Cable sheath damage monitoring method, chip, computer device, readable storage medium and program product

Abstract

The application relates to a cable sheath damage monitoring method, a chip, a computer device, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring an acoustic emission original signal sequence and a micro-leakage current original signal sequence carrying a synchronization timestamp, and acquiring compensation filter coefficients; wherein the compensation filter is constructed according to the acoustic transfer function of a propagation path between the cable sheath and a support; performing frame frequency domain transformation on the acoustic emission original signal sequence to obtain a first transformed signal sequence, and performing frequency domain compensation operation on the first transformed signal sequence according to the compensation filter coefficients to obtain a reconstructed acoustic emission signal sequence; performing filtering processing on the micro-leakage current original signal sequence according to the micro-leakage current original signal sequence to obtain a leakage current signal sequence; and determining the cable sheath state according to the reconstructed acoustic emission signal sequence and the leakage current signal sequence. The method can improve the timeliness of cable sheath damage monitoring.

Description

Technical Field

[0001] This application relates to the field of power facility monitoring technology, and in particular to a method for monitoring cable sheath damage, a chip, a computer device, a computer-readable storage medium, and a computer program product. Background Technology

[0002] With the large-scale construction of urban power infrastructure, the laying density of high-voltage cables in underground tunnels, utility corridors, and other scenarios continues to increase. As the first layer of insulation protection for cables, the cable sheath is subjected to the combined effects of various factors such as environmental stress, mechanical impact, temperature and humidity fluctuations, and stray currents over a long period of time. During operation, early damage such as microcracks, interface peeling, and dielectric fatigue will gradually occur. If the above damage is not identified in time at an early stage, the damage will continue to evolve, eventually leading to insulation breakdown and even serious accidents such as large-scale power outages and fires.

[0003] Related technologies for cable sheath condition detection mainly include partial discharge detection, infrared thermography, ultrasonic testing, and sheath insulation resistance measurement. A common limitation of these methods is that the physical quantities they can detect (discharge pulses, thermal anomalies, and decreased insulation resistance) are all consequence-type signals that appear only after the sheath damage has evolved to the mid-to-late stages and macroscopic insulation failure has occurred. These methods lack the ability to detect the early stages of damage, when only minor mechanical disturbances occur without accompanying electrical anomalies. This results in a significant early detection blind spot in the monitoring system, missing the optimal maintenance intervention window and failing to achieve effective detection in the early stages of cable sheath damage. Consequently, monitoring lag cannot meet the need for continuous perception and early warning throughout the entire sheath damage process. Summary of the Invention

[0004] Therefore, it is necessary to provide a cable sheath damage monitoring method, chip, computer equipment, computer-readable storage medium, and computer program product that can improve the timeliness of cable sheath damage monitoring in response to the above-mentioned technical problems.

[0005] In a first aspect, this application provides a method for monitoring cable sheath damage, including:

[0006] The original acoustic emission signal sequence and the original micro-leakage current signal sequence carrying synchronization timestamps are obtained, and the coefficients of the compensation filter are obtained; wherein, the compensation filter is constructed based on the acoustic transfer function of the propagation path between the cable sheath and the support.

[0007] The original acoustic emission signal sequence is subjected to frame-by-frame frequency domain transformation to obtain a first transformed signal sequence, and frequency domain compensation operation is performed on the first transformed signal sequence according to the compensation filter coefficients to obtain a reconstructed acoustic emission signal sequence;

[0008] The original micro-leakage current signal sequence is filtered according to the original micro-leakage current signal sequence to obtain the leakage current signal sequence.

[0009] The cable sheath condition is determined based on the reconstructed acoustic emission signal sequence and the leakage current signal sequence.

[0010] In one embodiment, obtaining the compensation filter coefficients includes:

[0011] Obtain the acoustic transfer function and regularization stability factor of the propagation path;

[0012] The acoustic transfer function is determined by obtaining the response signal at the sensor end after applying an excitation signal to the propagation path, and based on the excitation signal and the response signal; the regularization stabilization factor is obtained based on the noise power spectral density corresponding to the propagation path and the power spectral density of the acoustic emission source signal.

[0013] Based on the acoustic transfer function and the regularization stabilization factor, the compensation filter is constructed, and the compensation filter coefficients are obtained.

[0014] In one embodiment, the method further includes:

[0015] Obtain the ambient temperature of the cable sheath;

[0016] When the change in ambient temperature exceeds a preset temperature change threshold, the viscoelastic physical model parameters of the rubber pad in the propagation path are corrected according to the pre-established compensation relationship between temperature and elastic modulus.

[0017] The acoustic transfer function is updated based on the corrected viscoelastic physical model parameters, and the compensation filter coefficients are regenerated based on the updated acoustic transfer function.

[0018] In one embodiment, the step of filtering the original micro-leakage current signal sequence to obtain a leakage current signal sequence includes:

[0019] The original signal sequence of the micro-leakage current is processed by a digital frequency-locked loop to obtain the instantaneous frequency of the fundamental power frequency interference.

[0020] Based on the instantaneous frequency, generate a set of multiple harmonic orthogonal basis vectors synchronized with the power grid;

[0021] The harmonic interference estimate for each sampling time is determined based on the harmonic orthogonal basis vector set at each sampling time and the current weight vector.

[0022] Based on the sampled values ​​in the original signal sequence of the micro-leakage current and the estimated harmonic interference value, the net leakage current value at each sampling time is determined.

[0023] The weight vector is updated based on the net leakage current value and the harmonic orthogonal basis vector set to obtain the leakage current signal sequence.

[0024] In one embodiment, determining the cable sheath condition based on the reconstructed acoustic emission signal sequence and the leakage current signal sequence includes:

[0025] According to a preset time window, the reconstructed acoustic emission signal sequence is obtained to extract acoustic emission features and the first event time corresponding to each acoustic emission event;

[0026] Based on the time window, the leakage current signal sequence is obtained to extract microcurrent features and the second event time corresponding to each transient pulse;

[0027] A synchronization event pair is determined based on the first event time and the second event time;

[0028] The cable sheath condition is determined based on the acoustic emission characteristics, the microcurrent characteristics, and the synchronization events.

[0029] In one embodiment, determining the cable sheath state based on the acoustic emission characteristics, the microcurrent characteristics, and the synchronization event pair includes:

[0030] A time-series feature matrix is ​​constructed based on the acoustic emission characteristics and microcurrent characteristics of the current time window and several historical time windows;

[0031] The temporal feature matrix is ​​input into a pre-trained temporal inference model to obtain a probability assessment result of the current health status of the sheath.

[0032] Based on the probability assessment results and the synchronization event pair, the current health status level of the sheath is obtained.

[0033] Secondly, this application also provides a chip, which includes a processor and a data interface. The processor reads instructions stored in the memory through the data interface and can execute the steps of the method described in the first aspect.

[0034] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in the first aspect.

[0035] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0036] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in the first aspect.

[0037] The aforementioned cable sheath damage monitoring method, chip, computer equipment, computer-readable storage medium, and computer program product acquire original acoustic emission signal sequences and original micro-leakage current signal sequences carrying synchronization timestamps, and obtain compensation filter coefficients. The compensation filter is constructed based on the acoustic transfer function of the propagation path between the cable sheath and the support. The original acoustic emission signal sequence is subjected to frame-by-frame frequency domain transformation to obtain a first transformed signal sequence, and frequency domain compensation is performed on the first transformed signal sequence based on the compensation filter coefficients to obtain a reconstructed acoustic emission signal sequence. Since the compensation filter is constructed based on the acoustic transfer function of the propagation path, the frequency domain compensation operation can compensate for the amplitude attenuation and waveform distortion experienced by the acoustic emission signal in the multi-medium propagation path between the sheath and the support, making the energy and amplitude characteristics of the reconstructed acoustic emission signal sequence closer to the actual values ​​at the sheath. The physical state of the acoustic emission source is determined, thus solving the problem that the original acoustic emission signal cannot accurately reflect the early damage level due to path distortion. The original micro-leakage current signal sequence is filtered to suppress interference components mixed in with the original acquired signal, and the effective leakage current component reflecting the change of sheath insulation state is extracted to obtain the leakage current signal sequence. The cable sheath state is determined by reconstructing the acoustic emission signal sequence and the leakage current signal sequence. The reconstructed acoustic emission signal sequence reflects the mechanical damage information of the sheath, and the leakage current signal sequence reflects the electrical damage information of the sheath. Both carry synchronous timestamps, which can be correlated and analyzed in the time dimension. This enables the judgment of the entire process of sheath damage from early mechanical micro-damage to mid-to-late insulation conduction, solving the monitoring lag problem and improving the timeliness of cable sheath damage monitoring. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a diagram illustrating the application environment of a cable sheath damage monitoring method in one embodiment.

[0040] Figure 2 This is a flowchart illustrating a cable sheath damage monitoring method in one embodiment;

[0041] Figure 3 This is a schematic diagram of the acoustic emission signal path compensation and reconstruction process in one embodiment;

[0042] Figure 4 This is a schematic diagram of the adaptive harmonic cancellation process for microcurrent signals in one embodiment;

[0043] Figure 5 This is a schematic diagram of the overall processing architecture of the edge inference terminal in one embodiment;

[0044] Figure 6 This is a flowchart illustrating a cable sheath damage monitoring method in another embodiment;

[0045] Figure 7 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0047] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0048] The cable sheath damage monitoring method provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 is used to collect signals and send them to server 104 for processing and calculation. It can also display the information processed by server 104 and can communicate with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed on the cloud or other network servers. Terminal 102 can be an embedded smart terminal, integrating an embedded edge computing module and a built-in trained and optimized neural network model, enabling on-site data preprocessing, feature extraction, and fault classification to achieve "on-site early warning" and significantly shorten response time. Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0049] In one exemplary embodiment, such as Figure 2 As shown, a method for monitoring cable sheath damage is provided, which can be applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps S202 to S208. Wherein:

[0050] Step S202: Obtain the original acoustic emission signal sequence and the original micro-leakage current signal sequence carrying the synchronization timestamp, and obtain the compensation filter coefficients.

[0051] The acoustic emission (AE) raw signal sequence can be a raw digital signal sequence acquired by an AE sensor mounted on the cable support. This signal reflects the waveform of elastic waves radiated by mechanical damage events such as microcrack propagation, interface friction, and material fatigue during cable sheath operation, propagated through a multi-medium path to the sensor, and contains time-domain information of the sheath's mechanical damage state. The micro-leakage current raw signal sequence can be a raw digital signal sequence acquired by a micro-current sensor installed in the sheath grounding circuit. This signal reflects the weak leakage current and transient discharge pulses generated on the sheath surface or through the penetration path after the sheath insulation performance degrades, and contains time-domain information of the sheath's electrical damage state. Both signals carry synchronized timestamps, using the same clock source as a reference, and time-marking each sampling point at the acquisition time, thereby enabling precise time alignment and correlation analysis of the two signals in subsequent processes.

[0052] The compensation filter is constructed based on the acoustic transfer function of the propagation path between the cable sheath and the support. For example, the compensation filter coefficients refer to the set of coefficients of a digital filter that inversely compensates for the amplitude attenuation and waveform distortion experienced by the acoustic emission signal in the multi-medium propagation path of the sheath, rubber pad, and support. This set of coefficients can be pre-calculated offline and stored in the programmable logic device of the edge terminal for online real-time access.

[0053] In some embodiments, the server 104 can directly read the pre-fixed compensation filter coefficients from the storage area of ​​the programmable logic device, and simultaneously read the current sampling data from the high-speed ADC (Analog-to-Digital Converter) channel corresponding to the AE sensor and the high-precision ADC channel corresponding to the micro-current sensor. Both samplings are triggered by the same clock source, and each sampling point is automatically appended with a synchronization timestamp based on this unified clock during acquisition. The timestamp error is less than one microsecond, and then the server summarizes the data.

[0054] For example, server 104 can obtain the acoustic transfer function and regularization stabilization factor of the propagation path, and then construct a compensation filter based on the acoustic transfer function and regularization stabilization factor to obtain the compensation filter coefficients. The acoustic transfer function is determined by obtaining the response signal at the sensor end after applying an excitation signal to the propagation path, and based on the excitation signal and the response signal; the regularization stabilization factor is obtained based on the noise power spectral density corresponding to the propagation path and the power spectral density of the acoustic emission source signal.

[0055] Furthermore, the process of determining the acoustic transfer function is as follows. Let the original acoustic emission signal generated at the cable sheath be... The signal collected by the AE sensor on the bracket side is , The impulse response function for the "sheath-rubber pad-support" multi-media path. Considering environmental and circuit noise, the acquisition process in the time domain can be modeled as follows:

[0056]

[0057] in, This represents convolution. In the frequency domain, according to the convolution theorem:

[0058]

[0059] in, They are Fourier transform, It is the acoustic transfer function of the path, that is, a complete description of the frequency domain response of the path. The propagation path is obtained by applying a known excitation signal (e.g., standard lead breakage excitation or vibration table frequency sweep excitation) to the propagation path, while simultaneously acquiring the response signal at the sensor end, and then performing Fourier transforms on the excitation signal and the response signal respectively and taking the ratio of their spectra.

[0060] Furthermore, the regularization stabilization factor The physical meaning of NSR is the ratio of noise to signal power at that frequency point, i.e., the noise-to-signal ratio.

[0061]

[0062] in, It is the power spectral density of noise, which is obtained by collecting background signals and performing spectrum analysis in a silent state without any acoustic emission events; It is the prior power spectral density of the acoustic emission source signal, which is obtained by statistical analysis of the source signals of a large number of standard excitation experiments in the laboratory.

[0063] After obtaining the acoustic transfer function and regularization stability factor The server can then construct a regularized optimal compensation filter based on these two factors. Its frequency domain expression is:

[0064]

[0065] in for The complex conjugate, for The square of the amplitude. This compensation filter is effective at frequencies with high signal-to-noise ratio (i.e., ...). Much larger hour), Ignore Approximately equal to The reciprocal of the function is equivalent to a perfect inverse filter, achieving high-fidelity restoration of the source signal; at frequencies with low signal-to-noise ratio (i.e., ... (Notch point approaching zero), denominator from The bottom is used to prevent division by zero, while the numerator... Similarly, it approaches zero, making The overall signal-to-noise ratio (SNR) approaches zero, thus automatically reducing the noise gain at that frequency point to zero, rather than amplifying it infinitely. This fundamentally avoids the noise divergence problem at notch points caused by naive inverse filtering. It performs high-fidelity "inverse filtering" at high SNR locations and automatically switches to "attenuation filtering" at low SNR locations (notch points), fundamentally suppressing noise amplification. The server can then... Discretize the coefficients to generate a set of high-order FIR (Finite Impulse Response) digital filter coefficients, and store the set of coefficients in a lookup table or block memory of a programmable logic device to obtain the compensation filter coefficients.

[0066] Through the above steps, the synchronous acquisition of the two heterogeneous signals and the preparation of the compensation filter coefficients were completed. The two signals share a unified time reference, which ensures the accuracy of subsequent time correlation analysis. The compensation filter, through regularized optimal design, simultaneously takes into account high-fidelity reproduction in the high signal-to-noise ratio band and noise suppression in the low signal-to-noise ratio band.

[0067] Step S204: Perform frame-by-frame frequency domain transformation on the original acoustic emission signal sequence to obtain the first transformed signal sequence, and perform frequency domain compensation operation on the first transformed signal sequence according to the compensation filter coefficients to obtain the reconstructed acoustic emission signal sequence.

[0068] The first transformed signal sequence refers to the set of complex spectra corresponding to each frame obtained after dividing the original acoustic emission signal sequence into frames and performing frequency domain transformation. The complex spectrum of each frame is used to describe the amplitude and phase information of the acoustic emission signal at each frequency point within the time window. The reconstructed acoustic emission signal sequence refers to the signal sequence obtained after performing frequency domain compensation operation on the first transformed signal sequence and restoring it to the time domain through inverse transformation. This sequence has compensated for the amplitude attenuation and dispersion distortion of the acoustic emission signal in the above-mentioned "sheath-rubber pad-support" multi-medium propagation path, and its waveform energy and amplitude characteristics are closer to the real physical state of the acoustic emission source at the sheath.

[0069] For example, Figure 3 This diagram illustrates the acoustic emission signal path compensation and reconstruction process, showing the complete processing chain from signal generation to high-fidelity reconstruction: The acoustic emission signal generated by the micro-elastic wave source (including mechanical damage events such as crack propagation and interface friction) passes through a multi-medium propagation path (i.e., the path described by the transfer function) consisting of a sheath, rubber pad, and support, and is then acquired by an AE sensor to obtain a distorted acquisition signal. This acquisition signal is then sequentially processed by an FFT (Fast Fourier Transform) module to complete frame-by-frame frequency domain transformation, followed by a regularization compensation filter module to complete frequency domain compensation calculation, and finally by an IFFT (Inverse Fast Fourier Transform) module to complete inverse transformation and time-domain reconstruction, outputting a high-fidelity reconstructed waveform, i.e., the reconstructed acoustic emission signal sequence.

[0070] In some embodiments, the server can directly apply pre-calculated and fixed high-order FIR digital filter coefficients to the original acoustic emission signal sequence in the time domain and perform time-domain convolution operations, completing the effect of the compensation filter on the signal in one step in the time domain to obtain the reconstructed acoustic emission signal sequence. The time-domain convolution method is simple to implement and suitable for application scenarios with low real-time requirements or short signal frame lengths.

[0071] For example, the server can process the raw acoustic emission signal sequence. Frame segmentation is performed, dividing the continuous signal into several frames with a fixed frame length (e.g., 512 points). Adjacent frames maintain an overlap rate of no less than 50% to avoid signal discontinuity caused by frame switching. Each frame of data is multiplied by a Hanning window function to suppress spectral leakage.

[0072] Subsequently, the server can call the Fast Fourier Transform (FFT) module in the programmable logic device to perform STFT (Short-Time Fourier Transform) on the windowed data of each frame, and obtain the complex spectrum corresponding to that frame. This refers to the element corresponding to that frame in the first transformed signal sequence. The above framing, windowing, and transform operations are performed continuously on each frame to finally obtain the complete first transformed signal sequence.

[0073] After obtaining the first transformed signal sequence, the server can retrieve the fixed compensation filter coefficients from the storage area of ​​the programmable logic device. The complex spectrum of each frame in the first transformed signal sequence Parallel complex multiplication is performed at each frequency point to obtain the compensated spectrum for each frame. ,Right now:

[0074]

[0075] The frequency domain multiplication operation is executed in a highly parallelized manner within the programmable logic device (PLD), with complex multiplications at each frequency point occurring simultaneously. The computational latency is limited only by the PLD's clock cycle. After completing the frequency domain compensation operation, the server can invoke the inverse fast Fourier transform (IFFT) module within the PLD to process the compensated spectrum for each frame. An inverse transform is performed to restore the frequency domain compensation result to the time domain. The inverse transform results of adjacent frames are then overlapped and superimposed to obtain a continuous time-domain reconstructed signal sequence, i.e., a reconstructed acoustic emission signal sequence. The three operations described above—frame-based frequency domain transformation, frequency domain compensation multiplication, and inverse transformation—are executed in parallel in a pipelined manner within the programmable logic device, ensuring that the output delay of the reconstructed acoustic emission signal sequence does not exceed twice the length of a single frame, thus guaranteeing real-time processing capabilities.

[0076] Through the above steps, path distortion compensation and high-fidelity reconstruction of the original acoustic emission signal sequence were completed. Compared with the uncompensated original acoustic emission signal, the energy and amplitude of each event in the reconstructed acoustic emission signal sequence have compensated for the frequency-selective attenuation introduced by the strong viscoelastic damping of the rubber pad, allowing the event energy, which was originally severely compressed due to path distortion, to be restored to a level close to the true source signal. At the same time, since the regularized optimal compensation filter automatically returns the gain to zero at the notch point, it avoids the noise being infinitely amplified at the strong attenuation frequency point, ensuring the numerical stability of the reconstruction process. This allows the subsequent feature extraction steps to be performed on the physically accurate reconstructed signal, thereby improving the accuracy of early damage identification.

[0077] Step S206: Filter the original micro-leakage current signal sequence according to the original micro-leakage current signal sequence to obtain the leakage current signal sequence.

[0078] The leakage current signal sequence refers to the effective current signal sequence reflecting the degradation state of the sheath insulation performance, extracted after suppressing and canceling the power frequency and harmonic interference components in the original micro-leakage current signal sequence. It includes the DC leakage component and transient discharge pulse component generated by insulation degradation during the mid-to-late-stage damage phases of the sheath. The original micro-leakage current signal sequence is mixed with two main types of interference during acquisition: one is common-mode interference generated by the high-voltage-to-ground capacitance coupling of the cable, with amplitudes reaching the volt level, which has been eliminated in the hardware isolation circuit; the other is differential-mode interference of power frequency and its harmonics induced on the sensor by the main cable current, with amplitudes reaching the ampere level, far exceeding the microampere amplitude of the target micro-leakage current signal, which can be further eliminated in this step.

[0079] For example, the server can use a digital frequency-locked loop to process the original micro-leakage current signal sequence to obtain the instantaneous frequency of the power frequency interference fundamental wave; based on the instantaneous frequency, generate a set of multiple harmonic orthogonal basis vectors synchronized with the power grid; determine the harmonic interference estimate for the corresponding sampling time based on the harmonic orthogonal basis vector set at each sampling time and the current weight vector; determine the net leakage current value at each sampling time based on each sampling value and the harmonic interference estimate in the original micro-leakage current signal sequence; update the weight vector based on the net leakage current value and the harmonic orthogonal basis vector set to obtain the leakage current signal sequence.

[0080] The harmonic orthogonal basis vector set refers to a set of mutually orthogonal sine and cosine function vectors generated in real time within a programmable logic device based on a locked instantaneous fundamental frequency of the power grid. This set covers the fundamental frequency and selected odd harmonics, and is used for parametric modeling of the structure of power frequency harmonic interference. The weight vector is a vector in the adaptive filter that stores the estimated amplitude coefficients of each harmonic. Its length is equal to twice the selected harmonic order. Through adaptive iterative updates at each sampling point, the interference estimate continuously approximates the actual interference.

[0081] For example, Figure 4 This diagram illustrates the adaptive harmonic cancellation process for micro-current signals, showing the complete processing chain from acquisition to purification: The weak leakage current signal generated by the micro-current source is superimposed with strong harmonic interference from the power frequency power supply during transmission. The two are added together to form a superimposed current signal, which is then measured by a current sensor to obtain the current sensor measurement signal. This measurement signal enters a dual-core parallel processing module composed of a digital frequency-locked loop (FLL) and a parametric Fourier transform (LMS) algorithm. The FLL is responsible for locking the instantaneous fundamental frequency of the power grid in real time and generating a synchronous harmonic orthogonal basis. The parametric Fourier transform (LMS) algorithm uses the above basis to adaptively estimate the harmonic interference. Finally, the interference estimate is subtracted from the measurement signal through a cancellation operation, and the processed current signal without distortion, i.e., the leakage current signal sequence, is output.

[0082] In some embodiments, the server can directly apply a fixed-coefficient digital band-stop filter bank to the original micro-leakage current signal sequence, design notch filters for the power frequency and each harmonic frequency point, and execute them in series to filter out the signal components at each harmonic frequency point to obtain the leakage current signal sequence. This method is simple to implement and suitable for scenarios where the power grid frequency is stable and the harmonic amplitude changes slowly.

[0083] For example, after the original signal sequence of micro-leakage current is isolated by a hardware isolation circuit to eliminate common-mode interference, the signal acquired by the ADC is... Mathematically, this can be represented as the target microcurrent signal. Differential mode harmonic interference sum:

[0084]

[0085] Among them, differential mode harmonic interference The structure is a superposition of a series of time-varying harmonics:

[0086]

[0087] in, For the selected set of harmonic orders (e.g.) ), It is the fundamental angular frequency of the power grid (e.g., 49.9Hz~50.1Hz). and The unknown harmonic amplitude coefficient is a function that varies rapidly with the power grid load.

[0088] Furthermore, in order to... Extracting data in real time without distortion The server can implement a dual-core parallel processing structure in a programmable logic device to perform two tasks: baseband tracking and adaptive interference cancellation.

[0089] In terms of baseband tracking, the server can use the original signal sequence with micro-leakage current. It serves as the input, implementing a high-performance digital FLL (Frequency-Locked Loop) in a programmable logic device, which uses the most energetic fundamental frequency component in the signal to lock the current power grid's fundamental frequency in real time and with precision. The instantaneous value. Digital frequency-locked loops require no external reference source, relying entirely on the input signal itself to complete frequency tracking. This allows for locking accuracy controlled at the millihertz level, ensuring strict synchronization with the power grid even when the grid frequency fluctuates (e.g., within the range of 49.9 Hz to 50.1 Hz). The server can generate harmonic orthogonal basis vector sets in real time within the programmable logic device. :

[0090]

[0091] in, This is the real-time locking angular frequency of the FLL, and this vector group contains a total of The element covers the fundamental frequency up to the element. The in-phase and quadrature components of the subharmonics are all strictly synchronized with the grid base frequency locked in real time by the digital frequency-locked loop. They are generated entirely within the programmable logic device without the need for external reference signals.

[0092] In terms of adaptive interference cancellation, the server can run a parameterized least mean square adaptive filter, whose weight vector... For a short vector of length only 2N:

[0093]

[0094] Each element represents the current estimated value of the corresponding subharmonic amplitude coefficient. At each sampling time... The server can transmit the weight vector The harmonic orthogonal basis vector set at the current moment Perform the inner product operation to obtain the estimated harmonic interference value at the current sampling time. :

[0095]

[0096] Subsequently, the server can obtain the current sampled values ​​of the original signal sequence of the micro-leakage current. Subtracting the estimated value of harmonic interference The net leakage current value at the current sampling time is obtained. That is, error signal :

[0097]

[0098] This net leakage current value is the estimated value of the target micro-leakage current signal extracted at the current moment. After interference cancellation is completed, the server can utilize the error signal. Orthogonal basis vector set of the current harmonic Update the weight vector according to the standard least mean square algorithm. :

[0099]

[0100] in, The step size parameter controls the balance between the convergence speed and steady-state error of the weight vector. Through iterative updates at each sampling point, the weight vector continuously approximates the true harmonic amplitude coefficients, ensuring that the estimated harmonic interference keeps pace with the rapid changes in harmonic amplitude caused by grid load fluctuations within tens of milliseconds, maintaining a continuous and effective interference cancellation effect. The net leakage current value at each sampling time is also considered. By summarizing them in chronological order, we obtain the leakage current signal sequence.

[0101] Through the above steps, real-time adaptive cancellation of power frequency and harmonic differential mode interference in the original micro-leakage current signal sequence was achieved, suppressing background interference to a level far below the target signal amplitude. This allowed for the clear extraction of the micro-ampere to milliampere level true leakage current components, which were originally submerged in ampere-level harmonic noise background. The introduction of a digital frequency-locked loop enables the system to track grid frequency fluctuations in real time without an external reference source, fundamentally avoiding the harmonic cancellation mismatch problem caused by grid frequency deviation under a fixed frequency assumption. The parameterized least mean square adaptive filter utilizes the known structure of harmonic interference to compress the weight vector length to a minimum, significantly reducing the computational load. This allows the entire interference cancellation process to be executed in real time in a programmable logic device at a rate of one iteration per sampling point.

[0102] Step S208: Determine the cable sheath status based on the reconstructed acoustic emission signal sequence and leakage current signal sequence.

[0103] For example, the server can obtain the reconstructed acoustic emission signal sequence according to a preset time window to extract acoustic emission features and the first event time corresponding to each acoustic emission event; obtain the leakage current signal sequence according to the time window to extract micro-current features and the second event time corresponding to each transient pulse; determine the synchronization event pair according to the first event time and the second event time; and determine the cable sheath state according to the acoustic emission features, micro-current features and synchronization event pair.

[0104] The acoustic emission characteristics refer to the set of feature parameters extracted from the reconstructed acoustic emission signal sequence, used to quantify the degree of mechanical damage activity of the sheath within the current time window. These may include the acoustic emission event rate and average reconstructed energy calculated based on the reconstructed waveform, as well as the average reconstructed amplitude. Since these features are all calculated based on the reconstructed acoustic emission signal sequence, amplitude attenuation and dispersion distortion introduced by multi-medium propagation paths have been compensated for. The microcurrent characteristics refer to the set of feature parameters extracted from the leakage current signal sequence, used to describe the electrical damage state of the sheath within the current time window. These include at least the average DC leakage current, transient pulse count, and total transient pulse energy. The first event time refers to the precise occurrence time corresponding to the peak value of each valid acoustic emission event waveform in the reconstructed acoustic emission signal sequence, recorded with a synchronization timestamp as the reference, accurate to the microsecond level. The second event time refers to the precise occurrence time corresponding to the peak value of each valid transient pulse waveform in the leakage current signal sequence, also recorded with a synchronization timestamp as the reference. Synchronous event pairs refer to the pairing of acoustic emission events and microcurrent transient pulses within a preset synchronization time window, where the time difference between the first event moment and the second event moment is less than a preset synchronization judgment threshold. Their physical meaning is the synchronous response of the same sheath crack propagation event in the mechanical and electrical domains, and is the strongest physical evidence for determining that the sheath has entered a high-risk state of mid-to-late stage crack conduction.

[0105] Furthermore, the server can construct a timing feature matrix based on the acoustic emission characteristics and microcurrent characteristics of the current time window and several historical time windows; input the timing feature matrix into a pre-trained timing inference model to obtain a probability assessment result of the current sheath health status; and obtain the current sheath health status level based on the probability assessment result and the synchronization event pair.

[0106] The temporal feature matrix is ​​a matrix composed of acoustic emission features and microcurrent features of the current time window and several historical time windows arranged in chronological order. It is used to provide input to the temporal inference model that reflects the dynamic process of damage evolution, so that the model can perceive the trend of feature changes over time rather than relying solely on the instantaneous feature value at the current moment.

[0107] In some embodiments, the server can directly output the sheath health status level based on acoustic emission characteristics, microcurrent characteristics, and synchronization event pairs, according to pre-defined rules and logic, without using a timing inference model. It compares each characteristic parameter with a preset threshold one by one. This method has extremely low computational overhead and is suitable for lightweight deployment scenarios with strict limitations on computing resources.

[0108] For example, the server can perform event detection on the reconstructed acoustic emission signal sequence within a preset time window (e.g., one minute, configurable as needed). It identifies valid acoustic emission events by combining amplitude threshold detection with duration criteria, extracts acoustic emission features, and records the first event moment corresponding to the peak waveform of each valid acoustic emission event. Simultaneously, the server can perform low-pass filtering on the leakage current signal sequence within the current window, calculate the average value to obtain the DC leakage mean, detect transient pulses with amplitudes exceeding a preset threshold and durations shorter than a preset upper limit, count their number and total energy to obtain micro-current features, and record the second event moment corresponding to the peak value of each transient pulse.

[0109] After completing feature extraction and event timing recording, the server can perform time-paired retrieval on the first and second event timing lists. Using each first event timing as an anchor point, it searches for a corresponding second event timing within each preset synchronization threshold range before and after it. If a second event timing exists, the acoustic emission event and its corresponding transient pulse are recorded as a synchronization event pair, and the total number of synchronization event pairs within the current window is counted. The existence of synchronization event pairs indicates that the sheath is currently experiencing instantaneous crack propagation and synchronous local insulation conduction, providing direct evidence of the causal relationship between mechanical and electrical damage. This judgment cannot be made solely based on the signal of any one mode.

[0110] Furthermore, the server can concatenate acoustic emission characteristics and microcurrent characteristics from the current time window and several historical time windows in chronological order to construct a temporal feature matrix. This matrix is ​​then input into a lightweight temporal inference model, such as a GRU (Gated Recurrent Unit) network or an HMM (Hidden Markov Model), which is pre-trained and deployed on an edge terminal processor. This yields probability assessments of the current sheath's health status across various states. The temporal feature matrix enables the temporal inference model to perceive the historical evolution trend of features during inference, recognizing dynamic patterns of damage progressively from early to mid-to-late stages, rather than relying solely on instantaneous feature values ​​at the current moment. This significantly improves the ability to identify progressive damage in the mid-term.

[0111] Furthermore, the server can integrate the probability assessment results with the detection results of synchronous event pairs, and output the current sheath health status level according to the following hierarchical judgment logic. When both acoustic emission characteristics and microcurrent characteristics are within the health benchmark range and there are no synchronous event pairs, it is judged as a Level 1 healthy state. When a valid acoustic emission event occurs and the average reconstruction energy is lower than the early damage threshold, and there is no microcurrent abnormality, it is judged as a Level 2 early mechanical micro-damage state. At this stage, the sheath damage is limited to purely mechanical early micro-damage, and there are no electrical abnormalities. This is a unique sensing window of this method. When the average acoustic emission reconstruction energy continuously exceeds the early damage threshold or the acoustic emission event rate shows a continuous upward trend, but there is still no microcurrent transient pulse, it is judged as a Level 3 mid-term crack propagation state. When the probability assessment results show that the damage degree has exceeded the mid-term threshold and a valid synchronous event pair is detected within the current window, it is judged as a Level 4 mid-to-late stage high-risk state for crack conduction. This judgment condition requires that the acoustic emission event and the microcurrent pulse occur simultaneously within the synchronization time window, which is the strongest evidence of instantaneous crack conduction. When the average DC leakage rate consistently exceeds the penetration threshold, it is classified as a Level 5 severe fault, confirming that the sheath has experienced penetration insulation failure. The server can trigger corresponding warning commands based on the above levels and store the current window's feature vector, health status level, and warning timestamp in the local historical database, while simultaneously uploading them to the backend management platform via the communication module.

[0112] For example, Figure 5 This is a schematic diagram of the overall processing architecture of the edge inference terminal, illustrating the complete data processing flow: The left-side AE ​​sensor collects signals, which are then digitized by an ADC and sent to the AE signal reconstruction module in the FPGA (Field Programmable Gate Array) for path compensation reconstruction. The right-side MC (Micro-Current) sensor collects signals, which are then digitized by an ADC and sent to the MC signal cancellation module in the FPGA for harmonic interference cancellation. The two processing results are then fed into the AE feature extraction module and the MC feature extraction module, respectively. Both feature extraction modules are connected to the high-precision timestamp module in the FPGA, ensuring that the two features and their corresponding event times are marked with a unified time base. Finally, these features are integrated into a fusion judgment model running on an ARM (Advanced RISC Machine) core, outputting a sheath health status assessment result with five levels: H1 (healthy), H2 (attention), H3 (warning), H4 (alarm), and H5 (severe).

[0113] Through the above steps, continuous perception and graded early warning of cable sheath damage throughout its entire lifecycle are achieved. The collaborative analysis of reconstructed acoustic emission signal sequences and leakage current signal sequences enables the system to simultaneously cover both early-stage purely mechanical damage and mid-to-late-stage electrical damage, filling the blind spot of existing technologies in the early perception stage. Time-paired retrieval based on synchronous timestamps precisely correlates the two signals in the time dimension, enabling the system to identify high-risk events of simultaneous crack propagation and conduction that cannot be determined by a single mode, while effectively eliminating interference events generated independently by the two signals, significantly reducing the false alarm rate. The time-series reasoning model further enhances the ability to identify progressive damage through the historical evolution trend of perceived features, and the five-level graded judgment logic ensures that the early warning results have clear engineering operability.

[0114] In the aforementioned cable sheath damage monitoring method, the original acoustic emission signal sequence and the original micro-leakage current signal sequence carrying synchronization timestamps are acquired, and the compensation filter coefficients are obtained. The compensation filter is constructed based on the acoustic transfer function of the propagation path between the cable sheath and the support. The original acoustic emission signal sequence is subjected to frame-by-frame frequency domain transformation to obtain a first transformed signal sequence, and frequency domain compensation is performed on the first transformed signal sequence based on the compensation filter coefficients to obtain a reconstructed acoustic emission signal sequence. Since the compensation filter is constructed based on the acoustic transfer function of the propagation path, the frequency domain compensation operation can compensate for the amplitude attenuation and waveform distortion experienced by the acoustic emission signal in the multi-medium propagation path between the sheath and the support, making the energy and amplitude characteristics of the reconstructed acoustic emission signal sequence closer to the physical state of the actual acoustic emission source at the sheath, thereby solving the problem. This method addresses the problem that the original acoustic emission signal cannot accurately reflect the degree of early damage due to path distortion. It filters the original micro-leakage current signal sequence to suppress interference components mixed in with the original acquired signal, extracting the effective leakage current component reflecting changes in the sheath insulation state, thus obtaining a leakage current signal sequence. The cable sheath state is determined based on the reconstructed acoustic emission signal sequence and the leakage current signal sequence. The reconstructed acoustic emission signal sequence reflects the mechanical damage information of the sheath, while the leakage current signal sequence reflects the electrical damage information. Both sequences carry synchronous timestamps, allowing for correlation analysis in the time dimension. This enables the judgment of the entire process of sheath damage from early mechanical micro-damage to mid-to-late stage insulation continuity, solving the monitoring lag problem and improving the timeliness of cable sheath damage monitoring.

[0115] In one exemplary embodiment, such as Figure 6 As shown, the above method further includes steps S302 to S306. Wherein:

[0116] Step S302: Obtain the ambient temperature of the cable sheath.

[0117] Ambient temperature refers to the real-time temperature value at the installation location of the cable sheath and its support. This temperature directly affects the viscoelastic physical properties of the rubber pad, thereby altering the acoustic transfer function of the multi-medium propagation path of "sheath-rubber pad-support". When the ambient temperature changes significantly, the elastic modulus and viscosity coefficient of the rubber pad will change accordingly, causing a deviation between the previously calibrated and fixed compensation filter coefficients and the actual frequency response characteristics of the current propagation path, thus affecting the accuracy of reconstructing the acoustic emission signal sequence. Therefore, the server can continuously acquire the ambient temperature to determine whether the compensation filter coefficients need to be updated.

[0118] For example, the server can collect the current ambient temperature value by deploying a temperature sensor (such as a thermocouple sensor or a platinum resistance temperature sensor) near the cable bracket installation location, and record the temperature value along with the timestamp of the collection time, so as to compare it with a preset temperature change threshold in subsequent steps, thereby establishing the ability to continuously monitor the ambient temperature and ensuring the system's ability to perceive changes in the propagation path characteristics under temperature change conditions.

[0119] Step S304: When the change in ambient temperature exceeds the preset temperature change threshold, the viscoelastic physical model parameters of the rubber pad in the propagation path are corrected according to the pre-established compensation relationship between temperature and elastic modulus.

[0120] The viscoelastic physical model is a parameterized physical model describing the acoustic propagation characteristics of the rubber pad. Using elastic modulus and viscosity coefficient as core parameters, it accurately characterizes the amplitude attenuation and phase delay characteristics of the rubber pad on acoustic emission signals over a wide frequency range with a small number of physical parameters. This model is calibrated in the laboratory before shipment to determine initial parameter values, which are then fixed in the edge terminal. Elastic modulus is a physical parameter describing the elastic deformation capability of the rubber pad; its value decreases with increasing temperature, meaning the higher the temperature, the softer the rubber pad, and the more significant its impact on the propagation characteristics of the acoustic emission signal. Viscosity coefficient is a physical parameter describing the viscous damping characteristics of the rubber pad; it also changes with temperature, directly affecting the energy dissipation rate of the acoustic emission signal during propagation. The temperature-elastic modulus compensation relationship refers to a function relationship established in advance through laboratory measurements, describing the change in the elastic modulus of the rubber pad with temperature. This relationship is also fixed in the edge terminal during pre-shipment calibration. The preset temperature change threshold is the minimum temperature change that triggers the update of the compensation filter coefficients, used to avoid unnecessary computational overhead caused by frequent update operations due to small temperature fluctuations.

[0121] For example, the server can compare the acquired current ambient temperature with the reference temperature recorded during the last update of the compensation filter coefficients to calculate the temperature change. If the temperature change does not exceed a preset temperature change threshold, the server can maintain the current compensation filter coefficients unchanged and continue the reconstruction operation. If the temperature change exceeds the preset temperature change threshold, the server can calculate the corrected elastic modulus value of the rubber pad under the current temperature condition based on the pre-established compensation relationship between temperature and elastic modulus, using the current ambient temperature as input. Based on this, the server can simultaneously correct relevant parameters such as the viscosity coefficient in the viscoelastic physical model, obtaining the corrected viscoelastic physical model parameters. At the same time, the current ambient temperature is recorded as the new reference temperature for future comparisons. This achieves temperature-adaptive correction of the rubber pad viscoelastic physical model parameters, enabling the model parameters to dynamically reflect the true physical characteristics of the rubber pad under the current temperature condition, and avoiding the impact of propagation path characteristic shifts caused by temperature changes on the reconstruction accuracy of the acoustic emission signal.

[0122] Step S306: Update the acoustic transfer function according to the corrected viscoelastic physical model parameters, and regenerate the compensation filter coefficients according to the updated acoustic transfer function.

[0123] For example, the server can take the revised viscoelastic physical model parameters (including the updated elastic modulus and viscosity coefficient) as input, substitute them into the analytical expression of the viscoelastic physical model, and recalculate the acoustic transfer function of the propagation path under the current temperature conditions. Since the viscoelastic physical model is parameterized based on physical parameters, updating the acoustic transfer function only requires substituting a small number of corrected physical parameters into the analytical expression for recalculation. There is no need to repeat the time-consuming offline calibration experiments. The calculation process can be completed in milliseconds on an edge terminal processor.

[0124] After obtaining the updated acoustic transfer function, the server can apply it in the same way as the compensation filter, with the updated... and the current regularization stability factor As input, recalculate the regularized optimal compensation filter. :

[0125]

[0126] in, For the complex conjugate of the updated acoustic transfer function, This is the squared magnitude of the updated acoustic transfer function. The server can then recalculate the result. Discretization generates an updated set of higher-order FIR digital filter coefficients, which are then written to the storage area of ​​the programmable logic device, overwriting the original compensation filter coefficients, thus completing the online update of the compensation filter coefficients. After the update is complete, the frequency domain compensation operation will automatically retrieve the updated coefficients for execution, without interrupting the signal acquisition and processing flow.

[0127] Through the above steps, the compensation filter coefficients are dynamically and adaptively updated to the current temperature conditions. This allows the system to promptly correct the compensation filter to match the current true characteristics of the propagation path when the temperature changes significantly, ensuring the continuous accuracy of the reconstructed acoustic emission signal sequence under different temperature environments. Compared to methods that require re-performing offline calibration experiments, the analytical update method based on the viscoelastic physical model only requires correcting a small number of physical parameters and recalculating them analytically. This significantly reduces the time overhead and engineering complexity of temperature adaptive updates, enabling the system to operate stably for a long time in cable tunnel environments where the temperature varies significantly with the seasons or operating conditions.

[0128] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0129] In one exemplary embodiment, a chip is provided, which includes a processor and a data interface. The processor reads instructions stored in a memory through the data interface and can execute the steps of the above-described cable sheath damage monitoring method.

[0130] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 7As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and databases. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for monitoring cable sheath damage.

[0131] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0132] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0133] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0134] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0135] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0136] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0137] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0138] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for monitoring cable sheath damage, characterized in that, The method includes: The original acoustic emission signal sequence and the original micro-leakage current signal sequence carrying synchronization timestamps are obtained, and the coefficients of the compensation filter are obtained; wherein, the compensation filter is constructed based on the acoustic transfer function of the propagation path between the cable sheath and the support. The original acoustic emission signal sequence is subjected to frame-by-frame frequency domain transformation to obtain a first transformed signal sequence, and frequency domain compensation operation is performed on the first transformed signal sequence according to the compensation filter coefficients to obtain a reconstructed acoustic emission signal sequence; The original micro-leakage current signal sequence is filtered according to the original micro-leakage current signal sequence to obtain the leakage current signal sequence. The cable sheath condition is determined based on the reconstructed acoustic emission signal sequence and the leakage current signal sequence.

2. The method according to claim 1, characterized in that, The process of obtaining the compensation filter coefficients includes: Obtain the acoustic transfer function and regularization stability factor of the propagation path; The acoustic transfer function is determined by obtaining the response signal at the sensor end after applying an excitation signal to the propagation path, and based on the excitation signal and the response signal; the regularization stabilization factor is obtained based on the noise power spectral density corresponding to the propagation path and the power spectral density of the acoustic emission source signal. Based on the acoustic transfer function and the regularization stabilization factor, the compensation filter is constructed, and the compensation filter coefficients are obtained.

3. The method according to claim 2, characterized in that, The method further includes: Obtain the ambient temperature of the cable sheath; When the change in ambient temperature exceeds a preset temperature change threshold, the viscoelastic physical model parameters of the rubber pad in the propagation path are corrected according to the pre-established compensation relationship between temperature and elastic modulus. The acoustic transfer function is updated based on the corrected viscoelastic physical model parameters, and the compensation filter coefficients are regenerated based on the updated acoustic transfer function.

4. The method according to claim 1, characterized in that, The step of filtering the original micro-leakage current signal sequence to obtain a leakage current signal sequence includes: The original signal sequence of the micro-leakage current is processed by a digital frequency-locked loop to obtain the instantaneous frequency of the fundamental power frequency interference. Based on the instantaneous frequency, generate a set of multiple harmonic orthogonal basis vectors synchronized with the power grid; The harmonic interference estimate for each sampling time is determined based on the harmonic orthogonal basis vector set at each sampling time and the current weight vector. Based on the sampled values ​​in the original signal sequence of the micro-leakage current and the estimated harmonic interference value, the net leakage current value at each sampling time is determined. The weight vector is updated based on the net leakage current value and the harmonic orthogonal basis vector set to obtain the leakage current signal sequence.

5. The method according to any one of claims 1 to 4, characterized in that, The step of determining the cable sheath condition based on the reconstructed acoustic emission signal sequence and the leakage current signal sequence includes: According to a preset time window, the reconstructed acoustic emission signal sequence is obtained to extract acoustic emission features and the first event time corresponding to each acoustic emission event; Based on the time window, the leakage current signal sequence is obtained to extract microcurrent features and the second event time corresponding to each transient pulse; A synchronization event pair is determined based on the first event time and the second event time; The cable sheath condition is determined based on the acoustic emission characteristics, the microcurrent characteristics, and the synchronization events.

6. The method according to claim 5, characterized in that, The determination of the cable sheath condition based on the acoustic emission characteristics, the microcurrent characteristics, and the synchronization event includes: A time-series feature matrix is ​​constructed based on the acoustic emission characteristics and microcurrent characteristics of the current time window and several historical time windows; The temporal feature matrix is ​​input into a pre-trained temporal inference model to obtain a probability assessment result of the current health status of the sheath. Based on the probability assessment results and the synchronization event pair, the current health status level of the sheath is obtained.

7. A chip, characterized in that, The chip includes a processor and a data interface. The processor reads instructions stored in the memory through the data interface and can execute the steps of the method according to any one of claims 1 to 6.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images