Intelligent monitoring method and system for wires and cables

Abstract

The invention relates to the technical field of wires and cables, and discloses an intelligent monitoring method and system for wires and cables, and the method comprises the steps: injecting a high-frequency pulse signal into a cable, obtaining the signal propagation characteristics, and preliminarily determining the position of a potential defect point through a time domain reflection method; for a plurality of suspected defect points, time domain reflection waveforms are converted into frequency domain signals, the signal attenuation degree is judged and compensated, attenuation interference in long-distance monitoring is eliminated, signal propagation speed changes are calculated in combination with cable parameters, and the accurate positions of the defect points are determined; for a branch cable network, a network topology structure and signal injection parameters are obtained, propagation characteristics of signals in branch paths are analyzed, reflection characteristics of branch points are judged by adopting a graph theory algorithm, and the signal injection parameters are optimized to realize accurate positioning of defect points in the branch network. According to the invention, the defect positioning problem in a complex cable network can be effectively solved, and the positioning precision and reliability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wire and cable monitoring, and particularly to an intelligent monitoring method and system for wire and cable. Background Art

[0002] Currently, in the field of Prognostics and Health Management (PHM), the existing technologies mainly focus on data acquisition, condition monitoring, fault diagnosis, Remaining Useful Life (RUL) prediction, and maintenance decision optimization. Among them, wire and cable, as the core components of power transmission and communication systems, their operating status is directly related to grid safety and communication reliability. With the increasing demand for intelligent monitoring, ensuring the stable operation of the entire cable life cycle has become an important issue in the power and communication fields. Traditional monitoring methods mostly rely on manual inspection or simple voltage and current detection, which have problems such as low efficiency, poor accuracy, and inability to detect hidden defects in real time. Especially in complex environments, it is difficult to eliminate monitoring blind spots and dead corners, resulting in difficult fault location, increased maintenance costs, and system risks.

[0003] In the existing technologies, the existing monitoring technologies usually adopt methods such as time domain reflectometry, frequency domain reflectometry and swept frequency impedance method, distributed optical fiber sensing technology, partial discharge detection, multi-sensor fusion and artificial intelligence analysis, high-frequency traveling wave method, etc. However, these technologies generally face problems such as insufficient signal attenuation compensation, difficult separation of multipath interference, and inaccurate modeling of long-distance and branch networks, resulting in the detection of internal cable defects still remaining in the indirect inference stage, and it is difficult to achieve full coverage and high-precision real-time monitoring. For example, when there are insulation defects or breakpoints in the cable, it is difficult to accurately capture the change law of the signal propagation speed. This leads to limited accuracy in time difference calculation and waveform analysis. Especially in long-distance or branched cables, signal attenuation and interference further amplify the monitoring blind spots. The existence of monitoring blind spots makes the system unable to fully cover the cable network, and potential faults in some areas are ignored, thus affecting the accuracy and reliability of fault location.

[0004] Therefore, how to inject pulse signals and accurately analyze their propagation speed and reflection characteristics, overcome monitoring blind spots, and improve fault location accuracy has become a key issue in the field of cable intelligent monitoring. Summary of the Invention

[0005] The present invention provides an intelligent monitoring method and system for wire and cable to solve the problem of accurately locating monitoring blind spots.

[0006] In a first aspect, to solve the above technical problems, the present invention provides an intelligent monitoring method for wire and cable, including: Inject a high-frequency pulse signal into the cable through a preset pulse generator, obtain the propagation time and reflection waveform of the high-frequency pulse signal in the cable, and obtain the propagation characteristics of the initial signal; According to the propagation characteristics of the initial signal, analyze the time difference and amplitude change of the reflection waveform by using the time domain reflectometry method to determine the preliminary position of the defect point in the cable; If there is at least one defect point at the preliminary position, convert the reflection waveform into a first frequency domain signal by Fourier transform, obtain the frequency domain characteristics, and judge the signal attenuation information of the at least one defect point; According to the signal attenuation information, perform gain adjustment on the first frequency domain signal by using a preset signal compensation algorithm to obtain an enhanced second frequency domain signal; According to the second frequency domain signal, the geometric parameters and material characteristics of the cable, calculate the change law of the signal propagation speed, and determine the exact position of the at least one defect point according to the change law; If the exact position is located in the branch cable network, obtain the topological structure of the branch cable network and the injection frequency and amplitude of the pulse signal, analyze the propagation time and reflection coefficient of the pulse signal in the branch cable network, and obtain the initial propagation constraint conditions of the branch cable network; According to the initial propagation constraint conditions, analyze the propagation trajectory of the pulse signal in the branch cable network by using the graph theory algorithm, and combine the propagation time and the reflection coefficient to judge the reflection characteristics of the branch cable network; According to the reflection characteristics, adjust the injection frequency and amplitude of the pulse signal, obtain optimized reflection waveform data, and obtain the positioning result of the defect point in the branch cable network.

[0007] In an alternative embodiment, the step of analyzing the time difference and amplitude change of the reflection waveform by using the time domain reflectometry method according to the propagation characteristics of the initial signal to determine the preliminary position of the defect point in the cable includes: Obtain the emission time and reflection waveform data of the initial signal, and use a high-precision sampling device to record the signal frequency and amplitude change of the initial signal; Extract the waveform characteristics of the initial signal by Fourier transform to obtain an original data set containing the time difference and amplitude change; If the noise interference in the original data set exceeds a preset threshold, perform denoising processing on the reflection waveform of the initial signal by using wavelet transform, and pre-establish a signal propagation speed model according to the material characteristics of the cable; Calculate the corresponding relationship between the time difference and the propagation distance according to the speed model to determine the waveform feature set after denoising; Based on the denoised waveform feature set, combined with the propagation distance and the time difference, use the linear regression algorithm to calculate the propagation distance of the defect point relative to the emission point, and obtain the preliminary position coordinates.

[0008] In an alternative embodiment, if there is at least one defect point in the preliminary position, then convert the reflection waveform into a first frequency-domain signal through Fourier transform, obtain frequency-domain features, and judge the signal attenuation information of the at least one defect point, including: Obtain the time-domain data of the reflection waveform, convert the time-domain data into a frequency-domain signal by using Fourier transform, and extract the features of the signal frequency and amplitude change in the frequency domain to obtain a frequency-domain feature data set; According to the frequency-domain feature data set, for each suspected defect point, calculate the amplitude attenuation value of the frequency-domain signal, and use a preset threshold to judge whether the amplitude attenuation is abnormal to determine the signal attenuation information.

[0009] In an alternative embodiment, calculating the variation law of the signal propagation speed according to the second frequency-domain signal, the geometric parameters and material characteristics of the cable, and determining the exact position of the at least one defect point according to the variation law, includes: According to the second frequency-domain signal, query the geometric parameter database of the cable and calculate the signal propagation speed, where the signal propagation speed v is calculated by formula (1): , formula (1); Where, represents the signal propagation speed, represents the speed of light constant, represents the dielectric constant; If the deviation of the propagation speed from the preset standard value exceeds the preset threshold, then analyze the reflection coefficient by the time-domain reflectometry method, obtain the propagation delay time, and judge the preliminary position of the at least one defect point; Use a waveform analysis tool to extract the features of the amplitude and phase of the reflection signal, and combine the length and material characteristics of the cable to calculate the exact coordinates of the at least one defect point.

[0010] In an alternative embodiment, if the exact position is located in the branch cable network, then obtain the topological structure of the branch cable network and the injection frequency and amplitude of the pulse signal, analyze the propagation time and reflection coefficient of the pulse signal in the branch cable network, and obtain the initial propagation constraint conditions of the branch cable network, including: Obtain the network physical connection data of the cable, record the starting and ending coordinates of each branch path, generate a topological graph including node coordinates and connection relationships, and determine the branch network topology of the cable; Calculate the path lengths from the injection point to each branch path node according to the branch network topology and the injection position of the pulse signal, measure the propagation time of the pulse signal in each branch path using the time domain reflectometry method, and obtain the propagation time of the pulse signal; Based on the propagation time of the pulse signal and the signal frequency of the pulse signal, calculate the reflection amount of the pulse signal at the node according to the cable impedance characteristics using formula (2), and determine the signal reflection coefficient; R=(Z2-Z1) / (Z2+Z1), formula (2); where, Z1 is the incident path impedance and Z2 is the reflection path impedance; According to the reflection coefficient of the pulse signal and the branch network topology, combined with the length distribution of the branch paths, optimize the propagation constraint parameters using the support vector machine algorithm, generate constraint conditions including time delay and reflection intensity, and obtain the initial propagation constraint conditions.

[0011] In an optional implementation manner, according to the initial propagation constraint conditions, analyze the propagation trajectory of the pulse signal in the branch cable network using the graph theory algorithm, and combine the propagation time and the reflection coefficient to judge the reflection characteristics of the branch cable network, including: Obtain the node connections and path weights from the branch cable network, use the shortest path algorithm in the graph theory algorithm to calculate the propagation trajectory of the pulse signal from the starting node to each branch point, and obtain the propagation trajectory of the pulse signal; According to the propagation trajectory of the pulse signal, combined with the preset propagation speed, calculate the propagation time of the pulse signal at each branch point, and determine the time delay; If the time delay is greater than the preset threshold, then according to the signal attenuation and the reflection coefficient, use the linear regression algorithm to calculate the reflection intensity of each branch point, and judge the reflection characteristics of the branch cable network; Obtain the characteristics of each branch point through the network topology and the node connections, combine the reflection intensity of each branch point, update the path weights, and obtain the reflection characteristics at each branch point.

[0012] In an optional implementation manner, according to the reflection characteristics, adjust the injection frequency and amplitude of the pulse signal, obtain the optimized reflection waveform data, and obtain the positioning result of the defect point in the branch cable network, including: According to the network structure of the branch cable, obtain the signal propagation path, use the time domain reflectometer to generate an initial pulse signal, and obtain the first reflection waveform data; If the signal-to-noise ratio of the first reflected waveform data is lower than a preset threshold, the frequency components of the first reflected waveform data are extracted by Fourier transform, and the pulse signal frequency and amplitude are adjusted to obtain second reflected waveform data; Wavelet transform is used on the second reflected waveform data for waveform feature extraction to obtain a waveform feature vector and determine the preliminary position of the defect point; According to the preliminary position, through the support vector machine algorithm, combined with the network structure of the branch cable and the signal propagation path, the coordinates of the defect point are calculated to obtain the positioning result of the defect point.

[0013] In a second aspect, the present invention provides an intelligent monitoring system for electric wires and cables, including: A first acquisition module for injecting a high-frequency pulse signal into the cable through a preset pulse generator, acquiring the propagation time and reflected waveform of the high-frequency pulse signal in the cable, and obtaining the propagation characteristics of the initial signal; A first determination module for analyzing the time difference and amplitude change of the reflected waveform by using the time domain reflectometry according to the propagation characteristics of the initial signal, and determining the preliminary position of the defect point in the cable; A first judgment module for, if there is at least one defect point at the preliminary position, converting the reflected waveform into a first frequency domain signal by Fourier transform, acquiring frequency domain characteristics, and judging the signal attenuation information of the at least one defect point; A second acquisition module for performing gain adjustment on the first frequency domain signal by using a preset signal compensation algorithm according to the signal attenuation information to obtain an enhanced second frequency domain signal; A second determination module for calculating the change law of the signal propagation speed according to the second frequency domain signal, the geometric parameters and material characteristics of the cable, and determining the exact position of the at least one defect point according to the change law; A third acquisition module for, if the exact position is located in the branch cable network, acquiring the topological structure of the branch cable network and the injection frequency and amplitude of the pulse signal, analyzing the propagation time and reflection coefficient of the pulse signal in the branch cable network, and obtaining the initial propagation constraint conditions of the branch cable network; A second judgment module for analyzing the propagation trajectory of the pulse signal in the branch cable network by using a graph theory algorithm according to the initial propagation constraint conditions, and combining the propagation time and the reflection coefficient to judge the reflection characteristics of the branch cable network; A third determination module for adjusting the injection frequency and amplitude of the pulse signal according to the reflection characteristics, acquiring optimized reflected waveform data, and obtaining the positioning result of the defect point in the branch cable network.

[0014] In a third aspect, the present invention further provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the intelligent monitoring method described in any one of the above is implemented.

[0015] In a fourth aspect, the present invention further provides a computer-readable storage medium, which includes a stored computer program. Wherein, when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the intelligent monitoring method described in any one of the above.

[0016] Compared with the prior art, the present invention has the following beneficial effects: (1) By high-frequency sampling, the present invention captures the frequency and amplitude changes of the signal, ensuring the integrity of the waveform data and providing a reliable basis for subsequent feature extraction.

[0017] (2) By constructing the mapping relationship between the attenuation value and the defect type, the present invention further refines the diagnosis. Through this mapping, the system can not only locate the defect, but also infer the nature of the defect, providing support for maintenance decisions. This extended scheme improves the comprehensiveness of the diagnosis.

[0018] (3) By calculating the amplitude means of the data points before and after the interference point, the present invention fills in the abnormal fluctuations and generates smooth time-domain waveform data. In one test, the abnormal point of 0.95V at 42 μs was adjusted to 0.82V after interpolation, which is more coordinated with the surrounding waveforms. This smoothing process ensures the waveform continuity and facilitates subsequent defect location and analysis.

[0019] (4) By fast Fourier transform and high-pass filtering, the present invention ensures the purity of the frequency-domain signal. Database query and time-domain reflectometry provide a reliable basis for location, while waveform analysis further refines the results. This multi-faceted collaborative scheme design not only improves the monitoring efficiency, but also enhances the robustness of defect location.

[0020] (5) Through multi-dimensional data fusion, the present invention improves the accuracy of cable network analysis. The topology map visually presents the network structure, the path length and propagation time data provide a basis for reflection analysis, the reflection coefficient quantifies the node state, and the support vector machine optimizes the constraint conditions. This multi-faceted mutual support method ensures the logical rigor from data acquisition to constraint generation, helps to quickly locate network problems, and improves the maintenance efficiency.

[0021] (6) The weight design in the present invention facilitates the subsequent algorithm to accurately identify the priority path of signal propagation. When updating the path weight, the network topology can be adjusted according to the reflection intensity. This update can optimize the network model and make it closer to the real propagation environment.

[0022] (7) Through multi-step analysis, the present invention obtains defect location from the path and forms a complete technical chain. Each step supports each other. For example, Fourier transform optimizes the signal quality and provides a clear input for wavelet transform; wavelet transform extracts features and provides accurate data for SVM positioning. This progressive logic ensures the positioning accuracy. At the same time, the method is flexible and can adapt to scenarios with different network scales and complexities. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a schematic flowchart of an intelligent monitoring method for wire and cable provided by the first embodiment of the present invention; Figure 2 is a schematic structural diagram of an intelligent monitoring system for wire and cable provided by the second embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0025] Referring to Figure 1 , the first embodiment of the present invention provides an intelligent monitoring method for wire and cable, including the following steps: S11, injecting a high-frequency pulse signal into the cable through a preset pulse generator, obtaining the propagation time and reflection waveform of the high-frequency pulse signal in the cable, and obtaining the propagation characteristics of the initial signal; S12, according to the propagation characteristics of the initial signal, analyzing the time difference and amplitude change of the reflection waveform by using the time domain reflectometry method, and determining the preliminary position of the defect point in the cable; S13, if there is at least one defect point at the preliminary position, converting the reflection waveform into a first frequency domain signal by Fourier transform, obtaining frequency domain characteristics, and judging the signal attenuation information of the at least one defect point; S14, according to the signal attenuation information, performing gain adjustment on the first frequency domain signal by using a preset signal compensation algorithm to obtain an enhanced second frequency domain signal; S15, according to the second frequency domain signal, the geometric parameters and material characteristics of the cable, calculating the change law of the signal propagation speed, and determining the accurate position of the at least one defect point according to the change law S16. If the precise location is in the branch cable network, obtain the topological structure of the branch cable network and the injection frequency and amplitude of the pulse signal, analyze the propagation time and reflection coefficient of the pulse signal in the branch cable network, and obtain the initial propagation constraint conditions of the branch cable network; S17. According to the initial propagation constraint conditions, use graph theory algorithms to analyze the propagation trajectory of the pulse signal in the branch cable network, and combine the propagation time and the reflection coefficient to judge the reflection characteristics of the branch cable network; S18. According to the reflection characteristics, adjust the injection frequency and amplitude of the pulse signal, obtain optimized reflected waveform data, and obtain the positioning result of the defect point in the branch cable network.

[0026] As Figure 1 shown, a smart monitoring method for electric wires and cables provided by the first embodiment of the present invention may specifically include: In step S11, inject a high-frequency pulse signal into the cable through a preset pulse generator, obtain the propagation time and reflected waveform of the high-frequency pulse signal in the cable, and obtain the propagation characteristics of the initial signal.

[0027] Generate a pulse signal of a specific frequency through a preset pulse generator, inject it into the cable, and use a high-precision sensor to obtain the propagation time and reflected waveform data from the end of the cable to obtain an initial signal dataset. If the time resolution of the initial signal dataset is lower than a preset threshold, the initial signal dataset is interpolated by a digital signal processor, and a linear interpolation algorithm is used to improve the time resolution to obtain an enhanced signal dataset. The fast Fourier transform algorithm is used to perform frequency-domain analysis on the reflected waveforms of the enhanced signal dataset, and signal attenuation and impedance change characteristics are extracted from the frequency-domain features to determine the propagation characteristic parameters of the cable. Exemplarily, injecting a high-frequency pulse signal through a pulse generator in cable fault detection is a technique based on the time-domain reflectometry method. The pulse signal propagates in the cable and generates a reflected wave when it encounters an impedance change point such as an open circuit or a short circuit. By recording the time difference between the transmitted pulse and the received reflected wave, that is, the propagation time, the cable length or the location of the fault point can be calculated. This method is applicable to the maintenance of long-distance power cables or communication cables, can quickly locate faults, and reduce the troubleshooting time. In one possible implementation, the pulse generator is set to a device with an output frequency of 10 MHz, and a pulse is injected into a cable with a signal velocity of 2×10^8 m / s. Assuming the propagation time is recorded as 50 μs, exceeding the preset threshold of 40 μs, the cable length is calculated by the formula L = vt. Substituting the values, the cable length is approximately 10000 m. This method is simple and efficient, suitable for on-site rapid measurement, and can provide high-precision length data, which is helpful for subsequent fault location. It should be noted that the acquisition of the reflected waveform depends on a high-precision sensor. The sensor is installed at the cable terminal to capture the reflected waveform data in real time. The initial waveform may be distorted due to noise interference, such as electromagnetic interference or signal attenuation caused by cable aging. Using the wavelet transform algorithm to denoise the waveform can effectively filter out high-frequency noise and retain the characteristic information of the fault point.

[0028] For example, in a certain test, the initial waveform showed multiple clutter peaks. After wavelet transform, a reflected peak with an amplitude of 0.8 V and a time of 45 μs was clearly extracted. This denoising process improves the accuracy of feature extraction and provides reliable data for fault location.

[0029] Specifically, the extraction of characteristic parameters includes time and amplitude information. The time feature reflects the duration of the pulse propagating to the fault point and returning, and the amplitude feature is related to the degree of impedance change at the fault point.

[0030] In one embodiment, assuming the extracted features are time 45 μs and amplitude 0.8 V, combined with the signal velocity of 2×10^8 m / s, the distance of the fault point is calculated to be 4,500 m. This indicates that the fault point is located in the middle section of the cable, and it may be due to loose joints or damaged insulation. Compared with traditional manual troubleshooting, this method has high positioning accuracy, and the error is usually within 1 m, significantly shortening the repair time.

[0031] Preferably, the calculation of the impedance change point needs to consider the dielectric characteristics of the cable.

[0032] For example, the signal velocity of the cable is affected by the insulating material. The signal velocity in polyethylene dielectric is usually 2 / 3 of the speed of light. In actual tests, the velocity parameter needs to be calibrated to ensure accurate positioning. In one test, the calibrated velocity was 1.98×10^8 m / s, and the recalculated fault point distance was 4455 m, which was close to the actual measured value, verifying the reliability of the method.

[0033] It can be understood that the beneficial effects brought by this technology include high efficiency and accuracy. Quickly locating the fault point can reduce the power outage time and improve the power supply reliability; high-precision data supports preventive maintenance and extends the cable life.

[0034] For example, a certain power company adopted this method and completed the fault location of 10 km of cable within one day, saving 80% of the time compared with the traditional method. In addition, the application of feature extraction and denoising algorithms enables the system to adapt to complex environments, suitable for aging cables or high-noise scenarios, enhancing the robustness of the technology.

[0035] In step S12, according to the propagation characteristics of the initial signal, the time difference and amplitude change of the reflected waveform are analyzed by the time domain reflectometry method to determine the preliminary position of the defect point in the cable.

[0036] Obtain the emission time and reflected waveform data of the initial signal, and record the signal frequency and amplitude change of the initial signal using a high-precision sampling device; Extract the waveform characteristics of the initial signal through Fourier transform to obtain the original data set containing the time difference and amplitude change; If the noise interference in the original data set exceeds the preset threshold, wavelet transform is used to denoise the reflected waveform of the initial signal, and a signal propagation velocity model is established in advance according to the material characteristics of the cable; Calculate the corresponding relationship between the time difference and the propagation distance according to the velocity model to determine the denoised waveform feature set; Through the denoised waveform feature set, combined with the propagation distance and the time difference, a linear regression algorithm is used to calculate the propagation distance of the defect point relative to the emission point to obtain the preliminary position coordinates.

[0037] Exemplarily, in cable defect detection, obtaining the emission time of the initial signal and the reflected waveform data is a core step. The emission time records the moment when the pulse is injected into the cable, and the reflected waveform data contains the information returned after the signal encounters a defect point in the cable. By using a high-precision sampling device, such as an oscilloscope with a sampling rate of 100 MHz, the frequency and amplitude changes of the signal can be captured. This device ensures the integrity of the waveform data through high-frequency sampling, providing a reliable basis for subsequent feature extraction.

[0038] For example, in a test, the device records the emission time as 0 μs, and the reflected waveform shows a peak at 48 μs with an amplitude of 0.7 V, indicating the presence of a significant defect point.

[0039] In a possible implementation, waveform features are extracted through Fourier transform. Fourier transform converts the time-domain signal into the frequency domain, separating different frequency components to identify key features in the waveform, such as the main frequency and sub-harmonics. Suppose in a test, the Fourier transform shows that the main frequency is 8 MHz and there is obvious low-frequency noise. This analysis helps generate an original dataset containing time differences and amplitude changes, laying the foundation for subsequent processing.

[0040] It should be noted that the noise in the original dataset often exceeds the standard due to environmental electromagnetic interference or uneven cable material, and further denoising is required.

[0041] Specifically, if the noise interference exceeds a preset threshold, such as the signal-to-noise ratio is lower than 10 dB, wavelet transform is used for denoising. Wavelet transform decomposes the signal into sub-bands of different scales, filtering out high-frequency noise while retaining the characteristics of the defect points.

[0042] For example, in a detection, the original waveform shows multiple chaotic peaks. After wavelet transform, a clear reflected peak is extracted at 47 μs with an amplitude of 0.65 V. This denoising method improves the data quality and provides support for accurately calculating the defect location.

[0043] In an embodiment, according to the signal propagation speed model established in advance based on the cable material, the corresponding relationship between the time difference and the propagation distance is calculated. The insulating material of the cable, such as polyethylene, affects the signal speed, and usually takes 2 / 3 of the speed of light. Suppose the speed in the model is 2×10^8 m / s, combined with the time difference of 47 μs, the distance of the defect point can be initially estimated.

[0044] Preferably, the speed model is calibrated to improve the accuracy.

[0045] For example, after the measured speed is adjusted to 1.95×10^8 m / s, the recalculated distance is closer to the actual value.

[0046] It is understandable that, by using the denoised waveform feature set, combining the signal propagation speed and the time difference, the linear regression algorithm is used to calculate the position of the defect point. Linear regression fits the relationship between the time difference and the distance and outputs the preliminary position coordinates.

[0047] For example, multiple test data indicate that the time difference is between 45 μs and 50 μs, and the regression model predicts that the distance of the defect point is between 4000 m and 5000 m. This method improves the stability and accuracy of positioning.

[0048] For example, the system can be extended to multi-defect detection. By analyzing the multiple reflection peaks in the waveform, multiple defect points are identified. In one test, the waveform shows two peaks, corresponding to 46 μs and 52 μs respectively, indicating the existence of two defects. This extended solution enhances the system's diagnostic ability for complex cables and provides support for comprehensive maintenance.

[0049] In step S13, if there is at least one defect point in the preliminary position, the reflected waveform is converted into a first frequency-domain signal through Fourier transform, the frequency-domain features are obtained, and the signal attenuation information of the at least one defect point is judged.

[0050] Obtain the time-domain data of the reflected waveform, convert the time-domain data into a frequency-domain signal by using Fourier transform, and extract the features of the signal frequency and amplitude change in the frequency domain to obtain a frequency-domain feature data set; According to the frequency-domain feature data set, for each suspected defect point, calculate the amplitude attenuation value of the frequency-domain signal, and use a preset threshold to judge whether the amplitude attenuation is abnormal to determine the signal attenuation information.

[0051] Exemplarily, in cable defect detection, obtaining the time-domain reflected waveform data is a basic step in the analysis. The time-domain reflectometry generates waveform data containing time and amplitude information by recording the propagation and reflection of pulse signals in the cable.

[0052] For example, an oscilloscope with a sampling rate of 200 MHz can be used to capture high-frequency pulse signals to ensure the integrity of waveform details. In one test, after the device records the initial pulse emission, the waveform shows a significant peak at 50 μs with an amplitude of 0.8 V. This indicates that there may be an impedance change point in the cable, providing the original data for subsequent frequency-domain analysis.

[0053] In a possible implementation, Fourier transform is used to convert the time-domain signal into a frequency-domain signal to extract the frequency and amplitude features. The core of Fourier transform is to decompose the waveform in the time dimension into different frequency components, thereby revealing the spectral characteristics of the signal.

[0054] For example, in the test, the Fourier transform shows that the main frequency is 10 MHz, accompanied by a small amount of high-frequency harmonics. This frequency-domain feature dataset can clearly display the energy distribution of the signal at different frequencies, providing a basis for the identification of defect points.

[0055] It should be noted that frequency-domain analysis is more adept at separating noise and useful signals compared to time-domain analysis, especially in complex environments where it can highlight the main frequency characteristics of defect points.

[0056] Specifically, for the frequency-domain feature dataset, calculating the amplitude attenuation value of each suspected defect point is a key step. The amplitude attenuation value reflects the energy loss caused by defect points during signal propagation.

[0057] For example, by analyzing the frequency-domain data, it is found that the main frequency amplitude of a certain suspected defect point decays from 1.0 V to 0.4 V, with an attenuation ratio as high as 60%. This attenuation value can be quantified by comparing the amplitude difference between the initial signal and the reflected signal.

[0058] In one embodiment, assuming the preset threshold is 50% attenuation, if the attenuation value of a certain suspected defect point exceeds this threshold, it is determined as abnormal, indicating that there may be significant defects at this point, such as insulation damage or loose joints.

[0059] Preferably, the setting of the preset threshold needs to be combined with the cable material and the test environment.

[0060] For example, the normal attenuation range of polyethylene-insulated cables is usually between 20% - 40%. Therefore, it is reasonable to set 50% as the abnormal threshold. In a test, for three suspected defect points, the calculated amplitude attenuation values are 30%, 55%, and 65% respectively. According to the threshold judgment, the latter two are marked as abnormal. This method provides a preliminary assessment of the defect severity by quantifying the attenuation degree.

[0061] It can be understood that the determination of abnormal amplitude attenuation can also be combined with other indicators of frequency-domain features.

[0062] For example, if the main frequency of a certain defect point shifts, from the expected 10 MHz to 9 MHz, and is accompanied by high attenuation, it may indicate more serious structural damage. 6]

[0063] In one embodiment, the frequency-domain features of a defect point recorded by the system show that the main frequency shifts to 8.5 MHz and the amplitude attenuation reaches 70%. By comprehensively analyzing the frequency shift and the attenuation value, it is confirmed that this point is a high-risk defect. This multi-dimensional analysis enhances the reliability of defect determination.

[0064] For example, the expansion solution can further refine the diagnosis by constructing a mapping relationship between the attenuation value and the defect type. In a single test, the system compares historical data and finds that 60%-70% attenuation often corresponds to the aging of the insulation layer, while more than 80% attenuation is mostly related to broken cores. Through this mapping, the system can not only locate the defect but also infer the nature of the defect, providing support for maintenance decisions. This expansion solution improves the comprehensiveness of the diagnosis.

[0065] In one embodiment, for long-distance cables, frequency-domain analysis can be combined with the characteristics of multiple reflection waveforms. For example, the waveform shows two abnormal attenuations, at 45 μs and 55 μs respectively, with attenuation values of 58% and 62% respectively. By analyzing the frequency-domain characteristics of the two reflections, the independence of the two defects is confirmed. This method is applicable to complex cable networks to ensure accurate diagnosis in multi-defect scenarios.

[0066] In step S14, according to the signal attenuation information, a preset signal compensation algorithm is used to adjust the gain of the first frequency-domain signal to obtain an enhanced second frequency-domain signal.

[0067] Obtain frequency-domain signal data, and use a preset signal compensation algorithm to calculate the amplitude gain value for each suspected attenuation point to obtain the enhanced frequency-domain signal dataset. According to the enhanced frequency-domain signal dataset, extract a feature set containing frequency characteristics and amplitude changes, and use the inverse Fourier transform to convert the frequency-domain signal into a time-domain waveform to obtain the enhanced time-domain waveform data. If there are amplitude abnormal points in the enhanced time-domain waveform data, then use a preset threshold to determine whether the abnormal points belong to attenuation interference to determine the interference point position set. For the interference point position set, use a linear interpolation algorithm to smooth the enhanced time-domain waveform data to obtain the final smoothed time-domain waveform data.

[0068] Specifically, in the field of cable defect detection, obtaining frequency-domain signal data and performing subsequent processing are key steps.

[0069] Exemplarily, the frequency-domain signal data is converted from the time-domain waveform through the fast Fourier transform and contains frequency and amplitude information. To enhance the signal quality, a preset signal compensation algorithm needs to be used.

[0070] It should be noted that the signal compensation algorithm aims to offset the natural attenuation during propagation and highlight the characteristics of the defect points.

[0071] Specifically, a compensation algorithm can dynamically adjust the amplitude of each frequency component based on the attenuation characteristics of the cable material.

[0072] For example, it is found in the test that the polyethylene cable has a natural attenuation of 20% at a frequency of 5 MHz. The algorithm boosts the amplitude of this frequency component by 20% through a preset gain factor to generate an enhanced frequency-domain signal dataset.

[0073] In one embodiment, calculating the amplitude gain value for each suspected attenuation point is a core step. The amplitude gain value reflects the enhancement degree of the compensated signal.

[0074] For example, the original amplitude of a certain suspected attenuation point is 0.5 V, and it increases to 0.75 V after being processed by the compensation algorithm, with a gain value of 50%. This gain value is quantified by comparing the amplitude difference between the original and the compensated ones, ensuring that the signals at the defect points are more prominent.

[0075] Preferably, the gain value needs to be adjusted in combination with the cable length and the test environment to avoid introducing noise due to overcompensation. In one test, the gain values of three suspected attenuation points are 40%, 55%, and 60% respectively, providing clear data for subsequent analysis.

[0076] It can be understood that the enhanced frequency-domain signal dataset needs to extract the feature sets of frequency features and amplitude variations.

[0077] For example, the feature set can include the main frequency, harmonic distribution, and amplitude change rate. In one possible implementation, the test shows that the main frequency is 8 MHz, accompanied by a small amount of 12 MHz harmonics, and the amplitude change rate reaches 30%. These features are converted into enhanced time-domain waveform data through the inverse Fourier transform to restore the signal form in the time dimension.

[0078] For example, after the inverse transformation, the waveform has a peak value of 0.9 V at 40 μs, which is clearer than the original waveform and is conducive to detecting abnormal points. For the enhanced time-domain waveform data, it is necessary to identify the amplitude abnormal points and determine whether they are attenuation interferences.

[0079] Specifically, the preset threshold can be set such that an amplitude exceeding 0.85 V is considered abnormal.

[0080] In one embodiment, the waveform has a peak value of 0.95 V at 45 μs, exceeding the threshold. By analyzing its frequency-domain features, it is confirmed that it is the main frequency signal rather than noise, and it is determined as an attenuation interference point and recorded in the interference point position set.

[0081] For example, two interference points are recorded in the test, at 42 μs and 50 μs respectively. Finally, for the interference point position set, a linear interpolation algorithm is used to smooth the time-domain waveform.

[0082] For example, the interpolation algorithm fills abnormal fluctuations by calculating the amplitude means of data points before and after the interfering points, generating smooth time-domain waveform data. In a test, an abnormal point of 0.95V at 42 μs was adjusted to 0.82V after interpolation, which is more coordinated with the surrounding waveforms. This smoothing process ensures waveform continuity and facilitates subsequent defect location and analysis.

[0083] In step S15, according to the second frequency-domain signal, the geometric parameters and material properties of the cable, calculate the variation law of the signal propagation speed, and determine the exact positions of the at least one defect point according to the variation law.

[0084] According to the second frequency-domain signal, query the geometric parameter database of the cable and calculate the signal propagation speed, where the signal propagation speed v is calculated by formula (1): , formula (1); where, represents the signal propagation speed, represents the speed of light constant, represents the dielectric constant; If the deviation of the propagation speed from the preset standard value exceeds the preset threshold, analyze the reflection coefficient by the time-domain reflectometry method to obtain the propagation delay time, and judge the preliminary positions of the at least one defect point; Use a waveform analysis tool to extract the characteristics of the amplitude and phase of the reflection signal, and calculate the exact coordinates of the at least one defect point in combination with the length and material properties of the cable.

[0085] Specifically, the fast Fourier transform is an efficient algorithm for converting a time-domain signal into a frequency-domain signal to facilitate the analysis of the frequency components of the signal.

[0086] For example, in cable monitoring, the collected time-domain signal may contain voltage fluctuations or current changes. Through the fast Fourier transform, it can be converted into the frequency domain to obtain the amplitude and phase information of different frequency components.

[0087] Exemplarily, assume that the sampling rate of the collected time-domain signal is 10 kHz and the signal length is 1024 points. After the fast Fourier transform, 512 frequency points can be obtained, ranging from 0 to 5 kHz, revealing the main frequency components of the signal. This conversion facilitates subsequent filtering and feature extraction. A high-pass filter is used to remove low-frequency environmental noise and improve the clarity of the frequency-domain signal.

[0088] Specifically, low-frequency noise may be caused by temperature changes or power supply fluctuations, and the frequency is usually lower than 100 Hz.

[0089] In a possible implementation, a high-pass filter with a cut-off frequency of 200 Hz is designed to retain high-frequency signals and filter out low-frequency interference.

[0090] For example, the cable signal may contain 50 Hz power supply noise. After being processed by the high-pass filter, the useful signals above 200 Hz in the frequency-domain signal are retained, and the low-frequency noise is effectively suppressed. This processing ensures the accuracy of subsequent analysis. Based on the enhanced frequency-domain signal, querying the cable geometric parameter database to calculate the signal propagation speed is a key step. The cable geometric parameter database stores information such as the dielectric constant and conductor diameter of the cable.

[0091] Preferably, assume that the dielectric constant of a certain cable in the database is 2.25, and calculate the propagation speed v through the formula.

[0092] For example, combined with the speed of light c, the calculation result shows that the propagation speed is about 2 / 3 of the speed of light. This speed value is used to calibrate the signal propagation model and provide basic data for subsequent positioning.

[0093] It can be understood that the comprehensiveness of the database directly affects the reliability of the calculation results. If the propagation speed deviates significantly from the standard value, it is necessary to analyze the reflection coefficient by the time-domain reflectometry method to determine the defect location. The time-domain reflectometry method measures the propagation delay time by sending a pulse signal and receiving the reflected signal.

[0094] In one embodiment, assume that the cable length is 100 m and the propagation speed is 2×10^8 m / s. Under normal circumstances, the round-trip time of the pulse is about 1 μs. If the measured delay time is 1.2 μs, it can be inferred that the defect point is located at about 120 m. This method directly reflects the defect location through the time difference, which is simple and efficient. The waveform analysis tool is used to extract the amplitude and phase characteristics of the reflected signal, and combined with the cable length and material characteristics, the defect point is accurately located.

[0095] For example, a certain cable is made of polyethylene and has a length of 200 m. By analyzing the amplitude attenuation and phase shift of the reflected signal, it can be determined whether the defect point is an open circuit or a short circuit.

[0096] Specifically, the phase of the reflected signal of an open circuit usually shows a 180° inversion, while there is no obvious phase change for a short circuit. Combined with the cable length, the coordinates of the defect point are calculated to be 180 m. This feature extraction method improves the positioning accuracy.

[0097] It should be noted that the above-mentioned each technical topic is closely connected. From signal conversion to noise removal, then to speed calculation and defect location, a complete cable monitoring process is formed. The implementation methods of each step support each other, ensuring the logical consistency from signal acquisition to defect location.

[0098] For example, the fast Fourier transform and high-pass filtering ensure the purity of the frequency-domain signal, database query and time-domain reflectometry provide a reliable basis for positioning, and waveform analysis further refines the results. This multi-faceted collaborative scheme design not only improves the monitoring efficiency but also enhances the robustness of defect location.

[0099] In step S16, if the precise location is in the branch cable network, obtain the topological structure of the branch cable network and the injection frequency and amplitude of the pulse signal, analyze the propagation time and reflection coefficient of the pulse signal in the branch cable network, and obtain the initial propagation constraint conditions of the branch cable network.

[0100] Obtain the network physical connection data of the cable, record the start and end coordinates of each branch path, generate a topological graph including node coordinates and connection relationships, and determine the branch network topology of the cable; According to the branch network topology and the injection position of the pulse signal, calculate the path lengths from the injection point to the nodes of each branch path, and use time-domain reflectometry to measure the propagation time of the pulse signal in each branch path to obtain the propagation time of the pulse signal; Based on the propagation time of the pulse signal and the signal frequency of the pulse signal, and based on the cable impedance characteristics, use formula (2) to calculate the reflection amount of the pulse signal at the node and determine the signal reflection coefficient; R = (Z2 - Z1) / (Z2 + Z1), formula (2); where Z1 is the incident path impedance and Z2 is the reflected path impedance; According to the reflection coefficient of the pulse signal and the branch network topology, combined with the length distribution of the branch paths, use the support vector machine algorithm to optimize the propagation constraint parameters, generate constraint conditions including time delay and reflection intensity, and obtain the initial propagation constraint conditions.

[0101] Exemplarily, in the acquisition of the physical connection data of the cable network, the start and end coordinates of the cable branches can be collected through a distributed sensor network. Suppose an underground cable network contains three branch nodes, namely point A (0,0), point B (100,0), and point C (50,50), with the unit being meters. The sensor records the connection relationship of each section of the cable, such as the straight branch from A to B and another branch from B to C. Through these coordinate data, a two-dimensional topological graph is generated, where the nodes in the graph are represented by coordinates and the edges are represented by the actual cable paths. This method is convenient for visually displaying the network layout and provides a basis for subsequent analysis.

[0102] In a possible implementation, based on the above topological graph, determine the pulse injection location. For example, select point A as the injection point. Calculate the path lengths from point A to each node. For instance, the distance from A to B is 100 meters, and the distance from A to C through point B is 150 meters. The time domain reflectometry method injects a pulse signal into the cable and measures the signal return time. Assume that the propagation time of the signal along the path from A to B is 0.5 microseconds, and along the path from A to C is 0.75 microseconds. These time data reflect the propagation characteristics of the signal in different paths and provide key basis for subsequent analysis.

[0103] Specifically, for the signal propagation time and the pulse signal frequency, combined with the cable impedance characteristics, calculate the reflection coefficient at the node.

[0104] For example, the incident impedance of the path from A to B is 50 ohms, and the impedance of the reflected path is 75 ohms. Calculate the reflection coefficient through a formula. This method quantifies the signal reflection intensity through impedance differences and helps to identify the connection status or potential abnormalities at the node.

[0105] Exemplarily, if there is a looseness at the connection of point B, the reflection coefficient will increase significantly, prompting the maintenance personnel to check this node.

[0106] Preferably, when analyzing the reflection coefficient and the branch network topology, combined with the branch length distribution, adopt the support vector machine algorithm to optimize the propagation constraint parameters. Assume that the branch lengths are 100 meters and 50 meters respectively. The support vector machine can classify the normal and abnormal paths according to the reflection coefficient and the time delay data.

[0107] In one embodiment, the algorithm generates constraint conditions. For example, when the time delay is less than 0.6 microseconds and the reflection intensity is lower than a certain threshold, it is determined as a normal path. These constraint conditions provide an accurate basis for subsequent fault location.

[0108] It can be understood that the advantage of the above method is to improve the accuracy of cable network analysis through multi-dimensional data fusion.

[0109] For example, the topological graph visually presents the network structure. The path length and propagation time data provide a basis for reflection analysis. The reflection coefficient quantifies the node state, and the support vector machine optimizes the constraint conditions. This multi-faceted mutually supporting method ensures the logical rigor from data collection to constraint generation, helps to quickly locate network problems, and improves the maintenance efficiency.

[0110] In step S17, according to the initial propagation constraint conditions, use the graph theory algorithm to analyze the propagation trajectory of the pulse signal in the branch cable network, and combine the propagation time and the reflection coefficient to judge the reflection characteristics of the branch cable network.

[0111] Obtain the node connections and path weights from the branch cable network, and use the shortest path algorithm in graph theory algorithms to calculate the propagation trajectory of the pulse signal from the starting node to each branch point, thereby obtaining the propagation trajectory of the pulse signal; According to the propagation trajectory of the pulse signal and in combination with the preset propagation speed, calculate the propagation time of the pulse signal at each branch point to determine the time delay; If the time delay is greater than the preset threshold, then according to the signal attenuation and reflection coefficient, use the linear regression algorithm to calculate the reflection intensity at each branch point to judge the reflection characteristics of the branch cable network; Through the network topology and the node connections, obtain the characteristics of each branch point, and in combination with the reflection intensity of each branch point, update the path weights to obtain the reflection characteristics at each branch point.

[0112] Exemplarily, when obtaining the node connections and path weights in the branch cable network, a network model can be constructed through the cable connection data at the physical level. Node connection refers to the physical or logical connection relationship between cable branch points, and the path weight is usually related to the cable length, material, or impedance.

[0113] For example, in a power distribution network, assume there are 4 branch nodes A, B, C, and D. Node A is the signal injection point. The cable length from A to B is 100 meters, from A to C is 150 meters, and from C to D is 50 meters. The weight can be directly assigned according to the length, or adjusted in combination with the attenuation coefficient of the cable material.

[0114] It should be noted that this weight design facilitates the subsequent algorithm to accurately identify the preferred path of signal propagation.

[0115] In a possible implementation manner, the shortest path algorithm in graph theory (such as Dijkstra algorithm) is used to calculate the signal propagation trajectory. The signal is emitted from the starting node A, and the optimal paths to B, C, and D need to be determined.

[0116] Preferably, the algorithm will preferentially select the path with the smallest total weight based on the path weights.

[0117] For example, the path from A to D may directly pass through C (A→C→D, total length 200 meters), rather than passing through B. Such trajectory calculation can clearly reflect the physical route of signal propagation.

[0118] Specifically, the time delay is calculated in combination with the preset propagation speed. Assume that the propagation speed of the signal in the cable is 2×10^8 m / s, the path length from A to C is 150 meters, and the propagation time is 150 / (2×10^8) s, that is, 0.75 microseconds.

[0119] It is understandable that the calculation of time delay provides a time reference for subsequent reflection analysis.

[0120] For example, the path from A to D (200 meters) corresponds to a 1-microsecond delay. If the preset threshold is 0.8 microseconds, the delay from A to D exceeds the standard and further analysis is required.

[0121] In one embodiment, if the time delay exceeds the standard, the reflection intensity is calculated based on signal attenuation and reflection coefficient. A linear regression algorithm can be used to fit the relationship between reflection intensity and delay, attenuation.

[0122] For example, assuming that the signal attenuation of the path from A to D is 3 dB and the reflection coefficient is 0.4, the regression model can predict that the reflection intensity at point D is at a medium level. This analysis can determine whether the reflection characteristics of the branch point affect signal integrity.

[0123] For example, when updating the path weight, the network topology can be adjusted according to the reflection intensity. Assuming that the reflection intensity at point D is high, indicating an impedance mismatch, the weight can be increased (e.g., from 200 to 250) to reflect the actual difficulty of signal propagation.

[0124] Specifically, this update can optimize the network model to make it closer to the real propagation environment.

[0125] In one possible implementation, when obtaining the characteristics of the branch point, the topology and reflection intensity can be combined.

[0126] For example, the high reflection intensity at point D may be due to the type of load it is connected to (such as a terminal device), and the reflection can be reduced by adjusting the load impedance. This characteristic analysis provides a basis for network optimization and enhances the stability of signal transmission. In step S18, according to the reflection characteristics, the injection frequency and amplitude of the pulse signal are adjusted to obtain optimized reflected waveform data, and the positioning result of the defect point in the branch cable network is obtained.

[0127] According to the network structure of the branch cable, the signal propagation path is obtained, and an initial pulse signal is generated by a time domain reflectometer to obtain first reflected waveform data; If the signal-to-noise ratio of the first reflected waveform data is lower than a preset threshold, the frequency components of the first reflected waveform data are extracted by Fourier transform, and the pulse signal frequency and amplitude are adjusted to obtain second reflected waveform data; Waveform feature extraction is performed on the second reflected waveform data by wavelet transform to obtain a waveform feature vector, and the preliminary position of the defect point is determined; According to the preliminary position, by using a support vector machine algorithm, combined with the network structure of the branch cable and the signal propagation path, the coordinates of the defect point are calculated to obtain the positioning result of the defect point.

[0128] Exemplarily, a branched network structure generally refers to a multi-path network in signal propagation, such as a branched topology in a cable network or an optical fiber communication system. When obtaining the signal propagation path, a directed graph can be generated by scanning nodes and edges through a network topology analysis tool.

[0129] For example, in the scenario of cable fault detection, assuming the network contains 10 nodes and 15 branched paths, the tool can identify the length and connection relationship of each path to form a path matrix. This matrix records the path weights from the starting point to the ending point, such as path lengths of 100 meters, 200 meters, etc., providing a basis for subsequent analysis.

[0130] In one possible implementation, the working principle of a Time Domain Reflectometry (TDR) for generating an initial pulse signal is to inject a high-frequency pulse into the network and record the reflected waveform.

[0131] For example, a TDR device emits a rectangular pulse with a width of 10 nanoseconds. The pulse propagates along the cable and reflects when encountering impedance changes (such as branch points or defects). The reflected waveform data records the time and amplitude information, reflecting the physical characteristics of the network. Assume the first reflected waveform data contains multiple peaks, respectively corresponding to the reflected signals of different branch points.

[0132] Specifically, if the signal-to-noise ratio of the first reflected waveform data is lower than a preset threshold (such as 10 dB), it indicates that the noise interference is severe and further processing is required. Fourier transform can convert the time-domain waveform into the frequency domain to extract the main frequency components.

[0133] For example, after the transformation, it is found that the signal is mainly concentrated between 1 MHz and 5 MHz, and the noise is concentrated in the high-frequency band. Based on this, the frequency of the pulse signal can be adjusted to 2 MHz and the amplitude increased by 20% to generate the second reflected waveform data. This adjustment can enhance the signal strength and reduce the influence of noise.

[0134] Preferably, wavelet transform is used to extract waveform features from the second reflected waveform data. Wavelet transform can separate the waveform details of different frequencies through multi-scale decomposition.

[0135] For example, using the Daubechies wavelet basis to decompose the waveform into 4 layers to extract low-frequency and high-frequency features to form a feature vector. Assume the vector contains information such as peak time, amplitude, and waveform width, and the location of the defect point can be preliminarily judged, such as an abnormal peak appears at a certain branch point 150 meters away from the starting point.

[0136] For example, the Support Vector Machine (SVM) algorithm is used to accurately locate defect points. The SVM calculates the defect coordinates by constructing a classification hyperplane and combining network topology and feature vectors. Assuming that the training data contains the feature vectors and positions of 100 known defect points, the SVM model can predict the coordinates of new defect points.

[0137] For example, the model outputs that the defect point is located on branch path 3, about 145 meters from node A. Combining with the verification of the network topology, it is confirmed that this position is at the cable joint, and the positioning result is reliable.

[0138] It can be understood that the above method forms a complete technical chain from path acquisition to defect location through multi-step analysis. Each step supports each other. For example, the Fourier transform optimizes the signal quality and provides a clear input for the wavelet transform; the wavelet transform extracts features and provides accurate data for SVM positioning. This progressive logic ensures the positioning accuracy, and at the same time, the method is flexible and can adapt to scenarios with different network scales and complexities.

[0139] The above only lists some preferred embodiments of the present invention, but the present invention is not limited thereto, and many improvements and transformations can be made. As long as the improvements and transformations are made based on the basic principles of the present invention, they should be regarded as falling within the protection scope of the present invention.

[0140] In summary, the present invention discloses an intelligent monitoring method for wire and cable. By injecting high-frequency pulse signals into the cable, the signal propagation characteristics are obtained, and the time domain reflectometry is used to preliminarily determine the positions of potential defect points. For multiple suspected defect points, the time domain reflection waveform is converted into a frequency domain signal, the signal attenuation degree is judged and compensated to eliminate the attenuation interference in long-distance monitoring. Combining with the cable parameters, the change of the signal propagation speed is calculated to determine the exact positions of the defect points. For a branch cable network, the network topology structure and signal injection parameters are obtained, the propagation characteristics of the signal in the branch path are analyzed, the graph theory algorithm is used to judge the reflection characteristics of the branch points, and the signal injection parameters are optimized to achieve the accurate positioning of defect points in the branch network. The present invention can effectively solve the defect location problem in complex cable networks and improve the positioning accuracy and reliability.

[0141] Refer to Figure 2 , the second embodiment of the present invention provides a schematic structural diagram of an intelligent monitoring system for wire and cable, including: The first acquisition module 201 is used to inject high-frequency pulse signals into the cable through a preset pulse generator, obtain the propagation time and reflection waveform of the high-frequency pulse signals in the cable, and obtain the propagation characteristics of the initial signals; The first determination module 202 is configured to analyze the time difference and amplitude change of the reflected waveform by using the time domain reflectometry method according to the propagation characteristics of the initial signal, and determine the preliminary position of the defect point in the cable; The first judgment module 203 is configured to, if there is at least one defect point at the preliminary position, convert the reflected waveform into a first frequency domain signal through Fourier transform, obtain frequency domain characteristics, and judge the signal attenuation information of the at least one defect point; The second acquisition module 204 is configured to perform gain adjustment on the first frequency domain signal by using a preset signal compensation algorithm according to the signal attenuation information to obtain an enhanced second frequency domain signal; The second determination module 205 is configured to calculate the change law of the signal propagation speed according to the second frequency domain signal, the geometric parameters and material characteristics of the cable, and determine the accurate position of the at least one defect point according to the change law; The third acquisition module 206 is configured to, if the accurate position is located in the branch cable network, acquire the topological structure of the branch cable network and the injection frequency and amplitude of the pulse signal, analyze the propagation time and reflection coefficient of the pulse signal in the branch cable network, and obtain the initial propagation constraint conditions of the branch cable network; The second judgment module 207 is configured to analyze the propagation trajectory of the pulse signal in the branch cable network by using a graph theory algorithm according to the initial propagation constraint conditions, and combine the propagation time and the reflection coefficient to judge the reflection characteristics of the branch cable network; The third determination module 208 is configured to adjust the injection frequency and amplitude of the pulse signal according to the reflection characteristics, acquire optimized reflected waveform data, and obtain the positioning result of the defect point in the branch cable network.

[0142] It should be noted that an intelligent monitoring system provided in an embodiment of the present invention is used to execute all the process steps of an intelligent monitoring method in the above embodiment, and the working principles and beneficial effects of the two correspond one by one, so details are not described herein again.

[0143] An embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as an intelligent monitoring program. When the processor executes the computer program, the steps in the above-mentioned various embodiments of the intelligent monitoring method are implemented, such as Figure 1 the step S11 shown. Alternatively, when the processor executes the computer program, the functions of each module / unit in the above-mentioned various device embodiments are implemented, such as the first acquisition module.

[0144] Exemplarily, the computer program may be divided into one or more modules / units, and the one or more modules / units are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program in the electronic device.

[0145] The electronic device may be a computing device such as a desktop computer, a notebook, a palm computer, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device. It may include more or fewer components than the above, or combine certain components, or different components. For example, the electronic device may further include input / output devices, network access devices, a bus, etc.

[0146] The so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the electronic device and connects various parts of the entire electronic device through various interfaces and lines.

[0147] The memory can be used to store the computer program and / or module. By running or executing the computer program and / or module stored in the memory, and invoking the data stored in the memory, the processor can implement various functions of the electronic device. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store the operating system, application programs required for at least one function (such as the sound playback function, the image playback function, etc.); the data storage area can store the data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory can include high-speed random access memory, and can also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage devices.

[0148] Among them, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-mentioned embodiment methods of the present invention, it can also be completed by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, it can implement the steps of the above-mentioned various method embodiments. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0149] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the device embodiments provided by the present invention, the connection relationships between the modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0150] The specific embodiments described above further elaborate on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. In particular, for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent monitoring method for wire and cable, characterized in that, Including: Inject a high-frequency pulse signal into the cable through a preset pulse generator, obtain the propagation time and reflection waveform of the high-frequency pulse signal in the cable, and obtain the propagation characteristics of the initial signal; According to the propagation characteristics of the initial signal, analyze the time difference and amplitude change of the reflection waveform by using the time domain reflectometry method to determine the preliminary position of the defect point in the cable; If there is at least one defect point at the preliminary position, convert the reflection waveform into a first frequency domain signal through Fourier transform, obtain the frequency domain characteristics, and judge the signal attenuation information of the at least one defect point; According to the signal attenuation information, use a preset signal compensation algorithm to adjust the gain of the first frequency domain signal to obtain an enhanced second frequency domain signal; According to the second frequency domain signal, the geometric parameters and material characteristics of the cable, calculate the variation law of the signal propagation speed, and determine the exact position of the at least one defect point according to the variation law; If the exact position is located in the branch cable network, obtain the topological structure of the branch cable network and the injection frequency and amplitude of the pulse signal, analyze the propagation time and reflection coefficient of the pulse signal in the branch cable network, and obtain the initial propagation constraint conditions of the branch cable network; According to the initial propagation constraint conditions, use the graph theory algorithm to analyze the propagation trajectory of the pulse signal in the branch cable network, and combine the propagation time and the reflection coefficient to judge the reflection characteristics of the branch cable network; According to the reflection characteristics, adjust the injection frequency and amplitude of the pulse signal, obtain the optimized reflection waveform data, and obtain the positioning result of the defect point in the branch cable network.

2. The intelligent monitoring method according to claim 1, wherein The step of analyzing the time difference and amplitude change of the reflection waveform by using the time domain reflectometry method according to the propagation characteristics of the initial signal to determine the preliminary position of the defect point in the cable includes: Obtain the emission time and reflection waveform data of the initial signal, and use a high-precision sampling device to record the signal frequency and amplitude change of the initial signal; Extract the waveform characteristics of the initial signal through Fourier transform to obtain a raw data set containing time difference and amplitude change; If the noise interference in the raw data set exceeds the preset threshold, use wavelet transform to denoise the reflection waveform of the initial signal, and pre-establish a speed model for signal propagation according to the material characteristics of the cable; Calculate the corresponding relationship between the time difference and the propagation distance according to the speed model to determine the waveform feature set after denoising; Through the waveform feature set after denoising, combine the propagation distance and the time difference, and use the linear regression algorithm to calculate the propagation distance of the defect point relative to the emission point to obtain the preliminary position coordinates.

3. The intelligent monitoring method according to claim 1 or 2, characterized in that, The step of, if there is at least one defect point at the preliminary position, converting the reflection waveform into a first frequency domain signal through Fourier transform, obtaining the frequency domain characteristics, and judging the signal attenuation information of the at least one defect point includes: Obtain the time domain data of the reflection waveform, convert the time domain data into a frequency domain signal through Fourier transform, and extract the characteristics of the signal frequency and amplitude change in the frequency domain to obtain a frequency domain feature data set; According to the frequency-domain feature dataset, for each suspected defect point, calculate the amplitude attenuation value of the frequency-domain signal, and use a preset threshold to determine whether the amplitude attenuation is abnormal to determine the signal attenuation information.

4. The intelligent monitoring method according to any one of claims 1-3, characterized in that, The calculating the variation law of the signal propagation speed according to the second frequency-domain signal, the geometric parameters and material characteristics of the cable, and determining the precise position of the at least one defect point according to the variation law includes: According to the second frequency-domain signal, query the geometric parameter database of the cable and calculate the signal propagation speed, where the signal propagation speed v is calculated by formula (1): , formula (1); Among them, represents the signal propagation speed, represents the speed of light constant, represents the permittivity; If the deviation of the propagation speed from the preset standard value exceeds the preset threshold, analyze the reflection coefficient by the time-domain reflectometry method to obtain the propagation delay time, and judge the preliminary position of the at least one defect point; Use a waveform analysis tool to extract the amplitude and phase characteristics of the reflection signal, and combine the length and material characteristics of the cable to calculate the precise coordinates of the at least one defect point.

5. The intelligent monitoring method according to any one of claims 1-4, characterized in that, If the precise position is in the branch cable network, obtain the topological structure of the branch cable network and the injection frequency and amplitude of the pulse signal, analyze the propagation time and reflection coefficient of the pulse signal in the branch cable network, and obtain the initial propagation constraint conditions of the branch cable network, including: Obtain the network physical connection data of the cable, record the starting and ending coordinates of each branch path, generate a topological graph including node coordinates and connection relationships, and determine the branch network topology of the cable; According to the branch network topology and the injection position of the pulse signal, calculate the path lengths from the injection point to the nodes of each branch path, and use the time-domain reflectometry method to measure the propagation time of the pulse signal in each branch path to obtain the propagation time of the pulse signal; Based on the cable impedance characteristics, use formula (2) to calculate the reflection amount of the pulse signal at the node through the propagation time of the pulse signal and the signal frequency of the pulse signal to determine the signal reflection coefficient; R = (Z2 - Z1) / (Z2 + Z1), formula (2); where Z1 is the incident path impedance and Z2 is the reflected path impedance; According to the reflection coefficient of the pulse signal and the branch network topology, combined with the length distribution of the branch path, use the support vector machine algorithm to optimize the propagation constraint parameters, generate the constraint conditions including time delay and reflection intensity, and obtain the initial propagation constraint conditions.

6. The intelligent monitoring method according to any one of claims 1-5, characterized in that The analyzing the propagation trajectory of the pulse signal in the branch cable network according to the initial propagation constraint conditions, and combining the propagation time and the reflection coefficient to judge the reflection characteristics of the branch cable network includes: Obtain the node connections and path weights from the branch cable network, and use the shortest path algorithm in the graph theory algorithm to calculate the propagation trajectory of the pulse signal from the starting node to each branch point to obtain the propagation trajectory of the pulse signal; According to the propagation trajectory of the pulse signal, combined with the preset propagation speed, calculate the propagation time of the pulse signal at each branch point to determine the time delay; If the time delay is greater than a preset threshold, then according to the signal attenuation and reflection coefficient, a linear regression algorithm is used to calculate the reflection intensity of each branch point to determine the reflection characteristics of the branch cable network; Based on the network topology and the node connections, the characteristics of each branch point are obtained. Combining the reflection intensity of each branch point, the path weights are updated to obtain the reflection characteristics at each branch point.

7. The intelligent monitoring method according to any one of claims 1-6, characterized in that, According to the reflection characteristics, adjusting the injection frequency and amplitude of the pulse signal to obtain optimized reflected waveform data, and obtaining the positioning result of the defect point in the branch cable network, including: According to the network structure of the branch cable, the signal propagation path is obtained, and an initial pulse signal is generated by a time domain reflectometer to obtain the first reflected waveform data; If the signal-to-noise ratio of the first reflected waveform data is lower than a preset threshold, then the frequency components of the first reflected waveform data are extracted by Fourier transform, and the pulse signal frequency and amplitude are adjusted to obtain the second reflected waveform data; Wavelet transform is used on the second reflected waveform data to extract waveform feature vectors, and the preliminary position of the defect point is determined; According to the preliminary position, through the support vector machine algorithm, combining the network structure of the branch cable and the signal propagation path, the coordinates of the defect point are calculated to obtain the positioning result of the defect point.

8. An intelligent monitoring system for wire and cable, characterized in that, Including: A first acquisition module, configured to inject a high-frequency pulse signal into the cable through a preset pulse generator, obtain the propagation time and reflected waveform of the high-frequency pulse signal in the cable, and obtain the propagation characteristics of the initial signal; A first determination module, configured to analyze the time difference and amplitude change of the reflected waveform by using the time domain reflection method according to the propagation characteristics of the initial signal, and determine the preliminary position of the defect point in the cable; A first judgment module, configured to, if there is at least one defect point at the preliminary position, convert the reflected waveform into a first frequency domain signal by Fourier transform, obtain the frequency domain characteristics, and judge the signal attenuation information of the at least one defect point; A second acquisition module, configured to perform gain adjustment on the first frequency domain signal by using a preset signal compensation algorithm according to the signal attenuation information to obtain an enhanced second frequency domain signal; A second determination module, configured to calculate the variation law of the signal propagation speed according to the second frequency domain signal, the geometric parameters and material characteristics of the cable, and determine the precise position of the at least one defect point according to the variation law; A third acquisition module, configured to, if the precise position is located in the branch cable network, obtain the topological structure of the branch cable network and the injection frequency and amplitude of the pulse signal, analyze the propagation time and reflection coefficient of the pulse signal in the branch cable network, and obtain the initial propagation constraint conditions of the branch cable network; A second judgment module, configured to analyze the propagation trajectory of the pulse signal in the branch cable network by using a graph theory algorithm according to the initial propagation constraint conditions, and combine the propagation time and the reflection coefficient to judge the reflection characteristics of the branch cable network; A third determination module, configured to adjust the injection frequency and amplitude of the pulse signal according to the reflection characteristics, obtain optimized reflected waveform data, and obtain a positioning result of a defect point in the branch cable network.

9. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the intelligent monitoring method described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the intelligent monitoring method described in any one of claims 1 to 7.

Images