Rail construction cable fault online diagnosis and data intelligent analysis method and system

Abstract

The present application relates to the technical field of electrical measurement, and discloses a track construction cable fault online diagnosis and data intelligent analysis method and system, the method comprising: multi-dimensionally fusing multi-source operation data of the track construction cable to obtain fused data of the track construction cable; extracting fault features in the fused data to obtain abnormal characteristic values of the track construction cable; multi-level comparing the abnormal characteristic values with a historical fault threshold library to obtain a graded comparison result of the track construction cable; quantitatively scoring the graded comparison result to obtain abnormal indication information of the track construction cable; based on the abnormal indication information, dividing the abnormal state of the track construction cable to obtain a fault severity level of the track construction cable; and generating a diagnosis report of the track construction cable according to the fault severity level; the present application can improve the accuracy of track construction cable fault online diagnosis and data intelligent analysis.

Description

Technical Field

[0001] This invention relates to the field of electrical measurement technology, and in particular to a method and system for online diagnosis and intelligent data analysis of faults in railway construction cables. Background Technology

[0002] In the field of online fault diagnosis for railway construction cables, existing technologies lack scientific rigor in processing multi-source operational data and extracting fault features. Traditional methods, after collecting data such as cable voltage, current, and temperature, fail to determine the fusion order based on parameter importance or perform time alignment and standardization, resulting in dimensional fragmentation or temporal misalignment of the fused data. When extracting fault features, they fail to identify fault-sensitive patterns in conjunction with cable type parameters, only performing simple feature screening on the raw data. The generated abnormal feature values ​​have low correlation with the actual fault, making it difficult to accurately reflect the cable fault state and creating potential errors for subsequent fault diagnosis.

[0003] Existing technologies have significant shortcomings in the fault identification and diagnostic report generation stages. When comparing abnormal characteristic values ​​with historical fault thresholds, priority is not determined according to the comparison hierarchy and dispersion verification is not performed. Faults are determined solely based on a single-level threshold, which easily leads to misjudgments or omissions. In the quantitative scoring and fault level classification stages, the scoring benchmark is not dynamically adjusted based on real-time operating parameters, load status, and environmental conditions. Level classification is based solely on fixed standards, resulting in fault severity assessments that do not match actual scenarios. Furthermore, when generating diagnostic reports, key diagnostic elements are not integrated for multi-dimensional correlation analysis and spatiotemporal consistency verification. The report content lacks systematicity and accuracy, failing to provide reliable guidance for fault repair of track construction cables. Summary of the Invention

[0004] This invention provides a method and system for online diagnosis and intelligent data analysis of faults in railway construction cables to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for online diagnosis and intelligent data analysis of faults in railway construction cables, comprising:

[0006] S1. Based on the importance of the parameters of the track construction cable, the multi-source operation data of the track construction cable are fused in multiple dimensions to obtain the fused data of the track construction cable;

[0007] S2. Based on the type parameters of the track construction cable, extract the fault features from the fused data to obtain the abnormal feature values ​​of the track construction cable;

[0008] S3. Perform multi-level comparison between the abnormal feature values ​​and the historical fault threshold database to obtain the graded comparison results of the track construction cable;

[0009] S4. Based on the real-time operating parameters of the track construction cable, the graded comparison results are quantitatively scored to obtain the abnormal indication information of the track construction cable;

[0010] S5. Based on the abnormal indication information, classify the abnormal state of the track construction cable to obtain the fault severity level of the track construction cable;

[0011] S6. Generate a diagnostic report for the track construction cable based on the severity level of the fault.

[0012] In a preferred embodiment, the multi-source operational data of the track construction cable is fused in multiple dimensions based on the parameter importance of the track construction cable to obtain fused data of the track construction cable, including:

[0013] Simultaneously collect voltage, current, and temperature parameters from the track construction cables;

[0014] The voltage, current, and temperature parameters are used as multi-source operating data for the track construction cable.

[0015] The fusion order of the multi-source operation data is determined based on the importance of the parameters of the track construction cable.

[0016] The multi-source operational data is time-aligned to obtain time-aligned data of the multi-source operational data.

[0017] The time-aligned data is comprehensively organized according to a preset data standard to obtain the standard data for the track construction cable;

[0018] According to the fusion order, the standard data is combined into the fused data of the track construction cable.

[0019] In a preferred embodiment, the step of extracting fault features from the fused data based on the type parameters of the track construction cable to obtain abnormal feature values ​​of the track construction cable includes:

[0020] Based on the type parameters of the track construction cable, identify the fault sensitivity mode of the track construction cable;

[0021] Based on the fault sensitivity mode, fault features in the fused data are filtered to obtain the target fault data sequence of the track construction cable.

[0022] Feature value derivation is performed on the target fault data sequence to obtain the preliminary abnormal feature values ​​of the track construction cable;

[0023] The preliminary abnormal characteristic values ​​are verified in multiple dimensions to obtain the abnormal characteristic values ​​of the track construction cable.

[0024] In a preferred embodiment, the step of deriving feature values ​​from the target fault data sequence to obtain preliminary abnormal feature values ​​of the track construction cable includes:

[0025] Based on the fault sensitivity mode, extract the weight distribution sequence and reference baseline sequence of the target fault data sequence;

[0026] Based on the weighted distribution sequence, anomaly feature evolution is performed on the reference baseline sequence to obtain preliminary anomaly feature values ​​for the track construction cable. The calculation formula for the preliminary anomaly feature values ​​is as follows: ;

[0027] In the formula, The initial abnormal feature value, The number of data points in the target fault data sequence. The weight distribution sequence is the first... The weighting coefficients for each data point It is the hyperbolic tangent function. As a sensitivity modulator, The first in the target fault data sequence Data points, The first in the reference reference sequence A reference value, The first fault in the target fault data sequence The data point and the first data point in the reference reference sequence The absolute deviation between reference values.

[0028] In a preferred embodiment, the step of performing multi-level comparisons between the abnormal feature values ​​and the historical fault threshold database to obtain the graded comparison results of the track construction cable includes:

[0029] The comparison priority order of the abnormal feature values ​​is determined based on the comparison hierarchy in the historical fault threshold database;

[0030] Based on the comparison priority order, the discreteness of the abnormal feature values ​​is checked to obtain the feature deviation of the abnormal feature values;

[0031] Based on the feature deviation, the abnormal feature values ​​are subjected to multi-level refinement comparison to obtain the hierarchical deviation features of the abnormal feature values;

[0032] The hierarchical deviation features are mapped to the comparison hierarchy to obtain the hierarchical comparison results of the track construction cable.

[0033] In a preferred embodiment, the step of quantifying and scoring the graded comparison results based on the real-time operating parameters of the track construction cable to obtain abnormal indication information of the track construction cable includes:

[0034] Based on the load status and environmental conditions of the real-time operating parameters in the track construction cable, the hierarchical scoring benchmark in the hierarchical comparison results is determined;

[0035] Based on the hierarchical scoring benchmark, the hierarchical comparison results are weighted to obtain the hierarchical weight coefficients of the hierarchical comparison results.

[0036] Based on the hierarchical weight coefficients, the hierarchical comparison results are evaluated in multiple dimensions to obtain a preliminary quantitative score for the hierarchical comparison results.

[0037] The preliminary quantitative score is matched step by step with the real-time operating parameters to obtain the abnormal indication information of the track construction cable.

[0038] In a preferred embodiment, the step of performing a multi-dimensional evaluation of the hierarchical comparison results based on the hierarchical weight coefficients to obtain a preliminary quantitative score for the hierarchical comparison results includes:

[0039] Based on the hierarchical weight coefficients, determine the dynamic influence factors of the hierarchical comparison results;

[0040] Based on the dynamic influence factor, the classification comparison results are quantitatively evaluated to obtain a preliminary quantitative score for the classification comparison results. The calculation formula for the preliminary quantitative score is as follows:

[0041] ;

[0042] In the formula, For the aforementioned preliminary quantitative score, To compare the total number of levels, For the first Weighting coefficients for each level, It is the hyperbolic tangent function. For the first The quantified values ​​of the hierarchical comparison results at each level. For the first Stability adjustment factors at each level, For the first The gradient of changes in the hierarchical comparison results at each level.

[0043] In a preferred embodiment, the step of classifying the abnormal state of the track construction cable based on the abnormality indication information to obtain the fault severity level of the track construction cable includes:

[0044] Based on the inherent correlation characteristics of the abnormal indication information, a multi-dimensional status determination basis for the track construction cable is constructed;

[0045] Based on the multi-dimensional state determination criteria, the anomaly indication information is vectorized to obtain the state feature vector of the anomaly indication information.

[0046] Based on a pre-set cable feature library, multimodal feature matching is performed on the state feature vector to obtain the preliminary state classification result of the track construction cable;

[0047] Based on the preliminary state classification results, the state evolution trajectory analysis of the abnormal indication information is performed to obtain the fault severity level of the track construction cable.

[0048] In a preferred embodiment, generating a diagnostic report for the track construction cable based on the severity level of the fault includes:

[0049] Based on the severity level of the fault and the type parameter, determine the technical analysis dimensions and level of detail of the diagnostic report for the track construction cable;

[0050] Based on the aforementioned technical analysis dimensions, key diagnostic elements in the abnormal feature values, the hierarchical comparison results, and the abnormal indication information are determined.

[0051] A multi-angle correlation analysis was performed on the key diagnostic elements to obtain the core report content of the track construction cable;

[0052] Based on the level of detail described, the core report content is expanded hierarchically to obtain the structured report content for the track construction cable.

[0053] The structured report content is compared with the historical diagnostic records of the track construction cable using a multi-dimensional spatiotemporal consistency verification to obtain the diagnostic report of the track construction cable.

[0054] To address the aforementioned problems, this invention also provides an online fault diagnosis and intelligent data analysis system for railway construction cables, the system comprising:

[0055] The data fusion module is used to perform multi-dimensional fusion of multi-source operational data of the track construction cable based on the importance of the parameters of the track construction cable, so as to obtain the fused data of the track construction cable.

[0056] The fault identification module is used to extract fault features from the fused data based on the type parameters of the track construction cable, and obtain the abnormal feature values ​​of the track construction cable.

[0057] A multi-level comparison module is used to perform multi-level comparisons between the abnormal feature values ​​and the historical fault threshold database to obtain the graded comparison results of the track construction cable.

[0058] The quantitative scoring module is used to quantitatively score the graded comparison results based on the real-time operating parameters of the track construction cable, and obtain the abnormal indication information of the track construction cable.

[0059] The fault classification module is used to classify the abnormal state of the track construction cable based on the abnormal indication information to obtain the fault severity level of the track construction cable.

[0060] The diagnostic report module is used to generate a diagnostic report for the track construction cable based on the severity level of the fault.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] 1. This invention, through its online fault diagnosis and intelligent data analysis method and system for railway construction cables, significantly improves the accuracy of fault diagnosis and the effectiveness of intelligent data analysis. The technology determines the fusion order of multi-source operational data based on the importance of railway construction cable parameters, generates accurate fused data through time alignment and standardization, laying a reliable data foundation for subsequent fault feature extraction. It combines cable type parameters to identify fault-sensitive patterns, filters target fault data sequences, derives preliminary abnormal feature values ​​through formulas, and then obtains accurate abnormal feature values ​​through multi-dimensional verification. This makes fault feature extraction more closely match the actual fault characteristics of the cable, effectively reducing feature deviation.

[0063] 2. This invention generates accurate hierarchical comparison results by comparing abnormal feature values ​​with a historical fault threshold database at multiple levels according to comparison priority and combining discreteness verification to obtain hierarchical deviation features. Based on the load status and environmental conditions of real-time operating parameters, a scoring benchmark is determined, and the hierarchical comparison results are quantified and scored using a formula to obtain reliable abnormal indication information, thereby classifying fault severity levels that conform to actual scenarios. When generating a diagnostic report, key diagnostic elements are integrated for multi-angle correlation analysis, and spatiotemporal consistency verification is performed in conjunction with historical records to form a structured and highly accurate report, providing scientific guidance for cable fault repair and further ensuring the safe and stable operation of track construction cables. Attached Figure Description

[0064] Figure 1 This is a flowchart illustrating an embodiment of the online fault diagnosis and intelligent data analysis method for track construction cables provided by the present invention.

[0065] Figure 2 This is a functional module diagram of an online fault diagnosis and intelligent data analysis system for track construction cables provided in an embodiment of the present invention;

[0066] 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

[0067] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0068] This application provides a method for online diagnosis and intelligent data analysis of faults in railway construction cables. The executing 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 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 cloud server cluster. The server can be an independent server or a cloud server providing 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 (CDN), and big data and artificial intelligence platforms.

[0069] Reference Figure 1 The diagram shown is a flowchart illustrating an online fault diagnosis and intelligent data analysis method for track construction cables according to an embodiment of the present invention. In this embodiment, the online fault diagnosis and intelligent data analysis method for track construction cables includes:

[0070] S1. Based on the importance of the parameters of the track construction cable, the multi-source operation data of the track construction cable are fused in multiple dimensions to obtain the fused data of the track construction cable;

[0071] In this embodiment of the invention, the multi-source operational data of the track construction cable is fused in multiple dimensions based on the importance of the parameters of the track construction cable to obtain the fused data of the track construction cable, including:

[0072] Simultaneously collect voltage, current, and temperature parameters from the track construction cables;

[0073] The voltage, current, and temperature parameters are used as multi-source operating data for the track construction cable.

[0074] The fusion order of the multi-source operation data is determined based on the importance of the parameters of the track construction cable.

[0075] The multi-source operational data is time-aligned to obtain time-aligned data of the multi-source operational data.

[0076] The time-aligned data is comprehensively organized according to a preset data standard to obtain the standard data for the track construction cable;

[0077] According to the fusion order, the standard data is combined into the fused data of the track construction cable.

[0078] Specifically, when simultaneously collecting voltage, current, and temperature parameters in the track construction cable, data acquisition is initiated simultaneously through sensors installed on the cable. Voltage parameters are read in real time by voltage sensors to read the potential difference data at both ends of the cable. Current parameters are monitored in real time by current sensors to monitor the intensity of electron flow in the cable. Temperature parameters are obtained in real time by temperature sensors to acquire heat data on the cable surface and inside. During the acquisition process, the acquisition frequency of the three parameters is kept consistent to ensure that the corresponding parameter values ​​can be obtained within the same time interval.

[0079] Furthermore, when using voltage, current, and temperature parameters as multi-source operating data for track construction cables, the synchronously collected voltage, current, and temperature data will be classified and stored. Voltage data will be recorded in the format of "acquisition time - voltage value", current data in the format of "acquisition time - current value", and temperature data in the format of "acquisition time - temperature value". These three types of data together constitute multi-source operating data reflecting the operating status of track construction cables.

[0080] Furthermore, based on the importance of the parameters of the track construction cable, when determining the fusion order of multi-source operation data, the influence of three parameters on the safe operation of the cable will be analyzed first. The current parameter directly reflects the load condition of the cable, and overload will cause cable damage, so it has the highest importance. The voltage parameter affects the normal power supply of the cable, and abnormal voltage will affect the operation of the equipment, so it has the second highest importance. The temperature parameter reflects the heating state of the cable, and excessive temperature is a potential risk, so it has a relatively low importance. Therefore, the fusion order is determined to be current parameter, voltage parameter, and temperature parameter.

[0081] Furthermore, time alignment processing is performed on the multi-source operating data to obtain time-aligned data. Based on the acquisition time, the parameter values ​​corresponding to the same acquisition time in voltage, current, and temperature parameters are matched. If a certain type of parameter is missing at a certain acquisition time point, all parameter values ​​at that time point will be removed to ensure that each retained time point contains complete voltage, current, and temperature parameter values. The processed three types of parameter data correspond completely in the time dimension, which is the time-aligned data of multi-source operating data.

[0082] Furthermore, when the time-aligned data is comprehensively organized according to the preset data standards to obtain the standard data for track construction cables, the preset data standards include unified parameter units, unified data format, and numerical range verification. The parameter units are unified as volts for voltage, amperes for current, and degrees Celsius for temperature. The data format is unified as a table of "timestamp - voltage value - current value - temperature value". The numerical range verification will remove data that exceeds the reasonable range of voltage, current, and temperature during normal cable operation. The structured data formed after these processes is the standard data for track construction cables.

[0083] Furthermore, according to the fusion order, when combining the standard data into the fused data of the track construction cable, the parameter values ​​corresponding to each time point in the standard data will be arranged according to the determined fusion order of current parameters, voltage parameters, and temperature parameters. The fused data format for each time point is "current value - voltage value - temperature value". Then, the fused data of all time points will be arranged in chronological order of timestamps to form a dataset containing a complete time series and ordered parameter values. This dataset is the fused data of the track construction cable.

[0084] In summary, synchronously collecting voltage, current, and temperature parameters ensures that these three types of core operational data maintain time synchronization from the source of collection, avoiding data timing misalignment caused by differences in collection start time. This lays a "same source, same sequence" data foundation for subsequent multi-dimensional integration, ensuring the authenticity and relevance of the data in reflecting the cable's operating status.

[0085] In summary, defining voltage, current, and temperature parameters as multi-source operating data can systematically cover key dimensions such as the electrical and thermal characteristics of cables during operation. This avoids the problem of incomplete characterization of operating status caused by missing dimensions in traditional data acquisition, allowing subsequent analysis to fully capture potential fault signs in cables.

[0086] In general, determining the fusion order based on the importance of parameters, such as prioritizing the fusion of current parameters which have the greatest impact on cable safety, and then fusion of voltage and temperature parameters in turn, can highlight the weight of core parameters in the fused data, avoid the dilution of core fault information caused by traditional indiscriminate fusion, and make the fused data more accurately point to the critical operating status of the cable.

[0087] In summary, time alignment processing of multi-source operating data and elimination of time point data with missing parameters can ensure that each time point contains complete voltage, current, and temperature data, eliminate the interference of timing deviations on data correlation, make subsequent time-based fault feature analysis more reliable, and reduce misjudgments caused by incomplete data.

[0088] In summary, aligning data according to preset data standards, unifying parameter units and formats, and verifying numerical ranges can eliminate the interference of chaotic data formats, inconsistent units, and abnormal values ​​on the analysis. This transforms raw data into standardized data, reduces the complexity of subsequent data processing, and improves the efficiency and accuracy of fault feature extraction.

[0089] In summary, combining standardized data in the order of fusion to form fused data allows the fused data to retain the independent characteristics of each parameter while also reflecting the correlation between parameters through orderly combination. Compared with traditional scattered single-dimensional data, fused data better meets the needs of cable fault diagnosis for "multi-dimensional collaborative analysis" and provides structured and highly available data support for accurately extracting fault features.

[0090] S2. Based on the type parameters of the track construction cable, extract the fault features from the fused data to obtain the abnormal feature values ​​of the track construction cable;

[0091] In this embodiment of the invention, the step of extracting fault features from the fused data based on the type parameters of the track construction cable to obtain the abnormal feature values ​​of the track construction cable includes:

[0092] Based on the type parameters of the track construction cable, identify the fault sensitivity mode of the track construction cable;

[0093] Based on the fault sensitivity mode, fault features in the fused data are filtered to obtain the target fault data sequence of the track construction cable.

[0094] Feature value derivation is performed on the target fault data sequence to obtain the preliminary abnormal feature values ​​of the track construction cable;

[0095] The preliminary abnormal characteristic values ​​are verified in multiple dimensions to obtain the abnormal characteristic values ​​of the track construction cable.

[0096] The step of performing feature value derivation on the target fault data sequence to obtain preliminary abnormal feature values ​​for the track construction cable includes:

[0097] Based on the fault sensitivity mode, extract the weight distribution sequence and reference baseline sequence of the target fault data sequence;

[0098] Based on the weighted distribution sequence, anomaly feature evolution is performed on the reference baseline sequence to obtain preliminary anomaly feature values ​​for the track construction cable. The calculation formula for the preliminary anomaly feature values ​​is as follows:

[0099] ;

[0100] In the formula, The initial abnormal feature value, The number of data points in the target fault data sequence. The weight distribution sequence is the first... The weighting coefficients for each data point It is the hyperbolic tangent function. As a sensitivity modulator, The first in the target fault data sequence Data points, The first in the reference reference sequence A reference value, The first fault in the target fault data sequence The data point and the first data point in the reference reference sequence The absolute deviation between reference values.

[0101] Specifically, when identifying the fault-sensitive modes of track construction cables based on their type parameters, the system first reads the cable type parameters, including information such as cable material, cross-sectional specifications, rated voltage level, and laying method. Then, it retrieves a preset cable fault mode library, which stores typical fault modes corresponding to different types of cables. The system compares the current cable type parameters with the classification tags in the fault mode library one by one to match the completely corresponding typical fault modes. These modes are the fault-sensitive modes of the track construction cables.

[0102] Furthermore, when selecting fault features from the fused data based on fault sensitivity patterns to obtain the target fault data sequence for track construction cables, the change patterns of characteristic parameters corresponding to each fault sensitivity pattern are first analyzed. Then, the parameter values ​​at each time point in the fused data are traversed to check whether they conform to the characteristic patterns corresponding to the fault sensitivity patterns. The parameter values ​​at the time points that conform to the characteristic patterns are arranged in chronological order according to the timestamps, and the resulting continuous data sequence is the target fault data sequence for track construction cables.

[0103] Furthermore, when deriving feature values ​​from the target fault data sequence to obtain the preliminary abnormal feature values ​​of the track construction cable, features are extracted from the voltage, current, and temperature parameters in the target fault data sequence. The voltage difference, number of fluctuations and amplitude range are calculated, the frequency, duration and deviation from normal current of current peaks are statistically analyzed, and the rate of temperature rise, duration of high temperature and numerical range exceeding the normal range are analyzed. The set of indicators formed by integrating these specific values ​​is the preliminary abnormal feature value of the track construction cable.

[0104] Furthermore, the preliminary abnormal characteristic values ​​are verified in multiple dimensions. When obtaining the abnormal characteristic values ​​of the track construction cable, the verification is carried out from three dimensions: parameter correlation, historical data comparison, and threshold range. The logical correlation between characteristic values ​​is checked, compared with past fault characteristic values, and compared with safe operation thresholds. Abnormal values ​​that do not conform to the rules and exceed the reasonable range are eliminated, and the remaining characteristic values ​​are the abnormal characteristic values ​​of the track construction cable.

[0105] Specifically, when extracting the weight distribution sequence and reference benchmark sequence of the target fault data sequence according to the fault sensitivity mode, the influence of each parameter on the fault determination in the fault sensitivity mode is first analyzed. The current, voltage and temperature parameters of the target fault data sequence are assigned corresponding weight values ​​according to the degree of influence. The ordered set formed by sorting the weight values ​​according to the acquisition time is the weight distribution sequence. At the same time, the parameter sequence of the same type of track construction cable during normal operation is retrieved. The sequence containing the standard parameter values ​​at each time point is the reference benchmark sequence.

[0106] Furthermore, based on the weighted distribution sequence, when performing anomaly feature evolution on the reference benchmark sequence to obtain the preliminary anomaly feature values ​​of the track construction cable, each weight value of the weighted distribution sequence is associated with the standard value of the corresponding time point of the reference benchmark sequence. The actual value of the target fault data sequence at the corresponding time point is compared with the standard value, and the deviation is calculated. The deviation ratio is adjusted according to the weight value, with the parameter with a larger weight having a higher deviation ratio. The adjusted deviation is integrated, and key features such as the maximum deviation value and the duration of the deviation are extracted. The specific values ​​corresponding to these features are the preliminary anomaly feature values ​​of the track construction cable.

[0107] Specifically, in the parameter sources for the preliminary anomaly characteristic values, the number of data points is the total number of independent data points contained in the target fault data sequence. The weight coefficients come from the weight distribution sequence, which is related to the first... The weight values ​​corresponding to each data point. The sensitivity adjustment factor is a fixed value preset based on cable type parameters and fault sensitivity mode. The data point is the th data point in the target fault data sequence arranged chronologically. Each parameter value. The reference value is the one in the reference benchmark sequence that is related to the first... The standard parameter values ​​for each data point at the same time. The absolute deviation is the first... The data point and the first The non-negative value of the difference between the reference values.

[0108] Furthermore, the formula means that by combining the deviations of each data point in the target fault data sequence from the corresponding reference values ​​in the reference baseline sequence, and by using the weighting coefficient and sensitivity adjustment factor, the deviations are processed nonlinearly through the hyperbolic tangent function to calculate preliminary abnormality characteristic values ​​that can quantify the degree of cable abnormality, thus providing a basis for subsequent verification.

[0109] Furthermore, the formula shows that the greater the deviation between the target fault data sequence and the reference baseline sequence, the higher the weight of the key data points, or the larger the sensitivity adjustment factor, the larger the initial abnormal characteristic value; conversely, the smaller the deviation, the lower the weight, or the smaller the sensitivity adjustment factor, the smaller the initial abnormal characteristic value.

[0110] In summary, identifying fault-sensitive patterns based on the type parameters of railway construction cables allows for fault feature extraction that better reflects the actual characteristics of the cables. Cables with different types of parameters exhibit varying fault risks. For example, buried cables are prone to grounding faults due to soil corrosion, while overhead cables are susceptible to wire breakage due to wind and rain erosion. Accurately matching fault-sensitive patterns based on type parameters avoids the feature extraction bias caused by traditional generalized fault patterns, ensuring that the extracted fault patterns closely match the actual fault risks of the cables.

[0111] In summary, by filtering fault features from fused data based on fault sensitivity patterns to obtain target fault data sequences, it is possible to accurately locate fault-related feature data from massive fused data. The fused data contains complete information on both normal and abnormal cable operation. By filtering the feature patterns of fault sensitivity patterns, irrelevant normal operation data can be eliminated, focusing on data reflecting fault signs, reducing the interference of invalid data on subsequent analysis, and improving the efficiency and relevance of fault feature extraction.

[0112] In summary, deriving preliminary anomaly features from the target fault data sequence transforms discrete fault characteristic data into quantifiable indicators. By calculating derived indicators such as voltage fluctuation amplitude, current peak frequency, and temperature rise rate, abstract fault characteristics can be visualized. Compared to traditional analysis methods that rely solely on raw data, quantified preliminary anomaly features are easier to compare with historical fault thresholds, providing clear data support for fault determination.

[0113] In summary, multi-dimensional verification of preliminary abnormal characteristic values ​​can further improve their accuracy and reliability. Parameter correlation verification ensures that voltage, current, and temperature characteristic values ​​conform to the physical laws of cable faults; historical data comparison verification ensures that characteristic values ​​are within a reasonable fault range; and threshold range verification eliminates abnormal deviations. This multi-dimensional, layer-by-layer verification corrects deviations in preliminary characteristic values, avoids misjudgments caused by single-dimensional verification, and ultimately yields abnormal characteristic values ​​that accurately reflect the cable fault state.

[0114] In summary, extracting the weighted distribution sequence and reference benchmark sequence of the target fault data sequence based on the fault sensitivity mode makes feature value derivation more targeted and referential. The weighted distribution sequence assigns weights to each parameter in the fault sensitivity mode according to their influence on fault determination, highlighting the role of key parameters and avoiding the weakening of key fault information due to the indiscriminate participation of all parameters in the calculation. The reference benchmark sequence uses the normal operating parameters of similar cables as a standard, providing a precise reference for judging whether the target fault data sequence is abnormal, solving the "unstandardized judgment" problem that easily occurs in traditional non-benchmark analysis, and ensuring that there is a clear comparison standard for subsequent deviation calculation.

[0115] In summary, by performing anomaly feature evolution on the reference baseline sequence based on the weighted distribution sequence to obtain preliminary anomaly feature values, the difference between the target fault data and the normal baseline can be transformed into quantifiable feature indicators. By strengthening the influence of key parameter deviations through weighting coefficients and flexibly adapting the sensitivity adjustment factor to the fault sensitivity of different cable types, and then applying nonlinear processing to the deviation using a hyperbolic tangent function, excessive interference from extreme deviations can be avoided. This ensures that the preliminary anomaly feature values ​​accurately reflect fault differences and closely match the actual fault characteristics of the cable. Compared to simply calculating deviation values, the feature values ​​obtained in this method more precisely characterize the fault severity.

[0116] In summary, deriving preliminary anomaly characteristic values ​​using a calculation formula that includes factors such as the number of data points, weighting coefficients, and sensitivity adjustment factors makes the characteristic value derivation process more logical and rigorous. The formula integrates the biases of all data points through summation, ensuring that all fault data participates in characteristic value generation and avoiding the bias of analyzing a single data point. Simultaneously, the clear definition and synergistic effect of each factor allow characteristic value derivation to break free from the limitations of traditional subjective experience-based judgment, achieving standardized calculation and improving the consistency and comparability of preliminary anomaly characteristic values ​​across different cables and fault scenarios, laying a reliable foundation for subsequent multi-dimensional verification.

[0117] S3. Perform multi-level comparison between the abnormal feature values ​​and the historical fault threshold database to obtain the graded comparison results of the track construction cable;

[0118] In this embodiment of the invention, the step of performing multi-level comparisons between the abnormal feature values ​​and the historical fault threshold database to obtain the graded comparison results of the track construction cable includes:

[0119] The comparison priority order of the abnormal feature values ​​is determined based on the comparison hierarchy in the historical fault threshold database;

[0120] Based on the comparison priority order, the discreteness of the abnormal feature values ​​is checked to obtain the feature deviation of the abnormal feature values;

[0121] Based on the feature deviation, the abnormal feature values ​​are subjected to multi-level refinement comparison to obtain the hierarchical deviation features of the abnormal feature values;

[0122] The hierarchical deviation features are mapped to the comparison hierarchy to obtain the hierarchical comparison results of the track construction cable.

[0123] Specifically, when determining the comparison priority order of abnormal feature values ​​based on the comparison level in the historical fault threshold database, the historical fault threshold database is viewed. Its comparison level is divided into minor faults, moderate faults, and severe faults according to the severity of the fault. The priority order is determined according to the urgency of the handling as severe fault level, moderate fault level, and minor fault level. That is, the severe fault level threshold is compared first, and then the moderate and minor fault level thresholds are compared in turn.

[0124] Furthermore, based on the comparison priority order, the dispersion of abnormal feature values ​​is checked. When the feature deviation of abnormal feature values ​​is obtained, the difference between the abnormal feature values ​​and the threshold of each level is calculated in the order of severe, moderate and minor fault levels. The difference is compared with the fluctuation range of abnormal feature values ​​in the historical fault data of the same level to determine the current degree of deviation. The comprehensive index obtained by integrating the deviation degrees of each level is the feature deviation degree.

[0125] Furthermore, based on the feature deviation degree, multi-level refinement comparisons are performed on the abnormal feature values ​​to obtain the hierarchical deviation features of the abnormal feature values. The specific manifestation of the feature deviation degree is analyzed for each level in order of priority. For severe fault level analysis, it is determined whether the deviation is caused by multi-parameter or single-parameter anomalies and the duration. For moderate level analysis, it is determined whether the change trend is determined. For minor level analysis, it is determined whether it is related to specific working conditions. The feature set formed by organizing these results is the hierarchical deviation feature.

[0126] Furthermore, when mapping the hierarchical deviation features to the comparison hierarchy to obtain the hierarchical comparison results of the track construction cable, the detailed analysis results of each level in the hierarchical deviation features are matched with the standard features of the corresponding level in the historical fault threshold library to obtain the comparison results of each level. The results are then integrated in priority order to form a complete result containing whether each level reaches the threshold and the specific manifestation of the deviation. This result is the hierarchical comparison result.

[0127] In summary, prioritizing the comparison of abnormal feature values ​​based on the comparison hierarchy in the historical fault threshold database allows for fault comparisons that better align with actual maintenance needs. The comparison hierarchy in the historical fault threshold database is divided according to fault severity, prioritizing the comparison of severe faults followed by moderate and minor faults. This approach quickly identifies high-risk faults, avoiding the resource waste caused by traditional single-level comparisons, such as missing severe faults or prioritizing minor faults, thus improving the timeliness and targeted nature of fault response.

[0128] In summary, the dispersion of abnormal feature values ​​is verified based on the comparison priority order to obtain the feature deviation, which quantifies the degree of deviation between abnormal feature values ​​and historical fault thresholds. By comparing the fluctuation range of abnormal feature values ​​with the historical fault data at the same level, the current degree of deviation can be determined. This avoids the one-sidedness of judging solely based on whether the threshold is met, and accurately identifies special cases such as "not exceeding the threshold but with a significant deviation trend" or "exceeding the threshold but with a small deviation," providing a quantitative basis for subsequent detailed comparisons.

[0129] In summary, multi-level refinement and comparison of abnormal feature values ​​based on feature deviation can yield hierarchical deviation features, enabling in-depth analysis of the specific manifestations of faults. Analyzing at different levels whether deviations are caused by multi-parameter or single-parameter anomalies, the trend of deviation changes, and whether they are related to specific operating conditions breaks away from the traditional superficial judgment that only focuses on "whether a fault exists," obtaining deeper information such as the causes and development trends of faults. This makes fault analysis more comprehensive and provides detailed support for subsequent fault level classification.

[0130] In summary, mapping hierarchical deviation features to comparison levels yields tiered comparison results, generating structured fault comparison output. Matching the detailed analysis results of each level with their corresponding levels and integrating them according to priority clearly presents the specific details of faults of different severity. Compared to traditional fragmented comparison results, structured tiered comparison results facilitate subsequent quantitative scoring and anomaly indication information generation, improving the coherence and efficiency of the entire diagnostic process. Simultaneously, it provides maintenance personnel with clear data references for quickly grasping the full picture of the fault.

[0131] S4. Based on the real-time operating parameters of the track construction cable, the graded comparison results are quantitatively scored to obtain the abnormal indication information of the track construction cable;

[0132] In this embodiment of the invention, the step of quantifying and scoring the graded comparison results based on the real-time operating parameters of the track construction cable to obtain the abnormal indication information of the track construction cable includes:

[0133] Based on the load status and environmental conditions of the real-time operating parameters in the track construction cable, the hierarchical scoring benchmark in the hierarchical comparison results is determined;

[0134] Based on the hierarchical scoring benchmark, the hierarchical comparison results are weighted to obtain the hierarchical weight coefficients of the hierarchical comparison results.

[0135] Based on the hierarchical weight coefficients, the hierarchical comparison results are evaluated in multiple dimensions to obtain a preliminary quantitative score for the hierarchical comparison results.

[0136] The preliminary quantitative score is matched step by step with the real-time operating parameters to obtain the abnormal indication information of the track construction cable.

[0137] The step of evaluating the hierarchical comparison results in multiple dimensions based on the hierarchical weight coefficients to obtain a preliminary quantitative score for the hierarchical comparison results includes:

[0138] Based on the hierarchical weight coefficients, determine the dynamic influence factors of the hierarchical comparison results;

[0139] Based on the dynamic influence factor, the classification comparison results are quantitatively evaluated to obtain a preliminary quantitative score for the classification comparison results. The calculation formula for the preliminary quantitative score is as follows:

[0140] ;

[0141] In the formula, For the aforementioned preliminary quantitative score, To compare the total number of levels, For the first Weighting coefficients for each level, It is the hyperbolic tangent function. For the first The quantified values ​​of the hierarchical comparison results at each level. For the first Stability adjustment factors at each level, For the first The gradient of changes in the hierarchical comparison results at each level.

[0142] Specifically, when determining the tiered scoring benchmarks in the graded comparison results based on the load status and environmental conditions of the real-time operating parameters of the track construction cable, the real-time load status of the cable is first obtained, such as whether the current load is light, full, or overloaded, as well as the real-time environmental conditions, such as temperature, humidity, and vibration. If the load is overloaded and the ambient temperature is high, it indicates that the cable is under great operating pressure. At this time, the scoring benchmark for the severe fault tier will be increased, and the scoring benchmarks for the moderate and minor fault tiers will also be adjusted according to the corresponding degree of impact, forming a scoring benchmark for each tier that matches the real-time status.

[0143] Furthermore, based on the hierarchical scoring benchmark, the hierarchical comparison results are weighted and the hierarchical weight coefficients of the hierarchical comparison results are obtained. The weights are assigned according to the level of the scoring benchmark. The level with the higher scoring benchmark has a greater impact on the anomaly assessment and has a higher weight coefficient. For example, the severe fault level has the highest scoring benchmark and its weight coefficient is the largest, followed by the moderate fault level and the minor fault level has the smallest weight coefficient. These weight values ​​are bound to the corresponding level, which are the hierarchical weight coefficients of the hierarchical comparison results.

[0144] Furthermore, based on the hierarchical weight coefficient, the hierarchical comparison results are evaluated in multiple dimensions. When obtaining the preliminary quantitative score of the hierarchical comparison results, the comparison results of each level are evaluated from three dimensions: the degree of deviation from the threshold, the duration, and the scope of influence. The score is calculated by combining the weight coefficient of that level. The further the deviation from the threshold, the longer the duration, and the larger the scope of influence, the higher the score. The scores of each level are added together, and the sum is the preliminary quantitative score.

[0145] Furthermore, when the preliminary quantitative score is matched with the real-time operating parameters step by step to obtain the abnormal indication information of the track construction cable, the abnormal level corresponding to different scoring intervals is first set, and then the interval in which the preliminary quantitative score is located is checked to determine the corresponding abnormal level. At the same time, the specific abnormal parameters that caused the score in the real-time operating parameters are combined, such as excessive current or excessive temperature. The abnormal level and specific abnormal parameters are integrated to form information that includes the severity of the abnormality and the specific cause, which is the abnormal indication information.

[0146] Specifically, when determining the dynamic impact factor of the hierarchical comparison results based on the hierarchical weight coefficients, the actual impact corresponding to the weight coefficient of each level is first analyzed. If the weight coefficient of a certain level is high, it means that the comparison result of that level has a greater impact on the overall anomaly assessment. The impact factor is adjusted in combination with the degree of deviation from the threshold in the comparison result of that level. The more serious the deviation, the higher the dynamic impact factor, and vice versa. The weight coefficient of each level is combined with the corresponding degree of deviation to calculate the dynamic impact factor of the hierarchical comparison results.

[0147] Furthermore, based on the dynamic impact factor, the classification comparison results are quantitatively evaluated. When obtaining the preliminary quantitative score of the classification comparison results, the dynamic impact factor of each level is multiplied by the basic score of the comparison result of that level. The basic score is set according to whether the level has reached the fault threshold. If the threshold is reached, the basic score is high, and if it is not reached, it is low. Then, the scores calculated for all levels are added together, and the sum is the preliminary quantitative score of the classification comparison results.

[0148] Specifically, in the parameter sources for the preliminary quantitative scoring, the total number of comparison levels is the total number of comparison levels in the historical fault threshold database. The weight coefficient for each level comes from the weight value of the corresponding level in the level weight coefficient. The quantified value of the hierarchical comparison result at each level is a numerical conversion of the hierarchical comparison result at that level. The stability adjustment factor for each level is a fixed value preset based on the fluctuation of past fault data for that level. The gradient of the hierarchical comparison results at each level is the non-negative value of the difference between the quantized values ​​of two adjacent comparisons at adjacent levels.

[0149] Furthermore, the formula means that by combining the weight coefficients, quantization values, stability adjustment factors, and change gradients of each comparison level, the quantization values ​​are processed through the hyperbolic tangent function, then divided by the denominator containing the stability adjustment factor and change gradient, and the effects of each level are balanced before being added together to obtain a preliminary quantitative score that comprehensively reflects the degree of cable anomaly, providing a basis for generating anomaly indication information.

[0150] Furthermore, the trend of the formula is as follows: Increasing the weight coefficients or quantification values ​​at each level increases the initial quantification score; increasing the stability adjustment factor or change gradient decreases the initial quantification score; and vice versa. All level parameters jointly affect the score.

[0151] In summary, determining the hierarchical scoring benchmark based on the load status and environmental conditions of real-time operating parameters ensures that the scoring benchmark aligns with the actual operating scenarios of the cable. When the cable is under overload and the ambient temperature is high, raising the scoring benchmark for the severe fault level avoids misjudging the severity of faults caused by traditional fixed scoring benchmarks being out of touch with actual operating conditions. This ensures that the scoring benchmark matches the real-time operating pressure of the cable, providing a standard that conforms to the actual scenario for subsequent quantitative scoring.

[0152] In summary, assigning weights to the tiered comparison results based on tiered scoring benchmarks to obtain tiered weight coefficients can highlight the impact of high-risk tiers. Tiers with higher scoring benchmarks correspond to higher weight coefficients, giving them a larger share in the overall quantitative score. This avoids the weakening of the impact of high-risk faults due to indiscriminate weighting across tiers, allowing the quantitative score to more accurately reflect the threat level of faults at different tiers to cable safety.

[0153] In summary, a preliminary quantitative score is obtained by multi-dimensionally evaluating the hierarchical comparison results based on hierarchical weight coefficients, which can comprehensively consider multiple characteristics of the fault. The comparison results at each level are evaluated from dimensions such as the degree of deviation from the threshold, duration, and scope of impact. The scores are calculated and summed using the weight coefficients. Compared to single-dimensional evaluation, multi-dimensional evaluation allows the preliminary quantitative score to more comprehensively reflect the actual situation of the fault, avoiding the one-sidedness of single-indicator evaluation and improving the accuracy of the score.

[0154] In summary, matching the initial quantitative score with real-time operating parameters at each level to obtain anomaly indication information transforms the quantitative score into a concrete fault indication. By determining the anomaly level corresponding to the score through matching and clarifying the specific parameters causing the anomaly in conjunction with real-time operating parameters, this avoids the problem of traditional methods that only output scores without specific causes. The anomaly indication information includes both the severity of the fault and the identifiable cause, providing a direct and clear basis for subsequent fault level classification and maintenance.

[0155] In summary, determining the dynamic impact factor of the hierarchical comparison results based on hierarchical weight coefficients allows the impact factor to be closely linked to the actual importance of faults at each level. For levels with higher hierarchical weight coefficients, if their comparison results deviate more severely from the threshold, the dynamic impact factor will increase accordingly. This avoids the problem that traditional fixed impact factors cannot adapt to the differences in fault risk at different levels, ensuring that the impact of each level on the preliminary quantitative score accurately reflects its actual fault threat level and improving the relevance of the scoring.

[0156] In summary, a quantitative evaluation of the hierarchical comparison results based on dynamic influence factors, combined with a calculation formula incorporating elements such as the hyperbolic tangent function and stability adjustment factor, yields a preliminary quantitative score. This approach balances the influence of fault quantification values ​​and state stability on the score. The hyperbolic tangent function performs non-linear processing on the quantification values ​​of the hierarchical comparison results, avoiding excessive interference from extreme quantification values. The stability adjustment factor and variation gradient consider the fluctuations in the hierarchical comparison results. If a certain level of comparison result fluctuates significantly, its impact on the score is reduced through denominator adjustment. This ensures that the preliminary quantitative score reflects both the severity of the fault and the stability of the fault state, making it more comprehensive and objective than scoring solely based on quantification values.

[0157] In summary, the calculation formula integrates the evaluation results from all levels through summation, ensuring that all data from all comparison levels participate in the initial quantitative scoring, thus avoiding the bias caused by single-level analysis. Simultaneously, each element in the formula is aligned with the actual needs of cable fault diagnosis; for example, weight coefficients correspond to the importance of each level, and changing gradients reflect the fault development trend. This frees the scoring process from subjective experience dependence, achieving standardized calculations and improving the consistency and reliability of initial quantitative scoring for different cables and fault scenarios, laying the foundation for subsequently generating accurate anomaly indication information.

[0158] S5. Based on the abnormal indication information, classify the abnormal state of the track construction cable to obtain the fault severity level of the track construction cable;

[0159] In this embodiment of the invention, the step of classifying the abnormal state of the track construction cable based on the abnormal indication information to obtain the fault severity level of the track construction cable includes:

[0160] Based on the inherent correlation characteristics of the abnormal indication information, a multi-dimensional status determination basis for the track construction cable is constructed;

[0161] Based on the multi-dimensional state determination criteria, the anomaly indication information is vectorized to obtain the state feature vector of the anomaly indication information.

[0162] Based on a pre-set cable feature library, multimodal feature matching is performed on the state feature vector to obtain the preliminary state classification result of the track construction cable;

[0163] Based on the preliminary state classification results, the state evolution trajectory analysis of the abnormal indication information is performed to obtain the fault severity level of the track construction cable.

[0164] Specifically, when constructing a multi-dimensional status judgment basis for track construction cables based on the inherent correlation characteristics of abnormal indication information, the correlation between the abnormality level and specific abnormal parameters in the abnormal indication information will be analyzed. For example, whether excessive current is often accompanied by excessive temperature will jointly lead to a high abnormality level. At the same time, the relationship between the duration of the abnormality and the scope of influence will be analyzed. For example, the longer the duration and the wider the scope of influence, the more serious the abnormality. These correlations will be organized into a judgment standard that includes dimensions such as abnormal parameter combinations, duration, and scope of influence, which is the multi-dimensional status judgment basis.

[0165] Furthermore, based on the multi-dimensional state determination criteria, the abnormal indication information is vectorized into a feature vector. When obtaining the state feature vector of the abnormal indication information, the corresponding content in the abnormal indication information is converted into a feature value according to each dimension of the multi-dimensional state determination criteria. For example, the abnormal parameter combination corresponds to a specific value, the duration is divided into values ​​according to the length, and the influence range is determined by the size. The ordered set of these feature values ​​arranged in dimensional order is the state feature vector.

[0166] Furthermore, based on the preset cable feature library, multimodal feature matching is performed on the state feature vectors to obtain the preliminary state classification result of the track construction cable. The preset cable feature library stores standard feature vectors corresponding to different fault states. The state feature vectors are compared with the standard feature vectors in the library to find the standard feature vector with the highest similarity. The fault state corresponding to this vector is the preliminary state classification result.

[0167] Furthermore, based on the preliminary state classification results, the state evolution trajectory analysis of the abnormal indication information is performed to obtain the fault severity level of the track construction cable. The changes of the abnormal indication information over time are tracked to see whether the fault state corresponding to the preliminary state classification results is continuously deteriorating, remaining stable, or gradually alleviating. Combining the rate of change and trend, the fault state is mapped to the preset severe, moderate, and minor levels, which is the fault severity level.

[0168] In summary, constructing multi-dimensional status judgment criteria based on the inherent correlation characteristics of anomaly indication information can comprehensively cover the key dimensions of fault judgment. By analyzing the correlation between anomaly level and specific parameters, duration and impact range in anomaly indication information, these correlations are transformed into judgment criteria that include dimensions such as anomaly parameter combinations, duration, and impact range. This avoids misjudgment of fault status caused by traditional single-dimensional judgment criteria, making the judgment criteria more closely match the actual performance of cable faults and providing a scientific framework for accurately classifying anomaly states.

[0169] In summary, by vectorizing anomaly indication information based on multi-dimensional state determination criteria to obtain state feature vectors, abstract anomaly indication information can be transformed into a structured set of numerical values. By converting the corresponding content in the anomaly indication information into feature values ​​according to each dimension of the determination criteria and arranging them in an orderly manner, compared to traditional textual or fragmented information presentation, state feature vectors facilitate subsequent standardized feature matching, reduce information interpretation bias, and improve the efficiency and accuracy of anomaly state classification.

[0170] In summary, based on a pre-defined cable feature library, multi-modal feature matching of state feature vectors yields preliminary state classification results, which can improve the reliability of the classification by leveraging historical fault data. The pre-defined cable feature library stores standard feature vectors corresponding to different fault states. By comparing these vectors, the standard vector with the highest similarity to the current state feature vector and its corresponding fault state are found. This avoids the subjectivity of traditional state classification based on human experience, making the preliminary classification results more consistent with historical fault patterns and reducing the probability of misjudgment.

[0171] In summary, analyzing the state evolution trajectory of abnormal indication information based on the preliminary state classification results to obtain the fault severity level allows for a dynamic assessment of the fault's development trend. By tracking changes in abnormal indication information over time, it determines whether the fault state is worsening, stabilizing, or mitigating, and combines the rate and trend of change to determine the final severity level. This avoids the one-sidedness caused by traditional methods that only classify faults based on the current state, ensuring that the fault severity level not only reflects the current fault situation but also the fault's development trend, providing a more comprehensive basis for subsequent maintenance prioritization.

[0172] S6. Generate a diagnostic report for the track construction cable based on the severity level of the fault.

[0173] In this embodiment of the invention, generating a diagnostic report for the track construction cable based on the severity level of the fault includes:

[0174] Based on the severity level of the fault and the type parameter, determine the technical analysis dimensions and level of detail of the diagnostic report for the track construction cable;

[0175] Based on the aforementioned technical analysis dimensions, key diagnostic elements in the abnormal feature values, the hierarchical comparison results, and the abnormal indication information are determined.

[0176] A multi-angle correlation analysis was performed on the key diagnostic elements to obtain the core report content of the track construction cable;

[0177] Based on the level of detail described, the core report content is expanded hierarchically to obtain the structured report content for the track construction cable.

[0178] The structured report content is compared with the historical diagnostic records of the track construction cable using a multi-dimensional spatiotemporal consistency verification to obtain the diagnostic report of the track construction cable.

[0179] Specifically, when determining the technical analysis dimensions and level of detail in the diagnostic report for track construction cables based on the severity level and type parameters of the fault, the severity level is considered first. The higher the severity level, the more comprehensive the technical analysis dimensions should be, covering the causes of the fault, the scope of impact, and the development trend. The level of detail should also be higher, including specific parameter change curves and the impact of related equipment. At the same time, relevant analysis dimensions are added in combination with type parameters, such as cable material and laying method. For example, soil environmental impact analysis is added for buried cables. Finally, the technical analysis dimensions and level of detail that meet the requirements are determined.

[0180] Furthermore, when determining key diagnostic elements in abnormal feature values, hierarchical comparison results, and abnormal indication information based on technical analysis dimensions, corresponding feature parameters are extracted from abnormal feature values ​​for each technical analysis dimension. For example, abnormal current feature values ​​are extracted from the fault cause dimension. Relevant hierarchical comparison data are screened from the hierarchical comparison results. For example, the impact data of severe fault levels are screened from the impact range dimension. Related content is selected from the abnormal indication information. For example, abnormal change trend information is selected from the development trend dimension. The information extracted and screened are the key diagnostic elements.

[0181] Furthermore, when conducting multi-angle correlation analysis on key diagnostic elements to obtain the core report content for track construction cables, the correlation between abnormal characteristic values ​​and hierarchical comparison results will be analyzed from the perspective of causality, such as whether a certain abnormal characteristic value leads to the corresponding hierarchical comparison result. The correlation of changes in key diagnostic elements will be analyzed from the time dimension, such as whether the trend in abnormal indication information is consistent with the change of abnormal characteristic values. The conclusions of these correlation analyses will be integrated to form the core report content, which includes the core cause of the fault, the main impact, and the development trend.

[0182] Furthermore, based on the level of detail, the core report content is expanded hierarchically to obtain a structured report on track construction cables. In accordance with the level of detail requirements, supporting data is added to each core conclusion based on the core report content. For example, the cause of the fault is supplemented with the specific parameter values ​​that exceed the standard, the scope of impact is supplemented with the cable sections involved, and the development trend is supplemented with details of recent parameter changes. Then, the data is arranged in a logical order according to the technical analysis dimensions to form a structured report with clear hierarchy and complete content.

[0183] Furthermore, the structured report content is compared with the historical diagnostic records of the track construction cable in a multi-dimensional spatiotemporal consistency verification. When the diagnostic report of the track construction cable is obtained, the similarity between the current structured report content and the historical diagnostic records of the same period is compared from the time dimension, such as whether similar faults occur in the same season. From the spatial dimension, the correlation between the fault location and the historical records is compared, such as whether the faults occur frequently in the same section. This ensures that the current report content is consistent with historical patterns. The structured report content after verification is the diagnostic report of the track construction cable.

[0184] In summary, determining the technical analysis dimensions and level of detail in the diagnostic report based on the severity and type parameters of the fault makes the report more targeted and practical. Higher fault severity levels require more comprehensive analysis dimensions and more detailed content. Supplementing these with specific analysis dimensions based on type parameters avoids the problems of missing dimensions or redundant content found in traditional "one-size-fits-all" reports, ensuring that the report both aligns with the actual characteristics of the cable and meets the needs of fault handling.

[0185] In summary, by identifying key diagnostic elements from abnormal characteristic values, hierarchical comparison results, and abnormal indication information based on technical analysis dimensions, core information can be accurately extracted from data across multiple stages. By selecting corresponding key content from data at each stage according to dimensions such as fault causes and development trends, the report avoids becoming a mere accumulation of all data, allowing diagnostic elements to focus on the core of fault analysis. This provides precise data support for subsequent correlation analysis and report content construction, enhancing the report's professionalism and focus.

[0186] In summary, multi-faceted correlation analysis of key diagnostic elements yields core report content, revealing the inherent logic between fault data. By analyzing the correlation between abnormal characteristic values ​​and grading comparison results from a causal perspective, and analyzing the changes in key elements over time, scattered key elements are integrated into core content containing conclusions such as the core cause of the fault and its main impacts. This avoids the problem of traditional reports merely listing data without in-depth analysis, allowing the report to clearly present the essence of the fault and provide a clear direction for maintenance.

[0187] In summary, hierarchically expanding the core report content based on the level of detail yields a structured report, ensuring clear logic and complete content. Supplementing core conclusions with supporting data according to required levels of detail and sorting them by technical analysis dimensions transforms the core content into a well-structured, hierarchical format. This avoids the problems of disorganized and unfocused content in traditional reports, allowing maintenance personnel to quickly grasp key information and improving readability and efficiency.

[0188] In summary, performing multi-dimensional spatiotemporal consistency verification between structured report content and historical diagnostic records can improve the accuracy and reliability of the report. Comparing the similarity between the current report and historical records from the same period in time, and comparing the correlation between the fault location and historical records in space, ensures that the current report content conforms to the historical patterns of cable faults, corrects potential deviations, avoids misjudgments caused by isolated analysis of current data, and makes the final diagnostic report more credible, providing a reliable basis for fault repair.

[0189] like Figure 2 The diagram shown is a functional block diagram of an online fault diagnosis and intelligent data analysis system for track construction cables provided in an embodiment of the present invention.

[0190] The online fault diagnosis and intelligent data analysis system 100 for railway construction cables described in this invention can be installed in an electronic device. Depending on the functions implemented, the online fault diagnosis and intelligent data analysis system 100 for railway construction cables may include a data fusion module 101, a fault identification module 102, a multi-level comparison module 103, a quantitative scoring module 104, a grade classification module 105, and a diagnostic report module 106. 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.

[0191] In this embodiment, the functions of each module / unit are as follows:

[0192] The data fusion module 101 is used to perform multi-dimensional fusion of the multi-source operation data of the track construction cable based on the importance of the parameters of the track construction cable, so as to obtain the fused data of the track construction cable.

[0193] The fault identification module 102 is used to extract fault features from the fused data based on the type parameters of the track construction cable, and obtain the abnormal feature values ​​of the track construction cable.

[0194] The multi-level comparison module 103 is used to perform multi-level comparison between the abnormal feature value and the historical fault threshold database to obtain the graded comparison result of the track construction cable.

[0195] The quantitative scoring module 104 is used to quantitatively score the graded comparison results based on the real-time operating parameters of the track construction cable, and obtain the abnormal indication information of the track construction cable.

[0196] The classification module 105 is used to classify the abnormal state of the track construction cable based on the abnormal indication information to obtain the fault severity level of the track construction cable.

[0197] The diagnostic report module 106 is used to generate a diagnostic report for the track construction cable based on the severity level of the fault.

[0198] 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.

[0199] 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.

[0200] 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.

[0201] 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.

[0202] The embodiments of this application 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.

[0203] 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 track construction cable fault online diagnosis and data intelligent analysis method, characterized in that, The method comprises: S1. Based on the importance of the parameters of the track construction cable, multi-dimensional fusion is performed on the multi-source operation data of the track construction cable to obtain fusion data of the track construction cable; S2. Based on the type parameters of the track construction cable, extract the fault features in the fusion data to obtain the abnormal feature values of the track construction cable, including: According to the type parameters of the track construction cable, identify the fault sensitive mode of the track construction cable; Based on the fault sensitive mode, filter the fault features in the fusion data to obtain the target fault data sequence of the track construction cable; Derive the feature values of the target fault data sequence to obtain the preliminary abnormal feature values of the track construction cable; Multi-dimensional verification is performed on the preliminary abnormal feature values to obtain the abnormal feature values of the track construction cable; S3. Multi-level comparison is performed between the abnormal feature values and the historical fault threshold library to obtain the hierarchical comparison result of the track construction cable; S4. Based on the real-time operation parameters of the track construction cable, the hierarchical comparison result is quantitatively scored to obtain the abnormal indication information of the track construction cable, including: According to the load state and environmental conditions of the real-time operation parameters in the track construction cable, determine the hierarchical scoring benchmark in the hierarchical comparison result; Based on the hierarchical scoring benchmark, weight distribution is performed on the hierarchical comparison result to obtain the hierarchical weight coefficient of the hierarchical comparison result; Based on the hierarchical weight coefficient, the hierarchical comparison result is evaluated in multiple dimensions to obtain the preliminary quantitative score of the hierarchical comparison result, including: According to the hierarchical weight coefficient, determine the dynamic influence factor of the hierarchical comparison result; Based on the dynamic influence factor, the hierarchical comparison result is quantitatively evaluated to obtain the preliminary quantitative score of the hierarchical comparison result, wherein the calculation formula of the preliminary quantitative score is as follows: ； In the formula, For the aforementioned preliminary quantitative score, To compare the total number of levels, For the first Weighting coefficients for each level, It is the hyperbolic tangent function. For the first The quantified values ​​of the hierarchical comparison results at each level. For the first Stability adjustment factors at each level, For the first The gradient of changes in the hierarchical comparison results at each level; The preliminary quantitative score is matched with the real-time operation parameters step by step to obtain the abnormal indication information of the track construction cable; S5. Based on the abnormal indication information, the abnormal state of the track construction cable is divided to obtain the fault severity level of the track construction cable; S6. According to the fault severity level, a diagnostic report of the track construction cable is generated.

2. The method of claim 1, wherein the method further comprises: Based on the importance of the parameters of the track construction cable, the multi-dimensional fusion is performed on the multi-source operation data of the track construction cable to obtain the fusion data of the track construction cable, including: Synchronously collecting voltage, current and temperature parameters in the track construction cable; The voltage, current and temperature parameters are used as the multi-source operation data of the track construction cable; According to the importance of the parameters of the track construction cable, determine the fusion order of the multi-source operation data; Time alignment processing is performed on the multi-source operation data to obtain time alignment data of the multi-source operation data; The time alignment data is integrated according to the preset data standard to obtain the specification data of the track construction cable; According to the fusion order, the specification data is combined into the fusion data of the track construction cable.

3. The method of claim 1, wherein the method further comprises: determining the location of the fault in the track construction cable based on the data received from the plurality of sensors; and determining the type of the fault in the track construction cable based on the data received from the plurality of sensors. The eigenvalue derivation is performed on the target fault data sequence to obtain a preliminary abnormal eigenvalue of the track construction cable, including: According to the fault sensitive mode, a weight distribution sequence and a reference benchmark sequence of the target fault data sequence are extracted; Based on the weight distribution sequence, the reference benchmark sequence is subjected to abnormal feature evolution to obtain the preliminary abnormal eigenvalue of the track construction cable, wherein the calculation formula of the preliminary abnormal eigenvalue is as follows: ； In the formula, is the preliminary abnormal characteristic value, is the number of data points in the target fault data sequence, is the weight coefficient of the i-th data point in the weight distribution sequence, is the hyperbolic tangent function, is the sensitivity adjustment factor, is the i-th data point in the target fault data sequence, is the i-th reference value in the reference benchmark sequence, is the absolute deviation between the i-th data point in the target fault data sequence and the i-th reference value in the reference benchmark sequence, ​​​​​ 4. The method of claim 1, wherein the method further comprises: determining the location of the fault in the track construction cable based on the data received from the plurality of sensors; and determining the type of the fault in the track construction cable based on the data received from the plurality of sensors. The abnormal eigenvalue is compared with a historical fault threshold library in multiple levels to obtain a hierarchical comparison result of the track construction cable, including: According to the comparison level in the historical fault threshold library, an order of comparison priority of the abnormal eigenvalue is determined; Based on the order of comparison priority, a dispersion check is performed on the abnormal eigenvalue to obtain a feature deviation degree of the abnormal eigenvalue; Based on the feature deviation degree, a multi-layer refined comparison is performed on the abnormal eigenvalue to obtain a hierarchical deviation feature of the abnormal eigenvalue; The hierarchical deviation feature is mapped to the comparison level to obtain the hierarchical comparison result of the track construction cable.

5. The method of claim 1, wherein the method further comprises: determining the location of the fault in the track construction cable based on the data collected by the plurality of sensors; and determining the cause of the fault in the track construction cable based on the data collected by the plurality of sensors. Based on the abnormal indication information, an abnormal state of the track construction cable is divided to obtain a fault severity level of the track construction cable, including: According to the internal correlation characteristics of the abnormal indication information, a multi-dimensional state judgment basis of the track construction cable is constructed; Based on the multi-dimensional state judgment basis, a feature vectorization is performed on the abnormal indication information to obtain a state feature vector of the abnormal indication information; Based on a preset cable feature library, a multi-modal feature matching is performed on the state feature vector to obtain a preliminary state division result of the track construction cable; Based on the preliminary state division result, a state evolution trajectory analysis is performed on the abnormal indication information to obtain the fault severity level of the track construction cable.

6. The method of claim 1, wherein the method further comprises: The diagnostic report of the track construction cable is generated according to the fault severity level, including: According to the fault severity level and the type parameter, a technical analysis dimension and a content detail level of the diagnostic report of the track construction cable are determined; According to the technical analysis dimension, key diagnostic elements in the abnormal eigenvalue, the hierarchical comparison result and the abnormal indication information are determined; A multi-angle correlation analysis is performed on the key diagnostic elements to obtain core report content of the track construction cable; Based on the content detail level, the core report content is hierarchically expanded to obtain structured report content of the track construction cable; The structured report content is subjected to multi-dimensional spatiotemporal consistency check with historical diagnostic records of the track construction cable to obtain the diagnostic report of the track construction cable.

7. An online track construction cable fault diagnosis and data intelligent analysis system, configured to implement the online track construction cable fault diagnosis and data intelligent analysis method of claim 1, the system comprising: a data fusion module configured to perform multi-dimensional fusion on multi-source operation data of the track construction cable based on parameter importance of the track construction cable to obtain fused data of the track construction cable; The fault identification module is configured to extract a fault feature in the fusion data based on a type parameter of the track construction cable, and obtain an abnormal feature value of the track construction cable; The multi-stage comparison module is configured to perform multi-stage comparison between the abnormal feature value and a historical fault threshold library, and obtain a grading comparison result of the track construction cable; The quantitative scoring module is configured to perform quantitative scoring on the grading comparison result based on a real-time operation parameter of the track construction cable, and obtain abnormal indication information of the track construction cable; The grade division module is configured to divide an abnormal state of the track construction cable based on the abnormal indication information, and obtain a fault severity grade of the track construction cable; The diagnosis report module is configured to generate a diagnosis report of the track construction cable according to the fault severity grade.

Images