A method, apparatus, and medium for multi-channel leaky coaxial cable fault location

Abstract

The application discloses a kind of multi-channel leak coaxial cable fault location method, equipment and medium, it is related to health management technical field, including, the reference signal and detection signal of leak coaxial cable are collected, time delay compensation and time synchronization are carried out to reference signal and detection signal, and output after synchronization reference signal and detection signal;Synchronization is carried out to the detection signal and reference signal using phase error compensation function and phase compensation is carried out, and output after compensation detection signal and reference signal;The aging state of leak coaxial cable is comprehensively analyzed by the fault point positioning result after fusion, generates predictive maintenance suggestion, and the health index evaluation of the running state of leak coaxial cable is carried out, and output health state report.The application carries out time sequence arrangement and section mapping to the fault point positioning result after fusion, reduces shutdown risk and maintenance cost.

Description

Technical Field

[0001] This invention relates to the field of health management technology, and in particular to a method, device and medium for locating faults in multi-channel leaky coaxial cables. Background Technology

[0002] With the continuous development of communication technology, rail transit, industrial automation and security monitoring, leaky coaxial cables are widely used in wireless signal transmission systems in complex environments such as tunnels, subways, cable corridors and substations due to their advantages such as stable radiation characteristics along the line, controllable coverage and strong anti-interference ability. High-precision and real-time fault location and health assessment of the operating status of leaky coaxial cables has become an important research direction for ensuring the stable operation of communication systems and improving operation and maintenance efficiency.

[0003] Existing technologies still have the following shortcomings: they cannot solve the positioning deviation caused by the inconsistency of time references between different channels and the uneven distribution of noise. Traditional methods mainly focus on single-point detection of fault points and lack the ability to systematically analyze the overall health status of leaky coaxial cables as long-line facilities. Due to the lack of a joint compensation mechanism for phase error and time delay error, the reflection feature extraction process still suffers from drift and noise interference. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a fault location method for multi-channel leaky coaxial cables to solve the location deviation problem caused by inconsistent time references and uneven noise distribution between different channels.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for locating faults in a multi-channel leaky coaxial cable, comprising, The system acquires reference and detection signals from the leaky coaxial cable, performs time delay compensation and time synchronization on the reference and detection signals, and outputs the synchronized reference and detection signals. A phase error compensation function is used to perform phase compensation on the synchronized detection signal and reference signal, and the compensated detection signal and reference signal are output. The compensated detection signal and the reference signal are searched for correlation peaks to generate a time delay difference. Time domain correlation analysis is performed on the time delay difference to extract the signal reflection feature set. The reflected wave time delay localization analysis is performed on the signal reflection feature set to output the fault point localization information of each channel. The fault location information of each channel is weighted and fused to generate a fused location feature set. The fused location feature set is then corrected for error by weighted least squares fitting to generate the fused fault location result. The aging status of leaky coaxial cables is comprehensively analyzed by integrating the fault location results, predictive maintenance suggestions are generated, and the operating status of leaky coaxial cables is assessed by health index, outputting a health status report.

[0007] In a preferred embodiment of the multi-channel leaky coaxial cable fault location method of the present invention, the specific steps for outputting the synchronized reference signal and detection signal are as follows: Acquire reference and detection signals from the leaky coaxial cable, perform channel framing and timestamp alignment on the reference and detection signals, and generate an aligned frame set; Perform fixed inter-channel delay compensation and sampling jitter correction on the aligned frame set to generate a synchronized frame set; Amplitude normalization and DC bias elimination are performed on the synchronization frame set to generate synchronized reference and detection signals.

[0008] In a preferred embodiment of the multi-channel leaky coaxial cable fault location method of the present invention, the specific steps for outputting the compensated detection signal and reference signal are as follows: The phase features of the synchronized reference signal and the detection signal are extracted by phase analysis, the deviation between the phase features is calculated, and a phase deviation sequence is generated. The phase error compensation parameter set is generated by fitting the phase deviation sequence to a function and compensating the coefficients through the phase error compensation function. The phase of the synchronized reference signal and the detection signal is corrected using a set of phase error compensation parameters to generate compensated reference signal and detection signal.

[0009] In a preferred embodiment of the multi-channel leaky coaxial cable fault location method of the present invention, the step of performing correlation peak search on the compensated detection signal and the reference signal to generate a time delay difference includes the following specific steps. The compensated reference signal and the detection signal are truncated using a window function and normalized by energy to generate an analysis signal segment. Peak candidate filtering is performed on the analyzed signal segments to generate a peak candidate set; Fine-grained peak position interpolation and confidence evaluation are performed on the peak candidate set to generate delay difference.

[0010] In a preferred embodiment of the multi-channel leaky coaxial cable fault location method of the present invention, the specific steps for extracting the signal reflection feature set are as follows: Multi-scale time window splicing and outlier removal are performed on the time delay difference to generate a purified time delay difference sequence; A time-domain correlation analysis was performed on the purification delay difference sequence, and the reflection arrival point and polarity change point were extracted. Amplitude mapping is performed between the reflection arrival point and the polarity change point to generate a signal reflection feature set.

[0011] In a preferred embodiment of the multi-channel leaky coaxial cable fault location method of the present invention, the specific steps for outputting the fault location information of each channel are as follows: Perform reflected wave time delay localization analysis on the signal reflection feature set to generate a single-channel distance dataset; Perform connected segment merging and endpoint correction on the single-channel distance dataset, and output the fault location information for each channel.

[0012] In a preferred embodiment of the multi-channel leaky coaxial cable fault location method of the present invention, the specific steps for generating the fused fault point location result are as follows: The fault location information of each channel is weighted by the channel signal-to-noise ratio and the consistency coefficient, and a weighted coefficient set is generated. The fault location information of each channel is weighted and fused using a set of weighted coefficients to generate a fused location feature set; The weighted least squares fitting method is used to perform residual calculation and bias compensation on the fused positioning feature set to generate the fused positioning correction set. The weights of the fused localization correction set are adjusted and convergence is determined to generate the fused fault location result.

[0013] In a preferred embodiment of the multi-channel leaky coaxial cable fault location method of the present invention, the specific steps for generating predictive maintenance suggestions are as follows: The merged fault location results are processed by time series analysis and segment mapping to generate a fault segment sequence. Obtain inspection records and operation files of leaky coaxial cables, perform aging factor statistics on fault section sequences, inspection records and operation files, and generate an aging feature set; Multi-factor evaluation and trend analysis of aging feature sets are performed to generate predictive maintenance recommendations.

[0014] In a second aspect, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the multi-channel leaky coaxial cable fault location method as described in the first aspect of the present invention.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the multi-channel leaky coaxial cable fault location method as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: by performing channel framing and timestamp alignment on the reference signal and the detection signal, the time base of multi-channel signals is unified; by performing function fitting and coefficient compensation on the phase deviation sequence, the influence of random noise is effectively smoothed; by performing window function truncation and energy normalization, the interference of irrelevant segments on the analysis is reduced; by assigning weights to the channel signal-to-noise ratio and consistency coefficient, the quality quantification of different channels is achieved, thereby improving the reliability of fusion from the source; and by performing time series processing and segment mapping on the fused fault point location results, the downtime risk and maintenance cost are reduced. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart for a method to locate faults in multi-channel leaky coaxial cables.

[0019] Figure 2 This is a flowchart illustrating the generation of the synchronized reference signal and detection signal.

[0020] Figure 3 This is a flowchart of the phase error compensation process.

[0021] Figure 4 A flowchart for generating a time delay difference sequence for purification. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1-4As one embodiment of the present invention, this embodiment provides a method for locating faults in a multi-channel leaky coaxial cable, comprising the following steps: S1: Acquire reference and detection signals from the leaky coaxial cable, perform time delay compensation and time synchronization on the reference and detection signals, and output the synchronized reference and detection signals.

[0026] S1.1: Obtain the reference signal and detection signal of the leaky coaxial cable, perform channel framing and timestamp alignment on the reference signal and detection signal, and generate an aligned frame set.

[0027] The process involves acquiring reference and detection signals from a leaky coaxial cable. These signals are then input into sampling channels for raw signal acquisition, resulting in a raw signal data sequence containing continuous sampling points. This raw signal data sequence is then processed by channel framing. Specifically, the continuous sampling points are divided into multiple frame segments according to a fixed sampling time window. Each frame segment corresponds to a continuous sampling time interval. The fixed sampling time window limits the duration of each sampled signal. For example, when the signal sampling frequency is 1MHz, the fixed sampling time window is set to 1ms. In this case, each frame segment contains 1000 sampling points, ensuring that the reference and detection signals are aligned on the same time scale. Time stamp information is added to the beginning of each frame segment to obtain reference signal frame segments and detection signal frame segments. The time deviation between the reference and detection signal frame segments is calculated using the time stamp information, and the time deviation is corrected. The reference and detection signal frame segments are then time-aligned according to a unified time reference, ensuring that the corresponding sampling point positions of the reference and detection signals on the time axis are consistent. Finally, the time-aligned reference and detection signal frame segments are recombined in chronological order to generate an aligned frame set.

[0028] S1.2: Perform fixed delay compensation and sampling jitter correction between channels on the aligned frame set to generate a synchronized frame set.

[0029] The start times of the reference signal frame segments and detection signal frames for each channel in the alignment frame set are obtained. The time difference between the start times of the reference signal frame segments and the detection signal frame segments is calculated. This time difference is used as a fixed delay parameter between channels. The fixed delay parameter between channels is used to correct the time offset of the reference signal frame segments and detection signal frames for each channel in the alignment frame set. Specifically, the reference signal frame segments and detection signal frames are translated along the time axis to keep them synchronized on the time reference, thus completing the fixed delay compensation between channels. Sampling jitter correction is then performed on the reference signal frame segments and detection signal frames that have undergone fixed delay compensation between channels. Specifically, the instantaneous sampling interval between adjacent sampling points is calculated, the sampling time offset is detected and corrected, and the sampling points are redistributed to a uniform sampling interval position to obtain signal frame segments with stable sampling time. The reference signal frame segments and detection signal frames after fixed delay compensation and sampling jitter correction between channels are then recombined according to the channel number and time order to generate a synchronization frame set.

[0030] S1.3: Perform amplitude normalization and DC bias elimination on the synchronization frame set to generate synchronized reference signal and detection signal.

[0031] The amplitude value sequences of reference signal frames and detection signal frames for each channel in the synchronization frame set are obtained. Amplitude normalization processing is performed on the amplitude value sequences. The maximum and minimum amplitude values ​​of the reference signal frames and detection signal frames for each channel are calculated. An amplitude normalization coefficient is set based on the maximum and minimum amplitude values, usually in the range of 0-1. As the signal amplitude range increases, the amplitude normalization coefficient gradually approaches 0, and as the signal amplitude range decreases, the amplitude normalization coefficient gradually approaches 1. Using a value range of 0-1 can limit the amplitude distribution of the reference signal frames and detection signal frames to the effective range (usually 0-1), realize the amplitude standardization processing of signals from different channels, eliminate equipment gain differences, and suppress noise amplification effects.

[0032] An amplitude normalization coefficient is used to scale the amplitude values ​​of the reference signal frame and the detection signal frame proportionally. This compresses the amplitude value of each sampling point into the effective range according to a uniform scaling factor, thus limiting the amplitude range of the reference signal frame and the detection signal frame to a unified amplitude range. This eliminates amplitude differences between different channels, resulting in amplitude-normalized reference signal frame and detection signal frame. For example, when the amplitude normalization coefficient is 0.05, the amplitude of all sampling points will be scaled by a factor of 0.05, thereby maintaining the relative amplitude relationship between sampling points. The process involves performing DC bias cancellation on the amplitude-normalized reference signal frame segment and the detection signal frame segment. Specifically, by calculating the average amplitude value of each frame segment, the DC bias of the reference signal frame segment and the detection signal frame segment is detected, and the DC bias is canceled out from the amplitude value, so that the amplitude distribution of the reference signal frame segment and the detection signal frame segment is centered at zero, resulting in a signal frame segment without DC bias. The reference signal frame segment and the detection signal frame segment after amplitude normalization and DC bias cancellation are then recombined according to the channel number and time order to generate synchronized reference signal and detection signal.

[0033] S2: Use a phase error compensation function to perform phase compensation on the synchronized detection signal and reference signal, and output the compensated detection signal and reference signal.

[0034] S2.1: The phase features of the synchronized reference signal and the detection signal are extracted by phase analysis, the deviation between the phase features is calculated, and a phase deviation sequence is generated.

[0035] The instantaneous phase of the synchronized reference signal and the detection signal is calculated separately. Specifically, the analytic signal forms of the synchronized reference signal and the detection signal are obtained through Hilbert transform. The instantaneous phase is calculated using the imaginary and real parts of the analytic signal to obtain the instantaneous phase sequence of the reference signal and the instantaneous phase sequence of the detection signal. Phase expansion processing is performed on the instantaneous phase sequence of the reference signal and the instantaneous phase sequence of the detection signal to make the phase jumps in the range of [-π, π] continuous, eliminating the phase discontinuity problem caused by the periodic phase jumps, and obtaining a continuous phase sequence. Phase feature extraction processing is performed on the continuous phase sequence to extract phase feature parameters such as phase change rate, phase envelope, and phase extreme points. The phase feature parameters are arranged according to the sampling time order, and the time correspondence of the feature parameters is established according to the time index of the synchronized reference signal and the detection signal to generate the phase feature sequence of the reference signal and the phase feature sequence of the detection signal. The phase difference is calculated one-to-one between the phase feature sequence of the reference signal and the phase feature sequence of the detection signal to obtain the phase deviation value corresponding to each sampling point. The phase deviation values ​​are arranged in time order to generate the phase deviation sequence.

[0036] It should be noted that the instantaneous phase is usually obtained through the Hilbert transform and the arctangent function, and the principal value range of the arctangent function is limited to [-π, π]. Therefore, the representation of the instantaneous phase is also limited to the interval [-π, π].

[0037] S2.2: The phase deviation sequence is fitted and the coefficients are compensated by the phase error compensation function to generate a phase error compensation parameter set.

[0038] The phase deviation sequence is preprocessed by smoothing. Specifically, high-frequency noise components are eliminated using polynomial filtering to obtain a smoothed phase deviation sequence. The smoothed phase deviation sequence is then fitted with a phase error compensation function, which is a polynomial fitting function that describes the trend of phase deviation change. The theoretical phase deviation value of each sampling point is calculated using the polynomial fitting function and compared with the actual phase deviation value in the smoothed phase deviation sequence. The error value is used as the residual value, and the coefficients of the fitting function are corrected using the residual value to generate the phase error compensation function parameters after coefficient compensation. These parameters are then organized according to the sampling time sequence to form a phase error compensation parameter set.

[0039] The theoretical phase deviation value of the sampling point is expressed as: ; in, For the first The theoretical phase deviation value of each sampling point For the sampling point index, to avoid excessively large values ​​caused by high-order terms, it can be... Normalize based on the total number of sampling points, so that The value falls within the interval [0, 1]. For the polynomial fitting function in The fitting coefficients of order, Let be the order of the polynomial fitting function, and be the summation variable, ranging from 0 to . , This is the highest order of the polynomial fitting function, determined by the required fitting accuracy. For example... and It can describe the trend of instantaneous phase change.

[0040] S2.3: Use the phase error compensation parameter set to perform phase correction on the synchronized reference signal and detection signal to generate the compensated reference signal and detection signal.

[0041] The compensation parameter values ​​for each sampling point in the phase error compensation parameter set are obtained and matched with the sampling time indices of the synchronized reference signal and detection signal. Phase correction calculations are then performed on the synchronized reference signal and detection signal respectively. Specifically, the instantaneous phase of the synchronized reference signal and detection signal is corrected using the compensation parameter values ​​in the phase error compensation parameter set. The instantaneous phase value of each sampling point is superimposed with the corresponding compensation parameter value to obtain the phase-corrected instantaneous phase sequence. Phase reconstruction processing is then performed on the phase-corrected instantaneous phase sequence. Specifically, the phase reconstruction method is used to restore the corrected instantaneous phase sequence to the complex form of the time-domain signal, and the real part of the complex form is extracted as the phase-corrected reference signal waveform and detection signal waveform. The phase-corrected reference signal waveform and the phase-corrected detection signal waveform are then arranged into continuous time series to generate the compensated reference signal and detection signal.

[0042] S3: Perform correlation peak search on the compensated detection signal and the reference signal to generate time delay difference, perform time domain correlation analysis on the time delay difference, extract the signal reflection feature set, perform reflected wave time delay localization analysis on the signal reflection feature set, and output the fault point location information of each channel.

[0043] S3.1: Perform window function truncation and energy normalization on the compensated reference signal and detection signal to generate analysis signal segments. Then, perform peak candidate screening on the analysis signal segments to generate a peak candidate set.

[0044] The continuous time series of the compensated reference signal and the detected signal are obtained as time-domain waveform data. The continuous time series originates from the continuous sampling results of the compensated reference signal and the detected signal on the time axis, and is concatenated according to the time index order. The time-domain waveform data is divided into frames using the signal sampling frequency and a fixed sampling time window, generating the number of sampling points within each analysis window. The analysis window limits the time range of signal interception, and the fixed sampling time window means that the time length within each analysis window remains constant. For example, if the sampling time window is set to 1ms and the signal sampling frequency is 1kHz, then each analysis window contains 1000 sampling points. The number of sampling points is determined by multiplying the sampling frequency by the time window length. The window function length is determined based on the number of sampling points in each analysis window, and the window function type is selected as the Hanning window. The Hanning window can effectively reduce spectral leakage and maintain the main lobe energy concentration higher than that of the rectangular window during signal truncation. By setting the window overlap ratio, the continuous coverage characteristics of adjacent analysis windows on the time axis are ensured, and a window function parameter set is generated. The window overlap ratio is usually set by the application scenario. For example, if the window overlap ratio is set to 50%, then 50% of the sampling points of adjacent windows will overlap. 50% is a common setting value, which can provide sufficient time resolution while ensuring a smooth transition and avoiding spectral leakage caused by excessively small windows.

[0045] The time-domain waveform data is framed and windowed using a window function parameter set. The continuous time series is framed according to the analysis window length, resulting in reference and detection signal segments truncated by the window function. The difference between the reference and detection signal segments is used as the amplitude value. The amplitude values ​​are squared and summed to generate the total energy of the signal segment. The total energy of the signal segment represents the overall energy level of the signal segment within the current analysis window. The amplitude ratio of the reference and detection signal segments is adjusted based on the total energy of the signal segments to generate energy-normalized reference and detection signal segments. Energy normalization ensures that the signal energy scale is consistent across different channels and time windows, thereby guaranteeing the comparability of subsequent peak detection. The energy-normalized reference and detection signal segments are paired one-to-one according to the time index to generate the analysis signal segment.

[0046] Local maxima detection is performed on continuous time series by analyzing the amplitude change rate of signal segments to generate a preliminary peak set. An amplitude statistical threshold is set based on the amplitude distribution characteristics of the reference signal segment and the detected signal segment within the analysis window, with the value ranging from one to three times the average amplitude standard deviation. The optimal compromise range under the noise statistical distribution theory is also set, with the value range being between one and three times the average amplitude standard deviation, which will not miss weak effective reflection signals and can effectively remove background noise. The preliminary peak set is then filtered by the amplitude statistical threshold, retaining the maxima in the preliminary peak set whose amplitude is greater than the amplitude statistical threshold, thus generating a peak candidate set.

[0047] S3.2: Perform fine-grained peak position interpolation and confidence evaluation on the peak candidate set to generate time delay difference. Perform multi-scale time window splicing and outlier removal on the time delay difference to generate a purified time delay difference sequence.

[0048] Peak positions are processed using amplitude and time indices from the peak candidate set. Specifically, linear interpolation is performed on the peak positions using the amplitude variation trend between adjacent sampling points to generate a preliminary interpolated peak position sequence. A second interpolation process is then performed on this preliminary sequence, and parabolic interpolation is used to fit the local peak shape within the peak neighborhood, refining the peak position accuracy to the sub-sampling level of the sampling interval, generating a fine-grained peak position sequence. The peak stability is then assessed based on the amplitude characteristics of the fine-grained peak position sequence. Specifically, the amplitude distribution of the local signal in the peak neighborhood is used to perform amplitude stratification, and the local signal-to-noise ratio (SNR) of each interpolated peak position is obtained using the ratio of the average amplitude at the peak point to the neighborhood noise amplitude. The amplitude variation trend on both sides of the peak point is used to assess the peak shape. The amplitude symmetry of each interpolated peak is obtained by comparing the right half-width and using the difference between the left and right half-widths. The peak curvature is calculated by the amplitude change rate in the peak neighborhood. The neighborhood curvature value of each interpolated peak is obtained by using the ratio of the amplitude increment of adjacent sampling points to the sampling interval. The local signal-to-noise ratio, amplitude symmetry and neighborhood curvature value of each interpolated peak are weighted and summed to obtain the peak confidence coefficient. The peak confidence coefficient is normalized and limited to the interval [0,1]. The normalized peak confidence coefficient is matched with the fine-grained peak sequence according to the time index to generate a peak sequence with confidence label. The peak time index of the reference signal and the detection signal is used to pair the peak sequence with confidence label in time. The time index difference is used to generate the time delay difference.

[0049] Multi-scale time window splicing and outlier removal are performed on the time delay difference. Specifically, adjacent time delay differences are averaged and spliced ​​within a short time window to suppress instantaneous fluctuations, generating a short-scale splicing result. Within a long time window, the short-scale splicing result is smoothed and spliced ​​to keep the time delay change continuous over a larger time interval, generating a multi-scale spliced ​​sequence. By statistically analyzing the time delay difference distribution characteristics of the multi-scale spliced ​​sequence, outliers are removed from the time delay difference distribution characteristics of the multi-scale spliced ​​sequence using a three-standard-deviation outlier detection principle, generating a cleaned time delay difference sequence.

[0050] S3.3: Perform time-domain correlation analysis on the purification delay difference sequence, extract the reflection arrival point and polarity change point, perform amplitude mapping on the reflection arrival point and polarity change point, and generate a signal reflection feature set.

[0051] The time-domain correlation method is used to perform correlation operations on the cleaned delay difference sequence. The difference value of adjacent sampling points is calculated within the continuous sampling time range, and all difference values ​​are arranged in the order of sampling point index to generate a time-domain correlation result sequence. The time-domain correlation result sequence is used to reflect the similarity of the signal delay difference change over time. The signal reflection event is located by the waveform change trend of the time-domain correlation result sequence. Specifically, local peak points in the correlation result sequence are detected and regarded as reflection arrival points. The zero-crossing feature of the waveform is used to determine the signal polarity change. Specifically, by detecting the sign change between adjacent sampling points in the time-domain correlation result sequence, sampling points where the sign changes from positive to negative are marked as positive polarity reversal points, and sampling points where the sign changes from negative to positive are marked as reverse polarity reversal points.

[0052] By verifying the amplitude change trend before and after the polarity reversal point, a polarity change judgment threshold was set based on the local amplitude change characteristics of the purification delay difference sequence, with a value range of 5% to 15% of the amplitude statistical threshold. The value range was also set based on the zero-crossing smoothness and noise jump statistical characteristics. Using a value range of 5% to 15% of the amplitude statistical threshold can improve the accuracy of polarity reversal judgment and reduce the noise false positive rate. When the amplitude difference between adjacent sampling points is less than the polarity change judgment threshold, and the sign change direction is consistent with the waveform continuity, the sampling point where the sign change occurs is determined as the valid polarity change point. This is further verified by comparing the valid polarity change point with adjacent local peaks. The time index of the value point is compared, and the previous local peak point corresponding to the polarity change point is determined as the reflection arrival point. The reflection arrival point sequence and the polarity change point sequence are output. The signal amplitude distribution is mapped by the reflection arrival point sequence and the polarity change point sequence. Specifically, the amplitude of the cleanup delay difference sequence is read at the time index corresponding to the reflection arrival point and the polarity change point, and the amplitude is mapped to the time axis coordinate to generate an amplitude mapping sequence containing time, amplitude and polarity information. The arrival time, amplitude intensity and polarity change direction of each reflection event are matched one by one by the amplitude mapping sequence to generate a signal reflection feature set.

[0053] The expression for the difference between adjacent sampling points is: ; in, The difference between adjacent sampling points. Sampling point index, To clean up the time delay difference sequence at the sampling point The value, To clean up the time delay difference sequence at the sampling point The value of the previous sampling point.

[0054] S3.4: Perform reflected wave time delay localization analysis on the signal reflection feature set, generate a single-channel distance dataset, merge connected segments and correct endpoints on the single-channel distance dataset, and output the fault point location information for each channel.

[0055] The propagation path of the reflected wave is converted into a time delay by using the time index of the reflection arrival point in the signal reflection feature set and the signal propagation speed. The distance corresponding to the reflection arrival point is generated. All the distances corresponding to the reflection arrival points are arranged in index order to generate a distance value sequence. By sequentially arranging the distance value sequence and performing adjacent value difference analysis, duplicate reflections and stray reflection points are eliminated to generate a continuous reflection distance sequence. A proportional threshold is set according to the distribution characteristics of the distance change rate between adjacent reflection distance points, usually ranging from 1.2 to 1.8. The value range is set according to the principle of signal continuity and noise distribution characteristics. Using a value range of 1.2 to 1.8 can ensure the continuity of the reflected wave while effectively eliminating abnormal reflection points, thereby improving the stability of the reflection wave time delay localization analysis and the accuracy of fault point location. Discrete reflection points in the continuous reflection distance sequence whose distance change rate between adjacent sampling points exceeds the proportional threshold are removed to generate a time delay localization analysis result sequence. The reflection features of each channel are aggregated using the time delay localization analysis result sequence and organized according to the channel number to generate a single-channel distance dataset.

[0056] By merging connected segments in continuously distributed reflection distance regions using a single-channel distance dataset, specifically, a merging threshold is set based on the distance interval distribution characteristics of adjacent reflection events in the single-channel distance dataset. Typically, this threshold ranges from 1.5 to 3 times the average distance interval. This threshold is set based on the principle of spatial continuity of reflected waves. Using a value range of 1.5 to 3 times the average distance interval ensures the continuity of reflection segments while effectively distinguishing independent reflection events, improving the accuracy and stability of fault segment identification. The difference between adjacent reflection points in the single-channel distance dataset is used as the distance interval between adjacent reflection points, and this distance interval is then compared with... The merging threshold is compared. When the distance between adjacent reflection points is less than the merging threshold, the corresponding reflection points are assigned to the same continuous reflection segment. Reflection events belonging to the same reflection segment are processed by connected segment merging to generate a set of connected segments. Endpoint correction is performed on the start and end positions of each connected segment using the set of connected segments. Specifically, based on the amplitude change trend of the first and last reflection points in the connected segment, the time index of the reflection point with the largest amplitude is used to correct the endpoint position of the connected segment, generating the distance result after endpoint correction. The reflection events of each channel are organized and numbered using the distance result after endpoint correction, and the fault location information of each channel is output.

[0057] The distance expression corresponding to the reflection point is: ; in, The distance corresponding to the point of reflection. For signal propagation speed, For the time index of the reflection arrival point, This represents the sampling time interval.

[0058] S4: The fault location information of each channel is weighted and fused to generate a fused location feature set. The fused location feature set is then corrected for errors using a weighted least squares fitting method to generate the fused fault location result.

[0059] S4.1: The fault location information of each channel is weighted by the channel signal-to-noise ratio and the consistency coefficient to generate a weighted coefficient set.

[0060] The channel signal-to-noise ratio (SNR) is calculated by using the mean amplitude of the signal and the mean amplitude of the noise in each channel from the fault location information. The ratio of channel signal energy to background noise energy reflects the channel signal quality, and the channel SNR is generated. The channel stability is evaluated by the spatial distribution consistency of fault points in each channel from the fault location information. Specifically, the spatial deviation of fault points at the same location in different channels is statistically analyzed, and the standard deviation and mean deviation of the fault point location in each channel are calculated. The standard deviation and mean deviation values ​​are then normalized by reciprocal to generate the channel consistency coefficient. The importance of the channel is evaluated by weighting the channel SNR and the channel consistency coefficient. Specifically, the channel SNR and the channel consistency coefficient are weighted and averaged according to a proportional coefficient. The proportional coefficient is set in the range of 0.4 to 0.6 based on experimental statistics to ensure a balance between the weights of signal quality and spatial consistency. The weighted average result is used to assign corresponding weight values ​​to each channel, and the weight values ​​are sorted by channel number to generate a set of weighted coefficients.

[0061] S4.2: Use the weighted coefficient set to perform weighted fusion of the fault location information of each channel to generate a fused location feature set.

[0062] The weight values ​​are extracted from the weighted coefficient set according to the channel number. The weight value of each channel is matched with the corresponding fault point location information to establish a one-to-one correspondence between the weight value and the channel location information, generating a set of channel information to be fused. The fault point coordinate values ​​of each channel in the set of channel information to be fused are weighted and calculated. Specifically, the fault point coordinate values ​​are weighted and summed with the corresponding weight values, and normalization is performed to generate a preliminary fused coordinate sequence. The spatial continuity is smoothed and corrected using the preliminary fused coordinate sequence to generate a spatially calibrated fused coordinate sequence. The time series of fault point location information is uniformly aligned using the spatially calibrated fused coordinate sequence to generate a time-synchronized fused positioning result. The fault point location information of each channel is integrated using the time-synchronized fused positioning result to output a fused positioning feature set containing location, amplitude, and time information.

[0063] S4.3: The weighted least squares fitting method is used to perform residual calculation and bias compensation on the fused positioning feature set to generate the fused positioning correction set.

[0064] By matching the fused coordinate values ​​in the fused positioning feature set with the original channel fault point positioning information, a residual input sequence is generated. A weighted least squares fitting method is then used to fit the residual input sequence. Specifically, the residual input sequence is weighted according to the weight values ​​of each channel in the weighted coefficient set, so that the channels with weight values ​​have a corrective influence during the fitting process, generating a weighted residual sequence. The spatial deviation of the fused positioning feature set is estimated using the weighted residual sequence. Specifically, the average deviation and standard deviation of the weighted residual sequence are calculated to generate a fusion deviation estimation result. The fusion deviation estimation result is used to perform deviation compensation on the fused coordinate values ​​in the fused positioning feature set, correcting each fused coordinate value according to the deviation direction. The correction magnitude is proportional to the average deviation value in the weighted residual sequence, generating a deviation-compensated positioning coordinate sequence. The positioning coordinate sequence is then organized according to time index and channel number to generate a fused positioning correction set.

[0065] S4.4: Adjust the weights and determine convergence of the fused localization correction set to generate the fused fault location result.

[0066] The positioning deviation values ​​of each channel in the fused positioning correction set are matched with the weight values ​​in the weighted coefficient set. The reciprocal of the residual average is used as the weight adjustment factor to generate a weight adjustment sequence. The weighted coefficient set is iteratively updated using the weight adjustment sequence. Specifically, the product of the weight values ​​and the weight adjustment factor in the weighted coefficient set is normalized to generate a weight correction set updated in one iteration. The fused positioning correction set is then re-weighted using the weight correction set, and the difference between the weighted average positioning coordinates and the average positioning coordinates of the previous iteration is calculated to generate an iterative deviation sequence. The convergence state is determined using the iterative deviation sequence, specifically based on the fused positioning iteration process. The statistical distribution characteristics of the changes in positioning coordinates are used to set a convergence threshold, which is usually set in the range of 0.1%-0.5% of the average distance difference. The range is set by the principles of iterative stability and convergence. The range of 0.1%-0.5% of the average distance difference can achieve stable convergence in a few iterations. While ensuring positioning accuracy, it reduces the amount of computation and improves the stability and real-time performance of fused positioning. When the maximum deviation value in the iterative deviation sequence is lower than the convergence threshold, the iterative process is considered to have reached stability, and the fused fault point positioning result is output. When the maximum deviation value is higher than the convergence threshold, the weight update and reweighting operation continue to be performed until the convergence condition is reached, that is, the maximum deviation value is lower than the convergence threshold.

[0067] S5: Based on the integrated fault location results, a comprehensive analysis of the aging status of the leaky coaxial cable is performed to generate predictive maintenance recommendations. The operating status of the leaky coaxial cable is assessed by a health index, and a health status report is output.

[0068] S5.1: Perform time series processing and segment mapping on the fused fault location results to generate a fault segment sequence.

[0069] The time index and spatial coordinates of each fault point are extracted from the fused fault point location results. The fault points are then sorted in ascending order according to the time index to generate a location time series arranged in chronological order. The time interval between adjacent fault points is calculated using the location time series to obtain a time interval series. Fault points with a time interval less than 0.5 times the average time interval are considered as continuous events within the same time slice. Continuous fault points are grouped into fault event sets according to time slices. The minimum spatial position, maximum spatial position, and spatial span value within each time slice are aggregated using the fault event set to generate a mapping information set containing time slice indexes and spatial interval boundaries. The spatial intervals within the time slice are numbered using the mapping information set, and the spatial intervals are arranged in positional order. Intervals with a distance less than a preset distance threshold between adjacent spatial intervals are merged to generate a continuous spatial segment set. The time slice number corresponding to each spatial segment is marked using the continuous spatial segment set, and the fault segment sequence corresponding to the time and spatial dual indices is output.

[0070] S5.2: Obtain inspection records and operation files of leaky coaxial cables, perform aging factor statistics on fault section sequences, inspection records and operation files, and generate an aging feature set; perform multi-factor evaluation and trend analysis on the aging feature set, and generate predictive maintenance suggestions.

[0071] Historical operating information is extracted from the inspection records and operation files of leaky coaxial cables. Operating data, including operating time, operating current, operating temperature, ambient humidity, and maintenance frequency, is obtained from this historical information. This operating data and time index are encapsulated into an operating dataset. Spatiotemporal matching is performed between the fault segment sequence and the operating dataset. Specifically, based on the time index and spatial location of each segment in the fault segment sequence, operating data within the same time period and spatial interval is extracted to generate an operating information set for the corresponding segment. The operating information set is then used to statistically analyze the aging factors of each fault segment's operating status. Specifically, the average temperature, humidity variation, operating current fluctuation rate, and maintenance interval cycle within the corresponding time period for each fault segment are statistically analyzed. These factors are then used as aging influencing factors, standardized, and an aging factor statistical table is generated.

[0072] The weights for the impact of aging factor statistics on cable performance degradation are set according to the aging factor statistics table. For example, the weight range for temperature is set to 0.3-0.4, the weight for humidity is set to 0.2-0.3, the weight for current fluctuation is set to 0.2-0.3, and the weight for maintenance cycle is set to 0.1-0.2. Four types of parameters are extracted from the aging factor statistics table: temperature change value, humidity change value, current fluctuation rate, and maintenance interval cycle. The weighted sum of the products of the four types of parameters and their corresponding weights is calculated. The weighted sum reflects the comprehensive aging impact of the fault section and generates a single-section aging characteristic value. The aging characteristic values ​​of all sections are normalized according to the number of sections to make the aging characteristics of different sections comparable, generating a standardized aging characteristic sequence. Each standardized aging characteristic value is associated with the corresponding fault section number and time index to form an aging characteristic set containing the section number, time index, and aging characteristic value.

[0073] The time dimension is analyzed by using an aging feature set. Specifically, the rate of change and acceleration of aging feature values ​​in adjacent time periods are obtained, the upward trend of aging is analyzed, the aging rate and fluctuation cycle features are extracted, and an aging trend sequence is generated. Potential risk segments are identified by the aging trend sequence, and the corresponding fault segments are marked as potential aging risk segments to form a risk segment set. By comprehensively comparing the risk segment set and the aging feature set, maintenance priority is generated for segments with high aging rates, and corresponding predictive maintenance suggestions are generated based on the ranking results.

[0074] S5.3: Perform feature fusion and scoring prediction on the operating status of leaky coaxial cables to generate an operating health score set.

[0075] Aging characteristic values ​​of fault sections within the aging feature set are extracted over different time periods. Real-time operating parameters, including operating current, insulation resistance, signal attenuation coefficient, and operating temperature, are obtained from inspection records and operation files. These real-time operating parameters are then normalized from different sources to generate a normalized operating characteristic sequence. This normalized operating characteristic sequence is then used to fuse the operating parameters. Specifically, the normalized values ​​of aging characteristic values, operating current, insulation resistance, signal attenuation coefficient, and operating temperature are weighted according to set weights. For example, the weight of aging characteristic values ​​is 0.3–0.4, operating current is 0.2–0.3, insulation resistance is 0.2–0.3, signal attenuation coefficient is 0.1–0.2, and operating temperature is 0.1–0.2, thus generating a fused operating characteristic sequence.

[0076] The operational status of each time period is predicted by fusing operational feature sequences. Specifically, the operational status score is determined by the mean and standard deviation of the fused feature sequences. For example, when the mean is less than 0.3, it is judged as a deteriorated state; when the mean is in the range of 0.3 to 0.7, it is judged as a sub-healthy state; and when the mean is greater than 0.7, it is judged as a healthy state. The corresponding operational health score value is generated. By matching the operational health score value with the corresponding time index and segment number, an operational health score set containing segment number, time index and operational health score value is generated.

[0077] S5.4: The health score set is structured and organized through a visualization mechanism to output a health status report.

[0078] Extract the time index, segment number, and operational health score value from the operational health score set. Sort the operational health score values ​​in ascending order by time index to generate a health score sequence arranged in chronological order. Calculate the average score value and score volatility of each segment over different time periods using the health score sequence. Group the health score sequence by segment number. Statistically analyze the operational health score values ​​of the same segment under continuous time indexes. Specifically, take the arithmetic mean of all operational health score values ​​occurring within a selected time period, such as one operational cycle. This average score reflects the overall health level within the selected time period. The deviations between each operational health score value and the average score within the same segment within the selected time period are statistically analyzed. The statistical results reflect the degree of fluctuation of the score in the same segment over time, called the score volatility. The smaller the score volatility, the more stable the health status and the more stable the operating conditions. The larger the score volatility, the more significant the fluctuations in the health status and the potential for unstable operation. Combine the average score value and score volatility to generate a score statistics table.

[0079] The operational health score is categorized and organized using a scoring statistics table. Specifically, a health status grading threshold is set based on the distribution characteristics of the operational health score values, typically ranging from 0.3 to 0.7. This range is determined by the statistical characteristics of the score distribution and the physical laws of aging and degradation. Using a range of 0.3-0.7 ensures clear categorization of health status and improves the stability and engineering applicability of the health assessment results. When the operational health score is greater than the health status grading threshold, it is classified as a healthy segment; when the operational health score is less than the health status grading threshold, it is classified as a sub-healthy segment; and when the operational health score is less than the health status grading threshold, it is classified as a deteriorated segment. The healthy, sub-healthy, and deteriorated segments are then mapped to segment numbers to generate a health status classification system. The health status grading set visualizes and formats the distribution of health status across different time periods. Specifically, healthy, sub-healthy, and deteriorated segments are marked with different color codes, and a health status matrix is ​​generated according to a time index. The changing trends of each segment's status over time are then visualized, generating a structured health status table. This structured health status table summarizes and statistically analyzes the proportions and trends of healthy, sub-healthy, and deteriorated segments, generating a health ratio statistics table. By combining the health ratio statistics table and the structured health status table, the overall operational health status is summarized, generating a health status report that includes segment numbers, time indexes, health levels, scoring trends, and proportion statistics.

[0080] This embodiment also provides a computer device applicable to the fault location method for multi-channel leaky coaxial cables, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the fault location method for multi-channel leaky coaxial cables as proposed in the above embodiment.

[0081] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0082] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the multi-channel leaky coaxial cable fault location method proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0083] In summary, this invention achieves unified time reference for multi-channel signals by framing and aligning reference and detection signals with timestamps. It effectively smooths the effects of random noise by performing function fitting and coefficient compensation on the phase deviation sequence. It reduces interference from irrelevant segments by using window function truncation and energy normalization. It quantifies the quality of different channels by assigning weights to channel signal-to-noise ratio and consistency coefficients, thus improving the reliability of fusion from the source. Finally, it reduces downtime risk and maintenance costs by performing time series processing and segment mapping on the fused fault location results.

[0084] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for locating faults in a multi-channel leaky coaxial cable, characterized in that: include, The system acquires reference and detection signals from the leaky coaxial cable, performs time delay compensation and time synchronization on the reference and detection signals, and outputs the synchronized reference and detection signals. A phase error compensation function is used to perform phase compensation on the synchronized detection signal and reference signal, and the compensated detection signal and reference signal are output. The compensated detection signal and the reference signal are searched for correlation peaks to generate a time delay difference. Time domain correlation analysis is performed on the time delay difference to extract the signal reflection feature set. The reflected wave time delay localization analysis is performed on the signal reflection feature set to output the fault point localization information of each channel. The specific steps for performing correlation peak search between the compensated detection signal and the reference signal to generate the time delay difference are as follows: The compensated reference signal and the detection signal are truncated using a window function and normalized by energy to generate an analysis signal segment. Peak candidate filtering is performed on the analyzed signal segments to generate a peak candidate set; Fine-grained peak position interpolation and confidence evaluation are performed on the peak candidate set to generate delay difference; The specific steps for extracting the signal reflection feature set are as follows: Multi-scale time window splicing and outlier removal are performed on the time delay difference to generate a purified time delay difference sequence; A time-domain correlation analysis was performed on the purification delay difference sequence, and the reflection arrival point and polarity change point were extracted. Amplitude mapping is performed between the reflection arrival point and the polarity change point to generate a signal reflection feature set; The fault location information of each channel is weighted and fused to generate a fused location feature set. The fused location feature set is then corrected for error by weighted least squares fitting to generate the fused fault location result. The aging status of leaky coaxial cables is comprehensively analyzed by integrating the fault location results, predictive maintenance suggestions are generated, and the operating status of leaky coaxial cables is assessed by health index, outputting a health status report.

2. The fault location method for multi-channel leaky coaxial cables as described in claim 1, characterized in that: The specific steps for outputting the synchronized reference signal and detection signal are as follows. Acquire reference and detection signals from the leaky coaxial cable, perform channel framing and timestamp alignment on the reference and detection signals, and generate an aligned frame set; Perform fixed inter-channel delay compensation and sampling jitter correction on the aligned frame set to generate a synchronized frame set; Amplitude normalization and DC bias elimination are performed on the synchronization frame set to generate synchronized reference and detection signals.

3. The fault location method for multi-channel leaky coaxial cables as described in claim 1, characterized in that: The specific steps for outputting the compensated detection signal and reference signal are as follows. The phase features of the synchronized reference signal and the detection signal are extracted by phase analysis, the deviation between the phase features is calculated, and a phase deviation sequence is generated. The phase error compensation parameter set is generated by fitting the phase deviation sequence to a function and compensating the coefficients through the phase error compensation function. The phase of the synchronized reference signal and the detection signal is corrected using a set of phase error compensation parameters to generate compensated reference signal and detection signal.

4. The method for locating faults in a multi-channel leaky coaxial cable as described in claim 1, characterized in that: The specific steps for outputting fault location information for each channel are as follows: Perform reflected wave time delay localization analysis on the signal reflection feature set to generate a single-channel distance dataset; Perform connected segment merging and endpoint correction on the single-channel distance dataset, and output the fault location information for each channel.

5. The fault location method for multi-channel leaky coaxial cables as described in claim 1, characterized in that: The specific steps for generating the fused fault location results are as follows: The fault location information of each channel is weighted by the channel signal-to-noise ratio and the consistency coefficient, and a weighted coefficient set is generated. The fault location information of each channel is weighted and fused using a set of weighted coefficients to generate a fused location feature set; The weighted least squares fitting method is used to perform residual calculation and bias compensation on the fused positioning feature set to generate the fused positioning correction set. The weights of the fused localization correction set are adjusted and convergence is determined to generate the fused fault location result.

6. The fault location method for multi-channel leaky coaxial cables as described in claim 1, characterized in that: The specific steps for generating predictive maintenance suggestions are as follows. The merged fault location results are processed by time series analysis and segment mapping to generate a fault segment sequence. Obtain inspection records and operation files of leaky coaxial cables, perform aging factor statistics on fault section sequences, inspection records and operation files, and generate an aging feature set; Multi-factor evaluation and trend analysis of aging feature sets are performed to generate predictive maintenance recommendations.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the multi-channel leaky coaxial cable fault location method according to any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the multi-channel leaky coaxial cable fault location method according to any one of claims 1 to 6.

Images