A pipeline fault accurate positioning method and system based on signal gradient

Abstract

The application relates to the technical field of pipeline fault positioning, and discloses a pipeline fault accurate positioning method and system based on a signal gradient, which comprises the following steps: filtering and removing a direct current component of a direct current background voltage signal of a starting reference position of a target pipeline to obtain reference data; performing differential operation on induced current signals of preset discrete position points based on the data to obtain signal gradient characteristic values; constructing a signal gradient atlas according to the signal gradient characteristic values, and marking space regions corresponding to polarity reversal points and disorder starting points as fault reference intervals; detecting potential difference of interval voltage signals of continuous position points in the intervals to obtain extreme characteristic points; and finally, performing space coordinate mapping on the extreme characteristic points to output accurate positioning points of pipeline faults; and the application can improve the efficiency of the pipeline fault accurate positioning based on the signal gradient.

Description

Technical Field

[0001] This invention relates to the field of pipeline fault location technology, and in particular to a method and system for accurate pipeline fault location based on signal gradient. Background Technology

[0002] As a critical infrastructure for energy transmission and municipal supply, the operational status of pipelines directly affects transmission safety and supply stability. Pipeline fault location technology, by collecting characteristic data such as electrical and magnetic signals along the pipeline, identifies and pinpoints fault points. It is a core technical means to ensure the continuous and reliable operation of pipelines, and accurate and efficient fault location can significantly reduce repair costs and downtime losses.

[0003] Existing pipeline fault location methods mostly rely on single signal amplitude detection or simple differential calculations, lacking a systematic extraction and analysis of signal gradient features. This makes it difficult to accurately identify signal polarity reversals and amplitude disturbances, hindering the rapid identification of the fault reference range. Furthermore, the lack of integration with potential difference extreme value detection for precise coordinate mapping results in insufficient fault location accuracy and low efficiency, failing to meet the demands for rapid and accurate fault diagnosis in complex pipeline scenarios. Therefore, improving the accuracy and efficiency of pipeline fault location has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method and system for accurate pipeline fault location based on signal gradient, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for precise pipeline fault location based on signal gradient, comprising: Ⅰ. Filter the DC background voltage signal at the starting reference position of the target pipeline for background noise, and eliminate the DC component of the filtered data to obtain the reference reference data of the DC background voltage signal; II. Based on the aforementioned reference data, differential calculation is performed on the induced current signal at preset discrete locations of the target pipeline, and polarity constraint correction is applied to the calculation result to obtain the signal gradient characteristic value of the induced current signal. III. Construct a signal gradient map based on the preset discrete location points and signal gradient feature values, and mark the spatial regions corresponding to the polarity reversal points and disorder initiation points in the signal gradient map as fault reference intervals based on the consistency between the gradient slope of the signal gradient map and the continuous points. IV. Detect the potential difference of the interval voltage signal at continuous locations within the fault reference interval, and perform dual feature fusion determination by combining the gradient features of the signal gradient spectrum with the potential difference features of the interval voltage signal to obtain the extreme feature points of the interval voltage signal. V. Perform spatial coordinate mapping on the extreme value feature points to obtain the spatial coordinates of the extreme value feature points, and output the spatial coordinates as the precise location points of pipeline faults.

[0006] In a preferred embodiment, the step of filtering the DC background voltage signal at the starting reference position of the target pipeline for background noise and eliminating the DC component of the filtered data to obtain the reference reference data of the DC background voltage signal includes: The DC background voltage signal at the starting reference position of the target pipeline is subjected to power frequency notch filtering to obtain the first filtered signal after processing the first filtered signal; Based on the target pipeline burial depth and soil moisture, the high-pass filter cutoff frequency is positively and adaptively adjusted, and the first filtered signal is high-pass filtered to obtain a second filtered signal that eliminates low-frequency drift components. The second filtered signal is subjected to sliding window mean filtering to obtain the zero-mean signal of the second filtered signal, and the zero-mean signal is used as the reference data of the DC background voltage signal.

[0007] In a preferred embodiment, the step of performing differential calculations on the induced current signals at preset discrete locations of the target pipeline based on the reference data, and correcting the polarity constraints on the calculation results to obtain the signal gradient feature values ​​of the induced current signals, includes: Based on a preset fixed sampling interval, the induced current signal in the extension direction of the target pipeline is discretized and extracted to obtain the original sampling sequence of the induced current signal. The relative offset compensation is performed between the induced current value of the sampling point in the original sampling sequence and the reference data, and the spatial position is synchronously calibrated on the compensated data to obtain the corrected current value of the sampling point. The original difference value of the sampling point is obtained by performing a difference operation between the non-first sampling point in the original sampling sequence and the corrected current value of the adjacent sampling point. Based on the spatial order of the sampling points, the original difference values ​​are serialized and recombined to obtain the original difference sequence of the original sampling sequence.

[0008] In a preferred embodiment, the step of performing differential calculations on the induced current signals at preset discrete locations of the target pipeline based on the reference data, and correcting the polarity constraints on the calculation results to obtain the signal gradient feature values ​​of the induced current signals, includes: The original difference value in the original difference sequence and the fixed sampling interval are divided to obtain the unit spacing difference value of the sampling points; Based on spatial topological order, coordinate serialization mapping is performed on the unit spacing difference values ​​of the original difference sequence to obtain the unit spacing difference sequence of the original difference sequence; Based on the first unit spacing difference sequence Based on the unit interval difference value and polarity abruptness constraint, the weighted gradient correction value of the sampling point is calculated using the following formula: ; in, Indicates the sequential numbering of the sampling points. In the unit spacing difference sequence, the first... The unit interval difference value of each sampling point Indicates the first digit in the original difference sequence. The absolute value of the original difference values ​​of each sampling point This represents the maximum absolute value of all original difference values ​​in the original difference sequence. This represents the preset weight adjustment coefficient, which is a constant with a value greater than 0 and less than 1. Represents a symbolic function; The weighted gradient correction value is assigned a value to obtain the signal gradient feature value of the sampling point; The signal gradient feature values ​​of all sampling points in the unit spacing difference sequence are sequentially merged along the path according to the spatial distribution order to obtain the signal gradient feature values ​​of the induced current signal.

[0009] In a preferred embodiment, constructing the signal gradient map based on the preset discrete location points and signal gradient feature values ​​includes: The spatial coordinates of the preset discrete location points are interpreted to obtain the spatial location dataset of the preset discrete location points; The signal gradient feature values ​​are extracted to obtain the gradient feature value dataset of the signal gradient feature values. Based on the spatial location extension direction of the spatial location dataset, feature space coupling is performed on the spatial location dataset and the gradient feature value dataset to obtain a gradient location association pair sequence of the spatial location dataset and the gradient feature value dataset; The gradient position association sequence is mapped to a two-dimensional coordinate system to obtain the signal gradient spectrum of the spatial position dataset and the gradient feature value dataset. The spatial coordinates of the spatial position dataset are used as the horizontal axis scale and the gradient feature values ​​of the gradient feature value dataset are used as the vertical axis magnitude.

[0010] In a preferred embodiment, marking the spatial regions corresponding to polarity reversal points and disorder initiation points in the signal gradient spectrum as fault reference intervals based on the consistency of the gradient slope and continuous points in the signal gradient spectrum includes: The polarity reversal point set of the signal gradient spectrum is obtained by performing a sign traversal on the locations where the signs of the gradient feature values ​​in the signal gradient spectrum are reversed. Amplitude fluctuation detection is performed on the starting points where the amplitude of the gradient feature value in the signal gradient spectrum continuously fluctuates beyond a preset fluctuation threshold and the gradient change slope exceeds a slope threshold, thereby obtaining the set of disordered starting points of the signal gradient spectrum; Spatial coordinate interpretation is performed on the polarity reversal point set to obtain the reversal point spatial coordinate set of the polarity reversal point set; The spatial coordinates of the disorder initiation points and their adjacent reversal points in the disorder initiation point set are marked with continuous region boundary closure, and false abnormal regions are eliminated based on the consistency check of three consecutive points, so as to obtain the fault reference interval of the disorder initiation point set and the reversal point spatial coordinate set.

[0011] In a preferred embodiment, the step of detecting the potential difference of the interval voltage signal at consecutive locations within the fault reference interval, and performing dual-feature fusion determination of the gradient features of the signal gradient spectrum and the potential difference features of the interval voltage signal to obtain the extreme feature points of the interval voltage signal, includes: Boundary anchoring is performed on the starting and ending endpoints within the fault reference interval to obtain the interval boundary coordinates of the fault reference interval; Based on a preset spatial step interval, voltage signals along the line are picked up at continuous position points between the boundary coordinates of the interval to obtain a continuous voltage sampling sequence of the fault reference interval. Potential difference quantization is performed on adjacent sampling point pairs in the continuous voltage sampling sequence to obtain the potential difference value of the adjacent point pairs; Based on the sampling order of the continuous voltage sampling sequence, a situation sequence is constructed for the potential difference value to obtain the potential difference distribution sequence of the fault reference interval.

[0012] In a preferred embodiment, the step of detecting the potential difference of the interval voltage signal at consecutive locations within the fault reference interval, and performing dual-feature fusion determination of the gradient features of the signal gradient spectrum and the potential difference features of the interval voltage signal to obtain the extreme feature points of the interval voltage signal, includes: A global traversal is performed on the position indices of adjacent values ​​with opposite signs in the potential difference distribution sequence to obtain the extreme value candidate point index set of the potential difference distribution sequence; The absolute value of the potential difference corresponding to adjacent indices in the extreme value candidate point index set is selected by the magnitude of the potential difference, and the index with the largest magnitude is retained to obtain the extreme value candidate point index set maximum value index set. Using the spatial coordinates of the sampling points corresponding to the index in the maximum point index set as the index spatial position, a candidate set of extreme point spatial coordinates of the maximum point index set is obtained; The gradient feature value corresponding to the candidate set of extreme point spatial coordinates is fused with the potential difference feature of the interval voltage signal to obtain the true fault point set of the candidate set of extreme point spatial coordinates. The coordinates of the real fault points in the set are traced and merged to calculate the fault confidence of the candidate points in the set of real fault points. The calculation formula is as follows: ; in, For the first Fault confidence of each candidate point The potential difference is a quantitative value. This is the confidence level adjustment coefficient, 0 < <1, Represents a symbolic function; The point with unique spatial coordinates and the largest absolute value of the corresponding potential difference is taken as the extreme value feature point of the interval voltage signal.

[0013] In a preferred embodiment, the step of mapping the extreme feature points to spatial coordinates to obtain their spatial coordinates, and outputting these spatial coordinates as precise location points for pipeline faults, includes: Align the extreme feature points with the reference domain to obtain the reference system registration coordinates of the extreme feature points; The reference frame registration coordinates are globally translated to obtain the absolute spatial coordinates of the extreme feature points; The absolute spatial coordinates are arranged into information frames to obtain the localization result data of the extreme value feature points; The location result data is released through the output port to obtain the precise location point of the target pipeline fault.

[0014] To address the aforementioned problems, the present invention also provides a pipeline fault accurate location system based on signal gradient, the system comprising: The reference data generation module is used to filter the background noise of the DC background voltage signal at the starting reference position of the target pipeline, and to eliminate the DC component of the filtered data to obtain the reference reference data of the DC background voltage signal. The gradient feature extraction module is used to perform differential operations on the induced current signals at preset discrete locations of the target pipeline based on the reference data, and to perform polarity constraint correction on the operation results to obtain the signal gradient feature values ​​of the induced current signals. The fault interval marking module is used to construct a signal gradient map based on the preset discrete location points and signal gradient feature values, and mark the spatial regions corresponding to polarity reversal points and disorder initiation points in the signal gradient map as fault reference intervals based on the consistency between the gradient slope of the signal gradient map and the continuous points. The extreme value feature detection module is used to detect the potential difference of the interval voltage signal at continuous location points within the fault reference interval, and to perform dual feature fusion judgment by combining the gradient feature of the signal gradient spectrum with the potential difference feature of the interval voltage signal to obtain the extreme value feature points of the interval voltage signal. A precise positioning output module is used to perform spatial coordinate mapping on the extreme feature points to obtain their spatial coordinates, and output these spatial coordinates as precise positioning points for pipeline faults. Compared with the prior art, the present invention has the following advantages: 1. This invention obtains stable and reliable reference data by performing multi-stage filtering and DC component elimination on the DC background voltage signal. Combined with differential operation and weighted gradient correction of the induced current signal, the signal gradient feature value can be accurately extracted, effectively improving the accuracy and stability of pipeline fault feature identification and ensuring the data reliability of the fault location process.

[0015] 2. This invention relies on signal gradient spectrum to quickly mark fault reference intervals, and locks extreme feature points by detecting potential difference within the fault interval. Then, it completes the accurate output of fault points through spatial coordinate mapping, which significantly shortens the pipeline fault location time, improves location efficiency and accuracy, and can directly output the absolute spatial coordinates of the fault point, meeting the practical application needs of rapid pipeline fault diagnosis and accurate emergency repair. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a method for precise pipeline fault location based on signal gradient, provided in an embodiment of the present invention. Figure 2 A functional block diagram of a pipeline fault accurate location system based on signal gradient provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0018] This application provides a method for precise pipeline fault location based on signal gradients. The execution entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for precise pipeline fault location based on signal gradients can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cluster of cloud servers. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0019] Reference Figure 1 The diagram shown is a flowchart illustrating a method for precise pipeline fault location based on signal gradients according to an embodiment of the present invention. In this embodiment, the method for precise pipeline fault location based on signal gradients includes: Ⅰ. Filter the DC background voltage signal at the starting reference position of the target pipeline for background noise, and eliminate the DC component of the filtered data to obtain the reference reference data of the DC background voltage signal; In this embodiment of the invention, the step of filtering the DC background voltage signal at the starting reference position of the target pipeline for background noise and eliminating the DC component of the filtered data to obtain the reference reference data of the DC background voltage signal includes: The DC background voltage signal at the starting reference position of the target pipeline is subjected to power frequency notch filtering to obtain the first filtered signal after processing the first filtered signal; Based on the target pipeline burial depth and soil moisture, the high-pass filter cutoff frequency is positively and adaptively adjusted, and the first filtered signal is high-pass filtered to obtain a second filtered signal that eliminates low-frequency drift components. The second filtered signal is subjected to sliding window mean filtering to obtain the zero-mean signal of the second filtered signal, and the zero-mean signal is used as the reference data of the DC background voltage signal.

[0020] After acquiring the DC background voltage signal corresponding to the starting reference position of the target pipeline, the DC background voltage signal is subjected to power frequency notch filtering according to a fixed suppression frequency band specifically set for power frequency interference. The interference components that match the power frequency band in the signal are completely removed, while the signal components corresponding to the other frequency bands are completely retained without modification. After this filtering process, the first filtered signal of the DC background voltage signal is directly generated.

[0021] Using a pre-set high-pass filter cutoff frequency based on the low-frequency drift characteristics of pipeline voltage signals as a fixed boundary standard, all signal components in the first filtered signal with frequencies lower than the preset high-pass filter cutoff frequency are blocked and retained. Only signal components with frequencies higher than the preset high-pass filter cutoff frequency are allowed to pass normally and retain the signal. By filtering and blocking through this fixed frequency boundary, various low-frequency drift components contained in the signal are eliminated. After processing, a second filtered signal with eliminated low-frequency drift components is generated.

[0022] Using a pre-defined fixed-length data acquisition interval as the unified interception range of the sliding window, starting from the starting position of the second filtered signal, the corresponding signal data is intercepted segment by segment according to the defined interval. All signal data within each intercepted window is subjected to overall numerical equalization and integration processing. After completing the data integration operation of all window intervals, the processed signals of each segment are sequentially arranged to make the overall data value of the second filtered signal stably converge to a unified reference value interval to form a zero-mean state. After segment-by-segment window interception and overall numerical equalization and integration processing, the zero-mean signal of the second filtered signal is obtained. The zero-mean signal obtained through the complete processing flow is directly determined as the reference data of the DC background voltage signal.

[0023] The beneficial effects of this technical solution are that, through a layered and progressive signal purification process, various interference components in the DC background voltage signal are gradually removed, enabling precise optimization of the purity and accuracy of the voltage signal at the initial reference position of the target pipeline. Power frequency notch filtering can specifically remove power frequency interference in fixed frequency bands, completely avoiding the superimposed pollution of the original DC background voltage signal by power frequency clutter, and ensuring the authenticity of the basic filtered signal. By relying on a preset fixed high-pass filter cutoff frequency to screen and eliminate low-frequency drift components, the low-frequency signal offset problem caused by environmental factors and equipment aging can be eliminated, stabilizing the frequency reference of the second filtered signal and preventing signal reference deviation caused by low-frequency drift. Zero-mean correction of the signal is achieved through fixed sliding window mean filtering, unifying the signal numerical reference, completely eliminating the DC offset of the signal, and obtaining standardized reference data. This layered processing method specifically addresses the technical problems of mixed interference, reference offset, and numerical imbalance in pipeline DC background voltage signals. It adopts fixed preset processing standards throughout the process, eliminating subjective bias and ensuring high reproducibility. This effectively improves the accuracy, stability, and uniformity of pipeline background voltage reference data, providing accurate, reliable, and interference-free basic reference data for subsequent technical operations such as pipeline voltage detection and fault diagnosis.

[0024] II. Based on the aforementioned reference data, differential calculation is performed on the induced current signal at preset discrete locations of the target pipeline, and polarity constraint correction is applied to the calculation result to obtain the signal gradient characteristic value of the induced current signal. In this embodiment of the invention, the step of performing differential calculations on the induced current signals at preset discrete locations of the target pipeline based on the reference data, and correcting the polarity constraints on the calculation results to obtain the signal gradient feature values ​​of the induced current signals, includes: Based on a preset fixed sampling interval, the induced current signal in the extension direction of the target pipeline is discretized and extracted to obtain the original sampling sequence of the induced current signal. The relative offset compensation is performed between the induced current value of the sampling point in the original sampling sequence and the reference data, and the spatial position is synchronously calibrated on the compensated data to obtain the corrected current value of the sampling point. The original difference value of the sampling point is obtained by performing a difference operation between the non-first sampling point in the original sampling sequence and the corrected current value of the adjacent sampling point. Based on the spatial order of the sampling points, the original difference values ​​are serialized and recombined to obtain the original difference sequence of the original sampling sequence.

[0025] Based on the reference data, the induced current signal at preset discrete locations of the target pipeline is subjected to differential calculation, and the result of the calculation is corrected for polarity constraints to obtain the signal gradient feature value of the induced current signal, including: The original difference value in the original difference sequence and the fixed sampling interval are divided to obtain the unit spacing difference value of the sampling points; Based on spatial topological order, coordinate serialization mapping is performed on the unit spacing difference values ​​of the original difference sequence to obtain the unit spacing difference sequence of the original difference sequence; Based on the first unit spacing difference sequence Based on the unit interval difference value and polarity abruptness constraint, the weighted gradient correction value of the sampling point is calculated using the following formula: ; in, Indicates the sequential numbering of the sampling points. In the unit spacing difference sequence, the first... The unit interval difference value of each sampling point Indicates the first digit in the original difference sequence. The absolute value of the original difference values ​​of each sampling point This represents the maximum absolute value of all original difference values ​​in the original difference sequence. This represents the preset weight adjustment coefficient, which is a constant with a value greater than 0 and less than 1. Represents a symbolic function; The weighted gradient correction value is assigned a value to obtain the signal gradient feature value of the sampling point; The signal gradient feature values ​​of all sampling points in the unit spacing difference sequence are sequentially merged along the path according to the spatial distribution order to obtain the signal gradient feature values ​​of the induced current signal.

[0026] A pre-calibrated and fixed sampling interval is used as the sole spatial sampling scale standard. This fixed sampling interval is a constant spatial distance preset to adapt to the target pipeline detection scenario, and its value does not change throughout the process. Along the linear extension of the target pipeline, the continuously distributed induced current signal within the entire pipeline area is collected and extracted by fixed-point discretization. All sampling points are evenly distributed according to the fixed spatial interval to ensure that all sampling points are spatially uniform and the sampling scale is consistent. The real-time induced current data corresponding to each preset discrete position point is fully recorded. The original induced current data corresponding to all uniformly sampled position points are uniformly integrated and arranged according to the pipeline extension spatial order to generate an original sampling sequence of induced current signals with complete structure and corresponding time and spatial position.

[0027] The reference data of the DC background voltage signal obtained after multi-stage filtering and zero-mean correction is used as the reference correction basis for the whole domain. This reference data is a standardized background reference data without interference or offset. The original induced current value corresponding to each spatial sampling point contained in the original sampling sequence is matched and calibrated one by one to accurately cancel the background reference offset component mixed in the induced current value of each sampling point, and eliminate the numerical offset problem caused by background noise and reference deviation. All sampling points are corrected using the same set of reference standards to ensure that the correction scale of all sampling points is completely uniform. Finally, the corrected current value corresponding to all sampling points in the original sampling sequence is obtained.

[0028] Following the predetermined pipeline spatial arrangement order of the original sampling sequence, and taking the first sampling point of the sequence as the starting boundary, after excluding the first sampling point without preceding comparison data, the difference calculation operation is performed on all remaining target sampling points with adjacent preceding sampling points. The corrected current value corresponding to a single target sampling point is independently calculated with the corrected current value corresponding to the preceding spatial sampling point that is immediately adjacent to the sampling point. The difference calculation work of all valid sampling points is completed point by point, and the original difference value corresponding to each sampling point is accurately obtained.

[0029] Strictly adhering to the actual spatial arrangement of all sampling points on the target pipeline, and using the pipeline extension direction as a fixed sorting benchmark, all original difference values ​​obtained through point-by-point calculation are matched and bound to the corresponding spatial location information of the sampling points. The discrete difference data is then reorganized and arranged in an orderly manner according to the spatial distribution logic, so that the arrangement order of all original difference values ​​completely corresponds to the spatial distribution order of the sampling points, forming an original difference sequence of the original sampling sequence with matching position and orderly order.

[0030] For each original difference value corresponding to a sampling point within the original difference sequence, a uniform and constant fixed sampling interval spatial scale value is used to perform an equal division conversion operation. The fixed spatial interval is used as a unified denominator benchmark to achieve standardized conversion between discrete current differences and actual pipeline spatial distances, eliminate data differences caused by different spatial scales, unify the spatial gradient calculation benchmark for all sampling points, and complete the spatial normalization conversion of all original difference values ​​point by point to obtain the unit spacing difference value corresponding to each sampling point.

[0031] Based on the pre-defined spatial topology distribution rules of the target pipeline detection scenario, and combined with the unique spatial coordinate information of each sampling point in the pipeline global coordinate system, the unit spacing difference value corresponding to all sampling points in the original difference sequence is accurately associated and bound with the spatial coordinates of the corresponding sampling points. This completes the serialization mapping and matching of difference data and spatial coordinates, so that each set of unit spacing difference data has a unique corresponding spatial location attribute, and finally generates a unit spacing difference sequence of the original difference sequence that has both numerical information and spatial location information.

[0032] The system sequentially retrieves each unit spacing difference value arranged in spatial order within the unit spacing difference sequence. Using the unit spacing difference data of a single point as the core calculation basis, and combined with the preset gradient weighting correction judgment rules and correction scale, it performs targeted fine-grained weighting compensation and correction processing on the initial gradient data of the corresponding sampling points. The correction process is executed entirely according to preset fixed rules, without any subjective adjustment, which can effectively adapt to the subtle deviations in gradient data caused by the spatial distribution characteristics of pipelines. The gradient correction calculation of all sampling points is completed point by point to obtain the weighted gradient correction value corresponding to each sampling point.

[0033] Establish a unique binding relationship between sampling point identifiers and gradient data. The weighted gradient correction value obtained by standardization calculation for each independent sampling point is accurately assigned and bound to the data storage unit and data identifier bit of the corresponding sampling point. This completes the solidification assignment operation of gradient correction data for all sampling points, so that each spatial sampling point corresponds to a set of independent and accurate gradient values. Finally, the unique signal gradient feature value corresponding to each sampling point is determined.

[0034] Strictly following the spatial extension distribution order of the target pipeline from the start end to the end, and using the spatial arrangement along the pipeline as the merging benchmark, the signal gradient feature values ​​of all sampling points in the unit spacing difference sequence that have been assigned and determined are sorted point by point, arranged in an orderly manner, and integrated as a whole. All discrete single-point gradient feature data are connected and integrated into a continuous and complete set of gradient data along the pipeline, and the arrangement logic of the gradient data in the whole domain is unified. Finally, the signal gradient feature values ​​of the induced current signal that completely covers the detection range of the target pipeline are obtained.

[0035] The sequential numbering of sampling points is derived from all sampling points laid out along the extension direction of the target pipeline. They are generated sequentially according to the spatial distribution order from the start end to the end end. Each sampling point corresponds to only one unique number, and the numbering order is completely consistent with the spatial distribution order of the sampling points in the pipeline.

[0036] The unit spacing difference value corresponding to the numbered sampling point in the unit spacing difference sequence is obtained by dividing the original difference value of the same numbered sampling point in the original difference sequence by a preset fixed sampling interval. This operation uses the fixed sampling interval as a unified denominator to achieve standardized conversion between current difference and pipeline space distance, generating difference data with unit spatial scale. Then, according to the spatial topological order of the sampling points, all converted difference values ​​are mapped to coordinate sequence to form a unit spacing difference sequence. The value at the corresponding numbered position in the sequence is the parameter.

[0037] The absolute value of the original difference value corresponding to the numbered sampling point in the original difference sequence is obtained by performing a difference operation on the corrected current value of the same numbered sampling point (excluding the first one) and the adjacent sampling point in the original sampling sequence. The absolute value of this difference is then processed to obtain a value that does not distinguish between positive and negative directions. The original difference sequence is generated by recombining the original difference values ​​of all sampling points in spatial order. The absolute value of the difference at the corresponding numbered position in the sequence is the parameter.

[0038] The maximum value of the absolute values ​​of all original difference values ​​in the original difference sequence is obtained by iterating through all the original difference values ​​corresponding to all sampling points in the original difference sequence, taking the absolute value of each difference value in turn, and then comparing them one by one to determine the result with the largest value. This maximum value is the fixed extreme value of the absolute values ​​of all differences in the sequence, with no room for subjective adjustment, and is completely objectively determined by the data in the sequence.

[0039] The preset weight adjustment coefficient is a constant with a value greater than 0 and less than 1. It is based on historical detection data and error control experience in pipeline induced current detection scenarios. The coefficient is set according to the influence weight of the unit spacing difference value and the current difference ratio in the balanced gradient correction. The value is constant and maintains a uniform standard throughout the detection process, without changing with the sampling point or sequence data.

[0040] The weighted gradient correction value for the sampling point is generated by multiplying the unit interval difference value by the correction term in parentheses. The correction term consists of 1 minus the product of the weight adjustment coefficient and the normalized current difference ratio. The normalized current difference ratio is the ratio of the absolute value of the current difference of a single sampling point to the maximum absolute value of the difference in the sequence. This correction term dynamically adjusts the weight of the gradient data according to the relative size of the current difference. The larger the current difference, the smaller the value of the correction term, and the stronger the suppression effect on the gradient data. Conversely, the smaller the current difference, the weaker the suppression effect. Finally, a weighted gradient correction value adapted to the current change characteristics is generated.

[0041] This formula achieves precise adaptation of gradient data to pipeline current variation characteristics by dynamically weighting the unit spacing difference value based on the proportion of current difference. Sampling points with larger current differences are assigned lower weights to weaken the impact of local abnormal fluctuations on the overall gradient characteristics, while sampling points with smaller current differences retain higher weights to highlight stable current variation trends. The final weighted gradient correction value can truly reflect the actual state of current variation along the pipeline, providing accurate and controllable basic data for the subsequent calculation of signal gradient feature values ​​of induced current signals.

[0042] The beneficial effects of this technical solution are that it relies on standardized reference data to extract gradient features from induced current signals. Fixed-interval discrete sampling ensures the uniformity and standardization of pipeline current data acquisition, avoiding data deviation problems caused by random sampling and guaranteeing that the original sampling sequence has a unified spatial scale. Compensation for current value offsets at each sampling point is achieved using reference data, completely eliminating the interference of DC background reference deviation on induced current data, ensuring that the sampled current values ​​closely match the actual operating state of the pipeline and improving the accuracy of the data. Through a hierarchical processing method involving adjacent sampling point difference calculation, spatial sequence recombination, unit interval conversion, and weighted correction, spatial quantification of induced current variation characteristics is achieved, abandoning the coarse-grained approach of traditional overall signal analysis and accurately capturing subtle current variation differences along the pipeline. Spatial coordinate mapping and gradient weighted correction processing adapt to the spatial topology distribution characteristics of the pipeline, correcting data deviations caused by spatial distribution and ensuring that gradient data closely matches the actual spatial variation patterns of the pipeline. Finally, the gradient feature data of the entire domain are merged in an orderly manner to form a complete gradient feature result along the pipeline, which effectively improves the accuracy, continuity and spatial matching of the pipeline induced current gradient features, and provides high-precision and reproducible quantitative data support for subsequent target pipeline defect location and status assessment.

[0043] III. Construct a signal gradient map based on the preset discrete location points and signal gradient feature values, and mark the spatial regions corresponding to the polarity reversal points and disorder initiation points in the signal gradient map as fault reference intervals based on the consistency between the gradient slope of the signal gradient map and the continuous points. In this embodiment of the invention, constructing a signal gradient map based on the preset discrete location points and signal gradient feature values ​​includes: The spatial coordinates of the preset discrete location points are interpreted to obtain the spatial location dataset of the preset discrete location points; The signal gradient feature values ​​are extracted to obtain the gradient feature value dataset of the signal gradient feature values. Based on the spatial location extension direction of the spatial location dataset, feature space coupling is performed on the spatial location dataset and the gradient feature value dataset to obtain a gradient location association pair sequence of the spatial location dataset and the gradient feature value dataset; The gradient position association sequence is mapped to a two-dimensional coordinate system to obtain the signal gradient spectrum of the spatial position dataset and the gradient feature value dataset. The spatial coordinates of the spatial position dataset are used as the horizontal axis scale and the gradient feature values ​​of the gradient feature value dataset are used as the vertical axis magnitude.

[0044] Based on the consistency between the gradient slope and continuous points in the signal gradient spectrum, the spatial regions corresponding to polarity reversal points and disorder initiation points in the signal gradient spectrum are marked as fault reference intervals, including: The polarity reversal point set of the signal gradient spectrum is obtained by performing a sign traversal on the locations where the signs of the gradient feature values ​​in the signal gradient spectrum are reversed. Amplitude fluctuation detection is performed on the starting points where the amplitude of the gradient feature value in the signal gradient spectrum continuously fluctuates beyond a preset fluctuation threshold and the gradient change slope exceeds a slope threshold, thereby obtaining the set of disordered starting points of the signal gradient spectrum; Spatial coordinate interpretation is performed on the polarity reversal point set to obtain the reversal point spatial coordinate set of the polarity reversal point set; The spatial coordinates of the disorder initiation points and their adjacent reversal points in the disorder initiation point set are marked with continuous region boundary closure, and false abnormal regions are eliminated based on the consistency check of three consecutive points, so as to obtain the fault reference interval of the disorder initiation point set and the reversal point spatial coordinate set.

[0045] The entire spatial coordinate interpretation process is performed on all the pre-deployed discrete location points. Based on the actual laying topology and point layout rules of the target pipeline, the precise spatial coordinate information corresponding to each pre-deployed discrete location point is extracted. The spatial coordinate data of all points are uniformly collected and integrated to generate a spatial location dataset of pre-deployed discrete location points containing spatial information of all detection points.

[0046] The signal gradient feature values ​​of the complete induced current signal obtained after the completion of the process are standardized and extracted. The gradient feature values ​​corresponding to all discrete sampling points in the entire pipeline detection range are completely extracted. All gradient feature values ​​are classified, organized and integrated to generate a gradient feature value dataset containing the signal gradient feature values ​​of all points.

[0047] Using the spatial extension direction corresponding to the actual laying trajectory of the target pipeline as the sole coupling and matching benchmark, the spatial layout logic of the points within the spatial location dataset and the feature value layout logic within the gradient feature value dataset are associated and bound one-to-one. This ensures that each set of spatial location data is precisely matched with the gradient feature data of the corresponding point, completing the dimensional alignment and feature fusion of the two datasets. This achieves feature spatial coupling between spatial dimension data and feature dimension data, generating a gradient location association sequence of spatial location dataset and gradient feature value dataset where the location and feature are precisely matched.

[0048] Using a conventional two-dimensional rectangular coordinate system as the mapping carrier, the gradient position correlation pair sequence is processed by global coordinate mapping and assignment. All point spatial coordinate data contained in the spatial position dataset are uniformly set as the horizontal axis scale of the two-dimensional coordinate system, and all gradient feature values ​​contained in the gradient feature value dataset are uniformly set as the vertical axis amplitude of the two-dimensional coordinate system. The coordinate points of all correlation data are plotted and arranged and integrated according to the spatial distribution order of the points, generating a signal gradient map of both the spatial position dataset and the gradient feature value dataset, which have both spatial position attributes and gradient amplitude attributes.

[0049] The gradient feature values ​​corresponding to all coordinate points within the formed signal gradient map are verified point by point. Using zero as the criterion for determining positive and negative attributes, all position points where the gradient feature values ​​within the map change from positive to negative or vice versa are selected. All target points that meet the conditions for reversing positive and negative attributes are uniformly collected and organized to form a polarity reversal point set of the signal gradient map.

[0050] Using a fixed amplitude fluctuation threshold preset according to the pipeline inspection conditions as the sole criterion, the amplitude change status of all gradient feature values ​​within the signal gradient spectrum is continuously detected segment by segment. The first coordinate position where the continuous change amplitude of the gradient feature value within the spectrum exceeds the fixed preset fluctuation threshold is used as the criterion to screen out all starting coordinate points that meet the abnormal amplitude fluctuation conditions, and integrate all target points to form a set of disordered starting points of the signal gradient spectrum.

[0051] Accurate spatial coordinate interpretation is performed on all polarity reversal points included in the polarity reversal point set. The accurate two-dimensional spatial coordinate information corresponding to each polarity reversal point is extracted one by one. The coordinate data of all reversal points are uniformly collected and standardized to generate a reversal point spatial coordinate set that completely covers all polarity anomaly points.

[0052] Using all the disorder initiation points recorded within the disorder initiation point set as the initial boundary of the region, and combining the spatial coordinate information of adjacent polarity reversal points contained in the reversal point spatial coordinate set, the continuous spatial region between the two types of points is marked with a global boundary closure, locking the complete spatial range between the disorder fluctuation initiation position and the polarity reversal position, completing the precise delineation and boundary solidification of the abnormal region, and generating a fault reference interval that simultaneously covers the disorder initiation point set and the reversal point spatial coordinate set of the abnormal region and the polarity reversal abnormal region.

[0053] The beneficial effects are as follows: By interpreting the spatial coordinates of preset discrete locations, the spatial attribute information of pipeline detection points can be completely organized, forming a standardized spatial location dataset, providing a unified spatial benchmark for subsequent map construction. By extracting the eigenvalues ​​of signal gradient features, global gradient numerical information can be systematically collected, constructing a well-organized gradient feature value dataset, ensuring the integrity and orderliness of the feature data. Based on the pipeline's spatial extension direction, the feature space of the two datasets is coupled, achieving a precise one-to-one association between spatial location and gradient features, avoiding data mismatch problems, and forming a logically rigorous gradient location association sequence. A signal gradient map is constructed through a two-dimensional coordinate system mapping, with spatial coordinates as the horizontal axis and gradient feature values ​​as the vertical axis, intuitively and concretely presenting the gradient change pattern along the pipeline, facilitating intuitive identification of numerical change trends. By traversing and filtering the polarity reversal point set point by point and obtaining the disordered starting point set based on a preset fluctuation threshold, and using fixed judgment criteria to complete the identification of abnormal points, the bias caused by subjective judgment is avoided. Then, the spatial coordinates of the abnormal points are interpreted and the boundaries of the continuous area are closed to accurately delineate the fault reference interval, realize the automated and standardized delineation of the pipeline abnormal area, improve the standardization and reproducibility of fault interval location, and provide reliable map data and interval reference basis for subsequent accurate judgment of pipeline faults and investigation of hidden dangers.

[0054] IV. Detect the potential difference of the interval voltage signal at continuous locations within the fault reference interval, and perform dual feature fusion determination by combining the gradient features of the signal gradient spectrum with the potential difference features of the interval voltage signal to obtain the extreme feature points of the interval voltage signal. In this embodiment of the invention, the step of detecting the potential difference of the interval voltage signal at consecutive locations within the fault reference interval, and performing dual-feature fusion determination of the gradient features of the signal gradient spectrum and the potential difference features of the interval voltage signal to obtain the extreme value feature points of the interval voltage signal includes: Boundary anchoring is performed on the starting and ending endpoints within the fault reference interval to obtain the interval boundary coordinates of the fault reference interval; Based on a preset spatial step interval, voltage signals along the line are picked up at continuous position points between the boundary coordinates of the interval to obtain a continuous voltage sampling sequence of the fault reference interval. Potential difference quantization is performed on adjacent sampling point pairs in the continuous voltage sampling sequence to obtain the potential difference value of the adjacent point pairs; Based on the sampling order of the continuous voltage sampling sequence, a situation sequence is constructed for the potential difference value to obtain the potential difference distribution sequence of the fault reference interval.

[0055] The step of detecting the potential difference of the interval voltage signal at continuous locations within the fault reference interval, and performing dual-feature fusion determination of the gradient features of the signal gradient spectrum and the potential difference features of the interval voltage signal to obtain the extreme value feature points of the interval voltage signal includes: A global traversal is performed on the position indices of adjacent values ​​with opposite signs in the potential difference distribution sequence to obtain the extreme value candidate point index set of the potential difference distribution sequence; The absolute value of the potential difference corresponding to adjacent indices in the extreme value candidate point index set is selected by the magnitude of the potential difference, and the index with the largest magnitude is retained to obtain the extreme value candidate point index set maximum value index set. Using the spatial coordinates of the sampling points corresponding to the index in the maximum point index set as the index spatial position, a candidate set of extreme point spatial coordinates of the maximum point index set is obtained; The gradient feature value corresponding to the candidate set of extreme point spatial coordinates is fused with the potential difference feature of the interval voltage signal to obtain the true fault point set of the candidate set of extreme point spatial coordinates. The coordinates of the real fault points in the set are traced and merged to calculate the fault confidence of the candidate points in the set of real fault points. The calculation formula is as follows: ; in, For the first Fault confidence of each candidate point The potential difference is a quantitative value. This is the confidence level adjustment coefficient, 0 < <1, Represents a symbolic function; The point with unique spatial coordinates and the largest absolute value of the corresponding potential difference is taken as the extreme value feature point of the interval voltage signal.

[0056] For the fault reference interval whose boundaries have been delineated, the exclusive spatial start and end points within the spatial coverage area of ​​the interval are accurately located. In strict accordance with the standardized spatial coordinate calibration specifications corresponding to the predetermined laying trajectory of the target pipeline, the coordinate information of the two endpoints is accurately analyzed and located and solidified. The independent spatial coordinate attributes of the two endpoints are completely preserved. The analyzed start and end point coordinates are uniformly collected and organized according to the interval range definition logic to generate the interval boundary coordinates of the fault reference interval with a clear spatial definition range.

[0057] Using a pre-calibrated spatial step spacing specifically adapted to the refined fault detection scenario of the target pipeline and maintained constant throughout, as the benchmark for the uniform travel sampling scale across the entire domain, the system strictly follows the actual linear laying path of the target pipeline. Within the closed space strictly defined by the boundary coordinates of the interval, continuous detection position points are sequentially and equidistantly deployed according to the predetermined fixed step size. Simultaneously, in-situ voltage signal acquisition and pickup are performed at each deployed continuous position point. Throughout the process, the acquisition rules, sampling scale, and pickup standards are completely consistent, and the original voltage signal data corresponding to all continuous position points within the fault reference interval are fully collected. Then, all acquired data are systematically connected and integrated according to the spatial extension arrangement of the pipeline from the start end to the end end, generating a continuous voltage sampling sequence of the fault reference interval with a regular arrangement and complete point correspondence.

[0058] Following the natural spatial arrangement of sampling points within the continuous voltage sampling sequence, adjacent sampling points are selected in pairs to form fixed adjacent sampling point pairs. Using the same reference voltage standard as the basis for numerical comparison, the original voltage signal values ​​of each pair of adjacent sampling points are precisely compared and the potential difference is quantified and verified to fully characterize the actual difference in voltage potential between adjacent points. The same set of quantization judgment standards is used throughout the process without any artificial adjustments. The potential difference quantization process is completed sequentially for all adjacent sampling point pairs that meet the pairing conditions within the sequence. After the verification is completed, the potential difference value of each pair of adjacent points is obtained.

[0059] Strictly adhering to the fixed sampling sequence of the continuous voltage sampling sequence extending from the beginning to the end of the fault reference interval, all potential difference values ​​obtained through individual quantization are sequentially arranged according to the actual spatial distribution of their corresponding sampling points on the pipeline. This fully restores the continuous trend and evolution of potential differences along the fault reference interval, completing the orderly organization, logical sorting, and overall situation reconstruction of various types of potential difference data in discrete states, forming a potential difference distribution sequence of the fault reference interval that can fully reflect the potential change pattern of the interval.

[0060] According to the predetermined order of the values ​​within the potential difference distribution sequence, the two sets of potential difference values ​​arranged immediately before and after each value are retrieved and their positive and negative attributes are compared and verified. The change of the positive and negative attributes of the values ​​is used as a fixed and unique criterion for point selection. All positions in the entire potential difference distribution sequence are thoroughly traversed and screened without omission. All position identifiers that meet the condition of opposite signs of adjacent position values ​​are accurately selected. All qualified position identifiers are uniformly collected, classified and arranged in an orderly manner, and a set of extreme value candidate point indexes covering all potential extreme value positions of the potential difference distribution sequence is formed.

[0061] For adjacent position identifiers with related arrangements within the extreme value candidate point index set, the potential difference value corresponding to each position identifier is retrieved one by one. Pure numerical amplitude discrimination and selection work is carried out without the influence of positive and negative attributes. The actual numerical magnitude without distinguishing direction is used as the sole selection criterion. Among the potential difference values ​​corresponding to each group of mutually adjacent candidate identifiers, the position identifier with the first position in the pure numerical amplitude level is fixed and retained. According to the unified selection rules, the comparison, discrimination and selection of all extreme value candidate identifiers are completed group by group. After regularization and integration, the extreme value candidate point index set and the maximum value point index set are generated after screening and purification.

[0062] The spatial coordinate information of the continuous voltage sampling sequence sampling points corresponding to each location identifier within the maximum point index set is retrieved one by one. The obtained spatial coordinates of the sampling points are directly established as the exclusive index spatial position of the corresponding location identifier and the data binding association is completed. This ensures that each location identifier and spatial coordinate form a unique matching relationship. The coordinate association matching work of all identifier points in the maximum point index set is completed one by one. All matched and bound spatial coordinate data are uniformly collected, organized and archived in an orderly manner to generate a candidate set of extreme point spatial coordinates of the maximum point index set with complete point information and accurate coordinate matching.

[0063] Based on the linear trajectory extension path formed by the actual laying of the target pipeline, the coordinates of all spatial coordinate information included in the candidate set of extreme point spatial coordinates are merged and integrated along the same path. Redundant coordinate data with overlapping and repetitive positions are eliminated according to the spatial position overlap judgment criteria. Only the valid coordinate points that are spatially independent and non-repeating are retained. Within the range of the unique spatial coordinate corresponding points after deduplication and screening, the magnitude of the pure value after removing the positive and negative attributes of the corresponding potential difference is used as the final judgment benchmark. The point with the highest pure value level is selected as the final judgment result and is standardized as the extreme feature point of the interval voltage signal that can characterize the core features of the abnormal change of interval voltage.

[0064] The beneficial effects are as follows: By anchoring the endpoints of the fault reference interval and generating interval boundary coordinates, the spatial range of fault detection can be accurately locked, providing a fixed boundary benchmark for subsequent voltage signal acquisition and avoiding data omissions or redundancy caused by arbitrary delineation of the sampling range. Relying on a preset constant spatial step interval to pick up voltage signals along the line ensures a uniform and regular distribution of sampling points within the fault interval, constructing a continuous voltage sampling sequence corresponding to time sequence and spatial height, improving the standardization and reproducibility of the basic sampling data. Potential difference quantization processing of adjacent sampling point pairs accurately quantifies the potential difference between points, objectively restoring the degree of voltage potential change within the interval, and providing standardized numerical basis for subsequent sequence construction. Reconstructing the potential difference distribution sequence according to the sampling order can completely reproduce the potential change trend along the fault interval, intuitively presenting the potential fluctuation pattern, and providing an orderly data carrier for extreme value point screening. By traversing the entire domain and filtering the inverted sign positions to obtain the extreme value candidate point index set, potential extreme value points are locked with a unified judgment standard, avoiding subjective bias caused by manual identification. A maximum value index set is obtained by selecting and refining the candidate indexes based on their amplitude, effectively eliminating invalid and redundant candidates and accurately focusing on core points with prominent amplitudes. The spatial coordinates of the sampling points are correlated and matched to form a candidate set of extreme value spatial coordinates, achieving precise binding between numerical features and spatial locations. By merging and deduplicating traces and selecting the optimal amplitude points to determine extreme value feature points, the core abnormal locations of voltage signal changes in the fault area can be accurately located, providing reliable feature data and spatial location support for in-depth analysis of pipeline faults and precise location of potential hazards.

[0065] V. Perform spatial coordinate mapping on the extreme value feature points to obtain the spatial coordinates of the extreme value feature points, and output the spatial coordinates as the precise location points of pipeline faults; In this embodiment of the invention, the step of mapping the extreme value feature points to spatial coordinates to obtain the spatial coordinates of the extreme value feature points, and outputting the spatial coordinates as precise location points for pipeline faults, includes: Align the extreme feature points with the reference domain to obtain the reference system registration coordinates of the extreme feature points; The reference frame registration coordinates are globally translated to obtain the absolute spatial coordinates of the extreme feature points; The absolute spatial coordinates are arranged into information frames to obtain the localization result data of the extreme value feature points; The location result data is released through the output port to obtain the precise location point of the target pipeline fault.

[0066] Using the unified reference area established by the previous pipeline inspection as a fixed alignment standard, the extreme feature points of the determined interval voltage signal are included in the reference area to complete the position matching calibration. According to the preset reference system matching rules, the relative position of the extreme feature points is corrected and the position is rectified so that the position information of the extreme feature points is completely consistent with the predetermined reference domain. After the registration correction is completed, the reference system registration coordinates of the extreme feature points are generated.

[0067] Based on the global spatial reference system predefined in the target pipeline laying scenario, and using the relative position information carried by the reference system registration coordinates as the basis, the relative position information is converted into the corresponding fixed spatial position in the whole domain according to the unified global coordinate conversion standard. The reference attribute of the position information of the limited local reference system is converted into the unified position information attribute of the whole domain. After the global coordinate calibration of the position information is completed, the absolute spatial coordinates of the extreme feature points are obtained.

[0068] In accordance with the standard information frame format arrangement specifications for pipeline fault location data, the absolute spatial coordinates corresponding to extreme value feature points are incorporated into the fixed data frame structure to complete the regular arrangement and information classification. The coordinate information is encapsulated and structured in an orderly manner by strictly following the preset data field arrangement rules, forming extreme value feature point location result data with unified format, regular fields, and direct readability and parsing.

[0069] The packaged and arranged positioning result data is connected to a pre-set dedicated data output port. The positioning result data is stably pushed and completely output according to a fixed data release and transmission standard. The output location information directly corresponds to the actual spatial location of the pipeline. After being released and output through the port, it forms a precise location point that can be directly used to determine the fault of the target pipeline.

[0070] The beneficial effects are as follows: By performing reference domain alignment processing on extreme feature points, the location information of feature points acquired in local detection scenarios can be uniformly adapted to a standard reference benchmark, eliminating the position offset problem caused by local detection coordinate system deviation and ensuring the benchmark uniformity and standardization of feature point location information. Through global coordinate translation processing of reference system registration coordinates, local relative coordinate information can be accurately converted into globally unified absolute spatial coordinates, realizing the accurate mapping of fault feature points from local relative positions to the actual physical positions of pipelines, and completely solving the technical problem that local positioning cannot match global laying scenarios. By standardizing information frame arrangement of absolute spatial coordinates, discrete coordinate data can be organized into structured positioning result data with regular structure and uniform format, standardizing the data output format and improving the readability, completeness, and universality of positioning data. By standardizing the release of positioning result data through a dedicated output port, accurate fault location information can be output stably and completely, ultimately forming a high-precision accurate location point for target pipeline faults. This hierarchical coordinate transformation and data output process adopts a fixed and unified transformation standard and output specification throughout the process, which is highly reproducible and effectively improves the accuracy and stability of pipeline fault location. It avoids the defects of inconsistent coordinates, position offset and data disorder in traditional positioning methods, and provides standardized and high-precision positioning data support for accurate pipeline fault inspection and source tracing.

[0071] like Figure 2 The diagram shown is a functional block diagram of a pipeline fault accurate location system based on signal gradient provided in an embodiment of the present invention.

[0072] The pipeline fault precise location system 100 based on signal gradient described in this invention can be installed in an electronic device. Depending on the functions implemented, the pipeline fault precise location system 100 based on signal gradient may include a reference data generation module 101, a gradient feature extraction module 102, a fault interval marking module 103, an extreme value feature detection module 104, and a precise location output module 105. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0073] In this embodiment, the functions of each module / unit are as follows: The reference data generation module 101 is used to filter the background noise of the DC background voltage signal at the starting reference position of the target pipeline, and to eliminate the DC component of the filtered data to obtain the reference reference data of the DC background voltage signal. The gradient feature extraction module 102 is used to perform differential operation on the induced current signal at a preset discrete location point of the target pipeline based on the reference data, and to perform polarity constraint correction on the operation result to obtain the signal gradient feature value of the induced current signal. The fault interval marking module 103 is used to construct a signal gradient map based on the preset discrete location points and signal gradient feature values, and based on the consistency between the gradient slope of the signal gradient map and the continuous points, mark the spatial regions corresponding to the polarity reversal points and disorder start points in the signal gradient map as fault reference intervals. The extreme value feature detection module 104 is used to detect the potential difference of the interval voltage signal at continuous position points within the fault reference interval, and to perform dual feature fusion judgment on the gradient feature of the signal gradient spectrum and the potential difference feature of the interval voltage signal to obtain the extreme value feature points of the interval voltage signal. The precise positioning output module 105 is used to perform spatial coordinate mapping on the extreme value feature point to obtain the spatial coordinates of the extreme value feature point, and output the spatial coordinates as the precise positioning point of the pipeline fault.

[0074] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0075] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0076] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0077] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0078] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0079] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for precise pipeline fault location based on signal gradient, characterized in that, The method includes: Ⅰ. Filter the DC background voltage signal at the starting reference position of the target pipeline for background noise, and eliminate the DC component of the filtered data to obtain the reference reference data of the DC background voltage signal; II. Based on the aforementioned reference data, differential calculation is performed on the induced current signal at preset discrete locations of the target pipeline, and polarity constraint correction is applied to the calculation result to obtain the signal gradient characteristic value of the induced current signal. III. Construct a signal gradient map based on the preset discrete location points and signal gradient feature values, and mark the spatial regions corresponding to the polarity reversal points and disorder initiation points in the signal gradient map as fault reference intervals based on the consistency between the gradient slope of the signal gradient map and the continuous points. IV. Detect the potential difference of the interval voltage signal at continuous locations within the fault reference interval, and perform dual feature fusion determination by combining the gradient features of the signal gradient spectrum with the potential difference features of the interval voltage signal to obtain the extreme feature points of the interval voltage signal. V. Perform spatial coordinate mapping on the extreme value feature points to obtain the spatial coordinates of the extreme value feature points, and output the spatial coordinates as the precise location points of pipeline faults.

2. The method for accurate pipeline fault location based on signal gradient as described in claim 1, characterized in that, The process of filtering the DC background voltage signal at the initial reference position of the target pipeline for background noise and eliminating the DC component of the filtered data to obtain the reference reference data for the DC background voltage signal includes: The DC background voltage signal at the starting reference position of the target pipeline is subjected to power frequency notch filtering to obtain the first filtered signal; Based on the target pipeline burial depth and soil moisture, the high-pass filter cutoff frequency is positively and adaptively adjusted, and the first filtered signal is high-pass filtered to obtain a second filtered signal that eliminates low-frequency drift components. The second filtered signal is subjected to sliding window mean filtering to obtain the zero-mean signal of the second filtered signal, and the zero-mean signal is used as the reference data of the DC background voltage signal.

3. The method for accurate pipeline fault location based on signal gradient as described in claim 1, characterized in that, Based on the reference data, the induced current signal at preset discrete locations of the target pipeline is subjected to differential calculation, and the result of the calculation is corrected for polarity constraints to obtain the signal gradient feature value of the induced current signal, including: Based on a preset fixed sampling interval, the induced current signal in the extension direction of the target pipeline is discretized and extracted to obtain the original sampling sequence of the induced current signal. The relative offset compensation is performed between the induced current value of the sampling point in the original sampling sequence and the reference data, and the spatial position is synchronously calibrated on the compensated data to obtain the corrected current value of the sampling point. The original difference value of the sampling point is obtained by performing a difference operation between the non-first sampling point in the original sampling sequence and the corrected current value of the adjacent sampling point. Based on the spatial order of the sampling points, the original difference values ​​are serialized and recombined to obtain the original difference sequence of the original sampling sequence.

4. The method for accurate pipeline fault location based on signal gradient as described in claim 3, characterized in that, Based on the reference data, the induced current signal at preset discrete locations of the target pipeline is subjected to differential calculation, and the result of the calculation is corrected for polarity constraints to obtain the signal gradient feature value of the induced current signal, including: The original difference value in the original difference sequence and the fixed sampling interval are divided to obtain the unit spacing difference value of the sampling points; Based on spatial topological order, coordinate serialization mapping is performed on the unit spacing difference values ​​of the original difference sequence to obtain the unit spacing difference sequence of the original difference sequence; Based on the first unit spacing difference sequence Based on the unit interval difference value and polarity abruptness constraint, the weighted gradient correction value of the sampling point is calculated using the following formula: ； in, Indicates the sequential numbering of the sampling points. In the unit spacing difference sequence, the first... The unit interval difference value of each sampling point Indicates the first digit in the original difference sequence. The absolute value of the original difference values ​​of each sampling point This represents the maximum absolute value of all original difference values ​​in the original difference sequence. This represents the preset weight adjustment coefficient, which is a constant with a value greater than 0 and less than 1. Represents a symbolic function; The weighted gradient correction value is assigned a value to obtain the signal gradient feature value of the sampling point; The signal gradient feature values ​​of all sampling points in the unit spacing difference sequence are sequentially merged along the path according to the spatial distribution order to obtain the signal gradient feature values ​​of the induced current signal.

5. The method for precise pipeline fault location based on signal gradient as described in claim 1, characterized in that, The construction of the signal gradient map based on the preset discrete location points and signal gradient feature values ​​includes: The spatial coordinates of the preset discrete location points are interpreted to obtain the spatial location dataset of the preset discrete location points; The signal gradient feature values ​​are extracted to obtain the gradient feature value dataset of the signal gradient feature values. Based on the spatial location extension direction of the spatial location dataset, feature space coupling is performed on the spatial location dataset and the gradient feature value dataset to obtain a gradient location association pair sequence of the spatial location dataset and the gradient feature value dataset; The gradient position association sequence is mapped to a two-dimensional coordinate system to obtain the signal gradient spectrum of the spatial position dataset and the gradient feature value dataset. The spatial coordinates of the spatial position dataset are used as the horizontal axis scale and the gradient feature values ​​of the gradient feature value dataset are used as the vertical axis magnitude.

6. The method for accurate pipeline fault location based on signal gradient as described in claim 1, characterized in that, Based on the consistency between the gradient slope and continuous points in the signal gradient spectrum, the spatial regions corresponding to polarity reversal points and disorder initiation points in the signal gradient spectrum are marked as fault reference intervals, including: The polarity reversal point set of the signal gradient spectrum is obtained by performing a sign traversal on the locations where the signs of the gradient feature values ​​in the signal gradient spectrum are reversed. Amplitude fluctuation detection is performed on the starting points where the amplitude of the gradient feature value in the signal gradient spectrum continuously fluctuates beyond a preset fluctuation threshold and the gradient change slope exceeds a slope threshold, thereby obtaining the disordered starting point set of the signal gradient spectrum; Spatial coordinate interpretation is performed on the polarity reversal point set to obtain the reversal point spatial coordinate set of the polarity reversal point set; The spatial coordinates of the disorder initiation points and their adjacent reversal points in the disorder initiation point set are marked with continuous region boundary closure, and false abnormal regions are eliminated based on the consistency check of three consecutive points, so as to obtain the fault reference interval of the disorder initiation point set and the reversal point spatial coordinate set.

7. The method for accurate pipeline fault location based on signal gradient as described in claim 1, characterized in that, The step of detecting the potential difference of the interval voltage signal at continuous locations within the fault reference interval, and performing dual-feature fusion determination of the gradient features of the signal gradient spectrum and the potential difference features of the interval voltage signal to obtain the extreme value feature points of the interval voltage signal includes: Boundary anchoring is performed on the starting and ending endpoints within the fault reference interval to obtain the interval boundary coordinates of the fault reference interval; Based on a preset spatial step interval, voltage signals along the line are picked up at continuous position points between the boundary coordinates of the interval to obtain a continuous voltage sampling sequence of the fault reference interval. Potential difference quantization is performed on adjacent sampling point pairs in the continuous voltage sampling sequence to obtain the potential difference value of the adjacent point pairs; Based on the sampling order of the continuous voltage sampling sequence, a situation sequence is constructed for the potential difference value to obtain the potential difference distribution sequence of the fault reference interval.

8. The method for precise pipeline fault location based on signal gradient as described in claim 7, characterized in that, The step of detecting the potential difference of the interval voltage signal at continuous locations within the fault reference interval, and performing dual-feature fusion determination of the gradient features of the signal gradient spectrum and the potential difference features of the interval voltage signal to obtain the extreme value feature points of the interval voltage signal includes: A global traversal is performed on the position indices of adjacent values ​​with opposite signs in the potential difference distribution sequence to obtain the extreme value candidate point index set of the potential difference distribution sequence; The absolute value of the potential difference corresponding to adjacent indices in the extreme value candidate point index set is selected by the magnitude of the potential difference, and the index with the largest magnitude is retained to obtain the extreme value candidate point index set maximum value index set. Using the spatial coordinates of the sampling points corresponding to the index in the maximum point index set as the index spatial position, a candidate set of extreme point spatial coordinates of the maximum point index set is obtained; The gradient feature value corresponding to the candidate set of extreme point spatial coordinates is fused with the potential difference feature of the interval voltage signal to obtain the true fault point set of the candidate set of extreme point spatial coordinates. The coordinates of the real fault points in the set are traced and merged to calculate the fault confidence of the candidate points in the set of real fault points. The calculation formula is as follows: ； in, For the first Fault confidence of each candidate point The potential difference is a quantitative value. This is the confidence level adjustment coefficient, 0 < <1, Represents a symbolic function; The point with unique spatial coordinates and the largest absolute value of the corresponding potential difference is taken as the extreme value feature point of the interval voltage signal.

9. The method for accurate pipeline fault location based on signal gradient as described in claim 1, characterized in that, The step of mapping the extreme feature points to spatial coordinates, obtaining the spatial coordinates of the extreme feature points, and outputting the spatial coordinates as precise location points for pipeline faults includes: Align the extreme feature points with the reference domain to obtain the reference system registration coordinates of the extreme feature points; The reference frame registration coordinates are globally translated to obtain the absolute spatial coordinates of the extreme feature points; The absolute spatial coordinates are arranged into information frames to obtain the localization result data of the extreme value feature points; The location result data is released through the output port to obtain the precise location point of the target pipeline fault.

10. A pipeline fault precise location system based on signal gradient, characterized in that, The system for implementing the pipeline fault accurate location method based on signal gradient as described in claim 1 includes: The reference data generation module is used to filter the background noise of the DC background voltage signal at the starting reference position of the target pipeline, and to eliminate the DC component of the filtered data to obtain the reference reference data of the DC background voltage signal. The gradient feature extraction module is used to perform differential operations on the induced current signals at preset discrete locations of the target pipeline based on the reference data, and to perform polarity constraint correction on the operation results to obtain the signal gradient feature values ​​of the induced current signals. The fault interval marking module is used to construct a signal gradient map based on the preset discrete location points and signal gradient feature values, and mark the spatial regions corresponding to polarity reversal points and disorder initiation points in the signal gradient map as fault reference intervals based on the consistency between the gradient slope of the signal gradient map and the continuous points. The extreme value feature detection module is used to detect the potential difference of the interval voltage signal at continuous location points within the fault reference interval, and to perform dual feature fusion judgment by combining the gradient feature of the signal gradient spectrum with the potential difference feature of the interval voltage signal to obtain the extreme value feature points of the interval voltage signal. The precise positioning output module is used to perform spatial coordinate mapping on the extreme value feature points to obtain the spatial coordinates of the extreme value feature points, and output the spatial coordinates as the precise positioning points of pipeline faults.

Images