A method and apparatus for detecting faults in a cable line

Abstract

The application discloses a kind of cable line fault detection method and device, it is relevant to cable detection technical field, the method includes: obtaining the cable line to be detected;The terminal node on the cable line to be detected is identified, the cable line to be detected is divided according to terminal node, and N cable lines are output;Based on N cable lines, record N groups of terminal nodes;Terminal operation monitoring is carried out, and N groups of monitoring data sets are output;The probability of occurrence of failure of N cable lines is identified, and N failure occurrence probabilities are output;N failure occurrence probabilities are input into balanced segmentation model, and M cable lines are output;The cable line to be detected is segmented fault detection.It solves the technical problems that existing cable line detection cannot accurately and quickly locate cable line fault points, cannot comprehensively monitor and evaluate cable line faults, and achieves the technical effect of improving the efficiency and accuracy of cable line fault positioning.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cable detection, and particularly relates to a fault detection method and device for a cable line. BACKGROUND

[0002] In a power system, as a key component for power transmission and distribution, the stability and safety of the operation state of a cable line directly relate to the reliability and efficiency of the entire power grid. However, with the aging of the cable line, the erosion of environmental factors, and improper human operation, cable faults occur from time to time, which not only may cause power failure accidents, but also may cause damage to equipment and even endanger personnel safety. Traditional cable fault detection methods mostly rely on manual inspection and simple instrument testing, and have problems such as long detection period and low accuracy. In particular, in a complex cable network, only limited monitoring data is used, which is difficult to fully reflect the operation state of the cable line, and accurate positioning of the fault point is particularly difficult, which often requires a large amount of manpower and material resources, and it is difficult to quickly respond and handle faults, thereby affecting the detection efficiency of the cable line and the accuracy of the health state evaluation of the cable line.

[0003] Therefore, in the current cable line detection related technology, there is a technical problem that it is difficult to accurately and quickly locate the fault point of the cable line, and it is impossible to comprehensively monitor and evaluate the fault of the cable line. SUMMARY

[0004] The present application provides a fault detection method and device for a cable line, which uses terminal node identification, cable line division, fault occurrence probability evaluation, and balanced segmentation, solves the technical problem that the existing cable line detection is difficult to accurately and quickly locate the fault point of the cable line, and cannot comprehensively monitor and evaluate the fault of the cable line, and achieves the technical effect of improving the efficiency and accuracy of cable line fault positioning.

[0005] The application provides a cable line fault detection method, which comprises the following steps: obtaining a cable line to be detected; identifying terminal nodes on the cable line to be detected, dividing the cable line to be detected according to the terminal nodes, and outputting N cable lines; recording N groups of terminal nodes based on the N cable lines, wherein each group of terminal nodes comprises a first terminal node and a second terminal node, and the first terminal node and the second terminal node are nodes at two ends of a corresponding cable line; performing terminal operation monitoring according to the N groups of terminal nodes, and outputting N groups of monitoring data sets; identifying fault occurrence probabilities of the N cable lines according to the N groups of monitoring data sets, and outputting N fault occurrence probabilities; inputting the N fault occurrence probabilities into an equal segmentation model, and outputting M cable lines, wherein the sum of fault occurrence probabilities of the M cable lines is less than an expected variance; and performing segmented fault detection on the cable line to be detected according to the M cable lines.

[0006] In a possible implementation, the N fault occurrence probabilities are input into the equal segmentation model to output M cable lines, and the following processing is further performed: initializing equalization parameters, including initializing a total cable length, initializing a number of cable segments, and initializing a fault occurrence probability of each cable segment; and calculating a fault occurrence probability variance of each cable segment, and the expression is as follows:

[0007] ;

[0008] wherein, represents a fault occurrence probability variance from a kth segment to an ith segment, i represents an ending position of a current cable segment, and k represents a starting position of the current cable segment, represents a length of the cable segment, and outputs an average weight of all fault occurrence probabilities in the cable segment, represents a fault occurrence probability of an mth segment, represents a fault occurrence probability mean value from the kth segment to the ith segment, represents a sum of all fault occurrence probabilities from the kth segment to the ith segment; and the M cable lines are output by minimizing a difference between the fault occurrence probability variance of each cable segment and the expected variance.

[0009] In a possible implementation, the fault occurrence probability variance of each cable segment is calculated, and the following processing is further performed: obtaining a segmentation point k by iterative calculation, so that the sum of fault occurrence probability variances of each cable segment is minimum, and the iterative calculation formula is as follows:

[0010] ;

[0011] wherein, represents a minimum variance sum when the first i nodes are divided into j segments, denotes the selection of the minimum value to ensure the minimum sum of variance, denotes all possible split points k before node i, denotes the division of the first k nodes into the minimum sum of variance at the segment, denotes the variance of the failure occurrence probability of the segment from node k+1 to node i.

[0012] In a possible implementation, the terminal nodes on the cable line to be detected are identified, and the following processing is performed: the terminal nodes on the cable line to be detected are identified, P terminal nodes are identified; each terminal node in the P terminal nodes is analyzed to determine whether the number of devices connected to each terminal node is greater than or equal to 2, and P first-type terminal nodes with a device number greater than or equal to 2 and P second-type terminal nodes with a device number less than 2 are obtained; the P first-type terminal nodes are subjected to work fusion monitoring, and P fusion monitoring data sets are output.

[0013] In a possible implementation, after the N failure occurrence probabilities of the N cable lines are output, the following processing is further performed: the cable material type, the cable environmental condition, and the cable aging degree are identified; the probability adjustment weight is generated according to the cable material type, the cable environmental condition, and the cable aging degree; the calculation weight of the failure occurrence probability is dynamically adjusted by using the probability adjustment weight, and the N failure occurrence probabilities are updated.

[0014] In a possible implementation, after the N failure occurrence probabilities are output, the following processing is further performed: the cable material type, the cable environmental condition, and the cable aging degree are identified; the probability adjustment weight is generated according to the cable material type, the cable environmental condition, and the cable aging degree; the calculation weight of the failure occurrence probability is dynamically adjusted by using the probability adjustment weight, and the N failure occurrence probabilities are updated.

[0015] In a possible implementation, after the M cable lines are output, the following processing is further performed: the historical data set of the failure occurrence probability of the M cable lines is obtained; the historical data set is input into a failure prediction model for training, and a probability prediction model is output; the future failure occurrence probability of the M cable lines is predicted based on the probability prediction model, and the M cable lines are optimized according to the prediction result.

[0016] ​​​​The application also provides a cable line fault detection device, comprising: a to-be-detected cable line acquisition module, configured to acquire a to-be-detected cable line; a terminal node identification module, configured to identify terminal nodes on the to-be-detected cable line, divide the to-be-detected cable line according to the terminal nodes, and output N cable lines; a terminal node recording module, configured to record N groups of terminal nodes based on the N cable lines, wherein each group of terminal nodes comprises a first terminal node and a second terminal node, and the first terminal node and the second terminal node are nodes at two ends of a corresponding cable line; a monitoring data set output module, configured to perform terminal operation monitoring according to the N groups of terminal nodes, and output N groups of monitoring data sets; a fault occurrence probability identification module, configured to identify fault occurrence probabilities of the N cable lines according to the N groups of monitoring data sets, and output N fault occurrence probabilities; a cable line balanced segmentation module, configured to input the N fault occurrence probabilities into a balanced segmentation model, and output M cable line segments, wherein a variance of the sum of fault occurrence probabilities corresponding to the M cable line segments is less than an expected variance; and a segmented fault detection module, configured to perform segmented fault detection on the to-be-detected cable line according to the M cable line segments.

[0017] The application provides a cable line fault detection method and device, which acquire a to-be-detected cable line; identify terminal nodes on the to-be-detected cable line, divide the to-be-detected cable line according to the terminal nodes, and output N cable lines; record N groups of terminal nodes based on the N cable lines, wherein each group of terminal nodes comprises a first terminal node and a second terminal node, and the first terminal node and the second terminal node are nodes at two ends of a corresponding cable line; perform terminal operation monitoring according to the N groups of terminal nodes, and output N groups of monitoring data sets; identify fault occurrence probabilities of the N cable lines according to the N groups of monitoring data sets, and output N fault occurrence probabilities; input the N fault occurrence probabilities into a balanced segmentation model, and output M cable line segments, wherein a variance of the sum of fault occurrence probabilities corresponding to the M cable line segments is less than an expected variance; and perform segmented fault detection on the to-be-detected cable line according to the M cable line segments. The application solves the technical problems that existing cable line detection cannot accurately and quickly locate a cable line fault point and cannot comprehensively monitor and evaluate a cable line, and achieves the technical effects of improving the efficiency and accuracy of cable line fault location. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. Flowcharts are used in the present application to illustrate the operations performed by the device according to the embodiments of the present application. It should be understood that the foregoing or the following operations are not necessarily performed in sequence. On the contrary, various steps can be processed in reverse order or simultaneously according to needs. Meanwhile, other operations can be added to these processes, or one or more steps of operations can be removed from these processes.

[0019] Figure 1 A flowchart of a cable line fault detection method provided by an embodiment of the present application;

[0020] Figure 2 A structural schematic diagram of a cable line fault detection device provided by an embodiment of the present application.

[0021] Legend: cable line to be detected acquisition module 10, terminal node identification module 20, terminal node recording module 30, monitoring data set output module 40, fault occurrence probability identification module 50, cable line equalization segmentation module 60, segmented fault detection module 70. DETAILED DESCRIPTION

[0022] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, the content of the specification can be implemented, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described.

[0023] In order to make the purposes, technical solutions and advantages of the present application more clear, the present application will be further described in detail below with reference to the drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by those of ordinary skill in the art without making creative labor are within the scope of protection of the present application.

[0024] In the following description, "some embodiments" are referred to, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict, and the term "first\second" referred to only distinguishes similar objects, and does not represent a specific order for the objects. The terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, device, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0025] The embodiments of the present application provide a fault detection method for a cable line, as shown in the method comprises: Figure 1

[0026] Step S100, obtaining a cable line to be detected. Based on the aging degree of the cable line, historical fault records, operating environment conditions (such as high temperature, humidity, corrosion, etc.), load conditions and other factors, the cable line needing to be detected is comprehensively evaluated and determined as the cable line to be detected.

[0027] Step S200, identifying the terminal nodes on the cable line to be detected, dividing the cable line to be detected according to the terminal nodes, and outputting N cable lines. The terminal node is an important part of the cable line for connecting cables, devices or other electrical elements. The terminal nodes on the cable line to be detected are identified, which can include visually inspecting the direction and connection points of the cable line to identify the positions of the terminal nodes; or using professional detection equipment (such as a cable detector, a signal generator, a receiver, etc.) to detect the cable line and determine the specific positions of the terminal nodes through signal feedback; after identifying the terminal nodes, the cable line to be detected is divided into several independent detection units, i.e. N cable lines, according to the positions of the terminal nodes. Specifically, according to the physical layout and distribution of the cable line, the cable line between two adjacent terminal nodes is divided into a detection unit, and N independent cable line detection units are obtained after division, wherein N is a positive integer greater than 1, and then N cable lines are outputted, and the starting terminal node and the ending terminal node of each cable line and the length of the cable line are determined as basic information.

[0028] ​In a possible implementation, step S200 further includes step S210 of identifying terminal nodes on the cable line to be detected, and identifying P terminal nodes. All connection points or branch points on the cable line to be detected are identified and marked, and these connection points or branch points, i.e., terminal nodes, are important components in the cable line, connecting cables, devices, or other electrical elements, and ensuring transmission and distribution of electric energy. Specifically, the cable line is detected by using professional detection equipment such as a cable detector, a signal generator, a receiver, and the like, which can send and receive signals, and the specific positions and quantities of the terminal nodes are determined through signal feedback, and finally P terminal nodes are identified.

[0029] Step S220 includes analyzing each of the P terminal nodes, determining whether the number of devices connected to each terminal node is greater than or equal to 2, obtaining P first-type terminal nodes with a device quantity greater than or equal to 2, and P second-type terminal nodes with a device quantity less than 2. The P identified terminal nodes are analyzed, including checking the physical connection of each terminal node, and understanding the types and quantities of cables, devices, or other electrical elements connected thereto. Specifically, the wiring terminals, connectors, or other connection devices of each terminal node are checked, and the number of devices directly connected to the terminal node is counted, so as to determine whether the number of devices connected to the terminal node is greater than or equal to 2. According to the number of connected devices, the terminal nodes are divided into two types, P terminal nodes with a device quantity greater than or equal to 2 are divided into first-type terminal nodes, and P terminal nodes with a device quantity less than 2 are divided into second-type terminal nodes. The P terminal nodes are analyzed, including checking the physical connection of each terminal node, and understanding the types and quantities of cables, devices, or other electrical elements connected thereto. Specifically, the wiring terminals, connectors, or other connection devices of each terminal node are checked, and the number of devices directly connected to the terminal node is counted, so as to determine whether the number of devices connected to the terminal node is greater than or equal to 2. According to the number of connected devices, the terminal nodes are divided into two types, P terminal nodes with a device quantity greater than or equal to 2 are divided into first-type terminal nodes, and P terminal nodes with a device quantity less than 2 are divided into second-type terminal nodes. The P terminal nodes are analyzed, including checking the physical connection of each terminal node, and understanding the types and quantities of cables, devices, or other electrical elements connected thereto. Specifically, the wiring terminals, connectors, or other connection devices of each terminal node are checked, and the number of devices directly connected to the terminal node is counted, so as to determine whether the number of devices connected to the terminal node is greater than or equal to 2. According to the number of connected devices, the terminal nodes are divided into two types, P terminal nodes with a device quantity greater than or equal to 2 are divided into first-type terminal nodes, and P terminal nodes with a device quantity less than 2 are divided into second-type terminal nodes. The P terminal nodes are analyzed, including checking the physical connection of each terminal node, and understanding the types and quantities of cables, devices, or other electrical elements connected thereto. Specifically, the wiring terminals, connectors, or other connection devices of each terminal node are checked, and the number of devices directly connected to the terminal node is counted, so as to determine whether the number of devices connected to the terminal node is greater than or equal to 2. According to the number of connected devices, the terminal nodes are divided into two types, P terminal nodes with a device quantity greater than or equal to 2 are divided into first-type terminal nodes, and P terminal nodes with a device quantity less than 2 are divided into second-type terminal nodes.

[0030] Step S230 includes performing operation fusion monitoring on the P first-type terminal nodes, and outputting P fusion monitoring data sets. The operation fusion monitoring on the P first-type terminal nodes specifically refers to performing multi-parameter and multi-dimensional real-time monitoring on the P terminal nodes connected to multiple devices (i.e., with a device quantity greater than or equal to 2), i.e., synchronously collecting data of multiple key parameters such as current, voltage, temperature, load, and vibration, and outputting P fusion monitoring data sets, each corresponding to one terminal node. Step S230 includes performing operation fusion monitoring on the P first-type terminal nodes, and outputting P fusion monitoring data sets. The operation fusion monitoring on the P first-type terminal nodes specifically refers to performing multi-parameter and multi-dimensional real-time monitoring on the P terminal nodes connected to multiple devices (i.e., with a device quantity greater than or equal to 2), i.e., synchronously collecting data of multiple key parameters such as current, voltage, temperature, load, and vibration, and outputting P fusion monitoring data sets, each corresponding to one terminal node. Step S230 includes performing operation fusion monitoring on the P first-type terminal nodes, and outputting P fusion monitoring data sets. The operation fusion monitoring on the P first-type terminal nodes specifically refers to performing multi-parameter and multi-dimensional real-time monitoring on the P terminal nodes connected to multiple devices (i.e., with a device quantity greater than or equal to 2), i.e., synchronously collecting data of multiple key parameters such as current, voltage, temperature, load, and vibration, and outputting P fusion monitoring data sets, each corresponding to one terminal node.

[0031] ​​​​​​Step S300, based on the N cable lines, record N sets of terminal nodes, wherein each set of terminal nodes includes a first terminal node and a second terminal node, and the first terminal node and the second terminal node are nodes at both ends of the corresponding cable line. According to the divided N cable lines (each cable line is a cable part connected by two or more terminal nodes, with a clear starting point and ending point), the corresponding N sets of terminal nodes are recorded, each set of terminal nodes represents the starting and ending positions of the corresponding cable line segment, i.e. the nodes at both ends of the cable line, and each set of terminal nodes includes a first terminal node and a second terminal node. The node at the starting end of the cable line is usually referred to as the first terminal node, and the node at the ending end of the cable line is referred to as the second terminal node, and the two terminal nodes together define the spatial position and direction of the cable line. By clearly defining the starting and ending terminal nodes of each cable line segment, the specific cable segment can be quickly located, and targeted operation can be performed on the cable fault.

[0032] Step S400, terminal operation monitoring is performed according to the N sets of terminal nodes, and N sets of monitoring data sets are output. Terminal operation monitoring refers to real-time monitoring of terminal nodes on the cable line to obtain their working status, operating parameters and other key information. Specifically, a plurality of key parameters of the N sets of terminal nodes are monitored, including but not limited to current, voltage, terminal node temperature, load and vibration data passing through the terminal nodes, etc. Each set of terminal nodes includes a first terminal node and a second terminal node, and terminal operation monitoring is performed on the N sets of terminal nodes, and N sets of monitoring data sets are output. Each set of monitoring data set includes a first terminal node operation monitoring data set and a second terminal node operation monitoring data set. By monitoring and analyzing the monitoring data of the terminal nodes of the cable line, abnormal conditions of the terminal nodes can be found in time, potential faults of the cable line can be predicted, the operating state of the cable line can be evaluated, and the safe and stable operation of the cable line can be ensured.

[0033] Step S500, according to the N sets of monitoring data sets, identify the fault occurrence probability of the N cable lines, and output N fault occurrence probabilities.

[0034] Further, step S500 identifies the failure occurrence probability of the N cable lines according to the N sets of monitoring data sets, and outputs N failure occurrence probabilities, and further comprises: step S510, acquiring the N sets of monitoring data sets, each set of monitoring data sets comprising a job monitoring data set of the first terminal node and a job monitoring data set of the second terminal node; step S520, collecting a monitoring data sample set, and performing failure influence analysis on the cable line according to the monitoring data sample set to acquire an associated influence index; step S530, performing abnormal pattern recognition according to the N sets of monitoring data sets to acquire a terminal abnormal probability; and step S540, performing conditional probability calculation on the terminal abnormal probability according to the associated influence index, and outputting N failure occurrence probabilities.

[0035] Preferably, N sets of monitoring data sets corresponding to the terminal nodes of the N cable lines are acquired, each set of monitoring data sets comprising a job monitoring data set of the first terminal node and a job monitoring data set of the second terminal node, and each set of job monitoring data sets contains monitoring information such as current, voltage, temperature, load and vibration data of the terminal nodes; monitoring data sets are formed by acquiring monitoring data from the monitoring data sets, and failure influence analysis is performed on the cable line according to the collected monitoring data sample set. Specifically, first, the monitoring data sample set is preprocessed (including data cleaning, removing noise, outliers, missing values and other data, and standardization), then features possibly related to cable line failure are extracted from the monitoring data sample set, such as the change trend, fluctuation range, etc. of parameters such as current, voltage, temperature, load, vibration, etc., and the types of possible cable line failures are determined, such as short circuit, open circuit, insulation breakdown, poor contact, etc., based on historical cable line failure records, the cable line failures in the monitoring data sample set are labeled, and then the correlation between each monitoring parameter and the cable line failure is analyzed, i.e. by calculating the correlation coefficient, it is determined which parameters have a significant influence on failure occurrence, and according to the results of the correlation analysis, the monitoring parameters that have a significant influence on cable line failure are selected as the associated influence index, for example, temperature exceeding the standard, current fluctuation, insulation resistance drop, etc. These indexes can comprehensively reflect the operation state and failure risk of the cable line.

[0036] Preferably, the collected N groups of monitoring data are cleaned and standardized, key features are extracted from each group of data, which can include voltage fluctuations, current anomalies, temperature changes, insulation impedance, etc., and then an identification model is constructed based on a machine learning model (such as a support vector machine, neural network, etc.), the preprocessed data is used to train the identification model, an anomaly detection model is obtained, and the anomaly detection model is used to identify anomalies in the N groups of monitoring data sets. For data points identified as abnormal, an anomaly degree is given to quantify the deviation of the data point from the normal mode. This anomaly degree is converted into an anomaly probability, generally, the larger the anomaly degree, the higher the anomaly probability. Finally, for each terminal node in the N groups of monitoring data sets, a corresponding anomaly probability value, i.e. terminal anomaly probability, is obtained, reflecting the possibility of the terminal node being abnormal; the correlation influence indicators and the terminal anomaly probability are integrated using a Bayesian network, i.e. the dependency relationship between the terminal nodes and the correlation influence indicators is determined according to historical data, for example, current anomalies directly affect the probability of fault occurrence, and the established Bayesian network is used to perform conditional probability calculation on each terminal node, i.e. the anomaly probability of each terminal node is calculated based on the actual correlation influence indicators, the probability of fault occurrence of each terminal node under the current condition is output, and the probabilities of fault occurrence of the terminal nodes under multiple correlation influence indicators are weighted and summed to output the probabilities of fault occurrence of the N cable lines.

[0037] In a possible implementation, step S500 further includes step S550 of identifying cable material types, cable environmental conditions, and cable aging degrees. The material types of the cable include conductor materials (such as copper, aluminum, etc.), insulation materials (such as polyethylene, cross-linked polyethylene, etc.), and sheath materials (such as polyvinyl chloride, polyethylene, etc.), the environmental conditions of the cable include temperature, humidity, chemical corrosion, electromagnetic interference, mechanical stress, etc., and the aging degree of the cable refers to the degree of performance degradation and damage of the cable due to factors such as time and environment during use. The cable with high aging degree is more prone to failure. The aging degree of the cable is evaluated by observing the appearance of the cable (such as color change, cracks, deformation, etc.), measuring electrical performance (such as insulation resistance, leakage current, etc.), or performing non-destructive testing (such as infrared thermal imaging, ultrasonic testing, etc.).

[0038] Step S560, according to the cable material type, cable environmental conditions and cable aging degree, a probability adjustment weight is generated. According to the historical failure data of different cable material types, a relative weight is assigned to each material type, reflecting the contribution of the material type to the probability of cable failure. According to the specific influence of environmental conditions on the probability of cable failure (such as high temperature accelerating insulation aging, high humidity leading to corrosion, etc.), a corresponding weight is assigned to each environmental condition. According to the different stages of aging degree (such as initial, medium and final), the cable is assigned a corresponding weight. The weights of cable material type, environmental condition and aging degree are comprehensively evaluated to obtain a probability adjustment weight reflecting the overall failure risk of the cable.

[0039] Step S570, the probability adjustment weight is used to dynamically adjust the calculation weight of the failure probability, and the N failure probabilities are updated. The probability adjustment weight generated according to the cable material type, environmental condition and aging degree is used as an important parameter in the calculation of the probability of cable failure. With the change of the running state of the cable line (such as the aggravation of material aging, the deterioration of environmental conditions, etc.), the probability adjustment weight is dynamically adjusted. The probability adjustment weight is introduced into the calculation process of the failure probability, that is, the probability adjustment weight is combined with the original calculation factors (such as monitoring data, historical failure data, etc.), and the new failure probability is obtained by weighted summation. For N cable lines, their respective probability adjustment weights and real-time monitoring data are used to calculate and update their failure probabilities, and finally the updated N failure probabilities are obtained.

[0040] Step S600, input the N failure probabilities into the balanced segmentation model, and output M cable lines, wherein the variance of the sum of the failure probabilities of the M cable lines is less than the expected variance. The balanced segmentation model is an algorithm for optimizing the segmentation of cable lines. The goal is to divide the cable lines into several segments (M cable lines) while meeting the constraint condition (the variance of the sum of the failure probabilities is less than the expected variance). Specifically, the failure probabilities of the N cable lines are integrated into the balanced segmentation model as input data. The balanced segmentation model considers multiple factors such as failure probability, cable length, operating environment, etc. to determine the optimal segmentation scheme of the cable line. During the optimization calculation, the model pays special attention to the variance of the sum of the failure probabilities, which reflects the balance of the failure risk of each segment of the cable line. In the segmentation scheme determined by the balanced segmentation model, the variance of the sum of the failure probabilities of the M cable lines is less than the expected variance. The expected variance represents the balance of the failure probability of each segment of the cable line in the expectation. A smaller expected variance means that the failure risks of each segment of the cable line are closer. Finally, M cable lines are output, and the failure risk of each cable line is well balanced.

[0041] In a possible implementation, the step S600 further includes a step S610 of initializing the equalization parameters, including initializing the total cable length, initializing the number of cable segments, and initializing the failure occurrence probability of each cable segment.

[0042] In step S620, the failure occurrence probability variance of each cable segment is calculated, and the expression is as follows:

[0043] ;

[0044] wherein, denotes the failure occurrence probability variance from the kth segment to the ith segment, i denotes the end position of the current cable segment, and k denotes the start position of the current cable segment, denotes the length of the cable segment, and outputs the average weight of all failure occurrence probabilities in the cable segment, denotes the failure occurrence probability of the mth segment, denotes the average value of the failure occurrence probability from the kth segment to the ith segment, denotes the sum of all failure occurrence probabilities from the kth segment to the ith segment.

[0045] In step S630, the M cable segments are output by minimizing the difference between the failure occurrence probability variance of each cable segment and the expected variance. The optimization objective is to minimize the difference between the failure occurrence probability variance of each cable segment and the expected variance. The smaller the difference, the closer the actual divided M cable segments are to the expected equalization state in terms of failure risk.

[0046] In a possible implementation, the step S630 further includes obtaining the segmentation point k by iterative calculation, so that the sum of the failure occurrence probability variances of each cable segment is minimized, and the iterative calculation formula is as follows:

[0047] ;

[0048] wherein, denotes the minimum variance sum when the first i nodes are divided into j segments, denotes the minimum value selected to ensure the minimum variance sum, denotes all possible segmentation points k before the node i, denotes the minimum variance sum when the first k nodes are divided into segments, denotes the failure occurrence probability variance from the node k+1 to the node i segment.

[0049] In one possible implementation, step S600 further includes step S640 of obtaining a historical dataset of failure occurrence probability of the M-section cable line. The historical dataset of failure occurrence probability of the M-section cable line is collected, and specifically, the dataset can include failure records, maintenance records, monitoring data (such as temperature, humidity, current, voltage, etc.) of each section of cable in the past period of time, and other factors (such as cable material, aging degree, environmental factors, etc.) that can affect the failure occurrence probability.

[0050] Step S650, inputting the historical dataset into a failure prediction model for training, and outputting a probability prediction model. The prediction model is constructed based on machine learning (such as decision tree, random forest, neural network, etc.) or deep learning (such as convolutional neural network, recurrent neural network, etc.), the historical dataset is input into the selected failure prediction model for training, and how to predict the future failure occurrence probability of the cable line according to the current state and historical information of the cable line is learned, that is, a mapping relationship between the input cable line monitoring data features and the future cable line failure occurrence probability is obtained, and finally an accurate probability prediction model is obtained.

[0051] Step S660, predicting the future failure occurrence probability of the M-section cable line based on the probability prediction model, and optimizing the M-section cable line according to the prediction result. The trained probability prediction model is used to predict the future failure occurrence probability of the M-section cable line, the prediction result provides the probability of each section of cable to fail in the future period of time, then the prediction result is analyzed in depth, the failure probability distribution of each section of cable is viewed, it is identified which cable has a higher failure risk, the trend of failure probability over time is analyzed, it is judged whether there is a seasonal, periodic or other identifiable pattern, and the M-section cable line is optimized based on the analysis of the prediction result.

[0052] Step S700, segmenting the to-be-detected cable line according to the M-section cable line. The to-be-detected cable line is subjected to segmented fault detection according to the divided M-section cable line, that is, each section is subjected to fault detection and analysis, which is suitable for cable lines in long distance or complex environment, and improves the accuracy and efficiency of fault positioning. For example, time domain reflectometry (TDR), electrical spectrum analysis or infrared thermal imaging are used for fault detection, the to-be-detected cable line is detected, key data of each section of cable line is collected, and the cable line fault is identified and located.

[0053] In the foregoing, with reference to Figure 1 A cable line fault detection method according to an embodiment of the present application is described in detail. Next, with reference to Figure 2 A cable line fault detection device according to an embodiment of the present application is described.

[0054] The cable line fault detection device according to the embodiment of the present application is used to solve the technical problem that the existing cable line detection cannot accurately and quickly locate the fault point of the cable line, and cannot comprehensively monitor and evaluate the fault of the cable line, and achieves the technical effect of improving the efficiency and accuracy of cable line fault positioning. The cable line fault detection device comprises: a to-be-detected cable line acquisition module 10, a terminal node identification module 20, a terminal node recording module 30, a monitoring data set output module 40, a fault occurrence probability identification module 50, a cable line balanced segmentation module 60, and a segmented fault detection module 70.

[0055] The to-be-detected cable line acquisition module 10 is configured to acquire a to-be-detected cable line. The terminal node identification module 20 is configured to identify terminal nodes on the to-be-detected cable line, divide the to-be-detected cable line according to the terminal nodes, and output N cable lines. The terminal node recording module 30 is configured to record N groups of terminal nodes based on the N cable lines, wherein each group of terminal nodes comprises a first terminal node and a second terminal node, and the first terminal node and the second terminal node are nodes at two ends of the corresponding cable line. The monitoring data set output module 40 is configured to perform terminal operation monitoring according to the N groups of terminal nodes, and output N groups of monitoring data sets. The fault occurrence probability identification module 50 is configured to identify fault occurrence probabilities of the N cable lines according to the N groups of monitoring data sets, and output N fault occurrence probabilities. The cable line balanced segmentation module 60 is configured to input the N fault occurrence probabilities into a balanced segmentation model, and output M cable lines, wherein the M cable lines correspond to a variance of the sum of fault occurrence probabilities that is less than an expected variance. The segmented fault detection module 70 is configured to perform segmented fault detection on the to-be-detected cable line according to the M cable lines.

[0056] In the following, the specific configuration of the cable line balanced segmentation module 60 will be described in detail. The cable line balanced segmentation module 60 can further comprise: initialization of balanced parameters, including initialization of total cable length, initialization of cable segment number, and initialization of fault occurrence probability of each cable line segment;

[0057] The variance of the fault occurrence probability of each cable line segment is calculated, and the expression is as follows:

[0058] ;

[0059] wherein, represents the variance of the fault occurrence probability from the kth segment to the ith segment, i represents the end position of the current cable segment, and k represents the starting position of the current cable segment, represents the length of the cable segment, and outputs the average weight of all fault occurrence probabilities in the cable segment, represents the failure probability of the mth segment, represents the average failure probability from the kth segment to the ith segment, represents the sum of all failure probabilities from the kth segment to the ith segment; by minimizing the difference between the variance of the failure probability of each cable line and the expected variance, the M segments of the cable line are output.

[0060] Next, the specific configuration of the cable line equalization segmentation module 60 will be described in detail. The cable line equalization segmentation module 60 can further include: obtaining the division point k by iterative calculation, so that the sum of the variances of the failure probabilities of each cable line is minimized, and the iterative calculation formula is as follows:

[0061] ;

[0062] wherein, represents the minimum sum of variances when the first i nodes are divided into j segments, represents selecting the minimum value to ensure the minimum sum of variances, represents all possible division points k before node i, represents the minimum sum of variances when the first k nodes are divided into segments, represents the variance of the failure probability from node k+1 to node i segment.

[0063] Next, the specific configuration of the terminal node identification module 20 will be described in detail. The terminal node identification module 20 can further include: identifying the terminal nodes on the cable line to be detected, identifying P terminal nodes; analyzing each terminal node in the P terminal nodes, determining whether the number of devices connected to each terminal node is greater than or equal to 2, obtaining first type terminal nodes with a device quantity greater than or equal to 2, and second type terminal nodes with a device quantity less than 2; performing work fusion monitoring on the first type terminal nodes, and outputting fusion monitoring data sets.

[0064] Next, the specific configuration of the fault occurrence probability identification module 50 will be described in detail. The fault occurrence probability identification module 50 can further include: obtaining the N sets of monitoring data sets, each set of monitoring data sets including a job monitoring data set of the first terminal node and a job monitoring data set of the second terminal node; collecting a monitoring data sample set, performing fault impact analysis on the cable line according to the monitoring data sample set, and obtaining a correlation impact index; performing abnormal mode identification according to the N sets of monitoring data sets to obtain a terminal abnormal probability; and performing conditional probability calculation on the terminal abnormal probability with the correlation impact index to output N fault occurrence probabilities.

[0065] Next, the specific configuration of the fault occurrence probability identification module 50 will be described in detail. The fault occurrence probability identification module 50 can further include: identifying cable material types, cable environmental conditions, and cable aging degrees; generating a probability adjustment weight according to the cable material types, the cable environmental conditions, and the cable aging degrees; and dynamically adjusting a calculation weight of the fault occurrence probability with the probability adjustment weight to update the N fault occurrence probabilities.

[0066] Next, the specific configuration of the cable line equalization segmentation module 60 will be described in detail. The cable line equalization segmentation module 60 further includes: obtaining a historical data set of the fault occurrence probabilities of the M segments of the cable line; inputting the historical data set into a fault prediction model for training to output a probability prediction model; and predicting future fault occurrence probabilities of the M segments of the cable line based on the probability prediction model and optimizing the M segments of the cable line according to the prediction results.

[0067] The cable line fault detection device provided in the embodiments of the present application can perform the cable line fault detection method provided in any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0068] Although the present application makes various references to certain modules in the device according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or the server, and the various units and modules are only divided according to the functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of the functional units are only for easy mutual differentiation, and do not limit the protection scope of the present application.

[0069] The above specific embodiments do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method of fault detection of a cable line, characterized in that, The method comprises: acquiring a cable line to be detected; identifying terminal nodes on the cable line to be detected, dividing the cable line to be detected according to the terminal nodes, and outputting N cable lines; based on the N cable lines, recording N groups of terminal nodes, wherein each group of terminal nodes comprises a first terminal node and a second terminal node, and the first terminal node and the second terminal node are nodes at both ends of the corresponding cable line; carrying out terminal operation monitoring according to the N groups of terminal nodes, and outputting N groups of monitoring data sets; identifying the probability of failure occurrence of the N cable lines according to the N groups of monitoring data sets, and outputting N probabilities of failure occurrence; inputting the N probabilities of failure occurrence into an equal segmentation model, and outputting M cable lines, wherein the M cable lines correspond to variances of the probabilities of failure occurrence that are less than an expected variance; carrying out segmented fault detection on the cable line to be detected according to the M cable lines; wherein the method of inputting the N probabilities of failure occurrence into the equal segmentation model and outputting the M cable lines comprises: initializing equal parameters, including initializing a total cable length, initializing a number of cable segments, and initializing a probability of failure occurrence of each cable segment; calculating a variance of the probability of failure occurrence of each cable segment, and the expression is as follows: ； wherein, denotes the variance of the failure occurrence probability from the kth segment to the ith segment, i denotes the end position of the current cable segment, k denotes the start position of the current cable segment, i-k+1 denotes the total number of cable segments from the kth segment to the ith segment, denotes the failure occurrence probability of the mth segment, denotes the mean value of the failure occurrence probability from the kth segment to the ith segment, denotes the sum of all failure occurrence probabilities from the kth segment to the ith segment; outputting the M cable lines by minimizing the difference between the variance of the probability of failure occurrence of each cable segment and the expected variance; wherein a split point k is obtained by iterative calculation, so that the sum of the variances of the probabilities of failure occurrence of each cable segment is minimized, and the iterative calculation formula is as follows: ； where, denotes the minimum sum of variances when the first i nodes are divided into j segments, denotes the selection of the minimum value to ensure the minimum sum of variances, denotes all possible split points before node i denotes the minimum sum of variances when the first k nodes are divided into segments, denotes the variance of the failure occurrence probability of the segment from node k+1 to node i.

2. The method of claim 1, wherein, the method of identifying terminal nodes on the cable line to be detected further comprises: identifying P terminal nodes on the cable line to be detected; analyzing each of the P terminal nodes to determine whether the number of devices connected on each terminal node is greater than or equal to 2, obtaining a first type of terminal node with a number of devices greater than or equal to 2, and a second type of terminal node with a number of devices less than 2; analyzing each of the P terminal nodes to determine whether the number of devices connected on each terminal node is greater than or equal to 2, obtaining a first type of terminal node with a number of devices greater than or equal to 2, and a second type of terminal node with a number of devices less than 2; analyzing each of the P terminal nodes to determine whether the number of Performing fusion monitoring on the first type of terminal node, outputting a fusion monitoring data set.

3. The method of claim 1, wherein, the method of identifying the probability of failure occurrence of the N cable lines according to the N groups of monitoring data sets and outputting N probabilities of failure occurrence comprises: acquiring the N groups of monitoring data sets, wherein each group of monitoring data sets comprises a work monitoring data set of a first terminal node and a work monitoring data set of a second terminal node; collecting a monitoring data sample set, carrying out fault influence analysis on the cable line according to the monitoring data sample set, and obtaining an associated influence index; carrying out abnormal pattern recognition according to the N groups of monitoring data sets, and obtaining a terminal abnormal probability; carrying out conditional probability calculation on the terminal abnormal probability by using the associated influence index, and outputting N probabilities of failure occurrence.

4. The method of claim 3, wherein, after outputting the N probabilities of failure occurrence, the method further comprises: identifying a cable material type, a cable environmental condition, and a cable aging degree; generating a probability adjustment weight according to the cable material type, the cable environmental condition, and the cable aging degree; dynamically adjusting a calculation weight of the probability of failure occurrence by using the probability adjustment weight, and updating the N probabilities of failure occurrence.

5. The method of claim 1, wherein, after outputting the M cable lines, the method further comprises: acquiring a historical data set of the probability of failure occurrence of the M cable lines; inputting the historical data set into a fault prediction model for training, and outputting a probability prediction model; Based on the probability prediction model, future fault occurrence probabilities of the M cable lines are predicted, and the M cable lines are optimized according to the prediction results.

6. A fault detection device for a cable line, characterized in that The device is used to implement the cable line fault detection method of any one of claims 1-5, and the device comprises: a to-be-detected cable line acquisition module configured to acquire a to-be-detected cable line; a terminal node identification module configured to identify terminal nodes on the to-be-detected cable line, divide the to-be-detected cable line according to the terminal nodes, and output N cable lines; a terminal node recording module configured to record N groups of terminal nodes based on the N cable lines, wherein each group of terminal nodes comprises a first terminal node and a second terminal node, and the first terminal node and the second terminal node are nodes at two ends of a corresponding cable line; a monitoring data set output module configured to perform terminal operation monitoring according to the N groups of terminal nodes and output N groups of monitoring data sets; a fault occurrence probability identification module configured to identify fault occurrence probabilities of the N cable lines according to the N groups of monitoring data sets and output N fault occurrence probabilities; a cable line equalization segmentation module configured to input the N fault occurrence probabilities into an equalization segmentation model and output M cable lines, wherein a variance of the M cable lines corresponding to the fault occurrence probabilities is less than an expected variance; a segmented fault detection module configured to perform segmented fault detection on the to-be-detected cable line according to the M cable lines.

Images