Anti-interference detection method and device for online positioning of cable fault based on sparse optimization

Abstract

The application discloses a cable fault online positioning anti-interference detection method and device based on sparse optimization and belongs to the technical field of cable fault detection. The method comprises the following steps: an SSTDR detection signal is injected into a to-be-detected cable and echo is collected, a time delay dictionary matrix is constructed by cyclic time delay shifting on the detection signal; a sparse observation model is established by taking the echo as an observation vector, an optimization objective function is constructed, and sparse coefficients are solved; the time delay and amplitude of a reflection component are determined by the non-zero coefficients, near-end interference is identified according to a preset near-end time delay threshold and a detection threshold; after the interference is reconstructed and removed, the residual signal is cross-correlated with the detection signal, and cable fault positioning information is obtained. Through sparse optimization, the SSTDR echo is decomposed and the near-end interference is removed, so that cable fault online anti-interference positioning is realized.

Description

Technical Field

[0001] This invention relates to the technical field of cable fault detection, and in particular to a method and apparatus for online location and anti-interference detection of cable faults based on sparse optimization. Background Technology

[0002] With the development of information technology, electronic equipment in transportation vehicles such as airplanes, ships, and high-speed trains is increasing, leading to increasingly complex wiring harness networks. Wiring harnesses provide power and signals to electrical and electronic equipment, and their performance directly affects system safety and reliability. Wiring harness conductors are exposed to harsh environments such as vibration, friction, high and low temperatures, and high humidity for extended periods, which can easily cause conductor failures or loose connectors, resulting in signal or power supply abnormalities and seriously affecting operational safety. Therefore, testing the continuity, impedance, and insulation performance of cables, as well as rapid fault location, are crucial for the safe and reliable operation of equipment.

[0003] Short circuits, open circuits, and arcing faults can all be essentially classified as impedance change problems. Spread Spectrum Time Domain Reflectometry (SSTDR) can achieve real-time online location of impedance changes in cables. SSTDR utilizes correlation detection principles and features low test signal levels and high processing gain, making it suitable for online detection of cable faults.

[0004] However, SSTDR continuously transmits detection waveforms, and the received waveforms, in addition to cable fault echoes, also contain interference signals leaking from the transmitting circuit to the receiving circuit, as well as echoes caused by other impedance mismatches. Despite the design of a transmit / receive isolation circuit, some energy still leaks into the receiving circuit, affecting the detection of weak fault echoes and limiting detection performance. Summary of the Invention

[0005] This invention provides a sparse optimization-based online cable fault location and anti-interference detection method and device to solve the technical problem in the prior art where emission leakage and near-end impedance mismatch interference components in the SSTDR detection signal drown out weak fault echoes, making it difficult to detect weak fault echoes and limiting the cable fault detection distance.

[0006] In a first aspect, embodiments of the present invention provide an anti-interference detection method for online cable fault location based on sparse optimization, comprising: An SSTDR probe signal is generated and injected into the cable under test. The received echo signal returned by the cable under test is collected. A cyclic time delay shift operation is performed on the SSTDR probe signal to construct a time delay dictionary matrix. Each column of the time delay dictionary matrix corresponds to a discrete echo time delay signal template. Using the received echo signal as the observation vector, a sparse observation model is established by combining the time delay dictionary matrix. An optimization objective function with sparsity constraints is constructed. The dictionary atom with the greatest correlation to the current residual is selected and added to the support set. The optimization objective function is solved by combining the least squares method to obtain the sparse coefficient vector. Based on the non-zero element indices and values ​​of the sparse coefficient vector, the echo delay and signal amplitude corresponding to each reflected component in the received echo signal are determined. Based on the preset near-end delay threshold and detection threshold, the near-end interference component is identified from each reflected component. The time-domain interference signal is reconstructed using the time delay dictionary matrix and the sparse coefficients corresponding to the near-end interference components. The time-domain interference signal is then removed from the received echo signal to obtain the residual signal. Cross-correlation is performed on the residual signal and the SSTDR detection signal to determine the cable fault information.

[0007] Secondly, embodiments of the present invention also provide an online cable fault location and anti-interference detection device based on sparse optimization, comprising: The matrix construction module is used to generate SSTDR probe signals, inject the SSTDR probe signals into the cable under test, collect the received echo signals returned by the cable under test, perform cyclic delay shift operation on the SSTDR probe signals, and construct a delay dictionary matrix. Each column of the delay dictionary matrix corresponds to a signal template of a discrete echo delay. The function construction module is used to establish a sparse observation model by taking the received echo signal as the observation vector and combining it with the time delay dictionary matrix, construct an optimization objective function with sparsity constraints, select the dictionary atom with the greatest correlation to the current residual and add it to the support set, and solve the optimization objective function by combining the least squares method to obtain the sparse coefficient vector. The interference component identification module is used to determine the echo delay and signal amplitude corresponding to each reflection component in the received echo signal based on the index and value of the non-zero element of the sparse coefficient vector, and to identify the near-end interference component from each reflection component according to the preset near-end delay threshold and detection threshold. The fault information determination module is used to reconstruct the time-domain interference signal using the time delay dictionary matrix and the sparse coefficients corresponding to the near-end interference components, remove the time-domain interference signal from the received echo signal to obtain the residual signal, and perform cross-correlation operation on the residual signal and the SSTDR detection signal to determine the cable fault information.

[0008] Thirdly, embodiments of the present invention also provide a server, comprising: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the sparse optimization-based online cable fault location and anti-interference detection method provided in the above embodiments.

[0009] Fourthly, embodiments of the present invention also provide a medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform the sparse-optimized online cable fault location and anti-interference detection method provided in the above embodiments.

[0010] The cable fault online location and anti-interference detection method and device based on sparse optimization provided in this invention generates an SSTDR detection signal, injects it into the cable under test, collects the received echo signal returned by the cable under test, performs a cyclic delay shift operation on the SSTDR detection signal to construct a delay dictionary matrix, where each column of the delay dictionary matrix corresponds to a signal template of a discrete echo delay; using the received echo signal as the observation vector, a sparse observation model is established in conjunction with the delay dictionary matrix, and an optimization objective function with sparsity constraints is constructed, selecting the one most correlated with the current residual. Dictionary atoms are added to the support set, and the objective function is solved using the least squares method to obtain a sparse coefficient vector. Based on the indices and values ​​of the non-zero elements in the sparse coefficient vector, the echo delay and signal amplitude corresponding to each reflection component in the received echo signal are determined. According to the preset near-end delay threshold and detection threshold, near-end interference components are identified from each reflection component. The time-domain interference signal is reconstructed using the time delay dictionary matrix and the sparse coefficients corresponding to the near-end interference components. The time-domain interference signal is then removed from the received echo signal to obtain the residual signal. Cross-correlation is performed on the residual signal and the SSTDR detection signal to determine the cable fault information. By constructing a cyclic shift time delay dictionary and establishing a sparse optimization model to decompose each echo component, near-end emission leakage and stray reflection interference are reconstructed and removed, thereby improving the anti-interference performance of SSTDR cable online detection and enhancing the ability to locate and detect weak faults at long distances. Attached Figure Description

[0011] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0012] Figure 1 This is a flowchart of the anti-interference detection method for online cable fault location based on sparse optimization provided in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram illustrating the principle and interference source analysis of the SSTDR cable fault online location method based on sparse optimization and anti-interference detection provided in Embodiment 1 of the present invention. Figure 3This is a schematic diagram showing typical fault detection results and the impact of transmission interference on weak echo detection of the online cable fault location and anti-interference detection method based on sparse optimization provided in Embodiment 1 of the present invention. Figure 4 This is a simulation diagram illustrating the detection effect of the anti-interference echo detection algorithm of the online cable fault location and anti-interference detection method based on sparse optimization provided in Embodiment 1 of the present invention on near-range faults. Figure 5 This is a simulation diagram illustrating the detection effect of the anti-interference echo detection algorithm of the online cable fault location and anti-interference detection method based on sparse optimization provided in Embodiment 1 of the present invention on weak echoes of long-distance faults. Figure 6 This is a Monte Carlo simulation diagram of the anti-interference echo detection algorithm of the online cable fault location and anti-interference detection method based on sparse optimization provided in Embodiment 1 of the present invention; Figure 7 This is a schematic diagram of the actual test (frequency 10MHz) of the anti-interference echo detection algorithm of the online cable fault location and anti-interference detection method based on sparse optimization provided in Embodiment 1 of the present invention. Figure 8 This is a schematic diagram of the actual test (frequency 20MHz) of the anti-interference echo detection algorithm of the online cable fault location and anti-interference detection method based on sparse optimization provided in Embodiment 1 of the present invention. Figure 9 This is a schematic diagram of the structure of the cable fault online location and anti-interference detection device based on sparse optimization provided in Embodiment 2 of the present invention; Figure 10 This is a structural diagram of the server provided in Embodiment 3 of the present invention. Detailed Implementation

[0013] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present invention, and not all of the structures.

[0014] Example 1 Figure 1 This is a flowchart of the anti-interference detection method for online cable fault location based on sparse optimization provided in Embodiment 1 of the present invention. This embodiment can be applied to online fault detection and location scenarios of wiring harness networks in transportation vehicles such as aircraft, ships, and high-speed trains, and specifically includes the following steps: Step 110: Generate an SSTDR probe signal, inject the SSTDR probe signal into the cable under test, acquire the received echo signal returned by the cable under test, perform a cyclic delay shift operation on the SSTDR probe signal, and construct a delay dictionary matrix. Each column of the delay dictionary matrix corresponds to a signal template of a discrete echo delay.

[0015] Spread Spectrum Time Domain Reflectometry (SSTDR) is an online detection and location technology for impedance abrupt changes in cables. Its core principle is to use a pseudo-random sequence to spread-spectrum modulate a sinusoidal carrier wave to generate a probe signal. Then, through cross-correlation operations, echo features are extracted from the received waveform to identify, locate, and determine the type of fault. Compared to traditional time-domain reflection schemes, SSTDR features lower test signal levels, higher correlation processing gain, and stronger immunity to background noise. It can complete detection without interrupting normal power supply and signal transmission in the cable, making it widely applicable to online monitoring scenarios for complex equipment such as aircraft, ships, and high-speed trains.

[0016] Generating SSTDR probe signals, injecting them into the cable under test, and acquiring the received echoes are the physical prerequisites and data foundations for the entire anti-interference detection method based on sparse optimization. Only by outputting standard and controllable probe signals and fully acquiring the mixed received waveforms including transmission leakage, near-end spurious reflections, and fault echoes can subsequent algorithmic steps such as time delay dictionary construction, sparse optimization solution, interference component identification and removal, and residual correlation detection have processing objects and judgment criteria. This is the core link for transmitting detection information from the hardware physical layer to the signal processing layer.

[0017] For example, in a specific implementation, such as Figure 2 As shown, the entire testing process unfolds along the principles of online fault location and interference source analysis for SSTDR cables. The corresponding SSTDR cable fault detection system mainly consists of a processing unit, a digital-to-analog converter (DAC), an analog-to-digital converter (ADC), an isolator, and a coupler. The processing unit integrates a waveform generation module and an echo detection module, respectively undertaking the core functions of generating the detection signal and analyzing the echo characteristics. The isolator typically uses an RF circulator from the communication or radar fields, or a transformer bank with equivalent functionality. Its core functions are twofold: first, to ensure that the detection waveform generated by the DAC can be effectively injected into the cable under test, guaranteeing signal injection efficiency; second, to block the detection signal from directly entering the ADC receiver, suppressing the interference of transmission leakage on fault echo detection from a hardware perspective. The coupler is responsible for enabling bidirectional transmission of the detection waveform between the cable under test and the testing equipment, while isolating the cable's original operating signals outside the detection loop to prevent the detection signal from interfering with the cable's normal operations. Common implementation forms include capacitive coupling and inductive coupling.

[0018] For example, the SSTDR device is installed near the monitored device 1, designed to achieve uninterrupted online fault detection for the BC segment of the target cable. During system operation, the waveform generation module in the processing unit first generates an SSTDR probe signal, which is obtained based on pseudo-random m-sequence binary phase-shift keying (BPSK) modulation, with an initial growth rate of... Code element value A pseudo-random m-sequence, where each symbol corresponds to a chip width Therefore, the frequency is The duration is obtained by modulating a sinusoidal carrier wave. The single-cycle SSTDR detection signal is mathematically expressed as follows: ,in , The unit rectangular chip waveform satisfies To adapt to the continuous online detection mode of cables, the single-cycle detection signal is cyclically extended according to a period T to form a continuously transmitted SSTDR detection signal. The detection performance of this signal is determined by the autocorrelation characteristics of the m-sequence, with a length of... For the m-sequence, the normalized periodic autocorrelation function takes the value 1 at zero time delay, and the sidelobe amplitudes at other time delays are... The longer the sequence length, the higher the ratio of the main lobe to the side lobe levels, which is more conducive to the detection of weak echoes. Optionally, a 63-code-length m-sequence can be used with a 20MHz sinusoidal carrier to generate a single-cycle probe signal. In this case, the autocorrelation main peak is sharp and the side lobe level is low, which can provide sufficient processing gain for subsequent correlation detection.

[0019] The generated digital probe sequence is converted into an analog excitation waveform by a digital-to-analog converter (DAC) and then injected into the cable under test sequentially through isolators and couplers (such as capacitive couplers). During the transmission of the probe signal along the cable, reflected echoes are generated at all impedance mismatch locations. These reflected signals retrace along the original transmission path, pass through couplers and isolators again, and enter the receiving link. Finally, the DAC performs equally spaced digital sampling to obtain the complete received echo signal, which is then sent to the echo detection module of the processing unit for further processing. The received echo is a multi-component superposition, containing both valid fault echoes and various non-fault interference components. Its time-domain expression is as follows: ,in, It is a direct-transmit interference signal formed by the detection signal leaking into the receiving channel through the isolator. This is the amplitude of the component. It is a stray reflected signal generated by impedance mismatch at connection point A on the equipment side. To determine the total time delay of the detection signal arriving at node A and being reflected back to the receiver, For the corresponding amplitude, This is a stray reflected signal caused by impedance mismatch at connection point B between the device and the cable. To account for the round-trip time of the corresponding reflection path, For the corresponding amplitude, It is a valid fault echo generated at cable fault location D. For the round-trip time to the fault point, This corresponds to the amplitude.

[0020] In traditional detection methods, the echo detection module directly extracts echo features through a cross-correlation algorithm. After cross-correlation between the received signal and the local reference signal, the output results form correlation peaks at the time delay positions corresponding to each reflection component. The peaks corresponding to near-end transmission leakage and stray reflections at connection points are prone to causing false faults. Only when the peak corresponding to the far-end fault echo exceeds a preset detection threshold is the cable fault identified. Besides the main peak, the correlation sidelobes of each signal component collectively raise the detection floor. The sidelobe level increases with the amplitude of the near-end interference component, causing the ratio of the fault peak to the maximum sidelobe to continuously decrease, directly weakening the system's ability to detect weak faults at long distances. Although hardware devices such as isolators can reduce interference intensity at the physical level, they cannot completely eliminate the influence of residual leakage, continuously limiting the detection performance of weak faults. This is the core motivation for introducing sparse optimization in this embodiment to carry out algorithm-level anti-interference processing.

[0021] Constructing a time delay dictionary matrix based on cyclic time delay shifting is the core foundation for transforming the time-domain echo decomposition problem into a sparse optimization problem, and it is also an essential representational basis for building sparse observation models. Only by constructing a set of signal templates corresponding to the time delays of each discrete echo can the received echo containing multiple superimposed components be represented as a linear combination of a few time delay templates, and then the interference components and fault echoes with different time delays can be separated through optimization.

[0022] For example, to adapt to the implementation process of digital signal processing, the continuous SSTDR probe signal first needs to be discretized and sampled to obtain a length of... Discrete reference signal ,in , This represents the sampling length of a single-cycle SSTDR probe signal. Based on the periodicity of the m-sequence itself and the continuous cyclic transmission characteristic of the SSTDR system, a cyclic delay shift operation is performed using this discrete reference signal as a reference. Each shift is performed by one sampling interval, resulting in a total of [number missing]. Several cyclic shift sequences with different time delays are used. Considering that if the fault echo delay exceeds the sampling length of a single-cycle detection signal, the reflected echo will alias with the transmitted direct interference in the time domain, making subsequent separation difficult, each cyclic shift sequence needs to be cyclically extended to... The lengths of these elements are used as one column of atoms in the delay dictionary matrix, ultimately constructing a matrix with dimensions of [value missing]. The time delay dictionary matrix, i.e., the length of the corresponding observation vector. And the maximum delay grid points of the dictionary The specific form of the delay dictionary matrix is ​​as follows: The first column is the original reference signal with zero shift, the second column is the sequence shifted right by one sampling interval, and so on. The time delay corresponding to the atoms in the i-th column is... Echo signal templates for each sampling interval, wherein This fully covers all possible echo delay locations. The delay dictionary constructed in this way, with each column of atoms corresponding to a specific discrete echo delay, not only matches the periodic autocorrelation characteristics of the m-sequence, but also provides a complete signal basis for the sparse representation and optimization solution of the subsequent received echoes. This allows the mixed received signal to be decomposed into a linear superposition of a few dictionary atoms, providing mathematical support for the accurate separation of interference and fault components.

[0023] This system completes the entire basic processing workflow for SSTDR detection signal generation, cable injection, echo acquisition, and time delay dictionary construction. On one hand, it generates a detection signal with sharp autocorrelation characteristics through pseudo-random spread spectrum modulation. Leveraging the advantages of low level and high processing gain, it is suitable for online cable detection scenarios. Signal injection and echo acquisition are completed via the transceiver link to obtain raw mixed received data containing transmission leakage, near-end spurious reflections, and fault echoes. On the other hand, based on the periodicity of the m-sequence, a time delay dictionary matrix covering all discrete time delays is constructed through cyclic time delay shifting. This transforms the multi-component superposition relationship of time-domain echoes into a linear representation problem in dictionary space. Overall, it not only builds the physical data link for online detection but also completes the preliminary mathematical preparation for sparse anti-interference algorithms, which is the prerequisite for achieving interference separation and weak fault detection.

[0024] Step 120: Using the received echo signal as the observation vector, establish a sparse observation model in conjunction with the time delay dictionary matrix, construct an optimization objective function with sparsity constraints, select the dictionary atom with the greatest correlation to the current residual and add it to the support set, and solve the optimization objective function using the least squares method to obtain the sparse coefficient vector.

[0025] Establishing a sparse observation model and constructing an optimization objective function with sparsity constraints are the core steps in transforming the interference separation problem of time-domain echoes into a solvable mathematical optimization problem. Only by relying on the time delay sparsity characteristics of the echoes to complete the modeling and clarifying the optimization objective and sparsity constraints for signal fitting can a clear mathematical framework be provided for subsequent iterative solutions to the sparse coefficients, thereby achieving accurate separation of interference components and fault echoes.

[0026] For example, in the SSTDR cable fault detection scenario, the actual number of fault points and near-end interference nodes is far less than the sampling scale corresponding to the detectable delay range of the cable. That is, fault echoes and interference components occupy only a few positions in the delay domain. This physical characteristic determines that the received echoes possess a naturally sparse representation characteristic in the signal space constructed by the delay dictionary. Based on this characteristic, using the acquired received echo signal as the observation vector and the pre-constructed delay dictionary matrix as the sparse representation basis, the observation vector can be represented as a linear combination of the delay dictionary matrix, the vector of sparse coefficients to be determined, and the form of observation noise, thus establishing a sparse observation model. ,in The observation vector corresponds to the received echo signal. The time delay dictionary matrix serves as the representation basis for signal decomposition. Let be the sparse coefficient vector to be solved. The positions of its non-zero elements correspond to the time delay indices of each reflection component, and the non-zero values ​​correspond to the signal amplitudes of each reflection component. To observe noise, random interference components such as circuit thermal noise are included.

[0027] Since the number of dictionary columns is less than the length of the observation vector, this equation is underdetermined and has no feasible solutions. Therefore, sparsity constraints need to be introduced to limit the physical validity of the solution. Based on the above sparse observation model, the optimization objective is to minimize the sum of squared errors between the observation vector and the product of the time delay dictionary matrix and the sparse coefficient vector. The constraint is that the number of non-zero elements in the sparse coefficient vector does not exceed a preset sparsity. The expression is as follows: Among them, the target item This is used to measure the fitting accuracy of the dictionary linear combination to the actual received signal. A smaller value indicates a lower fitting error. (Constraints...) This represents the number of non-zero elements in the sparse coefficient vector. The preset maximum sparsity corresponds to the maximum number of possible reflection components in the scene, ensuring that the solution conforms to the physical law of sparse reflection points. The core objective of the optimization problem is to achieve accurate fitting of the received signal by using as few dictionary atoms as possible for linear combination. Finally, the time delay and amplitude information corresponding to each reflection component can be extracted from the sparse coefficient vector obtained from the solution. The process of solving the sparse coefficients itself completes the separation of interference and fault components in the sparse domain. Based on the correspondence between coefficients and time delay and amplitude, combined with the detection threshold, the interference term and the target echo can be directly determined. At the same time, in order to ensure that the detector output format is consistent with conventional correlation detectors, the time-domain interference signal can also be reconstructed based on the identified interference coefficients and removed from the received signal. Then, correlation processing is performed on the residual signal, taking into account both algorithm performance and compatibility with the detection architecture.

[0028] Due to the belt Norm-constrained sparse optimization problems cannot directly obtain the global optimal solution. Therefore, a greedy iterative approach of orthogonal matching pursuit is adopted to solve the optimization objective. The core idea is to gradually select the dictionary atom with the greatest correlation to the current residual during the iteration process, and perform least squares projection on the subspace spanned by the selected atom to gradually approximate the optimal sparse representation of the observed signal.

[0029] For example, the specific solution process can be divided into four stages: preprocessing, initialization, iterative solution, and termination output. First, dictionary preprocessing is completed, and each column of the time delay dictionary matrix is ​​processed. Normalization is performed to obtain a normalized dictionary, thereby eliminating the bias caused by the energy differences of different atoms in the correlation selection, ensuring that atom selection is determined solely by the degree of waveform matching. Subsequently, parameter initialization is completed, setting the initial residual to the complete received echo observation signal. The support set used to store the selected atom indices is initialized to an empty set. In the iterative process, each iteration first calculates the current residual. The inner product of the atom and all unselected atoms in the normalized dictionary is used. The larger the absolute value of the inner product, the higher the matching degree between the time delay template corresponding to that atom and the currently uninterpreted residual signal, and the stronger its interpretability of the residual. The atom with the largest absolute value of the inner product is selected, and its corresponding index is determined. Add support set This completes the current round of atomic screening, that is... After atomic filtering is completed, the sub-dictionary corresponding to the support set is extracted. The optimal sparsity coefficients for the current support set are solved using the least squares method, and the solution formula is as follows: This yields the optimal amplitude coefficients for all currently selected atoms. The current residual is then updated using the following formula: This means that the fitted signal of the selected component is subtracted from the original observed signal, and the remaining unexplained part is used as the processing object for the next iteration, ensuring that subsequent iterations always select matching templates for the signal components that have not yet been fitted.

[0030] Repeat the iterative loop of atomic screening, coefficient solving, and residual updating as described above. Continue until the preset maximum number of iterations is reached, or the L2 norm of the updated residual is reached. The iteration terminates when the residual is less than the preset residual threshold. Finally, the least squares coefficients corresponding to the support set are filled into a zero vector with the same length as the number of columns in the dictionary according to the index position, and the complete final sparse coefficient vector is obtained. The non-zero positions of this vector correspond to the time delay index of each reflection component, and the non-zero values ​​correspond to the signal amplitude of each reflection component.

[0031] This section is the core solution module of the entire sparse anti-interference detection algorithm. Its function is to transform the component separation problem of time-domain mixed echoes into a numerically solvable sparse optimization problem, and to achieve accurate analysis of each reflection component through efficient iterative solution. On the one hand, a sparse observation model is established based on the sparse characteristics of echo time delay, representing the superimposed received signal containing interference and faults as a linear combination of time delay dictionary and sparse coefficients, thus transforming the physical problem into a mathematical optimization problem. On the other hand, by introducing sparsity constraints to construct optimization objectives, while ensuring signal fitting accuracy, the solution results are ensured to conform to the physical law of the sparse number of cable reflection points, avoiding ambiguity of multiple solutions in underdetermined equations. Based on this, a solution mechanism of correlation-based atom selection and least squares coefficient update is adopted, which can quickly approximate the optimal sparse solution with low computational complexity, and ensure the accuracy of amplitude estimation of each component through least squares projection. The final output sparse coefficient vector can be directly mapped to the echo time delay and signal amplitude corresponding to each reflection component, realizing effective separation of near-end interference and far-end fault echoes in the sparse domain.

[0032] Step 130: Based on the non-zero element indices and values ​​of the sparse coefficient vector, determine the echo delay and signal amplitude corresponding to each reflection component in the received echo signal, and identify the near-end interference component from each reflection component according to the preset near-end delay threshold and detection threshold.

[0033] The sparse coefficient vector obtained by sparse optimization solution contains both non-fault interference components such as near-end emission leakage and stray reflections at connection points, and actual cable fault reflection components. If the two are not accurately distinguished, near-end interference will continue to raise the detection floor and decision threshold, causing weak fault echoes at long distances to be constantly submerged in interference sidelobes, thus failing to achieve the expected effect of anti-interference detection.

[0034] For example, according to transmission line theory, the signal transmitted in a cable can be represented as the superposition of the incident wave and the reflected wave, expressed as: ,in Let be the distance between a point on the cable and the signal source. Let be the propagation constant of the transmission line. and These represent the complex voltage amplitudes of the incident and reflected waves at the signal source, respectively. The signal attenuates exponentially along the cable distance; the farther the fault point is from the detection end, the more significant the round-trip attenuation of the fault echo and the lower the echo amplitude. However, the near-end transmitted leakage signal is unaffected by cable transmission attenuation and maintains a relatively stable intensity, which is a key factor limiting the long-distance weak fault detection capability of the SSTDR system. In traditional related detection results, such as... Figure 3As shown in the figure, the horizontal axis represents the cable distance, the vertical axis represents the normalized correlation amplitude, the blue curve represents the output result of cross-correlation detection, and the red dashed line represents the detection threshold as a function of distance. Three types of characteristics are clearly identifiable in the figure: an interference peak caused by leakage signals from the transmitting end appears near 0m, with a normalized amplitude of approximately 0.15; the echo peak at near-distance fault points is strong enough to exceed the detection threshold, but the echo peak at long-distance faults (around 260m) weakens significantly and is easily submerged in the sidelobes generated by near-end interference. Furthermore, to avoid false detections caused by near-end interference, the detection threshold needs to be maintained at a high level near the fault and remain relatively high over long distances, further compressing the detection margin between weak fault echoes and the threshold, resulting in missed detections of long-distance faults. Based on the sparse solution, physical quantity mapping and component determination are performed. First, the sparse coefficient vector is analyzed. Each non-zero element of the sparse coefficient vector corresponds to an independent reflection component in the received echo. The index of the non-zero element corresponds to the column number of the time delay dictionary matrix. Multiplying the index value by the duration of a single sampling interval yields the echo delay corresponding to that reflection component. The magnitude of the non-zero element directly corresponds to the echo amplitude of that reflection component, thus completing the conversion from the mathematical sparse solution to the physical reflection parameters.

[0035] Both preset parameters used in the determination were obtained through physical calibration, including the near-end delay threshold. The physical area near the detection equipment corresponds to the range where reflections originate from direct leakage at the transmitting end and impedance mismatch reflections at the connection points between the equipment and cables; these are known non-fault-related interference sources. Detection threshold function. Using the correlation detection peak value of the fault echo at 0m as the normalized reference benchmark, the signal attenuation law in the cable is obtained by calibrating the measurement data of faults at different locations, and the voltage distribution along the cable is derived. Furthermore, considering the sidelobe peak values ​​of interference components such as leakage signals in the relevant detection outputs... The result of the fitting is expressed as follows: ,in This approach balances cable transmission attenuation characteristics with interference baseline levels. It iterates through all non-zero reflection components in the sparse coefficient vector, performing a dual-condition check on each one. For any reflection component, if its echo delay satisfies... And the corresponding amplitude satisfies If the signal is positive, the component is identified as a near-end interference component, corresponding to non-fault components such as direct transmission leakage and stray reflections from near-end junctions. Otherwise, it is identified as a valid fault echo component. This judgment logic can accurately filter and label all near-end interference components from the mixed sparse solution, completing the initial separation of interference and fault in the sparse domain. The identified near-end interference components can be directly used for subsequent time-domain signal reconstruction and elimination, eliminating their sidelobe lifting effect on the relevant detection output, allowing the detection threshold to be lowered synchronously with interference suppression. Long-distance weak fault echoes that were originally submerged in interference sidelobes can be effectively detected, ultimately improving the maximum detection distance and the reliability of weak fault identification in the SSTDR system.

[0036] Step 140: Reconstruct the time-domain interference signal using the time delay dictionary matrix and the sparse coefficients corresponding to the near-end interference components. Remove the time-domain interference signal from the received echo signal to obtain the residual signal. Perform cross-correlation operation on the residual signal and the SSTDR detection signal to determine the cable fault information.

[0037] After the identification and labeling of near-end interference components are completed, only the mathematical component distinction is completed in the sparse coefficient vector. The original received echo signal still contains interference components such as transmission leakage and near-end connection point stray reflection. If cross-correlation detection is performed directly on the original received echo, the high sidelobes corresponding to the near-end interference will continue to raise the detection floor, causing the weak fault peaks at long distances to be submerged in the sidelobes, and the anti-interference gain brought by sparse decomposition cannot be utilized.

[0038] For example, specifically, firstly, extract all non-zero coefficients that are determined to be near-end interference components from the complete sparse coefficient vector, and then construct an interference sparse coefficient vector according to their original index positions. In the vector, the non-interference positions are uniformly set to zero. This interference sparse coefficient vector is then linearly combined with the original time delay dictionary matrix using the formula... Reconstruction yields time-domain interference signal Since each column of the time delay dictionary corresponds to a signal template with a defined time delay, the time-domain interference signal, which is identical in amplitude to the measured interference waveform, can be reconstructed after sparse coefficient weighting and superposition, ensuring the accuracy of subsequent removal. The reconstructed time-domain interference signal is then subtracted point-by-point from the original received echo signal to remove it. This residual signal has been stripped of near-end emission leakage and stray reflection components, mainly retaining the reflected echoes and background noise corresponding to each fault point in the cable, significantly reducing the detection substrate. Finally, a sliding cross-correlation operation is performed between the residual signal and the local SSTDR detection signal. Based on the cross-correlation output, the cable fault information is determined. The time delay value corresponding to the peak value of the cross-correlation output is taken and multiplied by the signal propagation speed in the cable to calculate the physical distance between the fault point and the detection end. Simultaneously, the fault type is determined based on the polarity and magnitude of the peak value. Different impedance change types correspond to different reflection phases. The peak polarity can distinguish fault categories such as short circuits and open circuits, while the peak amplitude can characterize the severity of the impedance change. After time-domain interference removal, the non-fault peak value and sidelobe level in the cross-correlation output are significantly reduced, and the detection threshold can be adjusted accordingly. The long-distance weak fault echoes that were originally submerged in the interference sidelobes can be clearly highlighted, effectively improving the maximum detection distance and weak fault identification reliability of the SSTDR system.

[0039] To verify the effectiveness and applicability of the method in cable fault detection, verification was conducted through both simulation analysis and real-world testing. In the simulation, the SSTDR system parameters were set as follows: PN code length 63, carrier frequency 20MHz, sampling rate 200MHz, and signal propagation speed in the cable. The received signal model includes a zero-delay leakage signal with an amplitude of 0.18, a fault echo varying with fault delay, and noise components. The received signal-to-noise ratio is 15 dB. Two typical fault echo propagation delays, 0.5 μs and 1.2 μs, were selected to correspond to near-range strong fault and far-range weak fault scenarios, respectively. The detection performance of the traditional cross-correlation method and the method in this embodiment were compared. The results are as follows: Figure 4 and Figure 5 As shown. Figure 4 As shown in (a), when the fault echo delay is 0.5 μs, the fault point is relatively close and the echo attenuation is small. In traditional cross-correlation detection, the fault peak value can exceed the detection threshold, enabling correct identification of the fault location. However, interference from near-end leakage signals raises the detection threshold. After processing by the method of this embodiment, as shown in (a), Figure 4 The normalized sparse characterization in (b) can accurately distinguish between leakage and fault components. After removing near-end interference, correlation processing is performed to obtain... Figure 4In the residual correlation results of (c), the near-end residual emission leakage and its sidelobes are further reduced, the correlation output background is more stable, the contrast between the fault peak and the background is improved, and the detection threshold can be lowered accordingly. These results demonstrate that, under conditions of small time delay, the proposed method does not destroy the original fault characteristics while improving the correlation output quality. Figure 5 As shown in (a), when the fault echo delay increases to 1.2 μs, the fault echo amplitude has significantly weakened. At this point, the fault response in the traditional cross-correlation method is significantly affected by the near-end residual emission leakage sidelobes and noise fluctuations. The fault peak is submerged in the sidelobes and cannot cross the threshold to be detected by the system. In contrast, after processing by the method of this embodiment, Figure 5 The normalized sparse representation in (b) can separate the near-end leakage and far-end fault components in the sparse domain, and after removing near-end interference, we obtain... Figure 5 As shown in the residual correlation results in (c), near-end dominant interference is effectively suppressed, and the levels of non-fault peaks and sidelobes in the correlation output are significantly reduced, making previously difficult-to-identify fault peaks re-emerge. By lowering the detection threshold, long-distance weak fault localization can be achieved. These results demonstrate that the proposed method's advantage in improving the detectability of weak fault echoes becomes more pronounced as propagation delay increases.

[0040] To further verify the robustness of the method in this embodiment under different noise environments, multiple Monte Carlo simulations were conducted. Random noise was independently added to each signal-to-noise ratio point, and the experiment was repeated 1000 times. The root mean square value of the normalized correlation output sidelobe of the traditional cross-correlation method and the method of this embodiment were calculated respectively. The statistical performance was characterized by the mean and standard deviation. The results are as follows: Figure 6 As shown. Figure 6 Figure (a) presents a comparison of the normalized sidelobe RMS of the two methods under different signal-to-noise ratios. It can be seen that as the signal-to-noise ratio increases, the sidelobe RMS of both methods gradually decreases, indicating that the influence of random noise on the correlated output gradually weakens. However, throughout the entire signal-to-noise ratio range, the method in this embodiment consistently exhibits a lower sidelobe RMS, indicating that this method can more effectively suppress residual emission leakage and its caused sidelobe interference. Furthermore, the overall error bar is small, demonstrating that the proposed method has good stability in repeated random experiments. Figure 6Figure (b) further shows the reduction ratio of the sidelobe RMS of the method in this embodiment compared with the traditional cross-correlation method. The results show that the reduction ratio increases continuously with the increase of the signal-to-noise ratio. This indicates that under low signal-to-noise ratio conditions, the system performance is constrained by both random noise and residual leakage. Therefore, the relative improvement of the method in this embodiment is limited. However, under higher signal-to-noise ratio conditions, the noise effect gradually weakens, and residual emission leakage gradually becomes the main factor limiting the system performance. At this time, the interference cancellation advantage of the method in this embodiment is more prominent. Under high signal-to-noise ratio conditions, the reduction ratio of sidelobe RMS can reach more than 80%, which further verifies the effectiveness of the proposed method in suppressing residual emission interference.

[0041] Based on this, an SSTDR experimental platform was built, with its hardware architecture and... Figure 2 The system principle model shown is consistent. The experimental injected signal amplitude is 1Vpp, the ADC sampling rate at the receiving end is 200MHz, and the test cable is a QLA11142-22-U type cable with a cross-sectional area of ​​0.35mm². 2 The maximum DC resistance is 46.9 Ω / km. All relevant results are normalized based on the autocorrelation value of the injection signal to ensure a consistent comparison benchmark between simulation and experimental results. Figure 7 The figure shows the experimental results at a carrier frequency of 10MHz and a fault echo delay of 1.03μs. At this carrier frequency, the signal attenuation along the cable is relatively small. Figure 7 As shown in Figure 7(a), the traditional cross-correlation method can observe a clear fault correlation peak, achieving basic fault identification. Figure 7(b) shows the normalized sparse representation result after processing by the method of this embodiment. Three types of signal components can be distinguished: the transmit leakage signal near time 0, which originates from the direct crosstalk of the probe signal caused by the limited transmit-receive isolation of the isolator; the immediate following near-end reflection interference, which is generated by reflection and foldback at near-end impedance mismatch nodes such as the connection between the equipment and the cable, with a time delay slightly greater than the leakage signal; and the target fault echo at 1.03μs. The results show that the method of this embodiment can effectively distinguish between transmit leakage, near-end reflection interference, and fault echo. After removing the above two types of near-end interference components, correlation processing is performed on the residual signal, and the results are as follows. Figure 7 As shown in (c), with Figure 7 Compared to (a), the near-end residual leakage and its sidelobes are further weakened after interference suppression, and the fault peak contrast is significantly improved, verifying that the proposed method can effectively improve the output quality of related detections. Figure 8 The figure shows the experimental results when the carrier frequency is increased to 20MHz and the fault echo propagation delay is 1.03μs. Increasing the carrier frequency exacerbates signal attenuation along the cable, thus further reducing the echo correlation peak value for the same fault distance. Figure 8As shown in Figure 8(a), when using the traditional cross-correlation method, the weak fault echo is masked by the strong interference peak and sidelobes of the near-end transmission leakage, making it difficult to effectively distinguish and thus unable to achieve reliable fault detection. Figure 8(b) shows the normalized sparse representation result after processing by the method of this embodiment. The algorithm can effectively separate three types of components from the received signal: transmission leakage, near-end reflection interference, and target fault echo. Based on this, two types of near-end interference components are removed, and correlation operations are performed on the remaining residual signal. The result after interference suppression is shown in Figure 8(a). Figure 8 As shown in (c), the dominant emission leakage interference and its sidelobes are significantly suppressed, and the fault echo peak is clearly highlighted. The experimental results are consistent with the simulation analysis conclusions above, indicating that the proposed method can not only improve the detectability of weak faults at long distances in simulation scenarios, but also has good engineering applicability in actual cable fault detection platforms. The combined simulation and experimental results show that residual emission leakage and its sidelobes are key factors limiting the long-distance weak fault detection capability of the SSTDR system. This embodiment utilizes the separability of leakage and fault echoes in the time delay support domain to achieve accurate reconstruction and effective suppression of strong near-end interference, reduce the non-fault peak and sidelobe levels of the relevant output, and significantly improve the detectability of weak fault echoes, providing an effective and feasible technical approach for improving the long-distance weak fault detection capability of the SSTDR system.

[0042] This embodiment generates an SSTDR probe signal, injects it into the cable under test, and collects the received echo signal returned by the cable. A cyclic delay shift operation is performed on the SSTDR probe signal to construct a delay dictionary matrix. Each column of the delay dictionary matrix corresponds to a signal template for a discrete echo delay. Using the received echo signal as the observation vector, a sparse observation model is established in conjunction with the delay dictionary matrix. An optimization objective function with sparsity constraints is constructed. The dictionary atom with the highest correlation to the current residual is added to the support set. The optimization objective function is solved using the least squares method to obtain a sparse coefficient vector. Based on the non-zero element indices and values ​​of the sparse coefficient vector, the echo delay and signal amplitude corresponding to each reflection component in the received echo signal are determined. According to the preset near-end delay threshold and detection threshold, near-end interference components are identified from each reflection component. The time-domain interference signal is reconstructed using the delay dictionary matrix and the sparse coefficients corresponding to the near-end interference components. The time-domain interference signal is then removed from the received echo signal to obtain the residual signal. Cross-correlation is performed on the residual signal and the SSTDR probe signal to determine the cable fault information. By constructing a cyclic shift delay dictionary and establishing a sparse optimization model to decompose each echo component, near-end emission leakage and stray reflection interference are reconstructed and eliminated, thereby improving the anti-interference performance of SSTDR cable online detection and enhancing the ability to locate and detect weak faults at long distances.

[0043] Example 2 Figure 9 This is a schematic diagram of the anti-interference detection device for online cable fault location based on sparse optimization provided in Embodiment 2 of the present invention, as shown below. Figure 9 As shown, the device includes: The matrix construction module 210 is used to generate an SSTDR probe signal, inject the SSTDR probe signal into the cable under test, acquire the received echo signal returned by the cable under test, perform a cyclic delay shift operation on the SSTDR probe signal, and construct a delay dictionary matrix. Each column of the delay dictionary matrix corresponds to a discrete echo delay signal template. The function construction module 220 is used to establish a sparse observation model by taking the received echo signal as the observation vector and combining it with the time delay dictionary matrix, construct an optimization objective function with sparsity constraints, select the dictionary atom with the greatest correlation to the current residual and add it to the support set, and solve the optimization objective function by combining the least squares method to obtain the sparse coefficient vector. The interference component identification module 230 is used to determine the echo delay and signal amplitude corresponding to each reflection component in the received echo signal based on the index and value of the non-zero element of the sparse coefficient vector, and to identify the near-end interference component from each reflection component according to the preset near-end delay threshold and detection threshold. The fault information determination module 240 is used to reconstruct the time-domain interference signal using the time delay dictionary matrix and the sparse coefficients corresponding to the near-end interference components, remove the time-domain interference signal from the received echo signal to obtain the residual signal, and perform cross-correlation operation on the residual signal and the SSTDR detection signal to determine the cable fault information.

[0044] The sparse optimization-based online cable fault location and anti-interference detection device provided in this embodiment generates an SSTDR detection signal, injects it into the cable under test, and collects the received echo signal returned by the cable. A cyclic delay shift operation is performed on the SSTDR detection signal to construct a delay dictionary matrix, where each column of the delay dictionary matrix corresponds to a signal template of a discrete echo delay. Using the received echo signal as the observation vector, a sparse observation model is established in conjunction with the delay dictionary matrix, constructing an optimization objective function with sparsity constraints. The dictionary with the highest correlation to the current residual is selected. Atoms are added to the support set, and the objective function is solved using the least squares method to obtain a sparse coefficient vector. Based on the indices and values ​​of the non-zero elements in the sparse coefficient vector, the echo delay and signal amplitude corresponding to each reflection component in the received echo signal are determined. According to the preset near-end delay threshold and detection threshold, near-end interference components are identified from each reflection component. The time-domain interference signal is reconstructed using the time delay dictionary matrix and the sparse coefficients corresponding to the near-end interference components. The time-domain interference signal is then removed from the received echo signal to obtain the residual signal. Cross-correlation is performed on the residual signal and the SSTDR detection signal to determine the cable fault information. By constructing a cyclic shift time delay dictionary and establishing a sparse optimization model to decompose each echo component, near-end emission leakage and stray reflection interference are reconstructed and removed, thereby improving the anti-interference performance of SSTDR cable online detection and enhancing the ability to locate and detect weak faults at long distances.

[0045] Based on the above embodiments, the matrix construction module includes: The single-cycle detection signal acquisition unit is used to perform binary phase-shift keying modulation on a sinusoidal carrier using a pseudo-random m-sequence to obtain a single-cycle SSTDR detection signal. A continuous transmission detection signal acquisition unit is used to cyclically extend the single-cycle SSTDR detection signal to obtain a continuously transmitted SSTDR detection signal; The discrete reference signal acquisition unit is used to discretize the SSTDR probe signal to obtain a discrete reference signal. ; The cyclic shift sequence acquisition unit is used to perform a cyclic time-delay shift operation on the discrete reference signal based on the discrete reference signal, shifting by one sampling interval each time, and obtaining a total of A cyclic shift sequence; The delay dictionary matrix construction unit is used to cyclically extend each cyclic shift sequence to 2^2. The length, as the atoms of each column of the delay dictionary matrix, is constructed with dimension . The delay dictionary matrix.

[0046] Based on the above embodiments, the function construction module includes: The sparse observation model establishment unit is used to establish a sparse observation model based on the characteristic that fault echoes and interference components in the received echo signal occupy only a few positions in the time delay domain, using the time delay dictionary matrix as a sparse representation basis, and representing the observation vector as a linear combination of the time delay dictionary matrix and the sparse coefficient vector to be determined plus the observation noise. The objective function construction unit is used to construct an optimization objective function with sparsity constraints based on the sparse observation model, with the objective function being to minimize the sum of squared errors between the observation vector and the product of the time delay dictionary matrix and the sparse coefficient vector, and with the constraint that the number of non-zero elements in the sparse coefficient vector does not exceed a preset sparsity. The normalized dictionary acquisition unit is used to normalize each column of the time delay dictionary matrix to obtain a normalized dictionary; An initialization unit is used to initialize the residual to the received echo signal and to initialize the support set to an empty set. The support set addition unit is used to iteratively calculate the inner product of the current residual and each column of atoms in the normalized dictionary, and select the index corresponding to the atom with the largest absolute value of the inner product to add to the support set; The sparse coefficient solving unit is used to extract the sub-dictionary corresponding to the support set, solve the sparse coefficients corresponding to the sub-dictionary using the least squares method, and update the current residual. The final sparse coefficient vector acquisition unit is used to stop iterating when the number of iterations reaches the preset maximum number of iterations or when the L2 norm of the current residual is less than the preset residual threshold, and outputs the final sparse coefficient vector.

[0047] Based on the above embodiments, the interference component identification module includes: The mapping unit is used to map the echo delay corresponding to each reflection component in the received echo signal according to the index position of each non-zero element in the sparse coefficient vector, and use the value of each non-zero element as the amplitude of the corresponding reflection component. The near-end interference component determination unit is used to determine the reflection component corresponding to each non-zero element in the sparse coefficient vector as a near-end interference component when the echo delay is less than the near-end delay threshold and the amplitude is less than the detection threshold.

[0048] Based on the above embodiments, the fault information determination module includes: The interference sparse coefficient vector is composed of units used to extract the coefficients corresponding to all non-zero elements of the near-end interference components and form the interference sparse coefficient vector. The time-domain interference signal reconstruction unit is used to perform a linear combination operation between the interference sparse coefficient vector and the time delay dictionary matrix to reconstruct the time-domain interference signal. The residual signal acquisition unit is used to subtract the time-domain interference signal from the received echo signal point by point to obtain the residual signal; The fault type determination unit is used to perform cross-correlation operation on the residual signal and the SSTDR detection signal, multiply the time delay value corresponding to the peak value output by the cross-correlation operation by the propagation speed of the signal in the cable, calculate the distance between the fault point and the detection end, and determine the cable fault type according to the polarity and magnitude of the peak value.

[0049] The sparse optimization-based online cable fault location and anti-interference detection device provided in this invention can execute the sparse optimization-based online cable fault location and anti-interference detection method provided in any embodiment of this invention, and has the corresponding functional modules and beneficial effects of the method.

[0050] Example 3 Figure 10 This is a schematic diagram of the structure of a server provided in Embodiment 3 of the present invention. Figure 10 A block diagram of an exemplary server 12 suitable for implementing embodiments of the present invention is shown. Figure 10 The server 12 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0051] like Figure 10 As shown, server 12 is presented in the form of a general-purpose computing server. The components of server 12 may include, but are not limited to: one or more processors or processing units 16, system memory 28, and bus 18 connecting different system components (including system memory 28 and processing unit 16).

[0052] Bus 18 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. For example, these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0053] Server 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by server 12, including volatile and non-volatile media, removable and non-removable media.

[0054] System memory 28 may include computer system readable media in the form of volatile memory, such as RAM 30 and / or cache 32. Server 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 10 Not shown; usually referred to as a "hard drive"). Although Figure 10 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. System memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0055] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 42 typically perform the functions and / or methods described in the embodiments of the present invention.

[0056] Server 12 can also communicate with one or more external devices 14 (e.g., keyboard, pointing server, display 24, etc.), and with one or more servers that enable users to interact with server 12, and / or with any server (e.g., network card, modem, etc.) that enables server 12 to communicate with one or more other computing servers. This communication can be performed via I / O interface 22. Furthermore, server 12 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 20. As shown, network adapter 20 communicates with other modules of server 12 via bus 18. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with server 12, including but not limited to: microcode, server drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0057] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing the anti-interference detection method for online cable fault location based on sparse optimization provided in the embodiments of the present invention.

[0058] Example 4 Embodiment 4 of the present invention also provides a medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform the sparse optimization-based online cable fault location and anti-interference detection method as described in any of the above embodiments.

[0059] The computer storage medium of this invention can be any combination of one or more computer-readable media. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0060] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0061] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including, but not limited to, wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0062] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0063] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A sparse optimization-based online fault location and anti-interference detection method for cables, characterized in that, include: An SSTDR probe signal is generated and injected into the cable under test. The received echo signal returned by the cable under test is collected. A cyclic time delay shift operation is performed on the SSTDR probe signal to construct a time delay dictionary matrix. Each column of the time delay dictionary matrix corresponds to a discrete echo time delay signal template. Using the received echo signal as the observation vector, a sparse observation model is established by combining the time delay dictionary matrix. An optimization objective function with sparsity constraints is constructed. The dictionary atom with the greatest correlation to the current residual is selected and added to the support set. The optimization objective function is solved by combining the least squares method to obtain the sparse coefficient vector. Based on the non-zero element indices and values ​​of the sparse coefficient vector, the echo delay and signal amplitude corresponding to each reflected component in the received echo signal are determined. Based on the preset near-end delay threshold and detection threshold, the near-end interference component is identified from each reflected component. The time-domain interference signal is reconstructed using the time delay dictionary matrix and the sparse coefficients corresponding to the near-end interference components. The time-domain interference signal is then removed from the received echo signal to obtain the residual signal. Cross-correlation is performed on the residual signal and the SSTDR detection signal to determine the cable fault information.

2. The method according to claim 1, characterized in that, The generation of the SSTDR probe signal includes: A single-cycle SSTDR probe signal is obtained by using a pseudo-random m-sequence to perform binary phase-shift keying modulation on a sinusoidal carrier. The single-cycle SSTDR detection signal is cyclically extended to obtain a continuously transmitted SSTDR detection signal.

3. The method according to claim 2, characterized in that, The step involves performing a cyclic delay shift operation on the SSTDR probe signal to construct a delay dictionary matrix. Each column of the delay dictionary matrix corresponds to a signal template for a discrete echo delay, including: The SSTDR probe signal is discretized to obtain a discrete reference signal. ,in For discrete sampling point numbers, The value range is 0 to integers, The sampling length of a single-cycle SSTDR probe signal; Using the discrete reference signal as a reference, a cyclic time-delay shift operation is performed on the discrete reference signal, shifting by one sampling interval each time, resulting in a total of A cyclic shift sequence; Extend each cyclic shift sequence cyclically to 2. The length, as the atoms of each column of the delay dictionary matrix, is constructed with dimension . The delay dictionary matrix, the delay dictionary matrix of the first... The time delay corresponding to each column of atoms is Echo signal templates for each sampling interval, wherein The time delay dictionary matrix is ​​represented as follows: 。 4. The method according to claim 1, characterized in that, The process involves using the received echo signal as the observation vector, combining it with the time delay dictionary matrix to establish a sparse observation model, and constructing an optimization objective function with sparsity constraints, including: Based on the characteristic that fault echoes and interference components occupy only a few positions in the time delay domain in the received echo signal, the time delay dictionary matrix is ​​used as a sparse representation basis. The observation vector is represented as a linear combination of the time delay dictionary matrix and the sparse coefficient vector to be determined, plus the observation noise. A sparse observation model is established, and the expression of the sparse observation model is as follows: in, For the observation vector, For the delay dictionary matrix, Let be the sparse coefficient vector to be solved. To observe noise; Based on the sparse observation model, the objective function is to minimize the sum of squared errors between the observation vector and the product of the time delay dictionary matrix and the sparse coefficient vector. An optimization objective function with sparsity constraints is constructed, with the constraint that the number of non-zero elements in the sparse coefficient vector does not exceed a preset sparsity. The expression of the optimization objective function is as follows: in, The number of non-zero elements in the sparse coefficient vector. To preset sparsity, For the observation vector, For the delay dictionary matrix, Let be the sparse coefficient vector to be solved.

5. The method according to claim 1, characterized in that, The dictionary atom with the highest correlation to the current residual is selected and added to the support set. The objective function is then solved using the least squares method to obtain a sparse coefficient vector, including: Normalize each column of the time delay dictionary matrix to obtain a normalized dictionary; The residual is initialized to the received echo signal, and the support set is initialized to an empty set. Iteratively calculate the inner product between the current residual and each column of atoms in the normalized dictionary, and select the index corresponding to the atom with the largest absolute value of the inner product to add to the support set; Extract the sub-dictionary corresponding to the support set, solve the sparse coefficients corresponding to the sub-dictionary using the least squares method, and update the current residual; The iteration stops when the number of iterations reaches the preset maximum number of iterations or when the L2 norm of the current residual is less than the preset residual threshold, and the final sparse coefficient vector is output.

6. The method according to claim 1, characterized in that, The step of determining the echo delay and signal amplitude corresponding to each reflected component in the received echo signal based on the non-zero element indices and values ​​of the sparse coefficient vector, and identifying the near-end interference component from each reflected component according to a preset near-end delay threshold and detection threshold, includes: Based on the index position of each non-zero element in the sparse coefficient vector, the echo delay corresponding to each reflection component in the received echo signal is mapped, and the value of each non-zero element is used as the amplitude of the corresponding reflection component. For each non-zero element in the sparse coefficient vector, when the echo delay is less than the near-end delay threshold and the amplitude is less than the detection threshold, the reflection component corresponding to the non-zero element at this time is determined as a near-end interference component.

7. The method according to claim 1, characterized in that, The process involves reconstructing the time-domain interference signal using the time delay dictionary matrix and the sparse coefficients corresponding to the near-end interference components, removing the time-domain interference signal from the received echo signal to obtain the residual signal, and performing a cross-correlation operation between the residual signal and the SSTDR detection signal to determine cable fault information, including: Extract the coefficients corresponding to all non-zero elements of the near-end interference component to form an interference sparse coefficient vector; The time-domain interference signal is reconstructed by performing a linear combination operation between the interference sparse coefficient vector and the time delay dictionary matrix. The time-domain interference signal is subtracted point by point from the received echo signal to obtain the residual signal; The residual signal and the SSTDR detection signal are cross-correlated. The time delay value corresponding to the peak value of the cross-correlation operation is multiplied by the propagation speed of the signal in the cable to calculate the distance between the fault point and the detection end. The cable fault type is determined according to the polarity and magnitude of the peak value.

8. A cable fault online location and anti-interference detection device based on sparse optimization, characterized in that, include: The matrix construction module is used to generate SSTDR probe signals, inject the SSTDR probe signals into the cable under test, collect the received echo signals returned by the cable under test, perform cyclic delay shift operation on the SSTDR probe signals, and construct a delay dictionary matrix. Each column of the delay dictionary matrix corresponds to a signal template of a discrete echo delay. The function construction module is used to establish a sparse observation model by taking the received echo signal as the observation vector and combining it with the time delay dictionary matrix, construct an optimization objective function with sparsity constraints, select the dictionary atom with the greatest correlation to the current residual and add it to the support set, and solve the optimization objective function by combining the least squares method to obtain the sparse coefficient vector. The interference component identification module is used to determine the echo delay and signal amplitude corresponding to each reflection component in the received echo signal based on the index and value of the non-zero element of the sparse coefficient vector, and to identify the near-end interference component from each reflection component according to the preset near-end delay threshold and detection threshold. The fault information determination module is used to reconstruct the time-domain interference signal using the time delay dictionary matrix and the sparse coefficients corresponding to the near-end interference components, remove the time-domain interference signal from the received echo signal to obtain the residual signal, and perform cross-correlation operation on the residual signal and the SSTDR detection signal to determine the cable fault information.

9. A server, characterized in that, The server includes: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the sparse optimization-based online cable fault location and anti-interference detection method as described in any one of claims 1-7.

10. A medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform the sparse-optimized online cable fault location and anti-interference detection method as described in any one of claims 1-7.

Images