Cable fault accurate positioning method and system for cable health monitoring

Abstract

The invention relates to the technical field of cable fault positioning, in particular to a cable fault accurate positioning method and system for cable health monitoring, and the method comprises the steps: obtaining a collection reflection section and an experiment reflection section; obtaining the matching degree of the experimental reflection section; other parameters except the noise intensity in the experiment parameter group are constant, and the noise intensity is quantitatively analyzed to obtain the comprehensive matching degree of the collection reflection section and the experiment reflection section; screening the parameter groups to obtain a plurality of matching parameter groups and a reflection initial search range; comprehensively analyzing the probability that the correlation degree of the reflected pulse signal and the matching parameter group reflects each moment as the starting moment, and selecting the starting moment of the reflected pulse signal; and obtaining the distance of the fault point by using the starting time of the reflected pulse signal. The invention aims to solve the interference of noise on the starting moment of the reflected pulse signal and improve the accuracy of the time domain reflection method.

Description

Technical Field

[0001] This invention relates to the field of cable fault location technology, and more specifically to a method and system for precise location of cable faults for cable health monitoring. Background Technology

[0002] During long-term operation, cables are susceptible to defects such as aging, cracking, and moisture absorption in their insulation layer due to factors like electrical, thermal, mechanical, and chemical stresses. These defects cause impedance changes at specific locations within the cable, leading to short circuits, open circuits, or high-resistance faults that affect power transmission and, in severe cases, safety accidents. The time-domain reflectometry method transmits pulse signals through the cable. When the pulse signal reaches the fault location, it is reflected due to the smooth conduction caused by the defect. By measuring the time difference between the emitted and reflected pulse signals and combining this with the propagation speed, the fault distance can be calculated, allowing for precise location of the fault.

[0003] The accuracy of the time-domain reflectometry results depends on the starting point of the reflected pulse signal. However, the starting point is easily affected by noise, which superimposes onto the real reflected pulse signal, causing the pulse waveform to be distorted. Furthermore, the rising edge, which serves as the starting point, overlaps with the noise fluctuations, severely interfering with the interpretation of the starting point of the reflected pulse signal. This leads to errors in the calculation of the fault distance and results in measurement errors. Summary of the Invention

[0004] This invention provides a method and system for precise location of cable faults for cable health monitoring, in order to solve existing problems.

[0005] The cable fault precise location method and system for cable health monitoring of the present invention adopts the following technical solution: In a first aspect, one embodiment of the present invention provides a method for precise location of cable faults for cable health monitoring, the method comprising the following steps: The reflected pulse signal is segmented to obtain the acquired reflection segment; using the defect type, defect location, defect severity and noise intensity as experimental parameter sets, an experimental model is constructed and different experimental parameter sets are tested to obtain the experimental reflection segment for each experimental parameter set. Analyze the pulse waveforms of the acquired reflection segment and each of the experimental reflection segments to obtain the matching degree between the acquired reflection segment and the experimental reflection segments of each experimental parameter group; keep the other parameters in the constant experimental parameter group except for noise intensity, and obtain the comprehensive matching degree between the acquired reflection segment and the experimental reflection segments of each experimental parameter group by quantitatively analyzing the smoothness of the influence of noise intensity on the matching degree. The experimental parameter groups are screened using the comprehensive matching degree to obtain several matching parameter groups, and the reflection start search range of the acquisition reflection segment is obtained; within the intercept range of each moment in the reflection start search range, the correlation between the reflected pulse signal and each matching parameter group is comprehensively analyzed to reflect the probability of each moment as the start time of the reflected pulse signal, and the start time of the reflected pulse signal is selected. The distance to the fault point is obtained by using the start time of the reflected pulse signal as the input parameter of the time-domain reflection method.

[0006] Preferably, the specific steps for acquiring the reflection segment include: Connect one end of the cable that needs to be located to the pulse transmitter and sampling unit of the cable fault locator. After the pulse transmitter emits a narrow pulse signal, the sampling unit continuously monitors the pulse signal in the cable at each moment through the equivalent time sampling method to obtain the reflected pulse signal. Obtain all maxima in the reflected pulse signal except for the first maximum. Record the slope of each maximum relative to the pulse amplitude at the previous moment as the rising slope of each maximum, and record the slope of each maximum relative to the pulse amplitude at the next moment as the falling slope of each maximum. If there exists a constant a such that the rising slope of the i-th maximum at the a-th time step forward is less than a predetermined proportion of the rising slope of the i-th maximum, then the a-th time step forward of the i-th maximum is recorded as the starting point of the reflection segment of the i-th maximum. If there exists a constant b such that the absolute value of the descending slope of the i-th maximum at the b-th time after it is less than a predetermined proportion of the absolute value of the descending slope of the i-th maximum, then the b-th time after the i-th maximum is recorded as the termination point of the reflection segment of the i-th maximum. The pulse signal between the starting point of the reflection segment of the i-th maximum value and the ending point of the reflection segment of the i-th maximum value constitutes the pulse segment of the i-th maximum value. Any pulse segment is denoted as a reflection acquisition segment.

[0007] Preferably, the specific steps for obtaining the matching degree include: The DTW distance between the acquired reflection segment and the experimental reflection segments of each experimental parameter group is obtained using a dynamic time programming algorithm. The matching degree between the acquired reflection segment and the experimental reflection segments of each experimental parameter group is obtained, and the matching degree is inversely proportional to the DTW distance.

[0008] Preferably, the specific steps for obtaining the comprehensive matching degree include: Each set of experimental data with identical parameters except for noise intensity is denoted as a set of quantitative experimental parameters. For each set of quantitative experimental parameters, a curve space is constructed with noise intensity on the horizontal axis and matching intensity on the vertical axis to obtain the noise intensity-matching intensity curve for each set of quantitative experimental parameters. In the noise intensity-matching intensity curve of each set of quantitative experimental parameters, the difference amplitude of the matching intensity under two adjacent noise intensities and the difference distribution of all matching intensities are analyzed to obtain the smoothing coefficient of each set of quantitative experimental parameters. The smoothing coefficient is used to correct the matching strength between the acquired reflection segment and the experimental reflection segment of each experimental parameter group in each quantitative experimental parameter group set, so as to obtain the comprehensive matching degree between the acquired reflection segment and the experimental reflection segment of each experimental parameter group.

[0009] Preferably, the specific steps for obtaining the smoothing coefficient include: For each set of quantitative experimental parameters, the slope of the matching degree between the j-th noise intensity and the (j-1)-th noise intensity is denoted as the smooth difference of the j-th noise intensity in each set of quantitative experimental parameters; where the minimum value of j is 2. The variance of the smoothed differences in all noise intensities for each set of quantitative experimental parameters is denoted as the degree of difference distribution for each set of quantitative experimental parameters. Obtain the smoothing coefficient for each set of quantitative experimental parameters. The smoothing coefficient is inversely proportional to the degree of difference distribution and directly proportional to the mean of the smoothed difference of all noise intensities in each set of quantitative experimental parameters.

[0010] Preferably, the step of using the comprehensive matching degree to screen the experimental parameter set to obtain several matching parameter sets, and obtaining the initial search range of reflection for the acquisition reflection segment, includes: After keeping the defect type and defect location constant, the mean of the comprehensive matching degree of all experimental parameter groups under each combination of defect type and defect location is obtained and denoted as the defect determination index of each combination of defect type and defect location. Among all experimental parameter groups that have the highest defect determination index for the combination of defect type and defect location, and among all experimental parameter groups that have the highest comprehensive matching degree for each defect severity and different noise intensities, all experimental parameter groups are denoted as matching parameter groups. The start and end ranges of the reflection segments of all matching parameter groups are denoted as the reflection start search range of the acquisition reflection segment.

[0011] Preferably, within the intercepted range of each moment in the reflection initiation search range, the correlation between the reflected pulse signal and each matching parameter group is comprehensively analyzed to reflect the probability that each moment is the starting moment of the reflected pulse signal. The selection of the starting moment of the reflected pulse signal includes: Preset capture range; Obtain the intercept range of each moment in the reflected pulse signal and the intercept range of each moment within the initial search range of the reflection; The cross-correlation between the pulse signal of the intercepted range at each moment within the initial search range of reflection for each matching parameter group and the intercepted range at each moment within the reflected pulse signal is obtained and denoted as the correlation coefficient between each matching parameter group at each moment within the initial search range of reflection and each moment within the reflected pulse signal. Based on the correlation coefficient and the comprehensive matching degree of each matching parameter group, the likelihood probability of each moment in the reflected pulse signal corresponding to each moment within the reflection start search range is obtained. The likelihood probability is directly proportional to the correlation coefficient and the comprehensive matching degree of each matching parameter group. A scatter plot is constructed based on the reflection segment start point and severity number of all matching parameter groups. A fitting function is obtained by fitting the scatter plot using a polynomial fitting method. The residual between the corresponding scatter point of the reflection segment start point and severity number of each matching parameter group in the scatter plot and the fitting function is calculated, and the standard deviation of the residual is calculated. A Gaussian distribution model is constructed based on the standard deviation of the residuals, and the prior probability at each time step within the initial search range of reflection is obtained from the Gaussian distribution model. The product of the prior probability at each moment within the initial search range of reflection and the likelihood probability at each moment in the corresponding reflected pulse signal at each moment within the initial search range of reflection is denoted as the posterior probability at each moment in the reflected pulse signal. The moment with the highest posterior probability in the reflected pulse signal is denoted as the start time of the reflected pulse signal.

[0012] Preferably, obtaining the likelihood probability of the reflected pulse signal at each moment based on the comprehensive matching degree of the correlation coefficient and each matching parameter group includes: The product of the overall matching degree of each matching parameter group and the correlation coefficient between each time step in the reflection start search range and each time step in the reflected pulse signal is denoted as the matching correlation degree between each matching parameter group and each time step in the reflection pulse signal at each time step in the reflection start search range. Obtain the mean normalized value of the correlation between all matching parameter groups and each time step in the reflected pulse signal at each time step within the initial search range of reflection. This value is denoted as the likelihood probability of each time step in the reflected pulse signal at each time step within the initial search range of reflection.

[0013] Preferably, the step of using the start time of the reflected pulse signal as the input parameter of the time-domain reflection method to obtain the distance to the fault point includes: The reflection time of the reflected pulse signal is calculated by the interval between the start time of the reflected pulse signal and the starting time of the reflected pulse signal. The speed at which a pulse wave propagates in a cable is measured and denoted as the cable pulse velocity. The product of the cable pulse velocity and the reflection time is recorded as the distance to the fault point.

[0014] Secondly, the present invention also proposes a cable fault precise location system for cable health monitoring, the system comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the above method.

[0015] The beneficial effects of the technical solution of the present invention are as follows: The present invention constructs an experimental environment that can simulate different defect types, different defect locations, different defect severity, and different noise intensities as experimental parameter groups, and uses experimental reflection segments with different experimental parameter groups to simulate the experimental reflection segments. Then, the reflected pulse signals are verified and analyzed, and experimental parameter groups that are similar to the pulse signals of the reflected pulse segments are selected. Then, by analyzing the probability of each moment as the starting moment in all similar experimental parameter groups, the moment with the highest probability is selected as the starting moment, so as to improve the accuracy of the time-domain reflection method in determining the fault point. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. 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.

[0017] Figure 1 This is a flowchart illustrating the steps of the cable fault accurate location method for cable health monitoring according to the present invention. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the cable fault precise location method and system for cable health monitoring proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0020] The following description, in conjunction with the accompanying drawings, details the specific scheme of the cable fault accurate location method and system for cable health monitoring provided by this invention.

[0021] Firstly, please refer to Figure 1 The diagram illustrates a flowchart of a method for precise location of cable faults for cable health monitoring, provided by an embodiment of the present invention. The method includes the following steps: Step S001: Segment the reflected pulse signal to obtain the acquired reflected segment; use the defect type, defect location, defect severity and noise intensity as experimental parameter groups to construct an experimental model and experiment with different experimental parameter groups to obtain the experimental reflected segment for each experimental parameter group.

[0022] The time-domain reflectometry method transmits pulse signals through cables. When the pulse signal reaches the fault location, it is reflected due to the smooth conduction caused by the defect. By measuring the time difference between the emitted and reflected pulse signals and combining the propagation speed, the fault distance can be calculated to accurately locate the fault point. However, in the precise location of cable faults for cable health monitoring, the accurate location of the cable fault is affected by the start time of the reflected pulse signal. In real-world scenarios, the collected reflected pulse signals are affected by natural noise, making it difficult to accurately locate the start time even if the peak value of the pulse wave is determined.

[0023] Based on the above, this embodiment proposes a method for precise fault location in cable health monitoring. After acquiring the reflected pulse signal and identifying the acquisition reflection segment exhibiting the reflected pulse peak, an experimental environment is constructed to simulate different defect types, locations, severity levels, and noise intensities as experimental parameter groups. Experimental reflection segments with different experimental parameter groups are simulated to verify and analyze the reflected pulse signal. Experimental parameter groups similar to the acquisition reflection segment in the reflected pulse signal are selected. Furthermore, by analyzing the probability of each moment in all similar experimental parameter groups as the starting moment, the moment with the highest probability is selected as the starting moment to improve the accuracy of fault point determination using the time-domain reflection method.

[0024] Therefore, this embodiment first needs to acquire the reflected pulse signal and segment it into a collection reflection segment. The specific steps for acquiring the collection reflection segment are as follows: Connect one end of the cable that needs to be located to the pulse transmitter and sampling unit of the cable fault locator. After the pulse transmitter emits a narrow pulse signal, the sampling unit continuously monitors the pulse signal in the cable at each moment through the equivalent time sampling method to obtain the reflected pulse signal. It should be noted that the cable fault locator, its pulse transmitter, and sampling unit described in this embodiment are existing devices, and this embodiment will not pursue them. The rise time range of the narrow pulse signal is 0.1ns-5ns. In this embodiment, a 2ns pulse signal is used as an example for transmitting the pulse signal. The reflected pulse signal is a time sequence curve with time as the horizontal axis and pulse amplitude as the vertical axis. The interval of the time is related to the equivalent sampling rate of the equivalent time sampling method. In this embodiment, the equivalent sampling rate of the equivalent time sampling method is 15gs / s as an example, and the corresponding time interval is 66.7 picoseconds.

[0025] Since the transmitted pulse signal is reflected after encountering a fault or obstacle, and thus a new pulse is detected by the acquisition unit, and there is a long gap between the pulse peaks, in order to determine the interval of the reflected pulse, this embodiment divides the reflected pulse signal to obtain the acquisition pulse segment.

[0026] Specifically, obtain all maxima in the reflected pulse signal except for the first maxima, and record the slope of each maxima relative to the pulse amplitude at the previous moment as the rising slope of each maxima, and record the slope of each maxima relative to the pulse amplitude at the next moment as the falling slope of each maxima. If there exists a constant a such that the rising slope of the i-th maximum at the a-th time step forward is less than a predetermined proportion of the rising slope of the i-th maximum, then the a-th time step forward of the i-th maximum is recorded as the starting point of the reflection segment of the i-th maximum. If there exists a constant b such that the absolute value of the descending slope of the i-th maximum at the b-th time after it is less than a predetermined proportion of the absolute value of the descending slope of the i-th maximum, then the b-th time after the i-th maximum is recorded as the termination point of the reflection segment of the i-th maximum. The pulse signal between the starting point of the reflection segment of the i-th maximum value and the ending point of the reflection segment of the i-th maximum value constitutes the pulse segment of the i-th maximum value. Any pulse segment is denoted as a reflection acquisition segment.

[0027] It should be noted that the preset ratio described in this embodiment is 20% as an example. Other embodiments may use other preset ratios, and this embodiment does not impose specific limitations. In particular, since the first maximum value is the transmitted pulse signal, the first maximum value is not selected in the reflection acquisition segment.

[0028] It should be noted that this embodiment constructs an experimental environment that can simulate different defect types, defect locations, defect severity, and noise intensities as experimental parameter groups. It then uses experimental reflection segments with different experimental parameter groups to simulate the experimental reflection segments and compares them with the collected reflection segments to determine the start time of the reflected pulse signal. Therefore, it is necessary to construct an experimental model first.

[0029] Specifically, the steps for constructing an experimental model and experimenting with different sets of experimental parameters using defect type, defect location, defect severity, and noise intensity as experimental parameter sets to obtain the experimental reflection segment for each set of experimental parameters are as follows: An experimental parameter set is constructed, comprising the elements of defect type, defect location, defect severity, and noise intensity. The defect type is the defect number; the defect location is the distance from the pulse transmitter; the defect severity is set based on the defect type; and the noise intensity, in this embodiment, is represented by the signal-to-noise ratio (SNR) of the reflected pulse signal. The values ​​for defect severity and SNR range from [value missing]. The larger the value, the higher the signal-to-noise ratio and the more severe the defect. Furthermore, by adjusting the values ​​of each element in the experimental parameter set, several experimental parameter sets are obtained; simulation experiments are conducted using the experimental parameter sets to obtain the reflected pulse signal of each experimental parameter set; For each experimental parameter group, the experimental reflection segment is obtained by segmenting the reflected pulse signal to obtain the reflection acquisition segment.

[0030] It should be noted that the experimental reflection segments of each experimental parameter group were conducted using the controlled variable method, and each experimental reflection segment of the experimental parameter group contained only one experimental reflection segment.

[0031] As an example, using defect type, defect location, defect severity, and noise intensity as experimental parameter sets, an experimental model is constructed and different experimental parameter sets are tested to obtain the specific method for the experimental reflection segment of each experimental parameter set: The experimental setup was arranged in the laboratory, including 160 meters of cable with a core radius of 3.5 mm, an insulation layer thickness of 5.8 mm, and a shielding layer radius of 9.4 mm; both ends of the cable were open circuits. As an example, the cable defect type is to create a pit in the cable insulation layer. The defect location is 30M, 70M, 110M and 150M. The defect severity is the depth of the pit. In this embodiment, the preset pit depth is 10%, 30%, 50%, 70% and complete penetration. As another example, the cable defect type is moisture defect, with defect locations at 40M, 80M, 110M and 150M. The severity of the defect is determined by the water level and immersion time. In this embodiment, the preset values ​​are: 1 / 4 water level for 12 hours, 1 / 2 water level for 12 hours, 3 / 4 water level for 12 hours, and 3 / 4 water level for 24 hours. As another example, the cable defect type is mechanical conductor damage, with defect locations at 30M, 70M, 110M, and 150M. The defect severity is the reduction in cable core radius. In this embodiment, the preset reduction in cable core radius is 5%, 10%, 25%, 50%, and 70%, respectively. In the experimental environment, experiments were conducted on the different parameters mentioned above to obtain the pure reflected pulse signals under each experimental parameter group. For the aforementioned pure reflected pulse signals, this embodiment adds preset noise to the pure reflected pulse signals, with noise intensity ratios of 0%, 30%, 60%, 90%, and 100%, respectively. The corresponding noise power is calculated according to the signal power calculation formula, and the generated noise sequence is then spliced ​​with the aforementioned pure reflected pulse signals to obtain reflected pulse signals for each experimental parameter group with added noise. For the reflected pulse signal of each experimental parameter group, the experimental reflection segment of each experimental parameter group is obtained by segmenting the reflected pulse signal to obtain the reflection acquisition segment.

[0032] It should be noted that splicing the noise sequence with the clean reflected pulse signal is a well-known technique, and will not be described in detail in this embodiment.

[0033] Step S002: Analyze the pulse waveforms of the acquired reflection segment and each of the experimental reflection segments to obtain the matching degree between the acquired reflection segment and the experimental reflection segments of each experimental parameter group; keep the other parameters in the constant experimental parameter group except for the noise intensity, and obtain the comprehensive matching degree between the acquired reflection segment and the experimental reflection segments of each experimental parameter group by quantitatively analyzing the smoothness of the influence of the noise intensity on the matching degree.

[0034] Since this embodiment simulates experimental reflection segments with different defect types, locations, severity levels, and noise intensities, and compares them with the acquired reflection segments to select an experimental environment similar to the actual acquired reflection segments, the starting time of the reflected pulse signal is determined. Therefore, the matching degree between the acquired reflection segments and the experimental reflection segments of each experimental parameter group is first obtained.

[0035] Preferably, the specific steps for analyzing the pulse waveforms of the acquired reflection segment and each of the experimental reflection segments to obtain the degree of matching between the acquired reflection segment and the experimental reflection segments of each experimental parameter group are as follows: The DTW distance between the acquired reflection segment and the experimental reflection segments of each experimental parameter group is obtained using a dynamic time programming algorithm. The matching degree between the acquired reflection segment and the experimental reflection segments of each experimental parameter group is obtained, and the matching degree is inversely proportional to the DTW distance.

[0036] As an example, the matching degree is obtained by: inversely normalizing the DTW distance between the acquired reflection segment and the experimental reflection segment of each experimental parameter group, and recording this as the matching degree between the acquired reflection segment and the experimental reflection segment of each experimental parameter group. In this embodiment, the matching degree is obtained by... The function is inversely proportionally normalized. It is an exponential function with the natural constant as its base.

[0037] It should be noted that the dynamic time programming algorithm for obtaining the DTW distance is a well-known existing technology, and will not be described in detail in this embodiment. Since the time lengths of the acquired reflection segment and the experimental reflection segment are not necessarily the same, and the dynamic time programming algorithm can analyze the trend similarity of sequences of different lengths and obtain the DTW distance, the smaller the DTW distance, the greater the trend similarity. Therefore, after inverse proportional normalization, the smaller the DTW distance, the greater the matching degree between the acquired reflection segment and the experimental reflection segment of each experimental parameter group.

[0038] It should be noted that the matching degree indicates the similarity between the experimental reflection segment and the acquired reflection segment obtained under different experimental parameters. If the acquired reflection segment and the experimental reflection segment are under the same defect scenario conditions, that is, the defect type, defect severity, and defect location are the same, then their pulse waveforms are similar and will only be affected by noise. Therefore, the matching degree will change for different noise intensities. By increasing or decreasing the noise intensity, the matching degree will only change slightly and smoothly. Based on this, this embodiment analyzes the smoothness of the influence of the matching degree on the noise intensity by using other parameters in the constant experimental parameter set except for the noise intensity, and corrects the matching degree to obtain the comprehensive matching degree between the collected reflection segment and the experimental reflection segment of each experimental parameter set.

[0039] Preferably, the specific steps for obtaining the comprehensive matching degree between the acquired reflection segment and the experimental reflection segment of each experimental parameter group, excluding noise intensity, by quantitatively analyzing the smoothness of the influence of noise intensity on the matching degree, are as follows: Each set of experimental data with identical parameters except for noise intensity is denoted as a set of quantitative experimental parameters. For each set of quantitative experimental parameters, a curve space is constructed with noise intensity on the horizontal axis and matching intensity on the vertical axis to obtain the noise intensity-matching intensity curve for each set of quantitative experimental parameters. In the noise intensity-matching intensity curve of each set of quantitative experimental parameters, the difference amplitude of the matching intensity under two adjacent noise intensities and the difference distribution of all matching intensities are analyzed to obtain the smoothing coefficient of each set of quantitative experimental parameters. The smoothing coefficient is used to correct the matching strength between the acquired reflection segment and the experimental reflection segment of each experimental parameter group in each quantitative experimental parameter group set, so as to obtain the comprehensive matching degree between the acquired reflection segment and the experimental reflection segment of each experimental parameter group.

[0040] Specifically, in the noise intensity-matching intensity curve of each quantitative experimental parameter set, the method for analyzing the difference amplitude of the matching intensity under two adjacent noise intensities and the difference distribution of all matching intensities to obtain the smoothing coefficient of each quantitative experimental parameter set is as follows: For each set of quantitative experimental parameters, the slope of the matching degree between the j-th noise intensity and the (j-1)-th noise intensity is denoted as the smooth difference of the j-th noise intensity in each set of quantitative experimental parameters; where the minimum value of j is 2. The variance of the smoothed differences in all noise intensities for each set of quantitative experimental parameters is denoted as the degree of difference distribution for each set of quantitative experimental parameters. Obtain the smoothing coefficient for each set of quantitative experimental parameters. The smoothing coefficient is inversely proportional to the degree of difference distribution and directly proportional to the mean of the smoothed difference of all noise intensities in each set of quantitative experimental parameters.

[0041] As an example, this embodiment calculates the product of the degree of difference distribution of each quantitative experimental parameter set after inverse proportional normalization and the mean of the smooth difference of all noise intensities in each quantitative experimental parameter set, and then normalizes the product by the maximum and minimum values ​​to obtain the smoothing coefficient of each quantitative experimental parameter set.

[0042] It should be noted that the above inverse proportional normalization in this embodiment is implemented using an exponential function with the natural constant as the base, and the maximum and minimum value normalization is implemented using a maximum and minimum value normalization algorithm, which will not be described in detail in this embodiment.

[0043] It should be noted that, since the experimental parameter set that conforms to the actual defect changes its matching degree with the gradual change of noise when adjusting the noise effect, the smoothness coefficient of the quantitative experimental parameter set is larger. Therefore, this embodiment corrects the matching degree by the smoothness coefficient to obtain the comprehensive matching degree between the collected reflection segment and the experimental reflection segment of each experimental parameter set.

[0044] As an example, the specific method for correcting the matching strength between the acquired reflection segment and the experimental reflection segment of each experimental parameter group in each quantitative experimental parameter set using the smoothing coefficient is as follows: The sum of the smoothing coefficient of each set of quantitative experimental parameters and 1 is denoted as the correction coefficient of each set of quantitative experimental parameters. The product of the correction coefficient of each set of quantitative experimental parameters and the matching strength between the acquired reflection segment and the experimental reflection segment of each set of experimental parameters is denoted as the comprehensive matching degree between the acquired reflection segment and the experimental reflection segment of each set of experimental parameters.

[0045] Step S003: Use the comprehensive matching degree to screen the experimental parameter groups to obtain several matching parameter groups, and obtain the reflection start search range of the acquisition reflection segment; within the intercept range of each moment in the reflection start search range, comprehensively analyze the correlation between the reflected pulse signal and each matching parameter group to reflect the probability of each moment as the start time of the reflected pulse signal, and select the start time of the reflected pulse signal.

[0046] It should be noted that after obtaining the overall matching degree between the acquisition reflection segment and the experimental reflection segment of each experimental parameter group, the overall matching degree is mainly determined based on the peak performance of the pulse signal. That is, the type and location of the defect are determined based on the overall performance of the acquisition reflection segment and the experimental reflection segment.

[0047] However, each defect in the cable has a certain development process, from non-existent to severe. The development of defects of different severity can determine the change in the starting point of the corresponding reflected pulse. The development of defects of different severity only affects its pulse performance. The pulse performance will change the position of the starting time, but its pulse performance is constant. Therefore, this embodiment determines the defect type, noise intensity and defect location by comprehensively matching degree, and then analyzes the constantness of the pulse performance at the starting time under different severity to obtain the starting time.

[0048] On the other hand, since the final requirement of this embodiment is to determine the start time of the reflected pulse signal, the pulse performance of the reflected pulse signal is relatively weak at the start time, which is easily confused with noise and leads to errors in the start time. Therefore, under constant defect type and defect location, this embodiment obtains several matching parameter groups by screening the experimental parameter groups with the largest noise intensity but different defect severity through comprehensive matching degree. The range of the reflection segment start point of these matching parameter groups is analyzed probabilistically to determine the start time of the reflected pulse signal.

[0049] Preferably, the specific steps for obtaining several matching parameter sets by screening the experimental parameter sets using the comprehensive matching degree, and obtaining the initial search range of the reflection acquisition segment are as follows: After keeping the defect type and defect location constant, the mean of the comprehensive matching degree of all experimental parameter groups under each combination of defect type and defect location is obtained and denoted as the defect determination index of each combination of defect type and defect location. Among all experimental parameter groups that have the highest defect determination index for the combination of defect type and defect location, and among all experimental parameter groups corresponding to different noise intensities under each defect severity, the experimental parameter group with the highest comprehensive matching degree is denoted as the matching parameter group.

[0050] As an example, considering cable defect types such as pits in the cable insulation, defect locations of 30M, 70M, 110M, and 150M, and defect severity based on pit depth, with preset pit depths of 10%, 30%, 50%, 70%, and complete penetration, and noise intensity signal-to-noise ratios of 0%, 30%, 60%, 90%, and 100%, for instance, the combination of a pit defect and a 30M defect location results in the highest defect certainty index. Therefore, at each defect severity level—pit depth of 10%, 30%, 50%, 70%, and complete penetration—the defect severity at a pit depth of 10% is considered high. In the case of a severe defect with a pit depth of 10%, the experimental parameter group with a signal-to-noise ratio of 30% has the highest overall matching degree. Therefore, in the case of a severe defect with a pit depth of 30%, the experimental parameter group with a signal-to-noise ratio of 70% has the highest overall matching degree. Therefore, in the case of a severe defect with a pit depth of 30%, the experimental parameter group with a signal-to-noise ratio of 70% is recorded as a matching parameter group. And so on, all matching parameter groups are obtained under the combination of pit defect with the largest defect certainty index and defect location at 30M, which are the matching parameter groups described in this embodiment.

[0051] Furthermore, the start and end ranges of the reflection segments of all matching parameter groups are denoted as the reflection start search range of the acquired reflection segment.

[0052] It should be noted that after obtaining the reflection start search range of the acquisition reflection segment, since the simulated defect location and defect type are the same, and different severity levels will only change the starting point of the defect reflection segment, but will not change the pulse behavior of the defect, by analyzing the correlation between all the matching parameter segments within the intercept range of each moment of the reflection start range and each moment of the reflection pulse signal, the probability of each moment of the reflection pulse signal being the starting moment of the reflection pulse signal can be obtained, and then the starting moment of the reflection pulse signal can be selected.

[0053] It should be noted that this embodiment introduces Bayesian statistical concepts, namely posterior probability. ∝ Likelihood probability Prior probability By calculating the posterior probability, the moment with the highest probability in the reflected pulse signal can be selected as the start moment of the reflected pulse signal, and uncertainty can also be evaluated.

[0054] Preferably, within the intercepted range of each moment in the reflection initiation search range, the correlation between the reflected pulse signal and each matching parameter group is comprehensively analyzed to reflect the probability that each moment is the starting moment of the reflected pulse signal. The specific steps for selecting the starting moment of the reflected pulse signal are as follows: The preset truncation range is defined in this embodiment, with each moment as the center and three moments before and after as the truncation range for each moment. That is, the truncation range length for each moment is a total of seven moments including that moment. Obtain the intercept range of each moment in the reflected pulse signal and the intercept range of each moment within the initial search range of the reflection; The cross-correlation between the pulse signal of the intercepted range at each moment within the initial search range of reflection for each matching parameter group and the intercepted range at each moment within the reflected pulse signal is obtained and denoted as the correlation coefficient between each matching parameter group at each moment within the initial search range of reflection and each moment within the reflected pulse signal. Based on the correlation coefficient and the comprehensive matching degree of each matching parameter group, the likelihood probability of each moment in the reflected pulse signal corresponding to each moment within the reflection start search range is obtained. The likelihood probability is directly proportional to the correlation coefficient and the comprehensive matching degree of each matching parameter group. A scatter plot is constructed based on the reflection segment start point and severity number of all matching parameter groups. A fitting function is obtained by fitting the scatter plot using a polynomial fitting method. The residual between the corresponding scatter point of the reflection segment start point and severity number of each matching parameter group in the scatter plot and the fitting function is calculated, and the standard deviation of the residual is calculated. A Gaussian distribution model is constructed based on the standard deviation of the residuals, and the prior probability at each time step within the initial search range of reflection is obtained from the Gaussian distribution model. The product of the prior probability at each moment within the initial search range of reflection and the likelihood probability at each moment in the corresponding reflected pulse signal at each moment within the initial search range of reflection is denoted as the posterior probability at each moment in the reflected pulse signal. The moment with the highest posterior probability in the reflected pulse signal is denoted as the start time of the reflected pulse signal.

[0055] It should be noted that the fitting function represents the change in the starting position as the severity changes, the Gaussian distribution model is normally distributed, and the prior probability represents the probability of pulse behavior as the starting moment of reflection at each moment within the reflection initiation search range in the Gaussian distribution model; the prior probability is an existing parameter obtained from Bayesian theory, which will not be elaborated in this embodiment.

[0056] As an example, in this embodiment, the Pearson correlation coefficient between the pulse signal of the intercepted range at each time in the reflection start search range of each matching parameter group and the pulse signal of the intercepted range at each time in the reflected pulse signal is denoted as the correlation coefficient between each matching parameter group at each time in the reflection start search range and the reflected pulse signal at each time.

[0057] As an example, the specific method for obtaining the likelihood probability of the reflected pulse signal at each moment based on the correlation coefficient and the overall matching degree of each matching parameter group is as follows: The product of the overall matching degree of each matching parameter group and the correlation coefficient between each time step in the reflection start search range and each time step in the reflected pulse signal is denoted as the matching correlation degree between each matching parameter group and each time step in the reflection pulse signal at each time step in the reflection start search range. Obtain the mean normalized value of the correlation between all matching parameter groups and each time step in the reflected pulse signal at each time step within the initial search range of reflection. This value is denoted as the likelihood probability of each time step in the reflected pulse signal at each time step within the initial search range of reflection.

[0058] Step S004: Use the start time of the reflected pulse signal as the input parameter of the time-domain reflection method to obtain the distance to the fault point.

[0059] After obtaining the start time of the reflected pulse signal, the reflection time of the reflected pulse signal is calculated based on the start time and the origin of the reflected pulse signal. Based on the reflection time and the velocity of the pulse wave, the distance from the fault point to the origin of the reflected pulse signal can be obtained.

[0060] Specifically, the method for obtaining the distance to the fault point by using the start time of the reflected pulse signal as the input parameter of the time-domain reflection method is as follows: The reflection time of the reflected pulse signal is calculated by the interval between the start time of the reflected pulse signal and the starting time of the reflected pulse signal. The speed at which a pulse wave propagates in a cable is measured and denoted as the cable pulse velocity. The product of the cable pulse velocity and the reflection time is recorded as the distance to the fault point.

[0061] It should be noted that the embodiments used in this example The model only represents negative correlations and constraints. The model output results are in... Within the interval, This is the input to this model; in specific implementations, it can be replaced with other models that have the same purpose. This embodiment is merely an example. The description will be based on a model, without making any specific limitations.

[0062] Secondly, another embodiment of the present invention provides a cable fault precise location system for cable health monitoring, the system including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the above-described method.

[0063] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for precise fault location in cable health monitoring, characterized in that, The method includes the following steps: The reflected pulse signal is segmented to obtain the acquired reflection segment; using the defect type, defect location, defect severity and noise intensity as experimental parameter sets, an experimental model is constructed and different experimental parameter sets are tested to obtain the experimental reflection segment for each experimental parameter set. Analyze the pulse waveforms of the acquired reflection segment and each of the experimental reflection segments to obtain the matching degree between the acquired reflection segment and the experimental reflection segments of each experimental parameter group; keep the other parameters in the constant experimental parameter group except for noise intensity, and obtain the comprehensive matching degree between the acquired reflection segment and the experimental reflection segments of each experimental parameter group by quantitatively analyzing the smoothness of the influence of noise intensity on the matching degree. The experimental parameter groups are screened using the comprehensive matching degree to obtain several matching parameter groups, and the reflection start search range of the acquisition reflection segment is obtained; within the intercept range of each moment in the reflection start search range, the correlation between the reflected pulse signal and each matching parameter group is comprehensively analyzed to reflect the probability of each moment as the start time of the reflected pulse signal, and the start time of the reflected pulse signal is selected. The distance to the fault point is obtained by using the start time of the reflected pulse signal as the input parameter of the time-domain reflection method.

2. The method for precise location of cable faults for cable health monitoring according to claim 1, characterized in that, The specific steps for acquiring the reflection segment include: Connect one end of the cable that needs to be located to the pulse transmitter and sampling unit of the cable fault locator. After the pulse transmitter emits a narrow pulse signal, the sampling unit continuously monitors the pulse signal in the cable at each moment through the equivalent time sampling method to obtain the reflected pulse signal. Obtain all maxima in the reflected pulse signal except for the first maximum. Record the slope of each maximum relative to the pulse amplitude at the previous moment as the rising slope of each maximum, and record the slope of each maximum relative to the pulse amplitude at the next moment as the falling slope of each maximum. If there exists a constant a such that the rising slope of the i-th maximum at the a-th time step forward is less than a predetermined proportion of the rising slope of the i-th maximum, then the a-th time step forward of the i-th maximum is recorded as the starting point of the reflection segment of the i-th maximum. If there exists a constant b such that the absolute value of the descending slope of the i-th maximum at the b-th time after it is less than a predetermined proportion of the absolute value of the descending slope of the i-th maximum, then the b-th time after the i-th maximum is recorded as the termination point of the reflection segment of the i-th maximum. The pulse signal between the starting point of the reflection segment of the i-th maximum value and the ending point of the reflection segment of the i-th maximum value constitutes the pulse segment of the i-th maximum value. Any pulse segment is denoted as a reflection acquisition segment.

3. The method for precise location of cable faults for cable health monitoring according to claim 1, characterized in that, The specific steps for obtaining the matching degree include: The DTW distance between the acquired reflection segment and the experimental reflection segments of each experimental parameter group is obtained using a dynamic time programming algorithm. The matching degree between the acquired reflection segment and the experimental reflection segments of each experimental parameter group is obtained, and the matching degree is inversely proportional to the DTW distance.

4. The method for precise location of cable faults for cable health monitoring according to claim 1, characterized in that, The specific steps for obtaining the overall matching degree include: Each set of experimental data with identical parameters except for noise intensity is denoted as a set of quantitative experimental parameters. For each set of quantitative experimental parameters, a curve space is constructed with noise intensity on the horizontal axis and matching intensity on the vertical axis to obtain the noise intensity-matching intensity curve for each set of quantitative experimental parameters. In the noise intensity-matching intensity curve of each set of quantitative experimental parameters, the difference amplitude of the matching intensity under two adjacent noise intensities and the difference distribution of all matching intensities are analyzed to obtain the smoothing coefficient of each set of quantitative experimental parameters. The smoothing coefficient is used to correct the matching strength between the acquired reflection segment and the experimental reflection segment of each experimental parameter group in each quantitative experimental parameter group set, so as to obtain the comprehensive matching degree between the acquired reflection segment and the experimental reflection segment of each experimental parameter group.

5. The method for precise location of cable faults for cable health monitoring according to claim 4, characterized in that, The specific steps for obtaining the smoothing coefficient include: For each set of quantitative experimental parameters, the slope of the matching degree between the j-th noise intensity and the (j-1)-th noise intensity is denoted as the smooth difference of the j-th noise intensity in each set of quantitative experimental parameters; where the minimum value of j is 2. The variance of the smoothed differences in all noise intensities for each set of quantitative experimental parameters is denoted as the degree of difference distribution for each set of quantitative experimental parameters. Obtain the smoothing coefficient for each set of quantitative experimental parameters. The smoothing coefficient is inversely proportional to the degree of difference distribution and directly proportional to the mean of the smoothed difference of all noise intensities in each set of quantitative experimental parameters.

6. The method for precise location of cable faults for cable health monitoring according to claim 1, characterized in that, The process of using the comprehensive matching degree to filter experimental parameter sets yields several matching parameter sets, and the initial search range for the reflection of the acquisition reflection segment includes: After keeping the defect type and defect location constant, the mean of the comprehensive matching degree of all experimental parameter groups under each combination of defect type and defect location is obtained and denoted as the defect determination index of each combination of defect type and defect location. Among all experimental parameter groups that have the highest defect determination index for the combination of defect type and defect location, and among all experimental parameter groups that have the highest comprehensive matching degree for each defect severity and different noise intensities, all experimental parameter groups are denoted as matching parameter groups. The start and end ranges of the reflection segments of all matching parameter groups are denoted as the reflection start search range of the acquisition reflection segment.

7. The method for precise location of cable faults for cable health monitoring according to claim 1, characterized in that, Within the truncation range of each moment in the reflection start search range, the correlation between the reflected pulse signal and each matching parameter group is comprehensively analyzed to reflect the probability that each moment is the start moment of the reflected pulse signal. The selection of the start moment of the reflected pulse signal includes: Preset capture range; Obtain the intercept range of each moment in the reflected pulse signal and the intercept range of each moment within the initial search range of the reflection; The cross-correlation between the pulse signal of the intercepted range at each moment within the initial search range of reflection for each matching parameter group and the intercepted range at each moment within the reflected pulse signal is obtained and denoted as the correlation coefficient between each matching parameter group at each moment within the initial search range of reflection and each moment within the reflected pulse signal. Based on the correlation coefficient and the comprehensive matching degree of each matching parameter group, the likelihood probability of each moment in the reflected pulse signal corresponding to each moment within the reflection start search range is obtained. The likelihood probability is directly proportional to the correlation coefficient and the comprehensive matching degree of each matching parameter group. A scatter plot is constructed based on the reflection segment start point and severity number of all matching parameter groups. A fitting function is obtained by fitting the scatter plot using a polynomial fitting method. The residual between the corresponding scatter point of the reflection segment start point and severity number of each matching parameter group in the scatter plot and the fitting function is calculated, and the standard deviation of the residual is calculated. A Gaussian distribution model is constructed based on the standard deviation of the residuals, and the prior probability at each time step within the initial search range of reflection is obtained from the Gaussian distribution model. The product of the prior probability at each moment within the initial search range of reflection and the likelihood probability at each moment in the corresponding reflected pulse signal at each moment within the initial search range of reflection is denoted as the posterior probability at each moment in the reflected pulse signal. The moment with the highest posterior probability in the reflected pulse signal is denoted as the start time of the reflected pulse signal.

8. The method for precise location of cable faults for cable health monitoring according to claim 7, characterized in that, The step of obtaining the likelihood probability of the reflected pulse signal at each moment based on the comprehensive matching degree of the correlation coefficient and each matching parameter group includes: The product of the overall matching degree of each matching parameter group and the correlation coefficient between each time step in the reflection start search range and each time step in the reflected pulse signal is denoted as the matching correlation degree between each matching parameter group and each time step in the reflection pulse signal at each time step in the reflection start search range. Obtain the mean normalized value of the correlation between all matching parameter groups and each time step in the reflected pulse signal at each time step within the initial search range of reflection. This value is denoted as the likelihood probability of each time step in the reflected pulse signal at each time step within the initial search range of reflection.

9. The method for precise location of cable faults for cable health monitoring according to claim 1, characterized in that, The step of using the start time of the reflected pulse signal as the input parameter for the time-domain reflection method to obtain the distance to the fault point includes: The reflection time of the reflected pulse signal is calculated by the interval between the start time of the reflected pulse signal and the starting time of the reflected pulse signal. The speed at which a pulse wave propagates in a cable is measured and denoted as the cable pulse velocity. The product of the cable pulse velocity and the reflection time is recorded as the distance to the fault point.

10. A cable fault precise location system for cable health monitoring, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the cable fault precise location method for cable health monitoring as described in any one of claims 1-9.

Images