the strength of the most significant characteristic frequency activity, the waveform distortion rate describing the degree of signal harmonic pollution, and the dynamic fluctuation coefficient quantifying the relative severity of instantaneous signal fluctuations. These values ​​are arranged and encapsulated in a predetermined, consistent order. For example, they can be placed sequentially in fixed positions within a one-dimensional array or vector to form a set of feature parameters. This set of parameters constitutes a multi-dimensional digital profile of the current zero-sequence current signal state. Each dimension's value represents the signal's quantitative performance in a specific aspect; the four dimensions collectively cover the signal's "intensity," "characteristic frequency," "waveform quality," and "dynamic stability." Alternatively, a weighted summation or assignment can be performed using a preset weight set to generate the final set of feature parameters. No specific restrictions are imposed here, providing a direct, efficient, and information-rich input interface for the next stage of intelligent assessment and decision-making (such as inputting into the assessment model for fault area probability calculation).

[0054] S202. Obtain the network topology data of the target distribution network, which includes the electrical parameters and section identifiers of each line section.

[0055] Specifically, network topology data about the target distribution network is obtained from local storage, external maintenance databases, or through manual input. This data is a structured collection of information, the core of which is the physical wiring diagram and digital description of equipment parameters of the target distribution network. This data contains at least two key parts: first, the division of each line segment and its segment identification, that is, the distribution network is divided into segments according to certain rules (such as between adjacent towers, from the start to the end of a branch line), and each independent segment is assigned a unique identifiable code or name; second, the electrical parameters associated with these line segments. These parameters are quantitative data describing the physical characteristics of the line segment itself, such as the line length, conductor type, impedance per unit length, capacitance to ground, current carrying capacity, etc. After receiving this data, it is loaded and organized into an internal space-electrical information mapping table or topology structure. This data structure links each segment identifier with its corresponding spatial location range, electrical characteristics, and connection relationship with upstream and downstream segments, completing the "digital modeling" of the target distribution network. This lays an indispensable digital foundation for subsequent mapping of signal analysis results to specific spatial locations (such as determining which segment is suspected based on characteristics), calculating signal propagation laws (such as estimating normal attenuation based on impedance), and generating accurate positioning commands (such as directly outputting segment identifiers or coordinates).

[0056] Network topology data refers to the data set used to describe the physical structure and connection relationships of the distribution network, including the route, segments, connection points, and electrical attributes of each component. Each line segment refers to multiple independent, continuous line sections divided into the target distribution network according to certain rules (such as between adjacent poles, branch lines, etc.), and is one of the smallest spatial units for fault location. Electrical parameters refer to physical quantities related to each line segment that describe its electrical performance, such as line length, conductor type, resistance per unit length, reactance, susceptance, and capacitance to ground. Segment identifiers are unique identification codes or names assigned to each line segment, used to clearly distinguish and reference different line segments in the data.

[0057] S203. Analyze and process a set of feature-related network topology data through a preset evaluation model to generate signal analysis results. The signal analysis results are used to identify the potential leakage probability of each line section in the target distribution network.

[0058] Specifically, the set of feature parameters (a vector containing multiple values) obtained in the previous process is used as input to a pre-built, trained, and debugged evaluation model. This model is not a simple linear formula, but a complex algorithmic structure that internally embeds knowledge and judgment rules learned from a large amount of historical data (including normal operation data and various leakage fault data). Upon receiving the input feature parameters, it performs a comprehensive evaluation and nonlinear mapping of this data through its internal computational logic (such as weighted summation and activation functions in neural networks, and classification hyperplane calculation in support vector machines). This processing essentially calculates the degree of matching or similarity between the features of the current signal and the leakage feature patterns in the model, i.e., a specific probability value. This probability value is directly linked to the source of the signal—a specific line section in the target distribution network—thereby giving that section a clear risk label and accurately identifying the probability of leakage occurring. In low-voltage distribution networks, when leakage faults occur in different line sections, the zero-sequence current signals measured at the neutral point grounding lead of the distribution transformer exhibit distinguishable differences. These differences primarily stem from the following physical mechanisms: First, differences in line impedance. The electrical distance from each line section to the neutral point varies, resulting in differences in line impedance (including resistance and reactance). When different sections experience faults with the same leakage resistance, the amplitude of the zero-sequence current measured at the neutral point will differ due to the different total circuit impedance. Sections closer to the neutral point typically exhibit larger zero-sequence current amplitudes when leakage occurs, while sections farther away show relatively smaller amplitudes. Second, differences in harmonic characteristics. Different line sections are connected to loads with different types and characteristics. For example, some sections may be connected to numerous nonlinear loads (such as frequency converters and rectifiers). When leakage occurs in these sections, the leakage current carries harmonic components related to the load characteristics of that section, resulting in differences in the content of specific harmonics (such as the 3rd, 5th, and 7th harmonics) in the zero-sequence current signal measured at the neutral point. Third, there are differences in waveform dynamic characteristics. Different types of leakage faults (such as high-resistance grounding, low-resistance grounding, and intermittent arc grounding) will produce different waveform characteristics. Simultaneously, the line environmental conditions of different sections (such as mixed sections of overhead lines and cables, sections affected by tree obstructions, etc.) will also affect the dynamic fluctuation characteristics of the leakage current. These differences will be reflected in the waveform distortion rate and dynamic fluctuation coefficient of the zero-sequence current signal. Based on the above physical mechanisms, statistical analysis of a large number of historical fault cases can establish a statistical correlation between the combination of characteristic parameters and the faulty section. The preset evaluation model utilizes this statistical correlation, combining the currently measured characteristic parameters and the topology information of the target distribution network, to calculate the potential leakage probability of each line section.

[0059] The pre-defined evaluation model refers to a pre-established and configured mathematical or algorithmic model used to perform specific analysis or prediction tasks. It can be obtained by training historical data using machine learning algorithms (such as neural networks, support vector machines (SVM), and decision trees), or it can be a rule set or fuzzy logic system built based on expert knowledge and physical mechanisms. The potential leakage probability represents a value between 0 and 1 (or 0% to 100%), used to quantify the likelihood of a leakage event occurring in a certain line section under its current condition. A higher probability value means a greater risk of leakage in that section.

[0060] Based on the above embodiments, as an optional embodiment, step S203 can be implemented through steps S2031-S2033, which will be explained in detail below.

[0061] S2031. Input a set of feature parameters into the preset evaluation model. The preset evaluation model is an evaluation model trained based on historical fault data. The preset evaluation model includes a preset feature region association database, which stores the historical association weights of different feature parameter combinations and each line segment.

[0062] Specifically, the structured set of feature parameters obtained from the aforementioned steps is provided as input data to a pre-set software module with learning and reasoning capabilities—a preset evaluation model. The process of establishing the preset evaluation model includes: collecting historical fault case data of the target distribution network or similar distribution networks. Each case data includes zero-sequence current waveform data recorded at the neutral grounding down conductor when the fault occurred, as well as the actual fault section identifier confirmed on-site; secondly, the zero-sequence current waveform data of each historical case is processed with the same feature parameter extraction method as described in this article to obtain the feature parameter combination corresponding to the case; then, the correlation frequency between various feature parameter combinations and each line section is statistically analyzed, and the conditional probability is calculated, that is, the probability distribution of the fault being located in each line section when a certain feature parameter combination occurs; finally, the above statistical results are stored as historical correlation weights in the preset feature area correlation database. The essence of this model is a data-driven decision engine. Its internal knowledge comes from the analysis and summarization of a large amount of historical fault data. That is, through the training process, it learns the statistical correlation between feature parameter patterns and fault locations. The specific implementation architecture of this model includes a core component: a pre-set feature region association database. This database does not store raw waveform data, but is a refined knowledge base. It stores a large number of "if-then" association rules in a structured way. Specifically, it shows that there is a quantitative relationship between different combinations of feature parameters (i.e., various possible feature vector instances) and the fault probability of each line section in the distribution network. This relationship is recorded in the form of historical association weights. The weight values ​​reflect the frequency or confidence level of fault location in a specific line segment when a certain combination of feature parameters appears in historical data. When a new set of feature parameters is input, similarity matching or pattern lookup is performed in the feature-region association database to find the historical feature combination most similar to the current input features. The line segments associated with these similar historical combinations and their weights are retrieved and aggregated. Based on the aggregation results (e.g., weighted summation or probability calculation of all relevant weights for each line segment), an assessment output on the probability of faults in each line segment for the current input is generated. Therefore, the output of this step is not direct geographic coordinates, but an intermediate inference result containing spatial probability distribution, which directly prepares for the final generation of a list of potential leakage areas.

[0063] Historical fault data refers to a collection of verified and recorded case data accumulated from past operation and maintenance practices. Each case typically includes electrical characteristic data (similar feature parameters) at the time of the fault and the finally confirmed fault location (line segment). The pre-defined feature region association database is a data structure used within the evaluation model to store knowledge of feature pattern-spatial location associations. It can be a relational database table, lookup table, weight matrix, or other form of knowledge representation. Historical association weights are numerical values ​​assigned to each pair of feature parameter combinations and line segments in the pre-defined feature region association database. These values ​​quantify the strength or probability of their co-occurrence in historical data; a higher weight indicates a stronger association.

[0064] For example, the analysis and control module obtains a set of feature parameters [0.9A, 1800, 12%, 0.6]. This vector is input into a preset evaluation model. The preset feature region association database inside the model stores many records, for example: Record 1: Feature combination [0.85A, 1500, 10%, 0.5] → associated {segment A: weight 0.7, segment C: weight 0.2}; Record 2: Feature combination [0.95A, 2000, 15%, 0.7] → ​​associated {segment A: weight 0.3, segment B: weight 0.8}; The evaluation model calculates the similarity (e.g., Euclidean distance) between the current feature [0.9, 1800, 12, 0.6] and Record 1 and Record 2. If a segment is found to be similar to both, a weighted approach is used to aggregate the correlation information between the two. For example, a weight vector: [segment A: 0.65, segment B: 0.30, segment C: 0.05], indicates that based on the current features, segment A has the strongest fault correlation.

[0065] S2032. Based on the historical correlation weights and electrical parameters of each line segment in the network topology data, obtain the potential leakage probability corresponding to each line segment output by the preset evaluation model.

[0066] Specifically, based on the historical association weight set extracted from historical fault data and stored in the preset feature area association database, and the network topology data describing the physical structure and attributes of the target distribution network obtained through step S202, which includes electrical parameters (such as line length, impedance, and ground capacitance) of each line segment, the preset evaluation model no longer relies solely on a single feature-region statistical association when making probability outputs. Instead, it incorporates network topology data as an important correction factor or weighting coefficient into the calculation. The model retrieves matching historical association weights from the preset feature area association database based on the currently input feature parameters to obtain the baseline suspicion score for each line segment. Then, it reads the electrical parameters corresponding to each segment, such as the line insulation level, historical fault frequency, and ground capacitance, and uses these parameters as weight adjustment factors to correct the baseline suspicion score. The correction logic can be: for segments where electrical parameters indicate a high risk of insulation aging (such as abnormally increased ground capacitance), even if the historical association weight is not high, their suspicion score can be appropriately increased; conversely, for segments where electrical parameters indicate a good condition, their suspicion score can be appropriately decreased. After correction, the model normalizes the corrected scores and finally obtains and outputs a set of potential leakage probabilities corresponding to each line section. This probability value comprehensively reflects information from two dimensions: "the similarity between current electrical characteristics and historical fault modes" and "the inherent risk of the current section based on its physical characteristics," making the probability assessment more comprehensive and accurate.

[0067] S2033. The line section with a potential leakage probability greater than the preset probability threshold is taken as the potential leakage area, and the signal analysis results are obtained.

[0068] Specifically, a pre-set judgment criterion is invoked—a preset probability threshold. This threshold is a key threshold value, usually set based on operational experience, security level requirements, or historical false alarm rate analysis. The analysis and control module compares the potential leakage probability of each line segment in the list with the preset probability threshold one by one. For any line segment, if its potential leakage probability is greater than (or greater than or equal to) the preset probability threshold, the line segment is judged to have significant risk and is thus screened out as a potential leakage area. All the line segments that meet the criteria after screening constitute the set of suspected targets identified in this signal analysis. Finally, the module formats and encapsulates this set of suspected targets to obtain the final signal analysis result. This result is usually a structured data object, such as a list or array, which clearly lists the identifiers of all line segments judged to be potential leakage areas, and may be accompanied by their corresponding probability values ​​or other related information (such as geographical coordinate range). This signal analysis result no longer contains low-probability segment information below the threshold, thus achieving information focus and purification, marking the end of the "analysis" stage, and serving as the direct data source for "predictive information" to be passed to the subsequent "directional detection" step.

[0069] The preset probability threshold is a predefined numerical threshold used to distinguish between "high-risk sections that require attention" and "low-risk sections that can be temporarily ignored" before the judgment is made. It is a configurable parameter. Potential leakage areas refer to a set of one or more specific line sections that, after screening, are considered to have a high probability of leakage faults and require further key detection or investigation.

[0070] S103. When the signal analysis results indicate the existence of a potential leakage area, leakage prediction information is generated.

[0071] Specifically, the signal analysis results are parsed and judged to check whether they indicate the existence of at least one potential leakage area. If the check result is yes, that is, a suspected area requiring attention is confirmed, the specific information contained in the signal analysis results (such as the identifier of the suspected area, possible location description, associated probability value, etc.) is integrated and formatted into a clear instruction or data package, i.e., leakage prediction information is generated. This leakage prediction information is an actionable version of the signal analysis results, which not only includes a list of suspected areas, but may also include control parameter suggestions, priority indicators, or concise natural language descriptions for subsequent steps. Its purpose is to be directly received and used by the subsequent "directional collaborative detection" step, serving as a direct basis for initiating signal injection and directing the action of the sensing unit. If the judgment result is no, that is, the signal analysis results indicate that there are no potential leakage areas exceeding the threshold, then leakage prediction information is usually not generated, or a special message indicating "no abnormality" is generated, and the entire process may terminate at this point or enter a standby state.

[0072] Among them, leakage current prediction information refers to a data packet or instruction created by the analysis and control module. Its content is based on the signal analysis results and is used to directly guide the subsequent "signal injection" and "cooperative detection" steps. It usually includes a list of areas to be investigated and possible action parameters.

[0073] S104. Based on the leakage current prediction information, the analysis and control module controls the signal injection module to inject characteristic detection signals into the target power distribution network, and controls the high-altitude signal sensing unit 30 and the auxiliary environmental sensing unit 40 to work together to perform signal detection in the potential leakage current area to obtain leakage current detection data.

[0074] Specifically, the analysis and control module, acting as the command center, activates the signal injection module based on leakage current prediction information. This generates a characteristic detection signal with specific frequency, encoding, or waveform features, distinct from the normal power frequency signal of the power grid. This signal is then coupled and injected into the source end or designated node of the target distribution network. Immediately after injection, based on the geographical area indicated by the prediction information (i.e., the potential leakage area), the module precisely schedules and activates high-altitude signal sensing units 30 (such as sensors mounted on drones) and auxiliary environmental sensing units 40 on the ground or equipment deployed in that area. These different types of sensor units work collaboratively: the high-altitude units focus on capturing, non-contactly, the electromagnetic field characteristics generated by the injected signal propagating along the line and leaking at the leakage point over a large area; the auxiliary environmental units simultaneously collect key environmental parameters such as humidity, temperature, and sound in the area. The entire detection process is highly synchronized, meaning the sensor's acquisition window precisely matches the duration of the signal injection to ensure that the captured data is the response to the injected signal. Finally, all the raw information collected by the sensors is integrated to form a multi-dimensional, interconnected leakage current detection dataset.

[0075] Among them, the feature detection signal refers to an artificially designed electrical signal with unique identifiable characteristics, such as a sine wave of a specific frequency (e.g., 10kHz) or a specially coded pulse sequence, with the aim of being easily identified and extracted in complex power grid background noise. The leakage current detection data represents the collection of all raw data acquired in this active detection mission; it is a composite dataset containing electromagnetic field strength data from the high-altitude sensing unit, temperature and humidity data from the auxiliary environmental sensing unit 40, etc.

[0076] Based on the above embodiments, as an optional embodiment, for Figure 2 The step S104 shown, which controls the high-altitude signal sensing unit 30 and the auxiliary environmental sensing unit 40 to cooperate in signal detection within the potential leakage area to obtain leakage detection data, can be implemented through steps S301-S304, which will be explained in detail below.

[0077] S301, control the high-altitude signal sensing unit 30 to perform non-contact signal sensing on the overhead lines in the potential leakage area, and obtain the first characteristic signal strength data of each line section.

[0078] Specifically, based on the received leakage current prediction information, the geographical or electrical range of the potential leakage area to be detected is analyzed. A control command containing the area boundary, scanning path, or target line identifier is sent to the high-altitude signal sensing unit 30 via a wireless communication link. Upon receiving the command, the high-altitude signal sensing unit 30 (typically handheld by maintenance personnel or installed on auxiliary equipment) is operated to the designated potential leakage area and moves along the actual overhead line path within that area. Throughout the movement, the core sensing component of the unit (such as a clamp-on current transformer or a high-sensitivity magnetic field antenna) continuously senses the magnetic or electromagnetic field around the target line in a non-contact manner. It does not need to physically contact the conductor but detects the strength of the characteristic signal by sensing the alternating magnetic field generated by the flow of the characteristic detection current (previously injected by the signal injection module). The signal processing circuit inside the unit amplifies, filters, and demodulates the sensed raw signal, calculating the effective strength value of the characteristic signal in real time. Simultaneously, the sensing unit or its associated positioning system (such as GPS or Bluetooth beacon positioning) records or marks the location information corresponding to each strength reading. After traversing all relevant line segments within the area, the sensing unit integrates the collected time-location-intensity data and transmits it back to the analysis and control module via a wireless link. This data constitutes the first characteristic signal intensity data, essentially a dataset where each record corresponds to a detection point (or a small line segment), containing at least the line segment identifier and the characteristic signal intensity value measured at that point. This data depicts the distribution profile of the characteristic signal across the overhead lines in the entire suspected area, providing a basis for identifying signal abrupt changes (potentially indicating fault locations).

[0079] Overhead lines refer to power transmission conductors erected above the ground via poles, towers, or other supports. Non-contact signal sensing is a measurement technique where the sensing component does not have direct electrical or physical contact with the conductor being measured; instead, it acquires signal information by sensing the electromagnetic field in the space surrounding the conductor. The first characteristic signal strength data refers to the set of raw measurement data regarding the strength of the characteristic detection signal at various points on the overhead line, collected and reported by the high-altitude signal sensing unit 30 in this step.

[0080] S302. Based on the first feature signal intensity data, determine the target segment to be detected.

[0081] Specifically, the system analyzes the first characteristic signal strength data from the high-altitude signal sensing unit 30. This dataset typically contains characteristic signal strength values ​​measured at multiple detection points (corresponding to different line sections or locations) along the overhead line within the potential leakage area. Analyzing this data helps identify abnormal patterns in the signal strength distribution, particularly points where signal strength changes significantly and discontinuously. This can be achieved by calculating the signal strength difference or attenuation rate between adjacent detection points; observing the curve of signal strength change with location to identify steep falling or rising edges (abrupt change points); or comparing it with a preset background strength model to identify continuous sections with abnormally high or low strength. Through this analysis, one or a few narrow intervals where signal characteristics abruptly change can be located within the entire suspected area. These intervals are identified as the target section to be detected. The definition of this target section is more precise than the initial "potential leakage area," and may only be a small segment of the line, such as between two adjacent poles, with its range explicitly limited to the start and end points of the signal abrupt change. Once the target segment is identified, the analysis and control module generates a precise positioning command containing the segment's start and end locations, route markers, or geographic coordinates. This command directly guides the next step—ground-based near-field precision surveying performed by the auxiliary environmental sensing unit 40—ensuring that subsequent detection resources are deployed most efficiently to the smallest area with the highest probability of failure. Therefore, this step completes the refinement and convergence from "regional-level" suspicion to "segment-level" targets.

[0082] The target section to be detected refers to the specific line segment that has been identified after the analysis of the first characteristic signal strength data, and whose signal performance is abnormal (such as sudden change) and therefore selected for further refined and close-range detection.

[0083] S303, the control auxiliary environment sensing unit 40 performs near-field signal sensing on the ground environment objects associated with the target section and obtains the second characteristic signal intensity data of each ground environment object.

[0084] Specifically, based on the identified target section (e.g., "between pole 8 and pole 9"), ground environment objects physically closely related to that section are identified. These objects typically include: poles (pole body, crossarm) at both ends of the section, guy wires connecting the poles, grounding devices under the line, nearby trees, building walls, or any object that may come into electrical contact with the line due to insulation damage. Subsequently, control commands are sent to the auxiliary environmental sensing unit 40 via a wireless communication link. These commands contain a list of objects or spatial ranges to be detected. Maintenance personnel operate the auxiliary environmental sensing unit 40, sequentially approaching or contacting each designated or suspected ground environment object. The auxiliary environmental sensing unit 40 operates in a near-field signal sensing mode, where its sensing probe can come very close to, or even briefly contact, the surface of the object being measured, detecting with extremely high sensitivity the characteristic frequency electromagnetic field generated by leakage current flowing on the object or in the surrounding space. For each detected ground environment object, the sensing unit measures and records the characteristic signal intensity at that point, thereby obtaining a series of discrete, high-precision intensity readings. These readings constitute the second characteristic signal intensity data. Unlike the first characteristic signal strength data, which is continuously scanned along the line, the second characteristic signal strength data is a concentrated measurement of discrete points. It is characterized by a high signal-to-noise ratio and strong localization, and can clearly reflect the "convergence point" or "strongest radiation source" of the fault current in the ground environment. All data is transmitted back to the host 10 wirelessly, providing the most direct evidence for the final determination of the precise fault location.

[0085] The second characteristic signal intensity data refers to the set of intensity measurement data of the characteristic detection signal on or near various associated ground environmental objects, which is collected and reported by the auxiliary environmental sensing unit 40 in this step.

[0086] S304. Based on the first characteristic signal strength data and the second characteristic signal strength data, leakage current detection data is obtained.

[0087] Specifically, based on the first characteristic signal strength data describing the continuous spatial distribution of characteristic signals along overhead lines, and the second characteristic signal strength data describing the intensity concentration of characteristic signals at discrete key points on the ground, data fusion and structured encapsulation operations are performed. This involves associating each ground measurement point in the second data with its corresponding spatial line segment (derived from the geographical or logical identifier of the first data), establishing a mapping relationship between "spatial points" and "online sections." The two sets of data are then organized according to a unified format (such as timestamps, location identifiers, signal strength values, data source tags, etc.) and merged into a single overall data structure (such as a list, data table, or a specific object). Contextual information related to the detection may be added, such as the detection time, target distribution network identifier, and the frequency of the characteristic signal used. After this series of processing steps, a comprehensive dataset with a complete structure and complementary information is obtained—the leakage current detection data. This data not only includes the macroscopic attenuation trend of the signal along the "line" (used to locate the fault section) but also the microscopic intensity peak at the "point" (used to locate the specific fault object), with the two spatially correlated and mutually corroborating each other. This integrated dataset marks the end of the field data acquisition phase, encompassing complete detection results from high altitude to ground and from continuous to discrete data, providing a unique and sufficient input for performing the core analysis task of "determining the location of the leakage fault point".

[0088] Figure 3 This is another flowchart illustrating the low-voltage leakage fault location method provided in this application embodiment.

[0089] Based on the above embodiments, as an optional embodiment, see [link to embodiment]. Figure 3 ,exist Figure 2 Before step S104 shown, steps S401-S403 may also be included, which will be explained in detail below.

[0090] S401, control the high-altitude signal sensing unit 30 or the auxiliary environmental sensing unit 40 to perform background electromagnetic spectrum scanning in the area to be detected, and obtain the background noise intensity of each frequency point within the preset frequency range.

[0091] Specifically, based on the terrain features of the area to be detected, historical electromagnetic environment data, or user presets, a preset frequency range of interest is determined. Then, a control command is sent via a wireless communication link to either the high-altitude signal sensing unit 30 or the auxiliary environmental sensing unit 40. This command includes the task requirement of performing a background electromagnetic spectrum scan and the specified preset frequency range. Upon receiving the command, the sensing unit (e.g., prioritizing units with higher sensitivity or better positioning) enters scanning mode. At its current position and attitude, it ceases the detection of actively injected signals and instead acts as a high-sensitivity broadband receiver. Within the specified preset frequency range, it rapidly measures the signal strength of the environmental electromagnetic signals captured by its sensing elements, sequentially or in parallel, at a certain frequency resolution (step size). This measurement process aims to record the electromagnetic energy inherent in the environment, generated by non-injected signals, i.e., the background noise intensity. After completing the scan of the entire frequency band, the sensing unit packages each frequency point and its corresponding background noise intensity value into a data list and sends it back to the analysis and control module. The data acquired by the module is a "noise spectrum" depicting the distribution of environmental electromagnetic noise levels within the preset frequency range at a specific time and in a specific area to be detected.

[0092] The area to be detected refers to the physical space where leakage current location detection will be performed, which may coincide with or be related to the aforementioned "potential leakage current area". The preset frequency range refers to a continuous frequency interval that needs to be scanned and evaluated before the scan begins, such as 1kHz to 5kHz. A frequency point refers to a specific frequency value selected within the preset frequency range at certain intervals (e.g., 100Hz steps). Background noise intensity refers to the environmental electromagnetic signal intensity value measured at a specific frequency point without the injection of a feature detection signal, reflecting the inherent electromagnetic interference level at that frequency point.

[0093] S402. Based on the background noise intensity and the signal propagation efficiency coefficient at each frequency point, calculate the candidate score for each frequency point. The candidate score is negatively correlated with the background noise intensity and positively correlated with the signal propagation efficiency coefficient.

[0094] Specifically, the system acquires background noise intensity datasets for each frequency point, reflecting environmental interference levels, obtained through spectrum scanning, as well as signal propagation efficiency coefficient datasets for each frequency point, pre-stored in the system, reflecting the propagation performance of signals of different frequencies in typical distribution network lines. For each specific frequency point within the preset frequency range, the module reads its corresponding background noise intensity value (denoted as N). i ) and signal propagation efficiency coefficient (denoted as E) i Then, a pre-defined calculation formula or algorithm is applied, combined with these two parameters, to calculate the candidate score (denoted as S) for that frequency point. i The core principle of the calculation is clearly defined as: candidate score Si With background noise intensity N i Negative correlation means that the noise N i The larger the value, the higher the score (S). i It should tend to decrease; compared with the signal propagation efficiency coefficient E i A positive correlation means that the communication effectiveness E i The higher the score, the better. i The tendency should be towards increasing. A typical, simple implementation is a weighted linear combination, for example: S i =w1*f(N i )+w2*g(E i ), where the function f(N) i ) is N i A decreasing function (such as taking the reciprocal or negative value), function g(E) i ) is E i The increasing function (e.g., directly taking values) is used, with w1 and w2 as weighting coefficients. More complex models may consider nonlinear relationships. After completing the calculations for all frequency points, the analysis and control module obtains a candidate score list for each frequency point. This list assigns a single quantitative score to each frequency point, combining "anti-interference capability" (low noise) and "propagation advantage" (high efficiency). The score directly characterizes the expected overall performance of selecting that frequency point as the operating frequency, providing a clear and objective decision-making basis for the next step of selecting the optimal frequency point.

[0095] The signal propagation effectiveness coefficient is a dimensionless parameter, determined in advance or calculated through a model, characterizing the propagation capability of a current signal at a specific frequency in an overhead line of a target type of distribution network. A higher coefficient generally indicates less attenuation and stronger signal-to-noise ratio maintenance during propagation at that frequency. The candidate score is a quantitative value calculated for each frequency point, representing its suitability as a detection signal frequency, after a comprehensive evaluation of background noise intensity and the signal propagation effectiveness coefficient.

[0096] S403. Configure the frequency point corresponding to the maximum candidate score as the working frequency of the feature detection signal.

[0097] Specifically, the candidate score list is traversed and compared to find the element with the largest candidate score value, i.e., the maximum candidate score. The frequency corresponding to the maximum candidate score is then set as the operating frequency of the feature detection signal that the signal injection module will generate and inject into the power grid. The relevant parameters of the oscillator, frequency synthesizer, or waveform generator of the signal injection module will be adjusted to ensure that the fundamental frequency or center frequency of its output signal is precisely equal to the selected frequency. After configuration, this operating frequency will remain unchanged throughout the subsequent directional collaborative detection phase, serving as a "feature tag" for the entire system's identification and tracking.

[0098] Among them, the operating frequency refers to the frequency of the most important periodic changes in the feature detection signal, and is one of the core identification parameters of the signal.

[0099] Through the above embodiments, the optimal frequency point is fixed as the working frequency, providing a unified and clear signal tracking target for all subsequent detection units, ensuring the frequency consistency of the entire collaborative detection network, and avoiding detection failures caused by frequency mismatch.

[0100] S105. Analyze the leakage detection data through the analysis and control module to determine the location of the leakage fault point in the target distribution network.

[0101] Specifically, the system receives and integrates "leakage detection data" containing information such as high-altitude electromagnetic field intensity distribution and ground environmental parameters (e.g., temperature, humidity, sound). It processes the data transmitted from the high-altitude signal sensing unit 30, plots the energy intensity distribution of the characteristic detection signal on a geographic map of the potential leakage area, and automatically searches for intensity peak points or anomalous abrupt changes—these points are the locations most suspected of leakage. Simultaneously, it analyzes data from the auxiliary environmental sensing unit 40 to find environmental parameter anomalies consistent with leakage physical phenomena (e.g., partial discharge causing air ionization, temperature rise, or the generation of specific sounds). The system then overlays and compares these two or more analysis results spatially. By calculating the overlap or correlation between signal intensity peak points and environmental parameter anomalies in geographic location, false signals caused by environmental interference or single sensor errors can be largely eliminated. When evidence from multiple dimensions points to the same precise geographic coordinates, the system determines these coordinates with extremely high confidence as the final leakage fault point and outputs its precise location information.

[0102] A leakage fault point refers to the specific physical location on a power distribution line where current abnormally leaks to ground, such as a damaged insulator or a point where a tree touches a conductor. The location is used to indicate the precise coordinates of the fault point in geographic space, usually expressed in latitude and longitude, or in the form of "line XX, tower number XX".

[0103] Through the above embodiments, the fusion analysis of multi-source data enhances the reliability and anti-interference capability of diagnostic results, effectively avoiding misjudgments that may be caused by a single information source, ensuring the accuracy of conclusions, and precisely narrowing the scope of fault investigation from a "segment" of tens or even hundreds of meters to a "point" at the sub-meter level, achieving true "precise positioning" or "point-to-point elimination." This high-precision positioning result eliminates the need for maintenance personnel to conduct large-scale line inspections and investigations, allowing them to directly go to the fault point for handling, thereby greatly shortening repair time, reducing maintenance costs, quickly restoring the normal operation of the power grid, and ensuring the reliability of power supply.

[0104] Based on the above embodiments, as an optional embodiment, for Figure 2 The step S105 shown can be implemented through steps S501-S503, which will be explained in detail below.

[0105] S501. Based on the current flow direction of the line within the potential leakage area, sort the first characteristic signal intensity data corresponding to each detection point in the leakage detection data.

[0106] Specifically, from the identified potential leakage areas, the current flow direction of overhead lines within that area is extracted or deduced. This flow direction information is usually pre-determined based on the topology of the distribution network, the location of the power source (transformer), or the phase relationship of the lines. After clarifying the physical direction of the current flowing from "upstream" to "downstream," the module analyzes the integrated leakage detection data and specifically extracts the first characteristic signal strength data corresponding to each detection point collected by the high-altitude signal sensing unit 30. These data points may originally only have location identifiers (such as tower numbers). Location markers are mapped onto the line topology to determine their order along the current path. For current, it flows from the power source (upstream) to the load side (downstream). Along this direction, the module arranges all detection points sequentially. For example, if the current flows from pole 5 to pole 12, the detection points will be arranged in the order of pole 5, pole 6, pole 7... pole 12 (or in reverse order, depending on the sorting direction definition). After sorting, the previously scattered signal strength data, which was only bound to location, is organized into an ordered sequence arranged according to the physical order of signal propagation. This sequence visually demonstrates the continuous change trend of characteristic signal strength from the start point (upstream) to the end point (downstream) of the line, laying a direct and physically logical data foundation for the next step of identifying signal abrupt change points (before and after the fault point).

[0107] The current flow direction within the potential leakage area refers to the normal flow direction of electrical current within the previously designated area requiring focused monitoring, typically from the power source (transformer) to the load end. The detection point refers to the spatial location point corresponding to the measurement performed by the high-altitude signal sensing unit 30 in the leakage detection data.

[0108] S502. When the signal strength attenuation rate between adjacent detection points is greater than the preset attenuation threshold, the area between the two adjacent detection points corresponding to the signal strength attenuation rate is determined as the fault location interval.

[0109] Specifically, the analysis and control module has obtained an ordered sequence of first characteristic signal strength data sorted by current flow direction. It traverses this sequence in a step-by-step manner. For each pair of adjacent detection points, it calculates the signal strength attenuation rate between them. The attenuation rate can be calculated using the following formula: Attenuation rate = (Intensity of the previous point - Intensity of the next point) / Intensity of the previous point. The result is a percentage or decimal representing the degree of signal attenuation, reflecting the relative proportion of strength loss of the characteristic signal as it flows from the upstream point to the downstream point. Each calculated attenuation rate value is then compared with a predefined threshold value—a preset attenuation threshold—used to determine "significant attenuation." This threshold is set based on a large amount of attenuation data and experience from normal lines, used to distinguish between normal line losses and abnormally large attenuations caused by faults. When the module finds that the signal strength attenuation rate between a pair of adjacent detection points is greater than the preset attenuation threshold, it determines that an abnormal and drastic signal change has occurred at that point. This change indicates that the characteristic signal, during its flow from the previous detection point to the next, may have encountered significant "leakage" or "bypass" in the line segment between them, i.e., a fault point. Therefore, the module immediately identifies the region between two adjacent detection points corresponding to the signal strength attenuation rate (i.e., the line segment defined by these two detection points) as the fault location interval. For example, if the attenuation rate between point i and point i+1 exceeds the limit, then the physical line segment between "the location of point i" and "the location of point i+1" is marked as the final, highly suspected fault location interval. This interval is typically much smaller than the initial potential leakage area.

[0110] In this context, adjacent detection points refer to two data points that are immediately preceding and following each other in a sequentially arranged signal strength data sequence. Signal strength attenuation rate refers to the percentage decrease in characteristic signal strength value from one detection point to the next, usually expressed as a percentage. Preset attenuation threshold is a critical percentage value set before analysis to determine whether signal attenuation represents an abnormal abrupt change. Fault location interval refers to the specific line segment identified in this step as most likely containing the precise leakage fault point.

[0111] S503. The location of the ground environment object corresponding to the maximum intensity value among all second characteristic signal intensity data within the fault location interval is determined as the location of the leakage fault point.

[0112] Specifically, based on the fault location interval (e.g., "the line between poles 8 and 10"), all second characteristic signal strength data associated with the line located within this fault location interval are filtered from the complete leakage current detection data. This data is obtained by the auxiliary environmental sensing unit 40 through near-field scanning of suspected ground objects in the previous stage. The filtered data is then analyzed to find the maximum intensity value, which is achieved by iterating through and comparing the "intensity" field of each data point. Once the maximum intensity value is identified, the module reads the corresponding attribute field from the maximum value data record—that is, which ground environmental object (e.g., "the B-phase guy wire of pole 9") measured the intensity value. The location represented by this object is considered the strongest divergence or convergence point of the characteristic signal in the ground dimension. Finally, this specific and precise physical location (e.g., "the connection point between the B-phase guy wire and the crossarm of pole 9" or the GPS coordinates of the object) is determined as the location of the leakage current fault. This location is the final output of the entire process; it is no longer an interval or list, but a single, clear physical point that maintenance personnel can directly access and inspect.

[0113] The low-voltage leakage fault location system in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference needed]. Figure 4 , Figure 4 This is a schematic diagram of the low-voltage leakage fault location system provided in the embodiments of this application.

[0114] It should be noted that, Figure 4 The structure of the low-voltage leakage fault location system shown is merely an example and should not impose any limitations on the functionality and scope of application of the embodiments of the present invention.

[0115] like Figure 4 As shown, the low-voltage leakage fault location system includes a central processing unit 601, which can perform various appropriate actions and processes based on a program stored in a read-only memory 602 or a program loaded from a storage section 608 into a random access memory 603, such as performing the methods described in the above embodiments. The random access memory 603 also stores various programs and data required for system operation. The central processing unit 601, the read-only memory 602, and the random access memory 603 are interconnected via a bus 604. An input / output interface 605 is also connected to the bus 604.

[0116] The following components are connected to the input / output interface 605: an input section 606 including audio input devices, push-button switches, etc.; an output section 607 including a liquid crystal display (LCD) and audio output devices, indicator lights, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the input / output interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 610 as needed so that computer programs read from it can be installed into the storage section 608 as needed.

[0117] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 609, and / or installed from removable medium 611. When the computer program is executed by central processing unit 601, it performs the various functions defined in the present invention. It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0118] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, program segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those shown in the drawings.

[0119] Specifically, the low-voltage leakage fault location system of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, it implements the low-voltage leakage fault location method provided in the above embodiment.

[0120] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the low-voltage leakage fault location system described in the above embodiments; or it may exist independently and not assembled into the low-voltage leakage fault location system. The storage medium carries one or more computer programs, which, when executed by a processor of the low-voltage leakage fault location system, cause the low-voltage leakage fault location system to implement the low-voltage leakage fault location method provided in the above embodiments.

[0121] The above description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A method for locating low-voltage leakage faults, characterized in that, The method is applied to a low-voltage leakage current locator, which includes a main unit (10), a flexible zero-sequence current detection unit (20), a high-altitude signal sensing unit (30), and an auxiliary environmental sensing unit (40). The main unit (10) includes a signal injection module and an analysis and control module. The signal output terminal (21) of the flexible zero-sequence current detection unit (20) is detachably electrically connected to the signal input terminal (11) of the main unit (10) via an adapter cable. The high-altitude signal sensing unit (30) and the auxiliary environmental sensing unit (40) are wirelessly connected to the main unit (10) respectively. The method includes: The zero-sequence current signal at the grounding lead of the neutral line on the low-voltage side of the distribution transformer of the target distribution network is obtained through the flexible zero-sequence current detection unit (20). The zero-sequence current signal is analyzed by the analysis and control module to obtain the signal analysis results; When the signal analysis results indicate the existence of a potential leakage area, leakage prediction information is generated. Based on the leakage current prediction information, the analysis and control module controls the signal injection module to inject a characteristic detection signal into the target power distribution network, and controls the high-altitude signal sensing unit (30) and the auxiliary environmental sensing unit (40) to work together to perform signal detection in the potential leakage current area to obtain leakage current detection data. The analysis and control module analyzes the leakage detection data to determine the location of the leakage fault point in the target power distribution network.

2. The method according to claim 1, characterized in that, The analysis of the zero-sequence current signal by the analysis and control module to obtain signal analysis results includes: The zero-sequence current signal is filtered and its feature parameters are extracted by the analysis and control module to obtain a set of feature parameters characterizing the signal characteristics. Obtain the network topology data of the target distribution network, the network topology data including the electrical parameters and section identifiers of each line section; The set of feature parameters and the network topology data are analyzed and processed by a preset evaluation model to generate the signal analysis results. The signal analysis results are used to identify the potential leakage probability of each line section in the target distribution network.

3. The method according to claim 2, characterized in that, The analysis and control module filters and extracts feature parameters from the zero-sequence current signal to obtain a set of feature parameters characterizing the signal properties, including: The analysis and control module performs power frequency filtering and bandpass filtering on the zero-sequence current signal to obtain a power frequency component and at least one preset non-power frequency component in a non-power frequency band. Calculate the effective value of the current of the power frequency component, and calculate the frequency energy value of the highest signal energy among the non-power frequency components of the at least one preset non-power frequency band; Based on the waveform sampling data of the zero-sequence current signal, the waveform distortion rate and dynamic fluctuation coefficient of the zero-sequence current signal are calculated. The dynamic fluctuation coefficient represents the ratio of the average difference between the peak value and the valley value of the zero-sequence current signal to the effective value of the signal within a preset time period. Based on the effective value of the current, the energy value of the frequency band, the waveform distortion rate, and the dynamic fluctuation coefficient, the set of characteristic parameters is obtained.

4. The method according to claim 2, characterized in that, The step of analyzing and processing the set of feature parameters and the network topology data using a preset evaluation model to generate the signal analysis results includes: The set of feature parameters are input into the preset evaluation model, which is an evaluation model trained based on historical fault data. The preset evaluation model includes a preset feature region association database, which stores different combinations of feature parameters and historical association weights of each line segment. Based on the historical association weights and the electrical parameters of each line segment in the network topology data, the potential leakage probability corresponding to each line segment is obtained from the output of the preset evaluation model. The line section with a potential leakage probability greater than a preset probability threshold is designated as the potential leakage area, and the signal analysis result is obtained.

5. The method according to claim 1, characterized in that, The control unit (30) and the auxiliary environmental sensing unit (40) work together to detect signals within the potential leakage area, obtaining leakage detection data, including: The high-altitude signal sensing unit (30) is controlled to perform non-contact signal sensing on the overhead lines in the potential leakage area to obtain the first characteristic signal strength data of each line section; Based on the first feature signal intensity data, the target segment to be detected is determined; The auxiliary environment sensing unit (40) is controlled to perform near-field signal sensing on the ground environment objects associated with the target section, and to obtain the second characteristic signal intensity data of each ground environment object; The leakage current detection data is obtained based on the first characteristic signal strength data and the second characteristic signal strength data.

6. The method according to claim 5, characterized in that, The step of analyzing the leakage current detection data through the analysis and control module to determine the location of the leakage current fault point in the target distribution network includes: Based on the current flow direction of the line in the potential leakage area, the first feature signal intensity data corresponding to each detection point in the leakage detection data are sorted. When the signal strength attenuation rate between adjacent detection points is greater than a preset attenuation threshold, the area between the two adjacent detection points corresponding to the signal strength attenuation rate is determined as the fault location interval. The location of the ground environment object corresponding to the maximum intensity value among all the second characteristic signal intensity data within the fault location interval is determined as the location of the leakage fault point.

7. The method according to claim 1, characterized in that, Before the signal injection module injects the feature detection signal into the target distribution network, the method further includes: Control the high-altitude signal sensing unit (30) or the auxiliary environment sensing unit (40) to perform background electromagnetic spectrum scanning in the area to be detected, and obtain the background noise intensity of each frequency point within the preset frequency range; Based on the background noise intensity and the signal propagation efficiency coefficient of each frequency point, a candidate score is calculated for each frequency point. The candidate score is negatively correlated with the background noise intensity and positively correlated with the signal propagation efficiency coefficient. Configure the frequency point corresponding to the maximum candidate score as the working frequency of the feature detection signal.

8. A low-voltage leakage fault location system, characterized in that, The low-voltage leakage fault location system includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store calculations. The computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the low-voltage leakage fault location system to perform the method as described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that, When the instruction is executed on the low-voltage leakage fault location system, the low-voltage leakage fault location system performs the method as described in any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product is run on the low-voltage leakage fault location system, the low-voltage leakage fault location system performs the method as described in any one of claims 1-7.

Images