A detection system and method for cable fault location

Abstract

This invention discloses a detection system and method for cable fault location, relating to the field of cable fault location technology. It acquires nanosecond-level pulse propagation front-end change trajectory data and echo peak width change data to establish a high-frequency attenuation continuous-time distribution map; extracts pulse broadening growth segments and performs front-end compression control to form a compressed pulse front-end shape; rearranges the peak time position to separate low-amplitude reflection signals; tracks its amplitude trajectory and identifies local insulation anomaly segments by comparing with the peak width evolution trajectory; establishes a distance-back mapping and corrects the echo time coordinates to obtain the fault location interval. This invention improves the ability to identify early insulation defects by constructing a continuous-time distribution map and compressing pulse broadening segments, shortening the peak diffusion time, and separating low-amplitude reflection signals within the peak; simultaneously, it improves fault location accuracy and stability by establishing a distance-back mapping relationship and dynamically correcting the echo time coordinates.

Description

Technical Field

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

[0002] Cable fault location refers to a precise detection technology used in nuclear power plants' Class 1E nuclear safety cables during long-term operation. This technology addresses abnormal conditions such as insulation aging, core wire open circuits, short circuits, and localized damage caused by strong radiation, high temperature, high humidity, and vibration environments. It involves injecting high-speed, narrow-pulse signals into the cable and collecting the reflected echo characteristics during propagation. The position, amplitude, and morphological changes of the echo on the time axis are analyzed to determine the specific spatial location and type of the fault. In nuclear-grade applications, this process involves not only the effective separation of reflected signals in long-distance, multi-branch, and complex laying structures but also dynamic correction of propagation parameter changes caused by cable aging. This ensures a stable mapping between echo time and actual distance, ultimately achieving high-precision location that meets nuclear safety standards. This provides crucial technical support for the stable operation of instrumentation and control systems and the safety of nuclear power plants.

[0003] During the long-term operation of Class 1E nuclear safety cables in nuclear power plants, the cables are typically exposed to multiple coupled stress conditions, including strong gamma radiation, continuous thermal stress, high humidity, and mechanical vibration. These environmental factors continuously damage the molecular structure of the cable insulation material, leading to polymer chain breakage, enhanced oxidation reactions, and an increase in polar groups. This causes a gradual increase in the dielectric constant and the dielectric loss tangent. As aging progresses, the conductivity and loss parameters per unit length of the cable continuously change, and its frequency-dependent characteristics show a significant strengthening trend, especially exhibiting a noticeable additional attenuation effect in the high-frequency range.

[0004] When using nanosecond-level high-speed narrow pulse signals for fault location, the pulse signal inherently contains a large number of high-frequency spectral components. As cable aging leads to increased dielectric loss, these high-frequency components undergo stronger frequency-selective attenuation during propagation, causing the originally steep pulse rise edge to gradually become blunted, the main peak width to expand, and a noticeable tailing phenomenon to form in the time domain. Simultaneously, the aging process also alters the cable's distributed capacitance and distributed conductivity parameters, resulting in a slow drift in propagation speed and increased signal dispersion. The asynchronous superposition of the propagation phases of multiple frequency components further exacerbates the pulse self-broadening effect, causing blurred echo waveform edges, reduced peak value, and decreased signal-to-noise ratio.

[0005] As the width of the echo peak continues to increase, if the time interval between adjacent reflection events is less than the broadened pulse width, multiple reflected signals will overlap in the time domain, forming indistinguishable composite waveforms. For long-distance, multi-branch cable systems in the complex cabling structures of nuclear power plants, connection nodes, branch interfaces, and even slight impedance disturbances themselves generate multiple reflection signals. When resolution decreases, the system will struggle to accurately distinguish reflection sources from different locations, significantly reducing temporal resolution and thus weakening the ability to precisely locate fault positions.

[0006] More importantly, in the early stages of cable insulation aging, the impedance disturbances caused by localized microcracks, water trees, or changes in the dielectric microstructure are usually small, resulting in extremely low reflection coefficients and echo signal amplitudes much smaller than the main reflection peak. When the main peak broadens and is accompanied by energy tailing, weak fault reflections are easily covered or submerged by the energy at the tail of the main peak, making it impossible to effectively identify early insulation degradation characteristics. The detection system may only observe an overall increase in attenuation, failing to accurately locate localized potential defects, leading to the neglect of early hidden dangers and a significant risk of missed detection.

[0007] As aging continues, local insulation defects will accumulate, the space charge accumulation effect will intensify, and electric field distortion will worsen, potentially inducing dielectric breakdown or short circuits between conductors. Since Class 1E nuclear safety cables typically bear critical functions in instrumentation and control systems, reactor protection systems, and safety-related actuators, a sudden failure will directly impact the safe and stable operation of a nuclear power plant. Therefore, the decreased positioning accuracy caused by changes in propagation parameters, pulse self-broadening effects, and weak echo submersion phenomena under aging conditions is not only a technical challenge in terms of detection accuracy but also a core risk factor related to the long-term safety and reliability of nuclear-grade equipment.

[0008] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0009] The purpose of this invention is to provide a detection system and method for cable fault location. By constructing a continuous time distribution map and compressing the pulse broadening section, the main peak diffusion time is shortened, and the low-amplitude reflection signal inside the main peak is separated, thereby improving the ability to identify early insulation defects. At the same time, a distance-back mapping relationship is established and the echo time coordinate is dynamically corrected to improve the fault location accuracy and stability, so as to solve the problems in the background art.

[0010] To achieve the above objectives, the present invention provides the following technical solution: a detection method for cable fault location, comprising the following steps:

[0011] Acquire the leading edge trajectory data and corresponding echo peak width variation data of nanosecond-level pulses during propagation in aged nuclear-grade cables. Based on the leading edge trajectory data and echo peak width variation data, establish a continuous time distribution map of high-frequency attenuation along the propagation distance.

[0012] Based on the continuous time distribution map, the pulse broadening and growth segment is extracted, and the leading edge advancement rhythm of the nanosecond pulse is compressed and controlled according to the pulse broadening and growth segment to obtain the compressed pulse leading edge morphology with a shortened main peak diffusion time segment.

[0013] The time position of the echo main peak is rearranged based on the shape of the pulse leading edge after compression. The low-amplitude reflection signal that was inside the main peak before compression is separated into an independent time segment to obtain the stripped low-amplitude reflection signal.

[0014] The amplitude change trajectory of the low-amplitude reflected signal after stripping is tracked along the propagation distance, and the amplitude change trajectory is compared and analyzed with the main peak width evolution trajectory corresponding to the echo main peak width change data to identify local insulation abnormality sections.

[0015] Based on the local insulation anomaly section, a distance-back mapping relationship is established. Under the constraint of the distance-back mapping relationship, the echo time coordinate is dynamically corrected to obtain the fault location range.

[0016] Preferably, establishing a continuous-time distribution map of high-frequency attenuation along the propagation distance includes the following steps:

[0017] Nanosecond-level pulses are injected into one end of the aging nuclear-grade cable. The pulse front change trajectory data are collected at fixed time intervals under a unified time coordinate axis, and the time interval length between the start and end points of each echo peak is recorded simultaneously to form echo peak width change data.

[0018] The data on the trajectory of the frontal change and the data on the width of the echo main peak are paired and arranged in chronological order, and each time segment is converted into a corresponding propagation distance segment. The degree of frontal expansion and the value of the width of the echo main peak are jointly labeled in the same distance sequence.

[0019] The data on the trajectory of the leading edge and the width of the echo peak in each propagation distance segment are read sequentially along the propagation distance direction and continuously expanded to construct a continuous time distribution map with the propagation distance as the vertical axis and time as the horizontal axis.

[0020] Preferably, the data on the trajectory of the leading edge change and the data on the width change of the main peak of the echo form a one-to-one correspondence under a unified time coordinate axis, and the time order remains unchanged when converted into the propagation distance segment, thereby ensuring the consistency of the correspondence between the time dimension and the propagation distance dimension in the continuous time distribution map.

[0021] Preferably, obtaining the compressed pulse front morphology with a shortened main peak diffusion time segment includes the following steps:

[0022] In the continuous time distribution map, the echo peak width values ​​are read segment by segment along the propagation distance direction and continuously compared. At the same time, the leading edge change trajectory data in the corresponding time segment are retrieved. The pulse broadening growth segment is determined by the overlap between the continuous increase of the peak width and the time span of the rising segment of the leading edge.

[0023] Time sampling nodes are extracted with the time start and end points corresponding to the pulse broadening growth segment as the range. The time interval between adjacent sampling nodes is shortened and redistributed to each sampling node to form a compressed time segment.

[0024] The compressed time segment is sequentially arranged with the time segment preceding the pulse broadening and growth segment to construct a new time progression curve while keeping the sampling node order unchanged.

[0025] Based on the new time-progression curve, the time span between the start and end points of the echo main peak is recalibrated, forming a compressed pulse front morphology with a shortened main peak diffusion time segment.

[0026] Preferably, the pulse broadening growth segment is defined by the synchronous change relationship between the continuous increase in the width of the echo main peak and the time span of the rising segment of the leading edge in the trajectory of the leading edge change. The shortening of the time sampling node maintains the original sampling order, and the time segment of the echo main peak is redefined by the new time advancement curve.

[0027] Preferably, obtaining the low-amplitude reflection signal after stripping includes the following steps:

[0028] Using the time axis corresponding to the leading edge shape of the compressed pulse as a reference, the amplitude data is read point by point. The starting time position of the main peak of the echo is determined by entering the continuous rising section through the stable leading edge region. The ending time position of the main peak of the echo is determined by the peak node and the continuous falling section, forming the main peak time segment. A one-to-one correspondence between the time nodes before compression and the time nodes after compression is established.

[0029] By using the time node correspondence, each time segment within the main peak time zone is scanned to identify candidate segments of low-amplitude reflection signals that contain the complete amplitude rise and fall process. The sampling nodes of the candidate segments are then shifted to outside the main peak time zone to form independent time segments.

[0030] By keeping the positions of the leading edge segment and the main peak time segment unchanged in the compressed pulse leading edge morphology, the independent time segments formed by translation are connected and arranged in the original time sequence to obtain the stripped low amplitude reflection signal.

[0031] Preferably, the identification of candidate segments of low-amplitude reflected signals is based on the complete waveform process in which an independent amplitude rises and falls back to the amplitude level of the adjacent time segment within the main peak time segment, and the independent time segment is located outside the main peak time segment while keeping the original time interval of the sampling node unchanged.

[0032] Preferably, identifying sections with localized insulation anomalies includes the following steps:

[0033] Read the amplitude values ​​of the low-amplitude reflected signal at each time point after stripping, and convert them into propagation distance positions through the correspondence between time and propagation distance. Arrange them in order of propagation distance to form an amplitude change trajectory.

[0034] Retrieve the propagation distance position corresponding to each time node of the echo peak width change data to form the peak width evolution trajectory, and align it with the amplitude change trajectory on the same propagation distance coordinate;

[0035] Compare the amplitude change trajectory with the main peak width evolution trajectory segment by segment along the propagation distance direction, and mark the segments where the amplitude change direction is inconsistent with the main peak width change direction;

[0036] Adjacent marked sections with continuous amplitude variation trends are merged to determine the continuous propagation distance interval, forming local insulation abnormality sections.

[0037] Preferably, the marked segment is a segment whose amplitude change trajectory changes unidirectionally within the continuous propagation distance interval and whose main peak width evolution trajectory remains continuously changing. The continuous propagation distance interval is an interval formed by the uninterrupted connection of adjacent marked segments on the propagation distance coordinate.

[0038] Preferably, the dynamic correction of the echo time coordinates under the constraint of the distance-based mapping relationship includes the following steps:

[0039] Record the starting and ending propagation distances of the local insulation anomaly section, extract the corresponding echo time nodes, and form a distance-time corresponding data sequence in ascending order of propagation distance;

[0040] Based on the distance-time correspondence data sequence, the correspondence between propagation distance and time offset is constructed to form a distance back-mapping relationship covering the local insulation anomaly section;

[0041] Time nodes are read point by point along the echo time axis and converted into propagation distance positions. For propagation distance positions located in local insulation abnormality sections, time replacement processing is performed according to the distance back-mapping relationship to obtain corrected time nodes and convert them into propagation distance intervals, forming fault location intervals.

[0042] A detection system for cable fault location includes a continuous time sequence construction module, a broadening and compression control module, a main peak rearrangement and stripping module, an abnormal section identification module, and a mapping correction location module.

[0043] Continuous timing construction module: acquires the trajectory data of the leading edge change of nanosecond-level pulses in the cable during the propagation process and the corresponding echo main peak width change data, and establishes a continuous time distribution map of high frequency attenuation advancing along the propagation distance;

[0044] The pulse broadening and compression control module extracts the pulse broadening and growth segment based on the continuous time distribution map, and performs compression control on the leading edge advance rhythm of the nanosecond-level pulse according to the pulse broadening and growth segment, to obtain the compressed pulse leading edge shape with a shortened main peak diffusion time segment.

[0045] Main peak rearrangement and stripping module: The time position of the echo main peak is rearranged based on the shape of the pulse leading edge after compression, and the low amplitude reflection signal that was inside the main peak before compression and modulation is separated into an independent time segment to obtain the stripped low amplitude reflection signal.

[0046] Abnormal Section Identification Module: Tracks the amplitude change trajectory of the low-amplitude reflected signal after stripping along the propagation distance, and compares and analyzes the amplitude change trajectory with the main peak width evolution trajectory corresponding to the echo main peak width change data to identify local insulation abnormal sections;

[0047] Mapping Correction and Location Module: Based on the local insulation abnormality section, a distance back-calculation mapping relationship is established. Under the constraint of the distance back-calculation mapping relationship, the echo time coordinate is dynamically corrected to obtain the fault location range.

[0048] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0049] This invention establishes a continuous time distribution map of high-frequency attenuation along the propagation distance, and based on this, compresses and modulates the pulse broadening and growth segment, effectively shortening the main peak diffusion time segment and restoring the time resolution capability of nanosecond-level pulses. By time-rearranging and stripping the low-amplitude reflection signal within the main peak, the low-amplitude reflection caused by early insulation degradation can be presented in an independent time segment, thereby improving the ability to identify minute impedance disturbances, enhancing the detection sensitivity of early faults, and reducing the risk of hidden defects being covered and missed.

[0050] This invention establishes a distance-to-reverse mapping relationship based on local insulation anomaly sections, and dynamically corrects the echo time coordinates under the constraints of this mapping relationship to ensure consistency in the correspondence between propagation distance and echo time. By specifically correcting the changes in propagation parameters within the anomaly section, the spatial accuracy of fault location results is improved, and the location deviation caused by propagation speed drift is reduced, enabling a more stable determination of cable fault zones in complex laying environments. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0052] Figure 1 This is a flowchart of the cable fault location detection method of the present invention;

[0053] Figure 2 A flowchart illustrating the method for establishing a continuous time distribution map of high-frequency attenuation along the propagation distance according to the present invention;

[0054] Figure 3 This is a flowchart of the method for obtaining low-amplitude reflection signals after stripping according to the present invention;

[0055] Figure 4 This is a schematic diagram of the detection system for cable fault location according to the present invention. Detailed Implementation

[0056] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0057] This invention provides, for example Figures 1 to 3 The cable fault location detection method shown includes the following steps:

[0058] Step 1: Obtain the trajectory data of the leading edge of the nanosecond-level pulse during propagation and the corresponding echo peak width variation data in the aged nuclear-grade cable. Based on the leading edge trajectory data and the echo peak width variation data, establish a continuous time distribution map of high-frequency attenuation along the propagation distance.

[0059] The following steps are taken to establish a continuous time distribution map of high-frequency attenuation along the propagation distance:

[0060] Nanosecond-level pulses are injected into one end of an aging nuclear-grade cable, and their propagation status is continuously recorded throughout the entire process after the pulse enters the cable. The injected nanosecond-level pulse maintains a fixed starting time calibration point, and a unified time coordinate axis is established with this starting time as the zero point. During the pulse's propagation along the cable, the leading edge region is sampled segment by segment at equal time intervals. Each sampling point records the leading edge amplitude position and the time span of the rising segment at the corresponding moment, thus forming a time-sequential trajectory data of the leading edge change.

[0061] It should be noted that:

[0062] The amplitude position of the leading edge refers to the corresponding position of the amplitude on the time axis. More precisely, it is the state of signal amplitude change at a certain point in time corresponding to the pulse leading edge. Essentially, it reflects the distribution of amplitude changes over time and is used to describe the steepness and smoothness of the leading edge shape. The time span of the rising segment of the leading edge refers to the time length parameter, that is, the length of the time interval from the beginning of the rising segment to the point of stabilization or peak, and is used to characterize the duration of the rising process. The former focuses on the manifestation of amplitude change characteristics on the time axis and belongs to the description of the correspondence between amplitude and time; the latter focuses on the time range occupied by the rising process and belongs to the pure time scale parameter. Both are used to characterize the dynamic evolution of the pulse leading edge.

[0063] The leading edge trajectory data is stored sequentially from shortest to longest propagation time, ensuring that each time interval corresponds to a complete leading edge morphology description. Simultaneously, echo signals generated during pulse propagation are acquired synchronously within the same time coordinate system. The time interval between the start and end points of the main peak is located within each echo signal, and its length is recorded as the echo main peak width. Each echo main peak width record is bound to its corresponding propagation time, guaranteeing a one-to-one correspondence between the leading edge trajectory data and the echo main peak width variation data on the time axis. Through this continuous acquisition method, a leading edge trajectory data sequence covering the entire propagation process and an echo main peak width variation data sequence are obtained, both sets of data based on the same zero point.

[0064] After obtaining complete data sequences of leading edge change trajectory and echo peak width change, the two sets of data are synchronously arranged and combined. Starting from the moment the nanosecond-level pulse enters the cable along the time axis, the leading edge change trajectory data recorded in each time segment is paired with the corresponding echo peak width change data for that time segment, forming data pairs arranged in chronological order.

[0065] Subsequently, each time segment on the time axis is converted into a corresponding segment along the cable propagation distance according to a predetermined propagation speed calibration relationship, thus establishing a corresponding mapping between time segments and propagation distance segments. After mapping, in an arrangement structure with propagation distance as the longitudinal advancement direction, the leading edge change trajectory data and echo peak width change data within the corresponding propagation distance segment are jointly labeled, ensuring that each propagation distance segment simultaneously contains numerical information on the degree of leading edge expansion and the width of the echo peak within that segment. By arranging segment by segment, a distance advancement sequence covering the entire length of the cable is formed, while maintaining the time unfolding order within this sequence, ensuring that time and distance maintain a corresponding relationship within the same expression structure.

[0066] It should be noted that:

[0067] The established propagation speed calibration relationship refers to the mapping rule used to describe the correspondence between the time axis and the propagation distance. Its essence originates from the pre-determining process of pulse propagation speed in the cable. Specifically, this calibration relationship establishes a correspondence between time and actual propagation distance by injecting pulses into a cable segment of known length and recording the echo time position, thus obtaining the propagation distance increment per unit time. In subsequent processing, each time segment on the time axis can be converted into a corresponding propagation distance segment through this calibration relationship, giving the time information spatial location meaning. This relationship is expressed as a one-to-one pairing sequence of time nodes and propagation distance nodes, used to ensure consistency between the time unfolding process and the spatial progression process, so that the distribution of the echo signal on the time axis can be accurately mapped to the actual propagation path position in the cable.

[0068] The degree of pulse front extension refers to the quantitative result of the morphological change of the pulse front as it gradually flattens from an initial steep state during propagation. It is mainly used to describe the broadening of the pulse front along the time axis. Specifically, it is defined as the length of the time interval from the initial rise point to reaching a stable amplitude or peak value within a certain propagation distance, and the smoothness of the amplitude change during this rise. The degree of pulse front extension can be obtained by analyzing the amplitude distribution at multiple time points in the pulse front trajectory data, such as recording the time position differences corresponding to the starting point, middle rise, and near peak stage, thus reflecting the trend of the front transitioning from sharp to smooth. This parameter is used to characterize the degree of influence of high-frequency component attenuation on the pulse shape and, together with the echo peak width, is used in a continuous-time distribution plot to characterize the broadening evolution state during pulse propagation.

[0069] After arranging the data according to time and distance, the entire data structure is continuously expanded to construct a continuous time distribution map of high-frequency attenuation along the propagation distance. During the construction process, the leading edge change trajectory data and echo peak width change data corresponding to each propagation distance segment are read sequentially along the propagation distance direction. The leading edge rise time extension and peak width change within the same propagation distance segment are displayed side by side, so that the two types of data within the same propagation distance segment form a joint representation.

[0070] The rise-leader time extension refers to the change in the extent of the rise of a nanosecond-level pulse along the time axis as its leading edge expands from its initial rise to near-stable amplitude during propagation. Specifically, within each propagation distance segment, the time difference between the initial rise time of the leading edge and the time corresponding to the leading edge reaching its stable amplitude or peak value is determined using leading edge trajectory data. This time difference is taken as the length of the rising-leader process and compared with the rising-leader time length in the initial injection state, thus reflecting the degree of extension of the rising-leader process during propagation. This extension manifests as the stretching of the rising-leader segment along the time axis, describing the gradual slowing of the leading edge morphology due to the attenuation of high-frequency components as the propagation distance increases. It is an important temporal characteristic parameter for characterizing pulse broadening evolution.

[0071] Subsequently, all propagation distance segments were continuously stitched together in order from near to far, so that the data on the trajectory of the leading edge change showed a continuous trend as the propagation distance increased, and the data on the change of the width of the echo main peak also formed a continuous evolution trajectory as the propagation distance increased. Through this continuous unfolding method along the distance direction, a continuous time distribution map was obtained with the propagation distance as the vertical axis and time as the horizontal axis.

[0072] This continuous time distribution map fully illustrates the evolution of the leading edge broadening corresponding to the segmented attenuation of high-frequency components during the propagation of nanosecond-level pulses in aging nuclear-grade cables, as well as the change in the width of the echo peak as the propagation distance increases. Because the leading edge trajectory data and the echo peak width change data maintain a strict temporal and distance correspondence within the map, this continuous time distribution map presents both the leading edge evolution in the time dimension and the high-frequency attenuation progression in the distance dimension, providing a unified and continuous temporal reference structure for subsequent identification and processing of pulse broadening growth segments.

[0073] Step 2: Extract the pulse broadening and growth segment based on the continuous time distribution map, and compress and regulate the leading edge advance rhythm of the nanosecond pulse according to the pulse broadening and growth segment to obtain the compressed pulse leading edge shape with a shortened main peak diffusion time segment.

[0074] The specific steps to obtain the morphology of the compressed pulse front after the main peak diffusion time segment is shortened are as follows:

[0075] In the established continuous-time distribution map, the echo peak width values ​​corresponding to each propagation distance segment are read segment by segment, arranged vertically in the direction of increasing propagation distance, and then arranged into a continuous width sequence according to the propagation time. During the reading process, the peak width difference between two adjacent propagation distance segments is compared segment by segment. When the peak width values ​​of three or more consecutive propagation distance segments are greater than the peak width value of the previous segment, and the peak width difference is continuously increasing, the time position corresponding to the initial propagation distance segment is determined as the starting point of the pulse broadening growth segment; when the peak width value stops increasing continuously and enters a stable range, the corresponding propagation distance segment is determined as the end point of the pulse broadening growth segment.

[0076] Entering the stable interval refers to the transition of the echo peak width value along the propagation distance from a continuously increasing state to a state where the fluctuation amplitude is limited and no longer continues to increase. Specifically, the determination method is as follows: within several consecutive propagation distance segments, the change in the peak width value relative to the previous segment no longer maintains a positive increasing relationship, but instead exhibits small fluctuations between adjacent segments, with this fluctuation remaining within a preset time offset. Furthermore, there are no longer instances where the width value of multiple consecutive segments is greater than that of the previous segment. When the change in peak width changes from a continuously increasing state to this fluctuating state with no obvious growth trend and limited change, it can be determined that the stable interval has been entered, and this position is taken as the termination boundary of the pulse broadening growth process.

[0077] Simultaneously, the leading edge change trajectory data within the corresponding time segment is retrieved from the same continuous time distribution map. The time span of the rising segment of the leading edge is measured segment by segment. When the time span of the rising segment of the leading edge coincides with the main peak width growth interval on the time axis, this time segment is finally determined as the pulse broadening growth interval. Through this dual correspondence between the echo main peak width change data and the leading edge change trajectory data, the pulse broadening growth interval is accurately delineated.

[0078] After determining the pulse broadening and growth segment, the compression control start time is set at the beginning of this segment, and the compression control end time is set at the end of this segment. The leading edge progression rhythm of the nanosecond-level pulse within this time segment is then rearranged. Specifically, the original time sampling nodes within the pulse broadening and growth segment are extracted in chronological order, and the time interval between every two adjacent sampling nodes is recorded. This time interval is then shortened one by one to a fixed time span consistent with the leading edge time interval before the pulse broadening and growth segment. Subsequently, the shortened time intervals are redistributed to the original sampling nodes, compressing the position of each sampling node on the time axis forward. This shortens the overall length of the pulse broadening and growth segment on the time axis while maintaining the original sampling node order. By compressing and adjusting each sampling node one by one, the leading edge progression rhythm of the nanosecond-level pulse within the pulse broadening and growth segment is restored to the progression rhythm before entering the broadening state.

[0079] It should be noted that:

[0080] The leading edge advance rhythm refers to the density and variation of the time distribution among consecutive sampling nodes as the nanosecond-level pulse leading edge advances backward from the initial rising point on the time axis. Essentially, it reflects the time interval structure corresponding to the amplitude change per unit time during the leading edge rise process. The leading edge time interval is specifically defined as the time difference between two adjacent sampling nodes on the time axis in the leading edge change trajectory data. This time difference is directly obtained through sampling and recording, that is, a sequence of sampling nodes arranged in chronological order has been formed during the construction of the continuous time distribution map. Each node has a corresponding time mark, and the difference between the time marks of adjacent nodes is the leading edge time interval.

[0081] During implementation, a time segment with stable frontal changes is first selected before the pulse broadening and growth segment. The time difference between adjacent sampling nodes within this segment is statistically analyzed and taken as the reference time interval. Then, the time intervals of the original adjacent sampling nodes within the pulse broadening and growth segment are uniformly replaced with this reference time interval, and new time node positions are generated by re-accumulating them in chronological order. This achieves the compression and adjustment of the frontal advancement rhythm, ensuring that the time expansion structure within the broadening segment remains consistent with that before broadening.

[0082] After compressing and adjusting the time positions of the sampling nodes within the pulse broadening and growth segment, the compressed time segment is sequentially connected with the time segment preceding the pulse broadening and growth segment. Specifically, all sampling nodes preceding the pulse broadening and growth segment retain their original time positions. Then, the compressed and adjusted sampling nodes are arranged sequentially, ensuring that the starting node of the compressed segment immediately follows the ending node of the segment preceding the broadening and growth segment, and guaranteeing that there is no time overlap or intersection between the sampling nodes.

[0083] By continuously splicing the preceding and following segments, a new time progression curve is constructed, so that the leading edge of the nanosecond pulse presents an overall structure on the time axis as follows: the initial segment maintains the original rhythm, the middle segment compresses the rhythm, and the subsequent segment extends continuously. This structure maintains a monotonically increasing arrangement on the time axis and preserves the original propagation order.

[0084] After rearranging and connecting the time segments, the new time progression curve is expanded as a whole, and the time diffusion range of the echo main peak corresponding to the compressed nanosecond pulse is remeasured. In practice, the time span between the start and end points of the echo main peak is recalibrated based on the compressed time progression curve, and this time span is compared and recorded with the width of the main peak before compression.

[0085] Since the time expansion length of the pulse broadening growth segment has been shortened, the diffusion range of the echo peak on the time axis is reduced accordingly, resulting in a compressed pulse front shape with a shortened peak diffusion time segment. This compressed pulse front shape maintains a continuous arrangement on the time axis while completely preserving the original sampling node sequence, providing a compressed and regulated time structure for subsequent rearrangement around the echo peak time position.

[0086] Step 3: Rearrange the time position of the main peak of the echo based on the shape of the pulse leading edge after compression, and separate the low amplitude reflection signal that was inside the main peak before compression into an independent time segment to obtain the stripped low amplitude reflection signal.

[0087] The specific steps for obtaining the low-amplitude reflection signal after stripping are as follows:

[0088] After compression and modulation, the compressed pulse leading edge pattern is used as a reference. Amplitude data is read point-by-point along the time axis, starting from the initial time position of the compressed pulse leading edge, to precisely define the time segment where the echo peak is located. Specifically, when the amplitude on the time axis transitions from a stable leading edge region to a continuously rising region, the starting time node of this rising region is recorded as the starting time position of the echo peak. When the amplitude reaches its peak and begins to enter a continuously falling region, the time node corresponding to the peak value is recorded. When the amplitude decreases to the same level as the amplitude corresponding to the starting time position of the echo peak and enters a stable region, this time node is recorded as the ending time position of the echo peak. Through these three calibrations, a complete echo peak time segment is formed.

[0089] The leading-edge plateau region, the continuously rising region, the continuously falling region, and the plateau region are all signal characteristic intervals divided based on the amplitude change trend on the time axis.

[0090] The pulse front plateau region refers to the time segment before the pulse front begins or just as it enters propagation, during which the amplitude remains at the baseline level and the variation between multiple consecutive sampling nodes remains within a preset amplitude fluctuation range. The continuous rising region refers to the time segment where the amplitude increases unidirectionally from the pulse front plateau region, and the amplitude of each adjacent sampling node in this segment is greater than the amplitude of the previous node, without any reverse fluctuation. The continuous falling region refers to the time segment where the amplitude decreases unidirectionally over time after reaching the peak value, and the amplitude of each adjacent sampling node in this segment is less than the amplitude of the previous node. The plateau region refers to the stable state where the amplitude returns to a limited variation after the continuous falling region ends, and the amplitude variation between multiple consecutive sampling nodes remains within the same amplitude fluctuation range as the pulse front plateau region, thus forming a stable interval after the end of the main peak of the echo.

[0091] Subsequently, the original time axis before compression and regulation and the time axis after compression are mapped node by node. Specifically, a one-to-one time node correspondence table is established according to the sampling order, so that each sampling node inside the main peak before compression and regulation has a unique corresponding time position in the time axis after compression, thereby providing a clear time mapping relationship for subsequent time position rearrangement processing.

[0092] After identifying the main peak time segment and establishing the correspondence between time nodes before and after compression, the low-amplitude reflection signal within the main peak before compression is identified and time-shifted segment by segment. Specifically, within the main peak time segment of the pulse leading-out pattern after compression, segmented scanning is performed according to a fixed time step, and the amplitude change process within each time segment is fully recorded. When a complete waveform process occurs within a time segment where the amplitude rises to a local peak and then falls to the amplitude level of the adjacent time segment, that time segment is identified as a candidate segment for the low-amplitude reflection signal. Subsequently, using the aforementioned correspondence between time nodes before and after compression, the corresponding position of this candidate segment on the time axis before compression is found, confirming that it was within the main peak time segment before compression.

[0093] The amplitude level of adjacent time segments refers to the baseline amplitude state within the adjacent time segments on either side of the current time segment, serving as a reference boundary for determining whether an independent reflection waveform exists within that time segment. Specifically, on the time axis divided according to a fixed time step, each time segment contains several consecutive sampling nodes, and its amplitude can form a local range of variation. The amplitude level of adjacent time segments is taken from the representative values ​​of the relatively flat amplitude variation intervals within the two adjacent time segments before and after the current time segment, typically represented as the average amplitude or a stable amplitude interval at the boundary of that time segment.

[0094] When the amplitude within a certain time segment rises from the amplitude level of the adjacent time segment, forms a local peak within that time segment, and then falls back to near the amplitude level of the adjacent time segment, it indicates that there is a complete independent amplitude fluctuation process within that time segment, which can be used as a basis for determining low amplitude reflection signals.

[0095] After confirmation, all sampling nodes of the candidate segment in the compressed time axis are shifted backward in their original order. The shift distance is set to the length of the idle time segment reserved after the end time of the main peak, so that the candidate segment is completely separated from the main peak time segment and forms an independent time segment. During the shift, the time interval between each sampling node within the candidate segment remains unchanged, so that the original time structure of the low-amplitude reflected signal is preserved.

[0096] The reserved idle time interval refers to the length of the time range on the time axis following the end time of the main echo peak, which has not yet been occupied by any echo signal. This time range serves as the carrying interval for the migrated low-amplitude reflection signal. Specifically, after defining the main peak time interval, a continuous time interval with a stable baseline amplitude and no significant amplitude fluctuations is selected by reading from the end time of the main peak along the time axis. The time span of this interval is taken as the idle time interval length. This length should be sufficient to fully accommodate the overall translation of all sampling nodes within the candidate interval while maintaining the original time interval, ensuring that the migrated low-amplitude reflection signal is completely separated from the main peak time interval on the time axis, without overlapping with other subsequent signals, thus guaranteeing its independent representation and the continuity of subsequent analysis.

[0097] After the overall translation of the low-amplitude reflection signal segment is completed, the entire time axis is rearranged and continuously connected. The specific operation process is as follows: keeping the time positions of the leading edge segment and the main peak time segment in the compressed pulse leading edge shape unchanged, the translated low-amplitude reflection signal segment is arranged immediately after the end time position of the main peak according to its original time sequence, and appropriate time intervals are inserted between its subsequent unadjusted time segments so that the entire time axis is still arranged in a unidirectional increasing order from front to back.

[0098] Inserting an appropriate time interval refers to the transition time length artificially introduced between the low-amplitude reflection signal segment and its subsequent unadjusted time segment after the low-amplitude reflection signal segment has completed its overall translation and is immediately following the end time position of the main peak. This is done to ensure that there is no overlap or temporal disorder between signal segments in the entire time axis. Specifically, this time interval is achieved by extending the time interval between two adjacent segments on the time axis. The principle for selecting its length is to neither change the original time interval between each sampling node within the low-amplitude reflection signal segment, nor to ensure that there is a clear separation between the end time position of this segment and the start time position of the subsequent unadjusted time segment. This ensures that the entire time axis remains unidirectionally increasing and that the boundaries of each segment are clear after rearrangement, thereby guaranteeing the continuity and distinguishability of the signal structure.

[0099] After completing the above arrangement, the new timeline is read as a whole to confirm that the main peak time segment and the low-amplitude reflection signal time segment are completely separated on the timeline and that there is no overlap between them. Through this time position rearrangement, the low-amplitude reflection signal, which was located inside the main peak before compression and modulation, is successfully separated into an independent time segment, thus obtaining the stripped low-amplitude reflection signal. This stripped low-amplitude reflection signal has an independent expression on the timeline, with a clear and definite start and end point, providing stable time positioning conditions for subsequent continuous tracking of its amplitude change trajectory along the propagation distance.

[0100] Step 4: Track the amplitude change trajectory of the low-amplitude reflected signal along the propagation distance after stripping, and compare the amplitude change trajectory with the main peak width evolution trajectory corresponding to the echo main peak width change data to identify local insulation abnormality sections.

[0101] The specific steps for identifying sections with localized insulation abnormalities are as follows:

[0102] After time rearranging the low-amplitude reflected signal and obtaining the stripped low-amplitude reflected signal, each sampling node of the stripped low-amplitude reflected signal within an independent time segment is read point by point. The amplitude value corresponding to each sampling node is recorded in chronological order, and a previously established correspondence table between time and propagation distance is used to accurately convert each time node into its corresponding propagation distance position. For example, the propagation distance value corresponding to each fixed time unit on the time axis is looked up in the table and recorded one by one, so that each amplitude value corresponds to a specific propagation distance value.

[0103] After the conversion of all sampling nodes is completed, these amplitude values ​​are rearranged in ascending order of propagation distance, forming a continuous amplitude data sequence along the propagation distance direction. In this amplitude data sequence, each propagation distance position corresponds to a specific amplitude value, thus constructing the amplitude variation trajectory of the stripped low-amplitude reflected signal along the propagation distance direction. This amplitude variation trajectory completely covers the propagation distance range corresponding to the stripped low-amplitude reflected signal, while maintaining consistency between the time order and the distance order.

[0104] After forming the amplitude variation trajectory of the stripped low-amplitude reflected signal, the echo peak width variation data previously recorded in the continuous time distribution map is read. Following the same time-to-propagation distance correspondence as described above, each echo peak width value is converted into its corresponding propagation distance position. Specifically, for each time node recorded in the peak width variation data, a lookup table is performed to obtain the corresponding propagation distance value, and the peak width value is then bound to this propagation distance value.

[0105] After the conversion is completed, the peak width values ​​are arranged in ascending order of propagation distance, forming a continuous peak width evolution trajectory along the propagation distance direction. Subsequently, the peak width evolution trajectory and the amplitude change trajectory of the stripped low-amplitude reflected signal are aligned with the same propagation distance coordinates, so that corresponding amplitude values ​​and peak width values ​​exist simultaneously at the same propagation distance position, thereby establishing a correspondence between the two trajectories on a unified distance axis.

[0106] After aligning the two trajectories by propagation distance, starting from the initial position of the propagation distance, each propagation distance segment is analyzed segment by segment according to a fixed distance step size. Within each propagation distance segment, the amplitude value of the corresponding stripped low-amplitude reflected signal and the corresponding peak width value are read, and the amplitude value of the current segment is compared with the amplitude value of the previous propagation distance segment, and the peak width value of the current segment is also compared with the peak width value of the previous propagation distance segment. When the peak width value remains within a stable range of variation within several consecutive propagation distance segments, while the amplitude value shows continuous increases or decreases between adjacent segments, the consecutive segments are marked as amplitude variation abnormal segments. Furthermore, when the direction of amplitude change and the direction of peak width change show different trends within the same propagation distance segment, the segment is recorded separately to distinguish it from the overall attenuation caused only by changes in peak width.

[0107] The stable variation range refers to the specific boundary for determining the range within which the peak width value remains stable. It is determined based on constraints on the variation amplitude of the peak width within a continuous propagation distance segment. Specifically, when reading the peak width value segment by segment, a continuous propagation distance segment is selected, and the difference in peak width between adjacent segments is calculated. When this difference falls within a pre-defined time variation tolerance range for multiple consecutive segments, and there is no continuous unidirectional increasing or decreasing trend, this continuous segment can be defined as the stable variation range.

[0108] This time variation tolerance range is usually obtained by statistically analyzing the non-broadening section data collected in the early stage. It is used to reflect the natural fluctuation level of the main peak width under normal propagation conditions, so that the stability range has a clear quantitative criterion, rather than relying on subjective judgment.

[0109] By comparing segment by segment, the difference between the amplitude change trajectory and the main peak width evolution trajectory is clearly marked on the propagation distance axis.

[0110] After comparing all propagation distance segments segment by segment, the marked amplitude variation anomaly segments are continuously integrated. Specifically, adjacent anomaly segments with continuous amplitude variation trends are merged into a single complete interval, and the starting and ending propagation distance positions of this complete interval are recorded. When the propagation distance interval between multiple anomaly segments is less than a predetermined distance unit, these segments are merged to form a continuous propagation distance interval. After merging, each continuous interval is identified as a local insulation anomaly segment and marked on the propagation distance axis.

[0111] The predetermined distance unit is a spatial interval standard used to determine whether multiple abnormal segments need to be merged. It is set based on the relationship between the cable detection resolution and the sampling interval. Specifically, on the propagation distance axis, each distance segment corresponds to a fixed time sampling interval. By calibrating the relationship between time and propagation distance, the smallest resolvable distance unit can be obtained. This unit reflects the smallest spatial interval that the system can distinguish between two independent reflection events under the current sampling conditions.

[0112] During implementation, this minimum resolvable distance is used as a benchmark, and a distance threshold is set based on actual detection needs. When the interval between adjacent abnormal segments is less than this distance threshold, they are considered to belong to the same abnormal influence area in space, and thus a merging process is performed. This preset method ensures that the division of abnormal segments conforms to both signal resolution capability and consistent spatial judgment criteria.

[0113] Through the above continuous implementation process, the amplitude change trajectory of the low-amplitude reflected signal after stripping was tracked along the propagation distance direction. The amplitude change trajectory was compared and analyzed with the main peak width evolution trajectory corresponding to the echo main peak width change data. Finally, local insulation abnormality sections were identified, providing a clear propagation distance range basis for subsequent establishment of distance back-calculation mapping relationship and dynamic correction of time coordinate.

[0114] Step 5: Establish a distance-back mapping relationship based on the local insulation anomaly section, and dynamically correct the echo time coordinates under the constraints of the distance-back mapping relationship to obtain the fault location range;

[0115] The echo time coordinates are dynamically corrected under the constraint of the distance-based mapping relationship to obtain the fault location range. The specific steps are as follows:

[0116] Within the identified local insulation anomaly sections, the starting and ending propagation distances of each section are precisely recorded, and all corresponding echo time nodes within that propagation distance section are extracted one by one. Specifically, based on the previously established time-to-propagation distance correspondence table, the time node corresponding to the starting propagation distance of the local insulation anomaly section is located on the echo time axis and marked as the starting time node of the anomaly section; subsequently, the time node corresponding to the ending propagation distance is located and marked as the ending time node of the anomaly section. Next, between the starting and ending time nodes, each sampling time node is read in the original time sequence, and the propagation distance value corresponding to each time node is confirmed again, forming a set of distance-time corresponding data sequences arranged in ascending order of propagation distance.

[0117] For two adjacent propagation distance positions in the data sequence, the time interval between their corresponding time nodes is recorded, and these time intervals are listed one by one in order of distance. This allows the actual correspondence between propagation distance and echo time in the entire local insulation anomaly section to be fully expressed, thereby constructing the original data set for establishing the distance-back mapping relationship.

[0118] After obtaining the complete distance-time correspondence data sequence within the local insulation anomaly section, the sequence is continuously reconstructed to establish a mapping relationship between propagation distance and echo time that can be directly used for back-calculation. Specifically, the starting propagation distance position of the local insulation anomaly section is used as a reference starting point, and its corresponding time node is set as a zero-offset reference point. Then, the time nodes corresponding to each subsequent propagation distance position are read sequentially in ascending order of propagation distance. For each propagation distance position, the reference time node is subtracted from its corresponding time node to obtain the actual time offset of that propagation distance position relative to the starting propagation distance position. Subsequently, these time offsets are arranged one by one in ascending order of propagation distance, forming a curve showing the correspondence between propagation distance and time offset. This curve is then extended to the ending propagation distance position of the local insulation anomaly section, so that each propagation distance unit within this section corresponds to a rearranged time offset value, thereby constructing a complete distance back-calculation mapping relationship. This distance back-calculation mapping relationship maintains a unidirectional ascending arrangement on the propagation distance axis and is continuous and uninterrupted throughout the entire local insulation anomaly section.

[0119] After the distance-to-propagation mapping relationship is established, the entire echo time coordinate is dynamically corrected point by point. Specifically, each time node is read sequentially along the echo time axis, starting from the initial time node, and converted into its corresponding propagation distance position according to the time-to-propagation distance correspondence table. If the propagation distance position is within a local insulation anomaly zone, the time offset value corresponding to that distance position is found on the propagation distance axis according to the established distance-to-propagation mapping relationship, and this time offset value replaces the original time node, thus rearranging the time axis within that propagation distance zone according to the distance-to-propagation mapping relationship. If the propagation distance position is not within a local insulation anomaly zone, the original time node remains unchanged. By performing the above replacement and retention operations on all time nodes one by one, the echo time coordinates are orderly adjusted within the local insulation anomaly zone, while maintaining their original arrangement outside the anomaly zone.

[0120] After completing the dynamic correction process, the corrected time nodes are converted back into propagation distance positions, and the propagation distance range covered by the corresponding time segment is read to finally obtain the fault location interval. This fault location interval, based on the time coordinates constrained by the distance-based mapping relationship, reflects the true position of the local insulation abnormality section in the propagation distance direction, thus achieving accurate location of the cable fault.

[0121] This invention establishes a continuous time distribution map of high-frequency attenuation along the propagation distance, and based on this, compresses and modulates the pulse broadening and growth segment, effectively shortening the main peak diffusion time segment and restoring the time resolution capability of nanosecond-level pulses. By time-rearranging and stripping the low-amplitude reflection signal within the main peak, the low-amplitude reflection caused by early insulation degradation can be presented in an independent time segment, thereby improving the ability to identify minute impedance disturbances, enhancing the detection sensitivity of early faults, and reducing the risk of hidden defects being covered and missed.

[0122] This invention establishes a distance-to-reverse mapping relationship based on local insulation anomaly sections, and dynamically corrects the echo time coordinates under the constraints of this mapping relationship to ensure consistency in the correspondence between propagation distance and echo time. By specifically correcting the changes in propagation parameters within the anomaly section, the spatial accuracy of fault location results is improved, and the location deviation caused by propagation speed drift is reduced, enabling a more stable determination of cable fault zones in complex laying environments.

[0123] This invention provides, for example Figure 4 The cable fault location detection system shown includes a continuous time sequence construction module, a broadening and compression control module, a main peak rearrangement and stripping module, an abnormal section identification module, and a mapping correction location module.

[0124] Continuous timing construction module: acquires the trajectory data of the leading edge change of nanosecond-level pulses in the cable during the propagation process and the corresponding echo main peak width change data, and establishes a continuous time distribution map of high frequency attenuation advancing along the propagation distance;

[0125] The pulse broadening and compression control module extracts the pulse broadening and growth segment based on the continuous time distribution map, and performs compression control on the leading edge advance rhythm of the nanosecond-level pulse according to the pulse broadening and growth segment, to obtain the compressed pulse leading edge shape with a shortened main peak diffusion time segment.

[0126] Main peak rearrangement and stripping module: The time position of the echo main peak is rearranged based on the shape of the pulse leading edge after compression, and the low amplitude reflection signal that was inside the main peak before compression and modulation is separated into an independent time segment to obtain the stripped low amplitude reflection signal.

[0127] Abnormal Section Identification Module: Tracks the amplitude change trajectory of the low-amplitude reflected signal after stripping along the propagation distance, and compares and analyzes the amplitude change trajectory with the main peak width evolution trajectory corresponding to the echo main peak width change data to identify local insulation abnormal sections;

[0128] Mapping Correction and Location Module: Based on the local insulation abnormality section, a distance back-calculation mapping relationship is established. Under the constraint of the distance back-calculation mapping relationship, the echo time coordinate is dynamically corrected to obtain the fault location range.

[0129] The present invention provides a cable fault location detection method, which is implemented by the above-mentioned cable fault location detection system. The specific method and process of the cable fault location detection system are detailed in the above-mentioned embodiment of the cable fault location detection method, and will not be repeated here.

[0130] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A detection method for locating cable faults, characterized in that, include: Acquire the leading edge trajectory data of the pulse in the cable during the propagation process and the corresponding echo main peak width change data, and establish a continuous time distribution map of the high frequency attenuation along the propagation distance; Based on the continuous time distribution map, the pulse broadening and growth segment is extracted, and the leading edge advancement rhythm of the nanosecond pulse is compressed and controlled according to the pulse broadening and growth segment to obtain the compressed pulse leading edge morphology with a shortened main peak diffusion time segment. The time position of the echo main peak is rearranged based on the shape of the pulse leading edge after compression. The low-amplitude reflection signal that was inside the main peak before compression is separated into an independent time segment to obtain the stripped low-amplitude reflection signal. The amplitude change trajectory of the low-amplitude reflected signal after stripping is tracked along the propagation distance, and the amplitude change trajectory is compared and analyzed with the main peak width evolution trajectory corresponding to the echo main peak width change data to identify local insulation abnormality sections. Based on the local insulation anomaly section, a distance-back mapping relationship is established. Under the constraint of the distance-back mapping relationship, the echo time coordinate is dynamically corrected to obtain the fault location range.

2. The detection method for cable fault location according to claim 1, characterized in that, Establishing a continuous-time distribution map of high-frequency attenuation along the propagation distance includes the following steps: Nanosecond-level pulses are injected into one end of the aging nuclear-grade cable. The pulse front change trajectory data are collected at fixed time intervals under a unified time coordinate axis, and the time interval length between the start and end points of each echo peak is recorded simultaneously to form echo peak width change data. The data on the trajectory of the leading edge change and the data on the change of the width of the main peak of the echo are paired and arranged in chronological order, and each time segment is converted into a corresponding propagation distance segment; The data on the trajectory of the leading edge and the width of the echo peak in each propagation distance segment are read sequentially along the propagation distance direction and continuously expanded to construct a continuous time distribution map with the propagation distance as the vertical axis and time as the horizontal axis.

3. The detection method for cable fault location according to claim 2, characterized in that, The data on the trajectory of the leading edge change and the data on the width change of the main peak of the echo form a one-to-one correspondence under a unified time coordinate axis, and the time order remains unchanged when converted into the propagation distance segment.

4. The detection method for cable fault location according to claim 2, characterized in that, Obtaining the morphology of the compressed pulse front after a shortened main peak diffusion time segment includes the following steps: In the continuous time distribution map, the echo peak width values ​​are read segment by segment along the propagation distance direction and continuously compared. The pulse broadening growth segment is determined by the overlap between the continuous increase of the peak width and the time span of the rising edge. Time sampling nodes are extracted with the time start and end points corresponding to the pulse broadening growth segment as the range. The time interval between adjacent sampling nodes is shortened and redistributed to each sampling node to form a compressed time segment. The compressed time segment is sequentially arranged with the time segment preceding the pulse broadening and growth segment to construct a new time progression curve while keeping the sampling node order unchanged. The time span between the start and end points of the echo main peak is recalibrated based on the new time progression curve, forming the compressed pulse front shape.

5. The detection method for cable fault location according to claim 4, characterized in that, The pulse broadening growth segment is defined by the synchronous change relationship between the continuous increase in the width of the echo main peak and the time span of the rising segment of the leading edge in the trajectory of the leading edge change. The shortening of the time sampling node maintains the original sampling order, and the time segment of the echo main peak is redefined by the new time advancement curve.

6. The detection method for cable fault location according to claim 4, characterized in that, Obtaining the low-amplitude reflection signal after stripping includes the following steps: The starting time of the main echo peak is determined by entering the continuous rising section from the leading edge stable zone, and the ending time of the main echo peak is determined by the peak node and the continuous falling section, thus forming the main peak time segment, and establishing a one-to-one correspondence between the time nodes before compression and the time nodes after compression. By using the time node correspondence, each time segment within the main peak time zone is scanned to identify candidate segments of low-amplitude reflection signals that contain the complete amplitude rise and fall process. The sampling nodes of the candidate segments are then shifted to outside the main peak time zone to form independent time segments. By keeping the positions of the leading edge segment and the main peak time segment unchanged in the compressed pulse leading edge morphology, the independent time segments formed by translation are connected and arranged in the original time sequence to obtain the stripped low amplitude reflection signal.

7. The detection method for cable fault location according to claim 6, characterized in that, The identification of candidate segments for low-amplitude reflected signals specifically includes: using the complete waveform process in which an independent amplitude rises within the main peak time segment and then falls back to the amplitude level of the adjacent time segment as the judgment condition, and the independent time segment is located outside the main peak time segment while keeping the original time interval of the sampling node unchanged.

8. The detection method for cable fault location according to claim 6, characterized in that, Identifying sections of localized insulation abnormalities includes the following steps: Read the amplitude values ​​of the low-amplitude reflected signal at each time point after stripping, and convert them into propagation distance positions through the correspondence between time and propagation distance. Arrange them in order of propagation distance to form an amplitude change trajectory. Retrieve the propagation distance position corresponding to each time node of the echo peak width change data to form the peak width evolution trajectory, and align it with the amplitude change trajectory on the same propagation distance coordinate; Compare the amplitude change trajectory and the main peak width evolution trajectory segment by segment along the propagation distance direction, and mark the segments where the amplitude change direction and the main peak width change direction are inconsistent; Adjacent marked sections with varying amplitude trends are merged to determine the continuous propagation distance interval, forming local insulation abnormality sections.

9. The detection method for cable fault location according to claim 8, characterized in that, The marked segment is a segment whose amplitude change trajectory changes unidirectionally within the continuous propagation distance interval, and whose main peak width evolution trajectory remains unchanged. The continuous propagation distance interval is the interval formed by the uninterrupted connection of adjacent marked segments on the propagation distance coordinate.

10. The detection method for cable fault location according to claim 8, characterized in that, The dynamic correction of echo time coordinates under the constraint of distance inverse mapping includes the following steps: Record the starting and ending propagation distances of the local insulation anomaly section, extract the corresponding echo time nodes, and form a distance-time corresponding data sequence in ascending order of propagation distance; Based on the distance-time correspondence data sequence, the correspondence between propagation distance and time offset is constructed to form a distance back-mapping relationship covering the local insulation anomaly section; Time nodes are read point by point along the echo time axis and converted into propagation distance positions. For propagation distance positions located in local insulation abnormality sections, time replacement processing is performed according to the distance back-mapping relationship to obtain corrected time nodes and convert them into propagation distance intervals, forming fault location intervals.

11. A detection system for cable fault location, used to implement the detection method for cable fault location according to any one of claims 1-10, characterized in that, It includes a continuous time series construction module, a broadening and compression control module, a main peak rearrangement and stripping module, an abnormal segment identification module, and a mapping correction and localization module; Continuous timing construction module: acquires the trajectory data of the leading edge change of nanosecond-level pulses in the cable during the propagation process and the corresponding echo main peak width change data, and establishes a continuous time distribution map of high frequency attenuation advancing along the propagation distance; The pulse broadening and compression control module extracts the pulse broadening and growth segment based on the continuous time distribution map, and performs compression control on the leading edge advance rhythm of the nanosecond-level pulse according to the pulse broadening and growth segment, to obtain the compressed pulse leading edge shape with a shortened main peak diffusion time segment. Main peak rearrangement and stripping module: The time position of the echo main peak is rearranged based on the shape of the pulse leading edge after compression, and the low amplitude reflection signal that was inside the main peak before compression and modulation is separated into an independent time segment to obtain the stripped low amplitude reflection signal. Abnormal Section Identification Module: Tracks the amplitude change trajectory of the low-amplitude reflected signal after stripping along the propagation distance, and compares and analyzes the amplitude change trajectory with the main peak width evolution trajectory corresponding to the echo main peak width change data to identify local insulation abnormal sections; Mapping Correction and Location Module: Based on the local insulation abnormality section, a distance back-calculation mapping relationship is established. Under the constraint of the distance back-calculation mapping relationship, the echo time coordinate is dynamically corrected to obtain the fault location range.

Images