Intelligent online detection method and system for fatigue of electric wire and cable

Abstract

The invention belongs to the technical field of cable fatigue detection, and particularly discloses a wire and cable fatigue intelligent online detection method and system, and the method comprises the steps: laying an initial detection point and synchronously collecting a multi-dimensional state signal based on a cable line topological structure diagram and prior state information; calculating the fatigue degree of each point based on the signal, and generating a fatigue degree distribution heat map set continuously covering the full length of the line through spatial interpolation; based on the heat map set, identifying high-risk and low-risk sections through threshold comparison and section combination; the method comprises the following steps of: obtaining a detection point density of each section, calculating the detection point density of each section, respectively generating a detection point supplement scheme aiming at a high-risk section and a detection point simplification scheme aiming at a low-risk section, merging, executing reconfiguration to form an optimized network, continuously monitoring on the basis of the optimized network, and triggering optimization adjustment when a preset optimization triggering condition is met; according to the invention, global visualization of the fatigue state of the cable and dynamic and accurate optimization of a monitoring network are realized.

Description

Technical Field

[0001] This invention belongs to the field of cable fatigue testing technology, and relates to an intelligent online testing method and system for the fatigue of wires and cables. Background Technology

[0002] The long-term reliability of power cables is directly related to the safety of energy transmission and the stability of information transmission. In actual operation, cables are subjected to multiple stress coupling effects, including electrical, thermal, and mechanical stresses, which can lead to cumulative damage such as insulation aging, conductor fatigue, and deterioration of accessory connections, ultimately causing performance degradation. Therefore, real-time and accurate online monitoring and assessment of cable fatigue is crucial for achieving predictive maintenance and improving operational safety.

[0003] Current cable condition monitoring technology mainly utilizes online sensors for multi-parameter acquisition and data analysis. For example, Chinese invention patent CN116818510A discloses a fatigue testing system for electric wires and cables. This system collects multiple test data points for electric wires and cables, processes and analyzes the data, and combines it with environmental data to assess the safety and remaining service life of the cables, thereby reducing waste during use and improving their safety.

[0004] However, this type of wire and cable testing technology has the following obvious limitations: First, the method only evaluates the fatigue of wires and cables based on discrete data collected from a limited number of testing points. The results can only reflect the state of isolated points and cannot construct a spatial distribution map of fatigue that reflects the continuous changes of the entire cable. As a result, it is difficult for maintenance personnel to quickly locate specific risk sections and identify risk gradients, and at the same time, they cannot accurately grasp the overall health trend of the cable.

[0005] Second, the detection point layout of this method is pre-set and fixed. However, the fatigue deterioration of cables is a dynamic process. High-risk sections may shift or spread with time and operating conditions. The static network layout cannot respond to such changes, resulting in insufficient detection of high-risk areas and excessive detection resources for low-risk areas.

[0006] Therefore, there is an urgent need for an online detection method and system that can integrate multi-dimensional state information, visualize the fatigue state of the entire cable, and intelligently and dynamically optimize the layout of monitoring resources based on the evaluation results, in order to solve the above problems. Summary of the Invention

[0007] In view of this, in order to solve the problems mentioned in the background technology, a smart online detection method and system for the fatigue of wires and cables is proposed.

[0008] The objective of this invention can be achieved through the following technical solution: This invention provides an intelligent online detection method for the fatigue of electric wires and cables, comprising: setting up initial detection points based on the topology diagram and prior state information of the cable line, and simultaneously collecting multi-dimensional state signals of each initial detection point.

[0009] Based on the multidimensional state signal, a set of fatigue distribution maps of the cable line is calculated and generated.

[0010] Based on the fatigue distribution map set, high-risk and low-risk sections are identified, and an optimization and adjustment scheme is generated in combination with the layout of the initial detection points.

[0011] The detection points are reconfigured according to the optimization and adjustment scheme to form an optimized detection point network. Based on the optimized detection point network, data collection and fatigue analysis are continuously performed, and when the preset optimization trigger conditions are met, the optimization and adjustment of the detection point network is triggered.

[0012] The present invention also provides an intelligent online detection system for the fatigue of electric wires and cables, comprising: a signal acquisition module, which sets up initial detection points based on the topology diagram and prior state information of the cable line, and simultaneously acquires multi-dimensional state signals of each initial detection point.

[0013] The distribution map generation module calculates and generates a set of fatigue distribution maps of the cable line based on the multidimensional state signal.

[0014] The scheme optimization module identifies high-risk and low-risk sections based on the fatigue distribution map set, and generates an optimization and adjustment scheme in conjunction with the layout of the initial detection points.

[0015] The optimization and adjustment module reconfigures the detection points according to the optimization and adjustment scheme to form an optimized detection point network. Based on the optimized detection point network, it continuously collects data and performs fatigue analysis. When the preset optimization triggering conditions are met, it triggers the optimization and adjustment of the detection point network.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention generates a set of fatigue maps continuously distributed along the cable line by using multi-dimensional state signals based on each detection point, thereby enabling maintenance personnel to intuitively grasp the spatial gradient and evolution process of the fatigue state of the cable line, overcoming the limitation of only providing discrete point data, realizing the leap from point perception to line-surface collaborative perception, thereby improving the globality of state assessment and the efficiency of risk location.

[0017] (2) This invention uses a distribution map set to identify high-risk and low-risk sections in a time and space, effectively distinguishing between instantaneous fluctuations and continuous deterioration, and thus accurately delineating cable sections that need to be focused on. This solves the problem of false alarms or missed alarms caused by the single threshold judgment in traditional methods, and provides a reliable basis for the differentiated deployment of detection points.

[0018] (3) The present invention dynamically generates and executes a network optimization scheme for detection points that includes additions and simplifications based on the identified risk sections, so that the detection resources can actively adapt to the changes in the risk distribution of cables. While ensuring the coverage density of high-risk areas, it optimizes the resource allocation of low-risk areas, thereby solving the problem of monitoring blind spots and resource waste caused by static network layout. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram showing the connections between the steps of the method of the present invention.

[0021] Figure 2 This is a schematic diagram showing the connection steps for generating the optimized adjustment scheme of the present invention.

[0022] Figure 3 This is a schematic diagram showing the connections of the various modules in the system of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] This invention achieves non-uniform, precise online monitoring and optimization of cable fatigue state through intelligent dynamic sensing and closed-loop optimization mechanisms. Specifically, the method first intelligently deploys initial detection points based on topology and historical data to establish a monitoring network. Then, it generates a spatiotemporally continuous fatigue distribution map through multi-dimensional signal fusion calculation, achieving risk visualization. Next, based on the risk distribution identification results, it dynamically generates and executes supplementary and simplification schemes for the detection point network, intelligently focusing monitoring resources on high-risk areas. Finally, through continuous online data analysis and optimization triggering, a monitoring-evaluation-optimization closed loop is formed, enabling the detection network to adapt to changes in cable condition. This method solves the pain points of traditional fixed-point monitoring methods, such as low efficiency, delayed risk detection, and unreasonable resource allocation, achieving a transformation from static, uniform monitoring to dynamic, precise monitoring.

[0025] Please see Figure 1 As shown, the present invention provides an intelligent online detection method for the fatigue of electric wires and cables, comprising the following steps S1 to S4.

[0026] S1. Deploy initial detection points and collect multi-dimensional status signals to construct a preliminary monitoring network covering key locations and potential risk areas of the cable line, providing a data foundation for subsequent fatigue analysis and detection point optimization. Specifically, this includes the following steps: S1-1. Based on the aforementioned topology diagram, identify inherent structural key points in the cable line. These key structural points include, but are not limited to, cable joints, terminations, grounding boxes, and corner points in the line with bending radii less than a preset safety threshold. Mark all identified structural key points as the first type of initial detection points. The preset safety threshold refers to the minimum permissible bending radius specified according to the cable model and industry design specifications. For example, for a certain model of 110kV cross-linked polyethylene insulated cable, its minimum permissible bending radius is 20 times its outer diameter.

[0027] S1-2. Identifying Sensitive Sections Based on Prior State Information: Based on different types of prior state information, corresponding strategies are adopted to identify potential risk sections on cable lines that require special attention, i.e., sensitive sections.

[0028] (1) When the prior state information is historical operation data: extract historical fault records from the historical operation data, and identify the sections on the cable line where historical faults are spatially clustered and the number of fault occurrences exceeds the preset threshold as each sensitive section.

[0029] (2) When the prior state information is the design drawings and construction records: First, based on the natural markings in the design drawings or the significant changes in the laying environment, the cable line is divided into multiple logical segments.

[0030] For each segment, key parameters such as laying environment type and burial depth are extracted from its design drawings and construction records. The laying environment type includes, but is not limited to, direct burial, bridge overhead, pipe gallery, and waterlogged area.

[0031] Predefine one or more sets of sensitive parameter combinations to characterize high-risk conditions. For example, the following rules can be defined: Combination 1: The laying environment type is an overhead bridge and the burial depth is less than 0.8 meters. Combination 2: The laying environment type is a waterlogged area or a flood-prone area. Combination 3: The laying environment type is direct burial and the burial depth is less than 0.7 meters.

[0032] Iterate through all segments and identify the segments that simultaneously satisfy all conditions in any predefined combination of sensitive parameters as sensitive segments.

[0033] (3) When the prior state information is typical operating reference data of similar lines: obtain the operating reference data of similar lines that are the same or highly similar to this line in terms of cable type, voltage level, main laying method and other attributes.

[0034] Extract the total length of similar lines and this line, and record them as follows: and Based on the shared sequence of structural key points, the topology of the two lines is aligned, and the segment length ratio between adjacent key points is calculated.

[0035] If a fault-prone section of a similar line is located a short distance from the starting point Rice Between meters, the corresponding sensitive sections are mapped onto the route using overall proportional conversion, with their starting points... and the end point satisfy: , .

[0036] If there are multiple reference data for the same type of line, then after mapping them separately, the union of all mapped sensitive segments is taken as the final set of sensitive segments.

[0037] S1-3. Integrate detection points and complete deployment: For each sensitive section, determine whether a first type of initial detection point already exists within it. If it does, no more detection points will be added in the sensitive section. If it does not exist, detection points will be deployed in the sensitive section and marked as second type of initial detection points.

[0038] All the first type of initial detection points and the second type of initial detection points are combined to form an initial detection point set, thus completing the deployment of the initial detection points.

[0039] It should be noted that the multidimensional state signal mentioned in this invention refers to a set of physical quantities characterizing the electrical, thermal, and mechanical stress states and insulation performance of the cable, which are synchronously collected during cable operation. Specifically, the multidimensional state signal includes, but is not limited to, thermal stress signal dimensions, mechanical stress signal dimensions, electrical stress dimensions, and insulation state signal dimensions.

[0040] S2. Based on the multidimensional state signal, calculate and generate a set of fatigue distribution maps of the cable line.

[0041] For example, the calculation and generation of the fatigue distribution map set of the cable line includes: S2-1, obtaining the feature values ​​of each signal dimension at each time point in each initial detection point from the multi-dimensional state signal, forming a feature value time series of each signal dimension.

[0042] S2-2. Calculate the fatigue degree at each detection point, aiming to convert the time series of multidimensional state signals into a single quantitative index characterizing the cumulative damage degree of cable materials. Based on the eigenvalue time series, calculate the fatigue degree of each initial detection point at each time point, specifically including the following steps: S2-2-1. Extract data subsequence: Obtain the data subsequence of each initial detection point from the set start time point to the current time point from the eigenvalue time series of each signal dimension.

[0043] S2-2-2 Calculation of Preliminary Fatigue Index: Perform trend analysis on the data subsequences of each signal dimension and calculate their rate of change as the preliminary fatigue index for each signal dimension. Specifically, construct the characteristic value change curve for each signal dimension with time point as the x-axis and characteristic value as the y-axis, and then extract the slope of the change from the curve. This slope is used as the preliminary fatigue index for the corresponding signal dimension.

[0044] S2-2-3, Synthetic Comprehensive Fatigue Level: Select the maximum value from the preliminary fatigue level indices of all signal dimensions at each time point of the initial detection point, and use it as the fatigue level of each initial detection point at each time point.

[0045] S2-3. Map the comprehensive fatigue index of all initial detection points at each time point to the topology diagram of the cable line to obtain the discrete fatigue spatial distribution at each time point.

[0046] S2-4. Based on the discrete fatigue spatial distribution at each time point, fatigue is estimated at locations on the cable line where no detection points are set up through spatial interpolation, generating a fatigue distribution surface covering the entire length of the cable line, and thus obtaining the fatigue distribution map at each time point.

[0047] Specifically, the process of estimating the fatigue of locations on the cable line without detection points by spatial interpolation is as follows: using the topology diagram of the cable line as the spatial coordinate reference, the path length coordinates of each initial detection point on its respective cable segment are taken as the position coordinates, and the fatigue value of each detection point corresponding to the current time point is taken as the known sampling data point.

[0048] For any location on the cable line where no monitoring point is set, the fatigue estimate is obtained by calling a preset interpolation function based on the geometric distance between that location and each of the known sampling data points along the cable laying path. As a specific implementation method, an inverse distance weighted interpolation algorithm can be used. In this algorithm, the fatigue estimate for the location is a weighted average of the fatigue values ​​of all known data points, where the weight coefficient of each known point is a function of the reciprocal of its distance to the location to be estimated. By traversing all locations on the cable line and performing the above calculation, the discrete fatigue sampling data is reconstructed into a fatigue distribution curve continuously distributed along the cable line, thus forming a fatigue distribution surface covering the entire length of the line.

[0049] S2-5. Compile fatigue distribution maps at each time point in chronological order to form a set of fatigue distribution maps for cable lines.

[0050] S3. Based on the fatigue distribution map set, identify high-risk and low-risk sections, and generate an optimization and adjustment scheme in combination with the layout of the initial detection points.

[0051] Optionally, to improve the robustness of the system after initial setup and reconstruction, an initial stable observation period can be set before the first optimization and adjustment of the detection point network, or after the detection point network has undergone large-scale reconstruction. During this observation period, the system mainly accumulates monitoring data and establishes a state baseline. Even if the preset optimization trigger conditions are met during this period, physical addition or deletion of detection points will not be performed; only risk warnings will be issued. The formal optimization process will be started after the data quality stabilizes.

[0052] For example, identifying high-risk and low-risk segments includes: sorting the fatigue values ​​at each time point in the set from largest to smallest, and then selecting the fatigue value corresponding to the preset high percentile as the fatigue threshold.

[0053] It should be added that the preset high statistical percentile is a threshold parameter pre-calibrated based on the statistical characteristics of historical line operation data, used to statistically define abnormally high fatigue levels. Its calibration method includes: for each historical fault event, extracting the fatigue values ​​at all time points within a preset time period prior to its occurrence; calculating the percentile ranking of each value in the fatigue data set of all detection points across the entire network at the corresponding time point; compiling the percentile ranking data corresponding to multiple historical fault events, calculating the statistical quantile of the dataset, and determining this quantile value as the preset high statistical percentile. For example, the 75th quantile of the percentile ranking dataset can be taken as the calibration value of this parameter.

[0054] The cable line is divided into micro-segments, and the fatigue level of each micro-segment at each detection time point within the preset time period is compared with the fatigue level threshold. If all fatigue levels in the fatigue time series of a certain micro-segment are greater than the fatigue level threshold, the micro-segment is marked as a high-risk micro-segment; otherwise, it is marked as a low-risk micro-segment.

[0055] All adjacent high-risk micro-segments in the space are merged to form high-risk segments, and all adjacent low-risk micro-segments in the space are merged to form low-risk segments.

[0056] Please see Figure 2 As shown, the proposed optimization and adjustment scheme aims to achieve dynamic and precise reconstruction of the detection point network, focusing monitoring resources on high-risk areas and avoiding redundancy in low-risk areas. The scheme includes the following steps S3-1 to S3-6.

[0057] S3-1. Calculate the current detection point density of each segment to provide a quantitative benchmark for assessing the rationality of the monitoring network resource distribution. Based on the total spatial length of each high-risk and low-risk segment and the number of initial detection points contained within it, calculate the current detection point density for each segment. The specific calculation process is as follows: S3-1-1. Obtain segment geometric information and detection point count: The high-risk or low-risk segments are collectively referred to as target segments. Based on the cable line topology diagram, calculate the continuous length of the target segment along the cable laying path, and use this as the total spatial length of the target segment.

[0058] The total number of all initial detection points located entirely within the geographical area of ​​the target segment is counted and used as the initial detection point count within the target segment.

[0059] S3-1-2, Execution density calculation: The ratio of the initial number of detection points in the target section to its total spatial length is taken as the current detection point density of the target section. The density directly reflects the number of detection points distributed per unit cable length on average in the target section.

[0060] S3-2. Determine whether high-risk sections need to be supplemented and calculate the number of supplementary points. For each identified high-risk section, if its current detection point density is lower than the preset required detection density threshold for that type of section, it indicates insufficient monitoring coverage and supplementation is required. At this time, based on the threshold and the total length of the section space, calculate the total number of detection point targets required for that section, and then determine the number of detection points that need to be supplemented.

[0061] It should be noted that the preset required detection density threshold for high-risk sections is the minimum number of detection points that must be deployed per unit length of cable to ensure effective and thorough monitoring of high-risk cable sections without serious blind spots. This threshold constitutes a lower limit for density. If the current detection point density of a high-risk section is lower than this threshold, it is determined that monitoring resources are insufficient and a supplementary requirement is triggered. The specific method for obtaining this threshold is as follows: based on historical data, early fault cases that have been successfully alerted or detected within high-risk characteristic sections are selected; the spatial distance between the fault location and its nearest effective detection point in each case is statistically analyzed; a specific percentile of this distance statistical value is calculated, for example, the 90th percentile; finally, the reciprocal of this percentile distance value is determined as the required detection density threshold.

[0062] S3-3. Generate a supplementary plan for high-risk sections. Based on the layout of each initial detection point in the section and the number of detection points to be added, with the goal of optimizing the uniformity of detection coverage, determine the layout of the supplementary detection points and generate a supplementary plan.

[0063] Furthermore, determining the location of the supplementary detection points includes: S3-3-1, identifying the spacing of cable lines between all adjacent initial detection points in the high-risk section.

[0064] S3-3-2. Insert supplementary detection points between adjacent detection point pairs with the largest spacing in descending order of spacing, until the number of inserted supplementary detection points reaches the number of detection points to be added. When inserting supplementary detection points between adjacent detection points, the placement of the inserted detection points is determined by equal spacing.

[0065] S3-4. Determine whether low-risk sections need to be streamlined and calculate the number of streamlining points. For each identified low-risk section, if its current detection point density is higher than the preset allowable detection density threshold for that type of section, it indicates that the monitoring resources may be redundant and streamlining can be considered. At this time, based on the threshold and the total length of the section space, calculate the expected total number of detection point targets for that section, and then determine the number of detection points that need to be streamlined.

[0066] It should be noted that the allowable detection density threshold for low-risk sections refers to the maximum number of detection points allowed per unit length in low-risk cable sections to balance monitoring costs and effectiveness. This threshold is an upper limit. If the current detection point density in a low-risk section exceeds this threshold, it indicates potential redundancy in monitoring resources, which should be considered for simplification. Specifically, it is obtained by analyzing the spatial correlation between signals from adjacent detection points under low-risk conditions based on extensive historical or similar line monitoring data. When the distance between adjacent detection points increases to the point where their signal correlation coefficient falls below a certain empirical threshold, it indicates strong information independence. In this case, the reciprocal of the distance is used as the allowable detection density threshold for the low-risk section.

[0067] S3-5. Generate a simplification plan for low-risk sections. By analyzing the correlation of each initial detection point in the section on the multidimensional state signal, the information redundancy of each detection point is evaluated. Based on the number of detection points to be simplified and the level of information redundancy, the simplification targets are determined, and a simplification plan is generated. Specifically, this includes the following sub-steps: Further, the evaluation of the information redundancy of each detection point includes: S3-5-1. Calculating the average correlation between each signal dimension and all other initial detection points in the section in the corresponding dimension to obtain the average correlation of each signal dimension.

[0068] S3-5-2. Select the maximum value from the average correlation of all signal dimensions of the initial detection point as the information redundancy of each initial detection point.

[0069] Furthermore, determining the simplification target includes: after obtaining the information redundancy of each detection point, sorting all initial detection points in the segment from high to low according to the information redundancy.

[0070] Based on the calculated number of detection points that need to be reduced, select the corresponding number of detection points with the highest information redundancy from the sorted list as the objects to be reduced.

[0071] If multiple detection points in the sorting results have the same information redundancy value, affecting the order selection, then the final simplification target is determined from these detection points with the same information redundancy according to a preset secondary judgment rule. As one implementation, the secondary judgment rule is: preferentially select detection points that are spatially adjacent within the segment and have higher signal feature repeatability. Higher signal feature repeatability can be quantified by calculating the Euclidean distance or dynamic time warping distance of the time series of multidimensional state signal feature values ​​between two detection points; the smaller the distance value, the higher the signal feature repeatability.

[0072] For example, suppose there are 10 initial detection points in a low-risk area, and calculations show that 3 need to be eliminated. First, the information redundancy of each point is calculated and ranked. Assume that the redundancy values ​​of the top three points after ranking are 0.95, 0.92, and two points with a combined redundancy of 0.90. Then, the three points with the highest redundancy are selected, namely the detection points with values ​​of 0.95 and 0.92, and one detection point selected from the two points with a combined redundancy of 0.90. When a selection needs to be made from the combined detection points, the spatial distribution of these points is further compared, and the detection points with more concentrated spatial locations are removed first.

[0073] S3-6. Generate a comprehensive optimization and adjustment plan by merging the supplementary plans for each high-risk section with the simplified plans for each low-risk section.

[0074] S4. Based on the optimization and adjustment scheme, the detection points are reconfigured to form an optimized detection point network. Based on the optimized detection point network, data collection and fatigue analysis are continuously performed, and when the preset optimization triggering conditions are met, the optimization and adjustment of the detection point network is triggered.

[0075] For example, the preset optimization triggering conditions include at least one of the following: (1) a periodic triggering condition, that is, after the self-detection point network completes the last optimization configuration, its continuous running time reaches a preset optimization cycle threshold. Optionally, the optimization cycle threshold can be determined based on historical optimization experience or state evaluation cycle of similar lines. For example, the average time interval between significant spatial changes in cable fatigue distribution, such as migration of high-risk sections, under similar operating conditions can be statistically analyzed, and this interval can be used as the preset optimization cycle threshold.

[0076] (2) New risk triggering conditions, namely, when analyzing the latest fatigue distribution map set generated since the last optimization configuration, new high-risk sections are identified.

[0077] (3) Instantaneous abnormal triggering conditions, namely, based on the real-time acquisition of multi-dimensional state signals, identifying that the feature value of at least one signal dimension of any detection point has changed beyond its preset threshold.

[0078] Please see Figure 3 As shown, the present invention also provides an intelligent online detection system for the fatigue of electric wires and cables, which includes: a signal acquisition module, a distribution map generation module, a scheme optimization module, and an optimization adjustment module.

[0079] In the above, the distribution map generation module is connected to the signal acquisition module and the scheme optimization module, respectively, and the scheme optimization module is also connected to the optimization and adjustment module.

[0080] The signal acquisition module, based on the cable line topology diagram and prior state information, sets up initial detection points and simultaneously acquires multi-dimensional state signals from each initial detection point.

[0081] The distribution map generation module calculates and generates a set of fatigue distribution maps of the cable line based on the multidimensional state signal.

[0082] The optimization module identifies high-risk and low-risk sections based on the fatigue distribution map set and generates an optimization adjustment plan in conjunction with the layout of the initial detection points.

[0083] The optimization and adjustment module reconfigures the detection points according to the optimization and adjustment scheme to form an optimized detection point network. Based on the optimized detection point network, it continuously collects data and performs fatigue analysis. When the preset optimization triggering conditions are met, it triggers the optimization and adjustment of the detection point network.

[0084] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0085] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0086] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0087] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0088] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent online detection of fatigue in electric wires and cables, characterized in that: The method includes: Based on the cable line topology diagram and prior state information, initial detection points are set up, and multi-dimensional state signals of each initial detection point are collected simultaneously. Based on the multidimensional state signals, a set of fatigue distribution maps of the cable line is calculated and generated; Based on the fatigue distribution map set, high-risk and low-risk sections are identified, and an optimization and adjustment scheme is generated by combining the layout of the initial detection points. The detection points are reconfigured according to the optimization and adjustment scheme to form an optimized detection point network. Based on the optimized detection point network, data collection and fatigue analysis are continuously performed, and when the preset optimization trigger conditions are met, the optimization and adjustment of the detection point network is triggered.

2. The intelligent online detection method for fatigue of wires and cables according to claim 1, characterized in that: The initial detection points include: Identify key structural points in the cable line from the topology diagram and mark each key structural point as a first type of initial detection point; When the prior state information is historical operating data, historical fault records are extracted from the historical operating data, and sections on the cable line where historical faults are spatially clustered and the number of fault occurrences exceeds a preset threshold are identified as sensitive sections. For each sensitive section, determine whether a first type of initial detection point already exists within it. If it does, no more detection points will be added in the sensitive section. If it does not exist, detection points will be set up in the sensitive section and marked as second type of initial detection points. All the first type of initial detection points and the second type of initial detection points are combined to form an initial detection point set, thus completing the deployment of the initial detection points.

3. The intelligent online detection method for fatigue of wires and cables according to claim 1, characterized in that: The set of fatigue distribution maps for the calculated cable lines includes: The feature values ​​of each signal dimension at each time point in each initial detection point are obtained from the multidimensional state signal to form a time series of feature values ​​for each signal dimension; Based on the time series of the feature values, the fatigue level of each initial detection point at each time point is calculated; The comprehensive fatigue index of all initial detection points at each time point is mapped onto the topology diagram of the cable line to obtain the discrete fatigue spatial distribution at each time point. Based on the discrete spatial distribution of fatigue at each time point, fatigue is estimated at locations on the cable line where no detection points are set up through spatial interpolation, generating a fatigue distribution surface covering the entire length of the cable line, and thus obtaining fatigue distribution maps at each time point; The fatigue distribution maps of the cable line are compiled in chronological order at each time point.

4. The intelligent online detection method for fatigue of wires and cables according to claim 3, characterized in that: The calculation of fatigue at each initial detection point at each time point includes: Obtain the data subsequence of each initial detection point from the set start time point to the current time point from the feature value time series of each signal dimension; Trend analysis is performed on the data subsequences of each signal dimension, and their rate of change is calculated as a preliminary fatigue index for each signal dimension. The maximum value is selected from the preliminary fatigue indexes of all signal dimensions at each time point of the initial detection point, and this value is taken as the fatigue level of each initial detection point at each time point.

5. The intelligent online detection method for fatigue of wires and cables according to claim 1, characterized in that: The identification of high-risk and low-risk zones includes: The fatigue values ​​at each time point in the set are sorted from largest to smallest, and then the fatigue value corresponding to the preset high percentile is selected as the fatigue threshold. The cable line is divided into micro-segments, and the fatigue level of each micro-segment at each detection time point within the preset time period is compared with the fatigue level threshold. If all fatigue levels in the fatigue time series of a certain micro-segment are greater than the fatigue level threshold, the micro-segment is marked as a high-risk micro-segment; otherwise, it is marked as a low-risk micro-segment. All adjacent high-risk micro-segments in the space are merged to form high-risk segments, and all adjacent low-risk micro-segments in the space are merged to form low-risk segments.

6. The intelligent online detection method for fatigue of wires and cables according to claim 1, characterized in that: The generated optimization and adjustment scheme includes: The current detection point density is calculated based on the total spatial length of each high-risk and low-risk section and the number of initial detection points contained within it. For each identified high-risk segment, if the current detection point density of the high-risk segment is lower than the preset detection density threshold for that type of segment, the total number of detection point targets required for that segment is calculated based on the threshold and the total length of the segment space, thereby determining the number of additional detection points to be added. Based on the location of each initial detection point in the section and the number of detection points to be added, with the goal of optimizing the uniformity of detection coverage, the location of the additional detection points is determined and an addition plan is generated. For each identified low-risk segment, if the current detection point density of the low-risk segment is higher than the preset allowable detection density threshold for that type of segment, the total number of expected detection point targets for that segment is calculated based on the threshold and the total length of the segment space, thereby determining the number of simplified detection points. By analyzing the correlation of each initial detection point in the multidimensional state signal within the section, the information redundancy of each detection point is evaluated, and based on the number of detection points to be simplified and the level of information redundancy, the simplification targets are determined and a simplification scheme is generated. The supplementary plans for each high-risk section are combined with the simplified plans for each low-risk section to generate a comprehensive optimized adjustment plan.

7. The intelligent online detection method for fatigue of wires and cables according to claim 6, characterized in that: The determination of the location of the supplementary detection points includes: Identify the spacing of cable lines between all adjacent initial detection points within a high-risk section; Supplementary detection points are inserted sequentially between adjacent detection point pairs with the largest spacing, in descending order of spacing, until the number of inserted supplementary detection points reaches the number of detection points to be added. When inserting supplementary detection points between adjacent detection points, the placement of the inserted detection points is determined by an equidistant method.

8. The intelligent online detection method for fatigue of wires and cables according to claim 6, characterized in that: The assessment of information redundancy at each detection point includes: Calculate the average correlation between each signal dimension and all other initial detection points within the segment for each dimension, and obtain the average correlation of each signal dimension; The maximum value among the average correlations of all signal dimensions at the initial detection point is selected as the information redundancy of each initial detection point.

9. The intelligent online detection method for fatigue of electric wires and cables according to claim 1, characterized in that, The preset optimization triggering conditions include at least one of the following: After the self-detection point network completes its last optimization configuration, its continuous running time reaches the preset optimization cycle threshold. When analyzing the latest fatigue distribution map set generated since the last optimization configuration, new high-risk sections were identified. Based on real-time acquired multidimensional state signals, the system identifies a sudden change in the feature value of at least one signal dimension at any detection point that exceeds its preset threshold.

10. An intelligent online fatigue detection system for electric wires and cables, characterized in that: The system includes: The signal acquisition module, based on the cable line topology diagram and prior state information, sets up initial detection points and simultaneously acquires multi-dimensional state signals from each initial detection point. The distribution map generation module calculates and generates a set of fatigue distribution maps of the cable line based on the multidimensional state signal. The scheme optimization module identifies high-risk and low-risk sections based on the fatigue distribution map set, and generates an optimization and adjustment scheme in combination with the layout of the initial detection points; The optimization and adjustment module reconfigures the detection points according to the optimization and adjustment scheme to form an optimized detection point network. Based on the optimized detection point network, it continuously collects data and performs fatigue analysis. When the preset optimization triggering conditions are met, it triggers the optimization and adjustment of the detection point network.

Images