Cable fault detection method and device based on intelligent cable fault testing device

Abstract

The application provides a cable fault detection method and a testing device based on an intelligent cable fault testing device, and the method comprises the following steps: connecting a fault cable and a neighboring healthy cable synchronously through a cable fault tester, and capturing active reflection signals of the fault cable and passive coupling induction signals of the healthy cable synchronously when a high-voltage pulse is applied. By comparing the characteristics of the cross-cable signals, the real fault signals can be accurately identified, and the coarse fault distance can be calculated accordingly, so that the ambiguity of traditional waveform interpretation is overcome. According to the spectral characteristics of the real fault signals, the optimal filtering parameters are adaptively configured for the acoustic-magnetic synchronous fixed-point instrument, and the signal-to-noise ratio of accurate positioning is improved. The accurate positioning is carried out near the coarse distance, and the corresponding position of the healthy cable is verified again to complete the closed-loop positioning of the fault in the mode of confirming that there is no acoustic-magnetic signal. The application realizes accurate extraction and positioning guidance of fault characteristics, and greatly improves the accuracy and efficiency of cable fault detection.

Description

Technical Field

[0001] This invention belongs to the field of cable fault detection technology, specifically relating to a cable fault detection method and device based on an intelligent cable fault testing device. Background Technology

[0002] As the lifeline of energy supply in modern society, the safe and stable operation of urban underground cable networks is of paramount importance. During long-term operation, cables inevitably experience faults due to insulation aging, external damage, and joint defects. Once a fault occurs, quickly and accurately locating the fault point and carrying out emergency repairs is crucial for restoring power supply and minimizing economic losses. Currently, the high-voltage flashover method is the most widely used cable fault location technology. Its basic process involves first using a cable fault tester to roughly measure the fault distance, and then using an acoustic-magnetic synchronous locator to perform precise acoustic or magnetic field localization near the roughly measured distance.

[0003] In practical applications, the reflected waveforms generated by faulty cables under high-voltage pulses are often very complex. These waveforms not only include the reflected signal from the fault point itself, but also reflective signals from multiple impedance mismatch points such as cable joints, branches, and terminals, as well as various environmental electromagnetic interferences. Operators need to rely on personal experience to distinguish the true fault reflection wave from a group of similar waveforms, which is prone to misjudgment and leads to significant deviations in the given coarse measurement range. Secondly, the acoustic-magnetic signal generated by the discharge at the fault point is usually very weak and easily drowned out by environmental noise such as traffic and construction. Manually adjusting the filtering parameters of the locator yields inconsistent results, often resulting in the discharge sound being inaudible due to a low signal-to-noise ratio, leading to repeated fruitless searches within the coarse measurement range. Summary of the Invention

[0004] This invention provides a cable fault detection method and apparatus based on an intelligent cable fault testing device to solve the aforementioned technical problems.

[0005] In a first aspect, the present invention provides a cable fault detection method based on an intelligent cable fault testing device, which is applied to the cable fault testing device, including a cable fault tester and an acoustic-magnetic synchronous positioning instrument. The method includes the following steps:

[0006] Connect the high-voltage flashover unit and the first signal acquisition channel of the cable fault tester to the target fault cable, and at the same time connect the second signal acquisition channel to a nearby healthy cable laid in parallel with the target fault cable.

[0007] A high-voltage pulse is applied to the target faulty cable through a high-voltage flashover unit, and the active reflection signal of the target faulty cable and the passive coupling induction signal of the adjacent healthy cable are captured simultaneously using the first signal acquisition channel and the second signal acquisition channel.

[0008] The coarse fault distance of the fault point in the target fault cable is calculated based on the active reflection signal and the time-domain reflection method.

[0009] Identify the real fault signals in active reflection signals and passive coupling induction signals by comparing cross-cable signal features;

[0010] The optimal acoustic filtering parameters for the acoustic channel of the acoustic-magnetic synchronization locator are configured based on the spectral characteristics of the actual fault signal.

[0011] The configured acoustic-magnetic synchronous locator is used to perform precise locating near the coarsely measured fault distance of the target fault cable, and secondary verification is performed at the corresponding location of the adjacent healthy cable to confirm the absence of acoustic-magnetic signal, thus completing the closed-loop location of the fault point.

[0012] Optionally, the step of simultaneously capturing the active reflected signal of the target faulty cable and the passive coupling induction signal of the adjacent healthy cable using the first signal acquisition channel and the second signal acquisition channel includes the following steps:

[0013] Set synchronous acquisition trigger thresholds and pre-trigger data buffers in the first and second signal acquisition channels;

[0014] The signal amplitude in the first signal acquisition channel is monitored in real time. When the signal amplitude exceeds the synchronous acquisition trigger threshold, high-frequency synchronous data acquisition of the first and second signal acquisition channels is started.

[0015] Based on the preset acquisition time window, the original waveform data of the first signal acquisition channel and the second signal acquisition channel after the high voltage pulse is generated are recorded respectively.

[0016] The acquired raw waveform data is digitally filtered and normalized to generate the active reflection signal of the target faulty cable and the passive coupling induction signal of the adjacent healthy cable.

[0017] Optionally, the step of identifying the actual fault signal in the active reflection signal and the passive coupling induction signal by comparing cross-cable signal features includes the following steps:

[0018] The energy attenuation gradient analysis module of the cable fault tester is called to calculate and compare the energy attenuation gradient of the active reflected signal and the passive coupled induction signal to obtain the energy source determination result.

[0019] The initial phase offset verification module of the cable fault tester is called to calculate and verify the initial phase offset of the active reflection signal and the passive coupling induction signal, and the coupling path confirmation result is obtained.

[0020] The results of energy source determination and coupling path confirmation are subjected to a logical AND operation to form the signal identification result;

[0021] Based on the signal recognition results, the active reflection signal is marked as the real fault signal, and the spectral characteristics of the real fault signal are extracted. At the same time, the passive coupling induction signal is marked as the interference signal, and the interference signal is suppressed.

[0022] Optionally, the step of calling the energy attenuation gradient analysis module of the cable fault tester to calculate and compare the energy attenuation gradients of the active reflected signal and the passively coupled inductive signal to obtain the energy source determination result includes the following steps:

[0023] Fast Fourier transforms were performed on the active reflected signal and the passively coupled induced signal, respectively, to obtain the amplitude-frequency response curves of the two signals;

[0024] The broadband total energy is obtained by calculating the active reflection signal and the passive coupling induction signal, and a first energy ratio is generated;

[0025] The amplitude-frequency response curves of the two signals are divided into multiple pre-defined non-overlapping narrowband frequency bands.

[0026] Calculate the signal energy in each narrowband frequency band and generate the second energy ratio for each frequency band;

[0027] The energy decay gradient is obtained by analyzing the trend of the second energy ratio changing with frequency.

[0028] The first energy ratio and energy decay gradient are compared with the preset coupling model threshold, and the energy source determination result is output.

[0029] Optionally, the step of calling the initial phase offset verification module of the cable fault tester to calculate and verify the initial phase offset of the active reflection signal and the passive coupling induction signal to obtain the coupling path confirmation result includes the following steps:

[0030] The active reflection signal and the passively coupled induction signal are upsampled using an interpolation algorithm;

[0031] Calculate the first derivatives of the active reflected signal and the passively coupled induced signal after upsampling, and determine the starting time point of the signal waveform based on the first non-zero point of the first derivative;

[0032] Calculate the time difference between the starting time points of the active reflection signal and the passive coupling induction signal to obtain the absolute time value of the initial phase shift;

[0033] The absolute time value is converted into a phase angle based on the main frequency of the actual fault signal;

[0034] The phase angle is compared with the theoretical phase offset range set according to the cable spacing and medium parameters, and the coupling path confirmation result is output.

[0035] Optionally, configuring the optimal acoustic filtering parameters for the acoustic channel of the acoustic-magnetic synchronization locator based on the spectral characteristics of the actual fault signal includes the following steps:

[0036] The high-voltage flashover unit triggers multiple high-voltage flashovers within the area determined by the rough measurement of the fault distance, and the magnetic field signal and continuous on-site acoustic signal are synchronously collected by the acoustic-magnetic synchronization locator.

[0037] Based on the triggering time of each magnetic field signal, multiple synchronous acoustic samples are extracted from the continuous field acoustic signals. Each synchronous acoustic sample contains fault discharge sound and background noise.

[0038] Multiple asynchronous acoustic samples that do not contain fault discharge sounds are extracted from continuous on-site acoustic signals and used as background noise samples.

[0039] Calculate the average spectrum of the synchronous acoustic sample and the average spectrum of the background noise sample, respectively.

[0040] By comparing the average spectrum, the characteristic frequency bands where the signal-to-noise ratio of fault discharge sound is higher than that of background noise are identified, and the optimal acoustic filtering parameters for the acoustic channel of the acoustic-magnetic synchronization locator are dynamically generated based on the characteristic frequency bands.

[0041] Optionally, the step of identifying the characteristic frequency band where the signal-to-noise ratio of the fault discharge sound is higher than that of the background noise by comparing the average spectrum, and dynamically generating the optimal acoustic filtering parameters for the acoustic channel of the acoustic-magnetic synchronization locator based on the characteristic frequency band, includes the following steps:

[0042] The estimated spectrum of the pure fault sound is obtained by subtracting the average spectrum of the synchronous acoustic sample from the average spectrum of the background noise sample.

[0043] Calculate the signal-to-noise ratio (SNR) between the estimated spectrum of the clean fault sound and the average spectrum of the background noise sample, and generate an SNR spectrum map covering the entire frequency band;

[0044] Search the signal-to-noise ratio spectrum for all frequency bands whose signal-to-noise ratio values ​​exceed a preset threshold as feature frequency bands;

[0045] Analyze the frequency position, bandwidth and continuity of all characteristic frequency bands, and select the optimal filter type based on the analysis results. The filter types include high-pass filter, low-pass filter and band-pass filter.

[0046] The determined filter type and corresponding cutoff frequency are packaged to form the optimal acoustic filtering parameters for the acoustic channel of the acoustic-magnetic synchronization locator.

[0047] Optionally, the step of using the configured acoustic-magnetic synchronous locator to perform precise locating near the coarsely measured fault distance of the target faulty cable, and then performing secondary verification at the corresponding location of a nearby healthy cable to confirm the absence of acoustic-magnetic signals, thereby completing the closed-loop location of the fault point, includes the following steps:

[0048] The configured acoustic-magnetic synchronization locator receives the synchronized first acoustic signal and first magnetic field signal near the coarsely measured fault distance of the target fault cable, and calculates the signal time difference between the first acoustic signal and the first magnetic field signal in real time.

[0049] The acoustic-magnetic synchronization locator is moved based on the signal time difference until the signal time difference is minimized, and the physical location of the acoustic-magnetic synchronization locator is determined as the initial fault point of the target fault cable.

[0050] After secondary verification of the adjacent healthy cable at the location corresponding to the initial fault point using an acoustic-magnetic synchronization locator, the initial fault point is identified as the final fault point, thus completing the closed-loop location.

[0051] Optionally, after completing secondary verification using an acoustic-magnetic synchronization locator at the location of the initial fault point on a nearby healthy cable, locking the initial fault point as the final fault point and completing closed-loop localization includes the following steps:

[0052] Move the probe of the acoustomagnetic synchronization locator directly above the adjacent healthy cable;

[0053] The second acoustic signal and the second magnetic field signal originating from a nearby healthy cable were continuously monitored using an acoustic-magnetic synchronization pinpoint instrument.

[0054] Determine whether a valid second acoustic signal exists at the synchronization moment triggered by the second magnetic field signal;

[0055] If no valid second acoustic signal is detected within the preset verification time, the second verification is deemed to have passed.

[0056] If the secondary verification is successful, the initial fault point will be located as the final fault point using an acoustic-magnetic synchronization locator, and the geographical coordinates of the final fault point will be determined.

[0057] In a second aspect, the present invention also provides an intelligent cable fault testing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the cable fault detection method based on the intelligent cable fault testing device as described in the first aspect.

[0058] The beneficial effects of this invention are:

[0059] This invention solves the bottleneck problems of complex reflected signals, numerous interferences, and difficulty in identifying the true fault waveform in traditional cable fault testing. By introducing a healthy cable laid parallel to the faulty cable as a reference system, synchronous acquisition and feature comparison of cross-cable signals are achieved. This allows for the precise extraction of a unique, true fault signal from complex reflected waves, greatly improving the reliability of the initial fault measurement stage and eliminating subsequent errors caused by waveform misjudgment. This invention deeply integrates electrical measurement with precise acoustic positioning. Utilizing the spectral characteristics of confirmed true fault signals, it automatically configures the optimal filtering parameters for the acoustic-magnetic synchronous positioning instrument, giving it intelligent and targeted listening capabilities. This represents a leap from traditional experience-based blind listening and guesswork to targeted, precise listening, significantly improving the signal-to-noise ratio and success rate of precise positioning under strong environmental noise. The final secondary verification step constructs a closed-loop logic for positioning, completely eliminating the possibility of misjudgment and comprehensively improving the overall accuracy and efficiency of fault location. Attached Figure Description

[0060] Figure 1 This is a flowchart illustrating a cable fault detection method based on an intelligent cable fault testing device in one embodiment of this application.

[0061] Figure 2 This is a schematic diagram of the panel of a cable fault tester according to one embodiment of this application.

[0062] Figure 3 This is a schematic diagram of the connection for lightning current sampling in one embodiment of this application.

[0063] Figure 4 This is a schematic diagram illustrating the variation of the test waveform using the flashover method in one embodiment of this application.

[0064] Explanation of reference numerals in the attached figures:

[0065] 1. Power adapter charging socket; 2. Instrument indicator light; 3. Output socket; 4. Output amplitude; 5. Switch; 6. Display screen; 7. Reset button; 8. USB interface; 9. Touch mouse. Detailed Implementation

[0066] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0067] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0068] Figure 1 This is a flowchart illustrating a cable fault detection method based on an intelligent cable fault testing device in one embodiment. Figure 1 The cable fault detection method based on the intelligent cable fault testing device shown is applied to the cable fault testing device, which includes a cable fault tester (such as...). Figure 2 (As shown) and the acoustic-magnetic synchronization positioning instrument. It should be understood that, although Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps. For example Figure 1 As shown, the cable fault detection method based on an intelligent cable fault testing device disclosed in this invention specifically includes the following steps:

[0069] S101. Connect the high-voltage flashover unit and the first signal acquisition channel of the cable fault tester to the target fault cable, and at the same time connect the second signal acquisition channel to an adjacent healthy cable laid in parallel with the target fault cable.

[0070] First, the high-voltage flashover unit integrated within the cable fault tester is safely connected to the conductor core of the target faulty cable via a dedicated high-voltage output cable. Figure 3As shown, T1 is a 3KVA / 0.22KV voltage regulator, T2 is a 3KVA / 50KV AC / DC high-voltage transformer, D is a high-voltage rectifier silicon stack, C is a high-voltage pulse capacitor, V is a voltmeter, and B is a current sampler. Simultaneously, the first signal acquisition channel of the tester, a high-sensitivity signal sensor, is also connected in parallel at this connection point. Its task is to capture the signal subsequently returned by the target cable itself. Crucially, the second signal acquisition channel of the tester needs to be connected in the same way to a confirmed healthy cable that is laid adjacent to and parallel to the target faulty cable. This dual-channel parallel layout is the basis for subsequent signal comparison analysis and interference suppression. The adjacent healthy cable is chosen because when the high-voltage pulse propagates in the target cable and a fault discharge occurs, the resulting transient electromagnetic field will induce a characteristically relevant but weak signal in the adjacent healthy cable through spatial coupling. After completing all physical connections, the entire test system forms a differential detection structure, which can not only monitor the direct response of the faulty cable but also simultaneously record its electromagnetic impact on the surrounding environment.

[0071] S102. Apply a high-voltage pulse to the target faulty cable through the high-voltage flashover unit, and simultaneously capture the active reflection signal of the target faulty cable and the passive coupling induction signal of the adjacent healthy cable using the first signal acquisition channel and the second signal acquisition channel.

[0072] After the physical connection is completed, the cable fault tester is operated, activating its internal high-voltage flashover unit to apply one or more instantaneous high-voltage pulses to the target faulty cable. This pulse acts like an energy probe, propagating at high speed along the cable line. When it encounters a weak point or damage point in the cable insulation (i.e., the fault point), it breaks down the dielectric, generating an instantaneous discharge. This discharge process produces two key signals: first, within the target faulty cable, a portion of the pulse energy forms a reflected wave, returning along the original path to the tester; this is the active reflection signal (e.g., ...). Figure 4 (As shown in the image) Secondly, at the instant of fault discharge, the drastically changing current generates a strong electromagnetic field. This electromagnetic field radiates outward and induces a weak current pulse in the adjacent healthy cable; this is the passive coupling induction signal. To accurately capture these two almost simultaneous but distinct signals, the two signal acquisition channels are set to high-frequency synchronous acquisition mode. When the amplitude of the active reflection signal detected by the first channel exceeds the preset trigger threshold, the data recording function of both channels is immediately activated simultaneously, ensuring that the two signal waveforms are strictly aligned in time. The effect of this synchronous acquisition mechanism is to acquire two sets of raw waveform data, which have a precise correspondence on the time axis, and fully record the dynamic response of the target cable and adjacent cables under the action of the high-voltage pulse. This provides a high-quality raw data foundation for subsequent identification of the real fault signal through cross-cable signal comparison.

[0073] S103. The coarse fault distance of the fault point in the target fault cable is calculated based on the active reflection signal and the time-domain reflection method.

[0074] After obtaining two synchronization signals, the initial fault location is determined using the active reflected signal from the target faulty cable. This step employs the classic traveling wave method, also known as time-domain reflectometry (TDR). Its basic principle is similar to echolocation, calculating the distance by measuring the time difference between the transmitted detection signal and the reflected signal returning from the fault point. On the waveform diagram of the active reflected signal, two significant waveform features can be clearly observed: one is the initial high-voltage pulse wavefront, and the other is the reflected pulse wavefront generated by the fault point, appearing after a delay. Using a high-precision timing algorithm, the time interval between these two wavefronts on the time axis can be accurately measured. Since the propagation speed of electromagnetic waves in a specific type of cable is a basically known physical constant, the distance from the fault point to the test end can be calculated using a simple physical formula. The specific implementation method of this calculation is as follows: Where L represents the calculated coarse fault distance, v represents the propagation speed of electromagnetic waves in the medium of the cable under test, and Δt represents the time difference between the transmitted pulse recorded by the instrument and the first significant reflected pulse. Dividing by 2 in the formula is because the signal travels a round trip in the cable. The effect of this step is to quickly obtain a rough estimate of the fault location, i.e., the coarse fault distance.

[0075] S104. Identify the real fault signals in active reflection signals and passive coupling induction signals by comparing cross-cable signal characteristics.

[0076] After obtaining the coarse distance measurement, a crucial signal discrimination step is required to ensure that subsequent analysis and precise location are based on genuine fault information, rather than misleading interference. This step is accomplished by performing a deep feature comparison of the active reflected signal of the target cable and the passively coupled induced signal of a nearby healthy cable. Since one signal is a direct product of the fault and the other is an indirect product of its electromagnetic coupling, they inevitably differ fundamentally in energy distribution and phase relationship. The analysis process utilizes a built-in intelligent algorithm module to first discriminate from an energy perspective, calculating the energy attenuation gradient of the two signals. The source signal (active reflected signal) typically has higher energy, and its energy distribution characteristics across different frequency bands differ significantly from those of the coupled signal. Next, a phase verification is performed. Because electromagnetic coupling requires time, the wavefront of the passively coupled induced signal will inevitably have a slight delay compared to the wavefront of the active reflected signal. This delay can be calculated using a high-precision algorithm and converted into an initial phase offset. The conversion method is as follows: Here, φ is the calculated initial phase offset angle, f is the main frequency component of the signal, and Δτ is the absolute time difference between the starting points of the two signal waveforms. This calculated phase offset must conform to the theoretical coupling model determined by the physical spacing and insulation parameters of the two cables. Only when the results of the energy and phase dimensions match (logical AND operation) can the system finally confirm that the active reflected signal is a real fault signal. The effect of this step is to greatly enhance the accuracy of fault diagnosis, effectively eliminate complex electromagnetic interference in the field through cross-validation, and label the main signal as "real," providing reliable data input for subsequent spectrum analysis and acoustic filter configuration.

[0077] S105. Configure the optimal acoustic filtering parameters for the acoustic channel of the acoustic-magnetic synchronization locator based on the spectral characteristics of the actual fault signal.

[0078] Once a real fault signal is successfully identified, its spectral information is used to optimize the performance of the subsequent acoustic-magnetic synchronization locator. The high-voltage discharge at the fault point not only generates electrical signals but also weak acoustic signals, which is the physical basis for accurate locating. However, ambient noise (such as traffic, human voices, and wind noise) is often much louder than the fault discharge sound, making direct listening almost impossible to distinguish. To solve this problem, an optimal acoustic filter needs to be configured for the acoustic channel of the acoustic-magnetic synchronization locator, enabling it to "focus" on listening to the fault discharge sound while ignoring noise at other frequencies. The configuration is based on the spectral characteristics of the previously identified real fault signal, as the spectral characteristics of electrical signals and acoustic signals are highly correlated. By performing a Fast Fourier Transform (FFT) on the real fault signal, its energy distribution over the entire frequency domain, i.e., the spectrum, can be obtained. The frequency bands where energy is concentrated on the spectrum correspond to the frequency range where the fault discharge sound is most likely to occur. Based on this spectral characteristic, a set of optimal filter parameters can be dynamically generated, such as setting the center frequency and bandwidth of a bandpass filter, or the cutoff frequency of a high-pass or low-pass filter. This method, which uses electrical signal characteristics to preset parameters for acoustic detection, is an intelligent adaptive adjustment. Its effect is that the acoustic sensors of the acoustic-magnetic synchronization locator are pre-informed of the frequency characteristics of the sound to be searched, enabling them to extract weak but crucial fault discharge sound signals extremely effectively from noisy background noise. This results in a signal-to-noise ratio improvement of tens or even hundreds of times, creating the prerequisite for achieving high-precision on-site positioning.

[0079] S106. Using the configured acoustic-magnetic synchronous locator, perform precise locating near the coarsely measured fault distance of the target fault cable, and perform secondary verification at the corresponding location of the adjacent healthy cable to confirm the absence of acoustic-magnetic signal, thus completing the closed-loop location of the fault point.

[0080] After completing all preliminary preparations and parameter configurations, the final stage of precise on-site location and closed-loop verification begins. Personnel, carrying an acoustomagnetic synchronous locator configured with optimal acoustic filtering parameters, travel to the area near the previously roughly measured fault distance. A high-voltage flashover is triggered again directly above the cable path. At this time, the locator simultaneously receives two signals: one is the fault discharge magnetic field signal arriving almost instantaneously, and the other is the acoustic signal arriving with a slight delay after propagating through a medium (such as soil). The core function of the locator is to accurately measure the time difference between these two signals. Since the propagation speed of the magnetic field signal is close to the speed of light, its arrival time can be considered zero delay. Therefore, this time difference is essentially equal to the time required for the sound wave to travel from the underground fault point to the surface locator. According to the principle of sound speed, when the locator is directly above the fault point, the sound wave propagation distance is shortest, and the required time is also shortest. Therefore, the operator moves the locator a small distance along the cable path and continuously observes the acoustomagnetic time difference displayed on the screen. When this time difference reaches its minimum value, the physical location of the locator is preliminarily determined as the fault point. To ensure absolute certainty, a secondary verification, or "closed-loop" operation, is required. The operator moves the positioning instrument to the corresponding position directly above the adjacent healthy cable and performs the flashover test again. Since the fault exists only in the target cable, no discharge sound should be generated at the corresponding position on the healthy cable. If, within the preset verification time, the positioning instrument does not detect any valid acoustic signal synchronized with the magnetic field signal above the healthy cable, the secondary verification is successful.

[0081] In one embodiment, simultaneously capturing the active reflected signal of the target faulty cable and the passive coupling induction signal of a nearby healthy cable using a first signal acquisition channel and a second signal acquisition channel includes the following steps:

[0082] Set synchronous acquisition trigger thresholds and pre-trigger data buffers in the first and second signal acquisition channels;

[0083] The signal amplitude in the first signal acquisition channel is monitored in real time. When the signal amplitude exceeds the synchronous acquisition trigger threshold, high-frequency synchronous data acquisition of the first and second signal acquisition channels is started.

[0084] Based on the preset acquisition time window, the original waveform data of the first signal acquisition channel and the second signal acquisition channel after the high voltage pulse is generated are recorded respectively.

[0085] The acquired raw waveform data is digitally filtered and normalized to generate the active reflection signal of the target faulty cable and the passive coupling induction signal of the adjacent healthy cable.

[0086] In this implementation, the core of this step is to establish an intelligent watch and response mechanism in the two signal acquisition channels. Specifically, a clear synchronous acquisition trigger threshold is set for the system; this is a specific voltage value that serves as a threshold for determining whether a signal is a valid event. Only when the signal strength exceeds this threshold is it considered a potential fault signal worth recording. Simultaneously, to ensure the complete capture of the signal, especially the initial rise of the signal from nothing to something before it crosses the trigger threshold, the system allocates a memory area called a pre-trigger data buffer. This buffer acts like a continuously recording short-term memory, constantly overwriting the latest data before the actual acquisition is triggered. This design ensures that when the actual trigger event occurs, the waveform data stored in the buffer prior to the trigger moment is preserved, thus avoiding the risk of losing crucial waveform leading-edge information due to acquisition start delays.

[0087] After completing the preset steps, the system enters the real-time signal monitoring and synchronous trigger execution phase. During this phase, the high-speed analog-to-digital converter (ADC) inside the first signal acquisition channel (connected to the target faulty cable) continuously samples the input signal at an extremely high frequency and compares its voltage amplitude with the preset synchronous acquisition trigger threshold in real time. This process is a continuous judgment loop, with the system constantly monitoring the dynamic changes in the signal amplitude. If, at any moment, the monitored signal amplitude suddenly jumps and exceeds the set trigger threshold, for example, if the following condition is met: ,in It is the real-time input signal amplitude. This is a preset trigger threshold voltage. The instant this condition is met, the system immediately generates an internal trigger command. The key function of this command lies in its "synchronization"; it simultaneously activates the first and second signal acquisition channels (connected to the adjacent health cable) into high-speed data acquisition mode with millisecond precision. This strict synchronization mechanism is crucial, ensuring that the two subsequently recorded waveform data are perfectly aligned on the time base.

[0088] Once synchronous acquisition is triggered, the system begins recording raw waveform data within a preset time window. This acquisition time window is not infinitely long, but a precisely calculated finite duration, set based on the physical length of the cable under test and the inherent speed of electromagnetic waves propagating within it. The window duration must be long enough to ensure complete recording of the entire process from the emission of the high-voltage pulse, to the signal reaching the furthest fault point, and then to the reflected signal returning to the tester. The minimum required duration can be estimated as follows: .in, This represents the length of the data collection time window. It is the maximum possible length of the cable being tested, and This refers to the propagation speed of electromagnetic waves in the cable medium. After the trigger command is issued, the system will solidify the data temporarily stored in the pre-trigger buffer and continue to acquire and store data after the trigger until the entire time window is filled. This process is carried out synchronously in the first and second signal acquisition channels, and finally generates a discrete data point sequence representing the voltage change over time in their respective memories.

[0089] After acquiring the raw waveform data, it cannot be used directly for analysis because it contains various noises with varying amplitudes, requiring final purification and normalization. This step involves two key processes: digital filtering and normalization. First, digital filtering is performed on the two sets of acquired raw waveform data. Bandpass filters are typically used to filter out interference components unrelated to the fault signal characteristics, such as low-frequency power frequency interference introduced by the power supply system and high-frequency white noise generated by electronic components and the external electromagnetic environment. The filtering process significantly improves the signal-to-noise ratio, making subtle fault characteristics hidden in the noise (such as the inflection point of the reflected wave) clearly discernible. Next, the two filtered signal waveforms are normalized. Because the active reflection signal has strong energy while the passive coupling signal is very weak, their absolute amplitudes differ greatly, making direct comparison meaningless. Normalization aims to eliminate this amplitude difference, scaling the two waveforms to a uniform relative scale, such as the [-1, 1] interval. A common implementation is maximum value normalization: .in, It refers to each data point of the filtered signal. It is the maximum absolute value among all data points of the signal. These are the new data points after normalization. The final result of this step is the generation of two clean, clear signal curves with uniform amplitude scales, namely the active reflection signal of the target faulty cable and the passive coupling induction signal of the adjacent healthy cable, which are ultimately used for analysis. They provide high-quality, comparable, and ideal data for subsequent fault identification and feature extraction.

[0090] In one embodiment, identifying the true fault signal in the active reflection signal and the passive coupling induction signal by comparing cross-cable signal features includes the following steps:

[0091] The energy attenuation gradient analysis module of the cable fault tester is called to calculate and compare the energy attenuation gradient of the active reflected signal and the passive coupled induction signal to obtain the energy source determination result.

[0092] The initial phase offset verification module of the cable fault tester is called to calculate and verify the initial phase offset of the active reflection signal and the passive coupling induction signal, and the coupling path confirmation result is obtained.

[0093] The results of energy source determination and coupling path confirmation are subjected to a logical AND operation to form the signal identification result;

[0094] Based on the signal recognition results, the active reflection signal is marked as the real fault signal, and the spectral characteristics of the real fault signal are extracted. At the same time, the passive coupling induction signal is marked as the interference signal, and the interference signal is suppressed.

[0095] In this embodiment, the total energy of the actively reflected signal, which serves as the energy source, is necessarily much higher than that of the passively coupled signal indirectly induced by the spatial electromagnetic field. More importantly, the coupling process itself is not simply signal attenuation; the attenuation of high-frequency components is far more severe than that of low-frequency components. Therefore, the spectral shape of the passively coupled signal will be distorted, and the proportion of its high-frequency energy will decrease sharply. During analysis, a Fast Fourier Transform is first performed on both signals to obtain their respective amplitude-frequency characteristic curves. Then, the broadband total energy is calculated, forming a total energy ratio. Next, the entire spectrum is divided into multiple continuous narrowband frequency bands, and the energy ratio within each band is calculated. By observing the trend of these energy ratios changing with increasing frequency, the energy attenuation gradient can be quantified. A key criterion can be set as follows: .in It is the energy gradient factor. and These are the energies of active and passive signals in the high-frequency band, respectively. and It's their energy in the low-frequency range. If If the value is significantly greater than 1, it indicates that the coupling attenuation model is met.

[0096] After completing the energy dimension analysis, cross-validation is required from the time dimension, calling the initial phase offset verification module to confirm the causal relationship between the signals. The physical principle is clear: the passively coupled inductive signal, as a result, is necessarily generated later than the active reflected signal, as a cause, with an extremely short but measurable transmission delay between them. To accurately capture this minute delay, the original discrete signal data is first upsampled using an interpolation algorithm, which adds more detail to the time axis and improves the time resolution. Then, the first derivatives of the two upsampled signal waveforms are calculated. The moment when the derivative first becomes non-zero can be precisely defined as the physical starting point of the signal waveform. By calculating the time difference between these two starting points, an absolute time value of the initial phase offset can be obtained. To make it universally comparable, this absolute time value needs to be converted into a frequency-dependent phase angle. The conversion method is as follows: Here, It is the calculated initial phase offset angle. It is the center frequency of the signal, and This is the time difference between the starting points of the two signal waveforms. Finally, the calculated phase angle is compared with a theoretical phase offset range preset based on physical parameters such as cable spacing and insulation medium.

[0097] After obtaining independent judgment results for both energy and phase dimensions, a rigorous logical AND operation needs to be performed on these two results to form a final, high-confidence identification conclusion. The principle behind this comprehensive decision-making mechanism is that single-dimensional analysis may carry the risk of misjudgment. For example, a strong interference pulse with very similar energy characteristics might cause confusion in the energy analysis stage. Similarly, random noise that happens to closely follow the real signal in time sequence might slip through the phase verification. However, the probability of an external interference perfectly simulating the characteristics of the coupled signal in both the energy decay gradient and the initial phase shift—two completely independent physical characteristics—is extremely low. The logical AND operation requires that both the energy source judgment result and the coupling path confirmation result be "true" simultaneously for the final signal identification result to be judged as "true." This process can be described by a simple logical expression: the identification result is "passed" only when both the energy judgment and the phase judgment pass. The effect of this step is to greatly improve the robustness and reliability of signal identification, acting like a double insurance mechanism that combines the evidence chains of two different physical dimensions.

[0098] After the logical operations yield a clear signal identification result, the final stage of signal marking and feature utilization begins. This is a crucial step that bridges the gap between the previous and subsequent steps, aiming to transform the analysis conclusions into a direct basis for subsequent operations. Based on the signal identification result, if determined to be "true," the active reflection signal captured by the first signal acquisition channel originally connected to the target faulty cable will be officially marked as a real fault signal at the data level. This marking signifies that its data has the highest priority and reliability. Immediately afterwards, the key attributes of this real fault signal are extracted and stored, especially its spectral characteristics, i.e., the distribution of its energy at different frequencies. This spectral data will serve as the core basis for configuring the filter parameters of the acoustic-magnetic synchronization locator in the next stage. Simultaneously, the passive coupling induction signal captured by the second signal acquisition channel originally connected to the adjacent healthy cable is correspondingly marked as an interference signal. The most direct approach to handling this is to suppress or eliminate it in all subsequent analyses and displays to avoid visual or algorithmic interference for the operators.

[0099] In one implementation, the energy attenuation gradient analysis module of the cable fault tester is invoked to calculate and compare the energy attenuation gradients of the active reflected signal and the passively coupled inductive signal to obtain the energy source determination result, including the following steps:

[0100] Fast Fourier transforms were performed on the active reflected signal and the passively coupled induced signal, respectively, to obtain the amplitude-frequency response curves of the two signals;

[0101] The broadband total energy is obtained by calculating the active reflection signal and the passive coupling induction signal, and a first energy ratio is generated;

[0102] The amplitude-frequency response curves of the two signals are divided into multiple pre-defined non-overlapping narrowband frequency bands.

[0103] Calculate the signal energy in each narrowband frequency band and generate the second energy ratio for each frequency band;

[0104] The energy decay gradient is obtained by analyzing the trend of the second energy ratio changing with frequency.

[0105] The first energy ratio and energy decay gradient are compared with the preset coupling model threshold, and the energy source determination result is output.

[0106] In this embodiment, the active reflection signal and the passively coupled inductive signal are transformed from the time domain to the frequency domain, that is, the signal representation is changed from voltage variation over time to energy distribution at different frequencies. This transformation is achieved through Fast Fourier Transform (FFT). This algorithm can efficiently decompose any complex waveform into its basic sinusoidal components. After applying the FFT algorithm to the two time series data of the active reflection signal and the passively coupled inductive signal, two new sets of data are obtained. Each set of data describes the energy or amplitude intensity contained in the signal at various frequency points across the entire frequency spectrum from low to high frequencies. Plotting these data into a graph, with frequency on the horizontal axis and amplitude on the vertical axis, yields two intuitive amplitude-frequency characteristic curves. After obtaining the spectral fingerprints of the two signals, the first comparison to be made is to evaluate their overall energy intensity. This is a macroscopic, holistic comparison aimed at quickly determining which signal is the energy source. The physical principle is that the energy generated by fault discharge is mainly concentrated in the target faulty cable, with only a small portion leaking to the adjacent healthy cable through electromagnetic coupling. Therefore, the total energy of the active reflection signal must be much greater than that of the passively coupled inductive signal.

[0107] The method for calculating the total energy is to integrate its amplitude-frequency response curve. In discrete digital signal processing, this is equivalent to summing the squared values ​​of the energy at all frequency points. After calculating the broadband total energy of two signals, they are divided to generate a ratio, i.e., the first energy ratio, for ease of comparison. The calculation method is as follows: .in, The first energy ratio, Represents the broadband total energy of the actively reflected signal. This represents the broadband total energy of the passively coupled inductive signal. The effect of this step is to obtain a simple and direct quantitative indicator. The value is much greater than 1, for example, exceeding a preset empirical threshold of 10, which initially confirms the hypothesis that the active reflection signal is the energy source.

[0108] Simply comparing total energy is insufficient, as it cannot eliminate external interference with equally strong energy. For a more refined and discernible analysis, the two complete amplitude-frequency response curves need to be segmented. The principle behind this step is that electromagnetic coupling naturally has a stronger attenuation effect on high-frequency signals; this characteristic is key to distinguishing real coupling from random interference. In practice, the frequency axis (horizontal axis) of the two curves is divided into multiple pre-defined, non-overlapping continuous frequency bands from low to high. For example, the entire spectrum from 0 to 1 MHz can be divided into 10 narrowbands of 100 kHz each: 0-100 kHz, 100-200 kHz, and so on. The division can be based on the energy distribution characteristics of typical cable fault signals; finer divisions can be used in areas of concentrated signal energy, and coarser divisions in areas of sparse energy. After dividing the spectrum into multiple narrowbands, the signal energy can be calculated and compared in detail for each band. The purpose of this step is to quantify the energy difference between the two signals within each frequency range. The specific operation involves calculating the total energy of the active reflection signal and the passive coupling induction signal falling within the first narrowband (e.g., 0-100kHz) for each band. Then, dividing these two energy values ​​yields the energy ratio corresponding to that band. This same calculation process is repeated for the second, third, and finally the last narrowband. The original total energy ratio is expanded into a series of energy ratios corresponding to each band, forming the second energy ratio sequence. The energy ratio for the k-th band is calculated as follows: .in, This represents the second energy ratio of the k-th narrowband frequency band. The energy of the actively reflected signal in this frequency band. This represents the energy of the passively coupled induced signal in this frequency band.

[0109] After obtaining the sequence data describing the energy ratio change with frequency, the final step is to analyze the overall trend of this sequence and quantify it into a specific value, namely the energy decay gradient. The core principle is that, in the case of a true coupling relationship, because the high-frequency components attenuate more drastically, the energy of the passively coupled signal will decrease rapidly with increasing frequency, causing the energy ratio of the two signals (the second energy ratio) to increase significantly with increasing frequency. This steadily increasing trend is the manifestation of the energy decay gradient. The most common method for analyzing this trend is to use linear regression analysis, using a series of second energy ratios as the ordinate and the center frequency of each frequency band as the abscissa, fitting a straight line in the coordinate system that best represents the trend of these data points. The slope of this line accurately quantifies the energy decay gradient. This gradient value can be expressed as: Here, This is the final calculated energy decay gradient. This represents the center frequency of the k-th narrowband frequency band.

[0110] After calculating the first energy ratio, representing the overall energy difference, and the energy attenuation gradient, representing the trend of energy difference change, a final comprehensive evaluation can be performed to output the determination result of the energy source. The evaluation criteria are two pre-set coupling model thresholds: one for the total energy threshold used for the first energy ratio, and the other for the gradient threshold used for the energy attenuation gradient. The determination logic is very strict, requiring two conditions to be met simultaneously: the calculated first energy ratio must be greater than the total energy threshold, and the calculated energy attenuation gradient must also be greater than the gradient threshold. Only when both conditions are met will the system ultimately determine that the active reflection signal is the energy source. This dual verification mechanism ensures extremely high reliability; it not only requires the source signal to be stronger overall but also requires that its energy relationship with the coupled signal strictly follow the physical model of more severe high-frequency attenuation.

[0111] In one embodiment, the first energy ratio and the energy attenuation gradient are compared with a preset coupling model threshold, and the energy source determination result is output, specifically including the following steps:

[0112] Using the amplitude-frequency characteristic curve of the passively coupled sensing signal as a benchmark, a dynamic coupling attenuation fingerprint reflecting the current test environment is constructed.

[0113] Based on dynamic coupling attenuation fingerprint, the amplitude-frequency characteristic curve of the active reflection signal is scaled and shaped and fitted to generate an amplitude-frequency characteristic curve of the theoretically coupled prediction signal.

[0114] Calculate the mismatch between the actual amplitude-frequency response curve of the active reflection signal and the amplitude-frequency response curve of the theoretically coupled predicted signal;

[0115] If the mismatch exceeds the preset mismatch threshold, it is determined that the energy attenuation gradient of the active reflection signal does not conform to the coupling propagation characteristics.

[0116] Based on the combined determination that the first energy ratio and energy attenuation gradient do not conform to the coupling propagation characteristics, the final energy source determination result is output.

[0117] In this embodiment, to fundamentally reveal the intrinsic physical connection between the active reflection signal and the passively coupled inductive signal, it is necessary to first establish an accurate and dynamic attenuation model for the current test environment. Compared to other embodiments, this embodiment abandons the traditional method of using a fixed theoretical model, and instead uses the amplitude-frequency characteristic curve of the passively coupled inductive signal itself as a feature to describe the coupling attenuation characteristics under the current environment. The basic principle is that the passively coupled inductive signal itself is formed after the fault source signal passes through the medium (such as air or soil) between the target cable and the adjacent cable. Therefore, its spectral shape naturally and completely records all the attenuation information of different frequency signals for that specific path. In specific implementation, the amplitude-frequency characteristic curve is obtained by performing a fast Fourier transform on the passively coupled inductive signal. It is directly defined as a dynamically coupled attenuated fingerprint.

[0118] After constructing a dynamic coupling attenuation fingerprint that reflects the real environment, the next step is to perform a theoretical deduction based on this fingerprint, that is, to generate a theoretical coupling prediction signal. Specifically, using the dynamic coupling attenuation fingerprint as a template, the amplitude-frequency response curve of the active reflection signal is adjusted so that its attenuation ratio in the high-frequency part and the overall shape of the spectrum tend to conform to the characteristics of the fingerprint. This process can be abstracted as an application of a transfer function to generate the amplitude-frequency response curve of the theoretical prediction signal: .in, It is the actual amplitude-frequency response curve of the actively reflected signal, while This is based on dynamic coupling attenuation fingerprint An equivalent transfer function representing the characteristics of the coupled channel is established.

[0119] After generating the theoretical standard image, it needs to be precisely compared with the actual measured situation to quantify the degree of difference between them. This degree of difference is called the mismatch. The principle is that if the active reflection signal is indeed the fault source, then the predicted signal spectrum obtained through theoretical deduction should be highly similar to the spectrum of the actually measured passively coupled inductive signal. Conversely, if the active reflection signal and the passive signal have no direct physical coupling relationship, then the prediction based on it will inevitably differ from the actual situation. Calculating the mismatch is to express this degree of "similarity" or "dissimilarity" with a specific numerical value. A commonly used calculation method is to calculate the normalized mean square error between the two amplitude-frequency response curves. The specific calculation formula is:

[0120]

[0121] in, This is the degree of mismatch. It is the amplitude-frequency curve of the passively coupled signal as actually measured. This is the amplitude-frequency curve of the theoretically predicted signal generated in the previous step. This formula calculates the sum of squares of the differences between the two curves across the entire frequency domain and normalizes it using the total energy of the actual signal. The effect of this step is to condense the morphological differences between the two complex curves into a single, intuitive quantitative indicator. The closer the mismatch is to 0, the better the theory matches reality.

[0122] After calculating the mismatch value representing the difference between theory and reality, a clear logical judgment needs to be made based on this value. This step transforms the quantitative calculation result into a qualitative conclusion. This is achieved by comparing the calculated mismatch value with a pre-set mismatch threshold. This mismatch threshold is a critical value derived from extensive experimental data and empirical summaries; it represents the upper limit of the normal deviation between theory and reality that the system can tolerate due to random factors such as noise. The logic of the judgment is very straightforward: if the calculated mismatch value exceeds this pre-set mismatch threshold, it can be determined that there is a significant difference between the theoretical model and the actual situation, exceeding the normal error range. The conclusion drawn is that the initial assumption is invalid, i.e., the energy attenuation gradient of the active reflection signal does not conform to the expected coupling propagation characteristics. Finally, to ensure the final result of the energy source determination is foolproof, it is necessary to comprehensively consider evidence from both macroscopic and microscopic levels. This step integrates the previously calculated first energy ratio (representing macroscopic energy difference) and the judgment on whether the energy attenuation gradient conforms to coupling characteristics (representing the consistency of microscopic spectral morphology). The final determination of the energy source must pass the tests of both dimensions simultaneously. The determination logic is as follows: First, the total energy of the actively reflected signal must be significantly greater than that of the passively coupled inductive signal; that is, the first energy ratio must exceed a high threshold, ensuring that it is the source in terms of energy magnitude. Second, the spectral characteristics of the actively reflected signal must be able to accurately predict the spectral characteristics of the passively coupled signal; that is, the mismatch must be lower than a preset threshold, and the determination result is "conforming to the coupling propagation characteristics." Only when both conditions are met simultaneously will the system ultimately output the determination result that "the energy source is an actively reflected signal." If either condition is not met, it will be determined as a non-coupled relationship.

[0123] In one implementation, the initial phase offset verification module of the cable fault tester is invoked to calculate and verify the initial phase offset of the active reflection signal and the passive coupling induction signal to obtain the coupling path confirmation result, including the following steps:

[0124] The active reflection signal and the passively coupled induction signal are upsampled using an interpolation algorithm;

[0125] Calculate the first derivatives of the active reflected signal and the passively coupled induced signal after upsampling, and determine the starting time point of the signal waveform based on the first non-zero point of the first derivative;

[0126] Calculate the time difference between the starting time points of the active reflection signal and the passive coupling induction signal to obtain the absolute time value of the initial phase shift;

[0127] The absolute time value is converted into a phase angle based on the main frequency of the actual fault signal;

[0128] The phase angle is compared with the theoretical phase offset range set according to the cable spacing and medium parameters, and the coupling path confirmation result is output.

[0129] In this embodiment,

[0130] In one implementation, configuring the optimal acoustic filtering parameters for the acoustic channel of the acoustic-magnetic synchronization locator based on the spectral characteristics of the actual fault signal includes the following steps:

[0131] The high-voltage flashover unit triggers multiple high-voltage flashovers within the area determined by the rough measurement of the fault distance, and the magnetic field signal and continuous on-site acoustic signal are synchronously collected by the acoustic-magnetic synchronization locator.

[0132] Based on the triggering time of each magnetic field signal, multiple synchronous acoustic samples are extracted from the continuous field acoustic signals. Each synchronous acoustic sample contains fault discharge sound and background noise.

[0133] Multiple asynchronous acoustic samples that do not contain fault discharge sounds are extracted from continuous on-site acoustic signals and used as background noise samples.

[0134] Calculate the average spectrum of the synchronous acoustic sample and the average spectrum of the background noise sample, respectively.

[0135] By comparing the average spectrum, the characteristic frequency bands where the signal-to-noise ratio of fault discharge sound is higher than that of background noise are identified, and the optimal acoustic filtering parameters for the acoustic channel of the acoustic-magnetic synchronization locator are dynamically generated based on the characteristic frequency bands.

[0136] In this embodiment, to accurately measure the time difference between two signals at the microsecond or even nanosecond level, it is first necessary to improve the time resolution of the original digital signal. The original acquired signal consists of a series of discrete points at equal time intervals. The starting point of its true waveform is likely to fall between two sampling points, and directly searching for it will produce a large error. By upsampling the active reflection signal and the passively coupled inductive signal through interpolation algorithms, new data points can be calculated and inserted between the existing data points according to mathematical laws. Commonly used interpolation methods include linear interpolation, polynomial interpolation, or the more precise sinc interpolation. After obtaining the high-resolution signal waveform, an objective and accurate method is needed to define when the waveform begins. When a signal is in a silent state, its value does not change, and its rate of change is zero. At the instant the signal starts, its value begins to change rapidly, and the rate of change jumps from zero.

[0137] Using this principle, the instantaneous initiation of this change is captured by calculating the first derivative of the signal. The first derivative directly reflects the instantaneous rate of change of the signal. For discrete signal data points after upsampling, its first derivative can be approximated by calculating the difference between adjacent data points. The point in time when the first derivative curve first appears non-zero can be precisely defined as the starting point of the signal waveform. To avoid noise interference, the point in time when the derivative value first continuously exceeds a very small threshold ε is usually sought. That is, satisfying After determining the precise start times of the active reflected signal and the passively coupled inductive signal, calculating their timing relationship becomes straightforward. The purpose of this step is to quantify the delay time of the passively coupled signal relative to the active reflected signal. Because electromagnetic field propagation takes time, the coupling induction process from the faulty cable to the adjacent healthy cable will inevitably produce a small time lag; the start point of the passive signal will necessarily be later than that of the active signal.

[0138] The specific value of this delay time can be obtained by simply subtracting the start times of the two signals. This value is called the absolute time value of the initial phase shift. Its calculation method is as follows: .in, It is the calculated absolute time difference. It is the starting time point of the passively coupled inductive signal, and This is the starting time point of the actively reflected signal. After obtaining the absolute time difference in seconds, it needs to be converted into a more universal physical quantity, namely the phase angle. A fixed time difference represents completely different degrees of phase deviation for signals of different frequencies. Therefore, the absolute time value must be correlated with the signal's own period to objectively evaluate its delay. This step first requires determining the dominant frequency of the actual fault signal, that is, the frequency point where its energy is most concentrated, through spectrum analysis. Then, the time required for a complete cycle of the signal is calculated based on the dominant frequency. Finally, the absolute time difference is compared with this cycle time and then converted into 360-degree angles. The conversion formula is: In this formula, This is the final calculated phase angle. It is the absolute time difference calculated in the previous step, while It is the main frequency of the actual fault signal.

[0139] Finally, the measured and calculated phase angle is compared with a theoretical range based on physical laws to make a final determination on the authenticity of the coupling path. This theoretical phase offset range is not arbitrary but is a reasonable and acceptable phase delay interval calculated using electromagnetic field theory based on parameters such as the physical distance between the two cables in the field and the dielectric constant of the insulating medium between them. This interval defines a physical boundary of possibility. The comparison process is to determine whether the measured value falls within this theoretical range. If the measured and calculated phase angle meets the theoretical range, the measurement result is considered to be consistent with physical reality, and the coupling path is confirmed. Conversely, if the measured value falls outside the theoretical range, it indicates that the timing relationship between the two signals does not conform to the physical coupling laws.

[0140] In one embodiment, identifying the characteristic frequency band where the signal-to-noise ratio of the fault discharge sound is higher than that of the background noise by comparing the average spectrum, and dynamically generating the optimal acoustic filtering parameters for the acoustic channel of the acoustic-magnetic synchronization locator based on the characteristic frequency band, includes the following steps:

[0141] The estimated spectrum of the pure fault sound is obtained by subtracting the average spectrum of the synchronous acoustic sample from the average spectrum of the background noise sample.

[0142] Calculate the signal-to-noise ratio (SNR) between the estimated spectrum of the clean fault sound and the average spectrum of the background noise sample, and generate an SNR spectrum map covering the entire frequency band;

[0143] Search the signal-to-noise ratio spectrum for all frequency bands whose signal-to-noise ratio values ​​exceed a preset threshold as feature frequency bands;

[0144] Analyze the frequency position, bandwidth and continuity of all characteristic frequency bands, and select the optimal filter type based on the analysis results. The filter types include high-pass filter, low-pass filter and band-pass filter.

[0145] The determined filter type and corresponding cutoff frequency are packaged to form the optimal acoustic filtering parameters for the acoustic channel of the acoustic-magnetic synchronization locator.

[0146] In this embodiment, to extract the true nature of the fault discharge sound from the noisy ambient sound, a spectral subtraction operation is first required. The basic principle is that the synchronous acoustic sample contains both fault discharge sound and background noise, while the background noise sample contains only stable background noise. By subtracting the average spectrum of the background noise from the average spectrum containing the fault sound at each frequency point, the constant noise component can be theoretically eliminated, resulting in a relatively pure spectrum representing only the fault discharge sound. This process can be expressed as: Here, The estimated power spectrum of the pure fault sound. The average power spectrum of the synchronous acoustic samples, and This represents the average power spectrum of the background noise sample.

[0147] After obtaining the estimated spectrum of the clean fault sound, it is necessary to further quantify the significance of this fault sound signal relative to the background noise at each frequency point. This requires calculating the signal-to-noise ratio (SNR), which is the ratio of signal strength to noise strength. By calculating the ratio between the estimated spectrum of the clean fault sound and the average spectrum of the background noise point by point, a completely new SNR spectrum covering the entire frequency band can be generated. The vertical axis of this graph is no longer the absolute energy level, but rather the relative value of how many times stronger the signal is than the noise. The calculation method is as follows: .in, Let f be the signal-to-noise ratio (SNR) at frequency point f. Next, a comprehensive search needs to be performed on the SNR spectrum from low to high frequencies, marking all frequency ranges with continuously higher SNR values ​​than this threshold as characteristic frequency bands. For example, if all frequency points within a frequency band satisfy the condition... ,in If the preset signal-to-noise ratio threshold is used, then this frequency band is identified as a valid characteristic frequency band.

[0148] After identifying all characteristic frequency bands, the most efficient filter type needs to be selected based on their distribution across the spectrum. This is a rule-based decision-making process aimed at achieving the best noise reduction and fidelity using the simplest tools. The analysis focuses on the location, width, and continuity of all characteristic frequency bands. If all characteristic frequency bands are concentrated above a certain high frequency, a high-pass filter is most suitable. Conversely, if they are concentrated below a certain low frequency, a low-pass filter should be used. If the characteristic frequency bands are concentrated in a certain region of the spectrum, or if there is a dominant, most energy-concentrated core frequency band, then a band-pass filter will be the best choice. Once the optimal filter type is determined, the final step is to combine this decision with specific frequency parameters and encapsulate it into a set of executable instructions. This process translates the analysis results into the actual device configuration. Based on the analysis results from the previous step and the specific boundaries of the characteristic frequency bands, the cutoff frequency of the filter is determined. If a high-pass filter is selected, its cutoff frequency is usually set to the lowest frequency among all characteristic frequency bands. If a low-pass filter is selected, its cutoff frequency is set to the highest frequency. If a bandpass filter is selected, a lower cutoff frequency and an upper cutoff frequency need to be set. Finally, the determined filter type and its corresponding cutoff frequency values ​​are packaged together to form a complete and optimized set of acoustic filtering parameters.

[0149] In one embodiment, the configured acoustic-magnetic synchronous locator is used to perform precise locating near the coarsely measured fault distance of the target faulty cable, and secondary verification is performed at the corresponding location of a nearby healthy cable to confirm the absence of acoustic-magnetic signals, thus completing the closed-loop location of the fault point. This includes the following steps:

[0150] The configured acoustic-magnetic synchronization locator receives the synchronized first acoustic signal and first magnetic field signal near the coarsely measured fault distance of the target fault cable, and calculates the signal time difference between the first acoustic signal and the first magnetic field signal in real time.

[0151] The acoustic-magnetic synchronization locator is moved based on the signal time difference until the signal time difference is minimized, and the physical location of the acoustic-magnetic synchronization locator is determined as the initial fault point of the target fault cable.

[0152] After secondary verification of the adjacent healthy cable at the location corresponding to the initial fault point using an acoustic-magnetic synchronization locator, the initial fault point is identified as the final fault point, thus completing the closed-loop location.

[0153] In this embodiment, a pre-configured acoustic-magnetic synchronous locator is used for precise on-site positioning within the coarse fault area. When a high-voltage flashover triggers a fault discharge, a magnetic field signal traveling at the speed of light and an acoustic signal traveling at the speed of sound are simultaneously generated. The locator's magnetic field probe almost instantaneously captures the first magnetic field signal and uses it as a zero-time reference to start an internal high-precision timer. Simultaneously, the acoustic channel, filtered with optimal parameters, begins listening to the first acoustic signal. When the acoustic probe captures a sound wave with matching characteristics, the timer stops. The device calculates and displays the time difference between the two signals in real time. ,in For signal time difference, and These represent the arrival times of the acoustic and magnetic field signals, respectively. Since the acoustic-magnetic synchronization locator moves along the cable path on the ground, while the underground fault point is fixed, the sound wave travels the shortest distance when the locator is directly above the fault point, resulting in the smallest measured time difference between the acoustic and magnetic signals. Based on this principle, the locator can be controlled to move back and forth along the cable laying path near the approximate fault distance. Through repeated comparisons, the physical location where the time difference reading stably reaches its minimum value is found. This point is then identified as the initial fault point and marked.

[0154] To ensure absolute accuracy in location and eliminate any possibility of misjudgment, a secondary verification must be performed. The probe of the acoustic-magnetic synchronization locator is moved from the initial fault point to directly above a nearby healthy cable laid in parallel. The tester then applies a high-voltage flashover to the target faulty cable again. Since the fault exists only in the target cable, the healthy adjacent cable should not produce any discharge sound at this location. If, within the preset verification time, the locator fails to detect any valid acoustic signal at the adjacent healthy cable at the synchronization moment triggered by the magnetic field signal, the verification is successful. This step, through a rigorous comparative test, confirms the uniqueness of the signal source, completing the diagnostic loop. Once verification is successful, the initial fault point is locked as the final fault point, and the location work is considered complete.

[0155] In one implementation, after secondary verification is completed on a nearby healthy cable at the location corresponding to the initial fault point using an acoustic-magnetic synchronization locator, the initial fault point is identified as the final fault point, and closed-loop localization is achieved through the following steps:

[0156] Move the probe of the acoustomagnetic synchronization locator directly above the adjacent healthy cable;

[0157] The second acoustic signal and the second magnetic field signal originating from a nearby healthy cable were continuously monitored using an acoustic-magnetic synchronization pinpoint instrument.

[0158] Determine whether a valid second acoustic signal exists at the synchronization moment triggered by the second magnetic field signal;

[0159] If no valid second acoustic signal is detected within the preset verification time, the second verification is deemed to have passed.

[0160] If the secondary verification is successful, the initial fault point will be located as the final fault point using an acoustic-magnetic synchronization locator, and the geographical coordinates of the final fault point will be determined.

[0161] In this embodiment, the first step is to precisely laterally move the probe of the acoustic-magnetic synchronization locator from the initial fault location until it is directly overhead aligned with the adjacent healthy cable running parallel to the target fault cable. This lateral movement aims to change only the object being monitored while keeping all other test conditions (such as high-voltage flashover settings and ambient noise) constant. After moving the probe directly overhead to the adjacent healthy cable, the acoustic-magnetic synchronization locator resumes continuous monitoring. The tester continues to apply a high-voltage pulse to the initial target fault cable, meaning the magnetic field signal (second magnetic field signal) serving as the time reference will still be captured by the locator's magnetic field probe due to the spatial propagation of the electromagnetic field. However, the acoustic probe is now listening to the soil medium beneath the adjacent healthy cable. The locator continues to operate, synchronously acquiring the magnetic field signal and any possible acoustic signals (second acoustic signals). This process is identical to that used when locating the target cable, aiming to search for signals with the same standard.

[0162] After obtaining monitoring data from the location of the adjacent healthy cable, the core judgment process begins. This process utilizes a magnetic field signal as a precise synchronous trigger signal. Whenever the locator detects a second magnetic field signal generated by a high-voltage flashover of the target cable, its internal processor immediately checks the acoustic channel for a valid second acoustic signal input within a very short time window. A valid acoustic signal is defined as one whose energy and spectral characteristics conform to a preset fault discharge acoustic model. The logic behind this judgment is that if the fault source is indeed on the target cable, there should be no sound waves generated by discharge at the corresponding location on the adjacent healthy cable. The judgment condition can be expressed as follows: when the second magnetic field signal is triggered, if the measured energy of the second acoustic signal is less than a very small background noise energy threshold, then there is no valid signal. To increase the reliability of the verification conclusions and avoid the randomness of a single measurement, repeated confirmation is required within a preset verification time. This means that the judgment process for high-voltage flashover and synchronous monitoring will be repeated several times, for example, performing 5 to 10 flashover tests within one minute. If, throughout the entire verification period, each triggering of the second magnetic field signal fails to be accompanied by a valid second acoustic signal, then the secondary verification is considered passed. Once the secondary verification is deemed passed, it means the location of the initial fault point has been definitively confirmed, eliminating all reasonable doubt. At this point, the initial fault point is officially locked as the final fault point using an acoustic-magnetic synchronization locator. Modern locators typically integrate a Global Positioning System (GPS) module. Simultaneously with locking the fault point, the device automatically triggers the GPS module to accurately acquire and record the current location's geographic coordinates. This coordinate information, along with other test data, is stored to form a complete fault report.

[0163] The present invention also discloses an intelligent cable fault testing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the cable fault detection method based on the intelligent cable fault testing device described in any of the above embodiments.

[0164] The processor can be a central processing unit (CPU). Of course, depending on the actual use, it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc., and this application does not limit it.

[0165] The memory can be an internal storage unit of a computer device, such as a hard disk or RAM, or an external storage device, such as a plug-in hard disk, smart memory card (SMC), secure digital card (SD), or flash memory card (FC) provided on the computer device. Furthermore, the memory can be a combination of internal storage units and external storage devices of a computer device. The memory is used to store computer programs and other programs and data required by the computer device. The memory can also be used to temporarily store data that has been output or will be output. This application does not limit this.

[0166] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of protection of this application is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of one or more embodiments of this application as described above, which are not provided in detail for the sake of brevity.

[0167] One or more embodiments in this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of this application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments in this application should be included within the protection scope of this application.

Claims

1. A cable fault detection method based on an intelligent cable fault testing device, characterized in that, The method, applied to a cable fault testing device, which includes a cable fault tester and an acoustic-magnetic synchronous positioning instrument, comprises the following steps: Connect the high-voltage flashover unit and the first signal acquisition channel of the cable fault tester to the target fault cable, and at the same time connect the second signal acquisition channel to a nearby healthy cable laid in parallel with the target fault cable. A high-voltage pulse is applied to the target faulty cable through a high-voltage flashover unit, and the active reflection signal of the target faulty cable and the passive coupling induction signal of the adjacent healthy cable are captured simultaneously using the first signal acquisition channel and the second signal acquisition channel. The coarse fault distance of the fault point in the target fault cable is calculated based on the active reflection signal and the time-domain reflection method. The energy attenuation gradient analysis module of the cable fault tester is called to calculate and compare the energy attenuation gradient of the active reflected signal and the passive coupled induction signal to obtain the energy source determination result. The initial phase offset verification module of the cable fault tester is called to calculate and verify the initial phase offset of the active reflection signal and the passive coupling induction signal, and the coupling path confirmation result is obtained. The results of energy source determination and coupling path confirmation are subjected to a logical AND operation to form the signal identification result; Based on the signal recognition results, the active reflection signal is marked as the real fault signal, and the spectral characteristics of the real fault signal are extracted. At the same time, the passive coupling induction signal is marked as the interference signal, and the interference signal is suppressed. The high-voltage flashover unit triggers multiple high-voltage flashovers within the area determined by the rough measurement of the fault distance, and the magnetic field signal and continuous on-site acoustic signal are synchronously collected by the acoustic-magnetic synchronization locator. Based on the triggering time of each magnetic field signal, multiple synchronous acoustic samples are extracted from the continuous field acoustic signals. Each synchronous acoustic sample contains fault discharge sound and background noise. Multiple asynchronous acoustic samples that do not contain fault discharge sounds are extracted from continuous on-site acoustic signals and used as background noise samples. Calculate the average spectrum of the synchronous acoustic sample and the average spectrum of the background noise sample, respectively. By comparing the average spectrum, the characteristic frequency band with a signal-to-noise ratio higher than that of the background noise of the fault discharge sound is identified, and the optimal acoustic filtering parameters of the acoustic channel of the acoustic-magnetic synchronization locator are dynamically generated based on the characteristic frequency band. The configured acoustic-magnetic synchronous locator is used to perform precise locating near the coarsely measured fault distance of the target fault cable, and secondary verification is performed at the corresponding location of the adjacent healthy cable to confirm the absence of acoustic-magnetic signal, thus completing the closed-loop location of the fault point.

2. The cable fault detection method based on an intelligent cable fault testing device according to claim 1, characterized in that, The method of simultaneously capturing the active reflected signal of the target faulty cable and the passive coupling induction signal of the adjacent healthy cable using the first signal acquisition channel and the second signal acquisition channel includes the following steps: Set synchronous acquisition trigger thresholds and pre-trigger data buffers in the first and second signal acquisition channels; The signal amplitude in the first signal acquisition channel is monitored in real time. When the signal amplitude exceeds the synchronous acquisition trigger threshold, high-frequency synchronous data acquisition of the first and second signal acquisition channels is started. Based on the preset acquisition time window, the original waveform data of the first signal acquisition channel and the second signal acquisition channel after the high voltage pulse is generated are recorded respectively. The acquired raw waveform data is digitally filtered and normalized to generate the active reflection signal of the target faulty cable and the passive coupling induction signal of the adjacent healthy cable.

3. The cable fault detection method based on an intelligent cable fault testing device according to claim 1, characterized in that, The process of calling the energy attenuation gradient analysis module of the cable fault tester to calculate and compare the energy attenuation gradients of the active reflected signal and the passively coupled inductive signal to obtain the energy source determination result includes the following steps: Fast Fourier transforms were performed on the active reflected signal and the passively coupled induced signal, respectively, to obtain the amplitude-frequency response curves of the two signals; The broadband total energy is obtained by calculating the active reflection signal and the passive coupling induction signal, and a first energy ratio is generated; The amplitude-frequency response curves of the two signals are divided into multiple pre-defined non-overlapping narrowband frequency bands. Calculate the signal energy in each narrowband frequency band and generate the second energy ratio for each frequency band; The energy decay gradient is obtained by analyzing the trend of the second energy ratio changing with frequency. The first energy ratio and energy decay gradient are compared with the preset coupling model threshold, and the energy source determination result is output.

4. The cable fault detection method based on an intelligent cable fault testing device according to claim 1, characterized in that, The process of calling the initial phase offset verification module of the cable fault tester to calculate and verify the initial phase offset of the active reflection signal and the passive coupling induction signal to obtain the coupling path confirmation result includes the following steps: The active reflection signal and the passively coupled induction signal are upsampled using an interpolation algorithm; Calculate the first derivatives of the active reflected signal and the passively coupled induced signal after upsampling, and determine the starting time point of the signal waveform based on the first non-zero point of the first derivative; Calculate the time difference between the starting time points of the active reflection signal and the passive coupling induction signal to obtain the absolute time value of the initial phase shift; The absolute time value is converted into a phase angle based on the main frequency of the actual fault signal; The phase angle is compared with the theoretical phase offset range set according to the cable spacing and medium parameters, and the coupling path confirmation result is output.

5. The cable fault detection method based on an intelligent cable fault testing device according to claim 1, characterized in that, The process of identifying the characteristic frequency band where the signal-to-noise ratio of the fault discharge sound is higher than that of the background noise by comparing the average spectrum, and dynamically generating the optimal acoustic filtering parameters for the acoustic channel of the acoustic-magnetic synchronization locator based on the characteristic frequency band, includes the following steps: The estimated spectrum of the pure fault sound is obtained by subtracting the average spectrum of the synchronous acoustic sample from the average spectrum of the background noise sample. Calculate the signal-to-noise ratio (SNR) between the estimated spectrum of the clean fault sound and the average spectrum of the background noise sample, and generate an SNR spectrum map covering the entire frequency band; Search the signal-to-noise ratio spectrum for all frequency bands whose signal-to-noise ratio values ​​exceed a preset threshold as feature frequency bands; Analyze the frequency position, bandwidth and continuity of all characteristic frequency bands, and select the optimal filter type based on the analysis results. The filter types include high-pass filter, low-pass filter and band-pass filter. The determined filter type and corresponding cutoff frequency are packaged to form the optimal acoustic filtering parameters for the acoustic channel of the acoustic-magnetic synchronization locator.

6. The cable fault detection method based on an intelligent cable fault testing device according to claim 1, characterized in that, The process of using a configured acoustic-magnetic synchronous locator to perform precise location of the faulty cable near the coarsely measured fault distance, and then verifying the location at the corresponding position on a nearby healthy cable to confirm the absence of acoustic-magnetic signals, thus completing the closed-loop location of the fault point, includes the following steps: The configured acoustic-magnetic synchronization locator receives the synchronized first acoustic signal and first magnetic field signal near the coarsely measured fault distance of the target fault cable, and calculates the signal time difference between the first acoustic signal and the first magnetic field signal in real time. The acoustic-magnetic synchronization locator is moved based on the signal time difference until the signal time difference is minimized, and the physical location of the acoustic-magnetic synchronization locator is determined as the initial fault point of the target fault cable. After secondary verification of the adjacent healthy cable at the location corresponding to the initial fault point using an acoustic-magnetic synchronization locator, the initial fault point is identified as the final fault point, thus completing the closed-loop location.

7. The cable fault detection method based on an intelligent cable fault testing device according to claim 6, characterized in that, After completing secondary verification using an acoustic-magnetic synchronization locator on a nearby healthy cable at the location corresponding to the initial fault point, the initial fault point is locked as the final fault point, and the closed-loop localization is completed. This includes the following steps: Move the probe of the acoustomagnetic synchronization locator directly above the adjacent healthy cable; The second acoustic signal and the second magnetic field signal originating from a nearby healthy cable were continuously monitored using an acoustic-magnetic synchronization pinpoint instrument. Determine whether a valid second acoustic signal exists at the synchronization moment triggered by the second magnetic field signal; If no valid second acoustic signal is detected within the preset verification time, the second verification is deemed to have passed. If the secondary verification is successful, the initial fault point will be located as the final fault point using an acoustic-magnetic synchronization locator, and the geographical coordinates of the final fault point will be determined.

8. An intelligent cable fault testing device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the cable fault detection method based on the intelligent cable fault testing device as described in any one of claims 1 to 7.

Images