A method and system for identifying and processing abnormalities in optical cable laying

Abstract

The present invention provides a method and system for identifying and processing optical cable installation anomalies, relating to the field of power construction technology. The method comprises: acquiring multimodal sensor data of the optical cable at predetermined time intervals during installation; generating axial, sheath, and torsional characteristic vectors based on the multimodal sensor data; obtaining contribution weights of the axial, sheath, and torsional characteristic vectors to optical cable damage; performing a weighted summation of the characteristic vectors based on the contribution weights to obtain an installation risk index, which is used to quantify the risk of optical cable damage; performing analysis based on the installation risk index and obtaining analysis results, and performing closed-loop control of the traction equipment of the cable installation vehicle based on the analysis results. The present invention enables full-process, high-frequency, and seamless monitoring of the health of the optical cable.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power construction, and in particular to a method and system for identifying and processing abnormalities in optical cable laying. Background Art

[0002] In the early days, optical cable installation relied primarily on manual experience and simple mechanical assistance. Tension monitoring devices, such as tensiometers, provided crude control of the pulling force to prevent irreversible macroscopic damage to the cable core due to overstretching. Technological advances have led to the introduction of distributed fiber optic sensing technologies, such as those based on Brillouin and Rayleigh scattering, making it possible to perform static or quasi-static measurements of temperature and strain distribution along the cable.

[0003] In the prior art, the publication number is CN119738670A, and the name is a cable joint abnormal discharge monitoring system and method. The system includes a filling device, a communication optical fiber, a sensing optical fiber, and a distributed acoustic wave sensor host. The filling device is divided into a primary filling device and a secondary filling device, which is used to fill the cable joint with a sensitizing material. The filling device covers all the cable joints with the sensing optical fiber. One end of the communication optical fiber is pulled out by the distributed acoustic wave sensor host, and the other end is connected to the sensing optical fiber, which is used to monitor the optical signal transmission within the system. One end of the sensing optical fiber is connected to the communication optical fiber, and the other end is pulled out and laid on the cable and cable joint to sense the vibration of the cable joint. The distributed acoustic wave sensor host is connected to the communication optical fiber for signal detection, demodulation, and abnormal discharge identification. It can achieve high-precision abnormal discharge monitoring and intensity identification of cable joints, providing protection for cable power transmission safety.

[0004] However, existing technologies rely on post-layout testing. In the harsh environment of concrete ducts along high-speed railways, optical cables can experience localized stress concentration, sheath microcracks, and torsional stress accumulation when they navigate bends, joints, or rub against foreign objects within the duct. These initial defects, created at the moment of installation, can accelerate degradation under the long-term effects of the periodic vibrations of high-speed trains, significantly shortening the cable's useful life. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for identifying and processing abnormalities in optical cable laying, so as to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for identifying and processing abnormalities in optical cable laying, comprising the following steps:

[0008] During the optical cable laying process, multimodal sensing data of the optical cable is acquired at predetermined time intervals. The multimodal sensing data includes vibration data, internal axial strain data, external sheath integrity data, and internal torsional stress data. Static strain index, dynamic impact index, sheath damage index, and torsional stress index are extracted based on the multimodal sensing data.

[0009] generating an axial eigenvector, a sheath eigenvector, and a torsional eigenvector based on a static strain index, a dynamic impact index, a sheath damage index, and a torsional stress index;

[0010] Obtain the contribution weights of the axial eigenvector, sheath eigenvector, and torsional eigenvector to the optical cable damage, perform weighted summation of each eigenvector based on the contribution weights, and obtain the laying risk index, which is used to quantify the damage risk of the optical cable.

[0011] An analysis is performed based on the laying risk index and analysis results are obtained, and closed-loop control of the traction equipment of the cable laying engineering vehicle is performed based on the analysis results.

[0012] Furthermore, before laying the optical cable, remote sensing images of the laid line at multiple time points are obtained, a growth starting point is calibrated in the optical cable laying area of ​​each remote sensing image, the growth starting point is extended based on a region growing algorithm to divide the line area where the optical cable is laid in each remote sensing image, the line area where the optical cable is laid is gridded along the line area where the optical cable is laid to obtain multiple regional grids, the regional grids are divided into a first type and a second type based on backscattering intensity, wherein the first type of regional grid has no accumulated water and the second type of regional grid has accumulated water, a line area of ​​a remote sensing image is selected, and the second type of regional grid is located therein, the first type of regional grid at the same position is located in other remote sensing images, the second type of regional grid is restored based on the first type of regional grid to obtain a complete line area;

[0013] Conduct risk analysis on the line area, identify high-risk points in each regional grid, obtain the geographic coordinates of the high-risk points, and acquire multimodal sensing data if the predetermined time interval is not reached when the optical cable is laid through the high-risk points.

[0014] Furthermore, extracting the static strain index, dynamic impact index, sheath damage index, and torsional stress index based on the multimodal sensing data includes the following steps:

[0015] Phase-sensitive optical time-domain reflectometry is used to collect the phase change information of the backscattered Rayleigh light along the internal sensing optical cable. The phase change information is processed by short-time Fourier transform and low-pass filtering, and the dynamic impact index representing high-frequency impact events and the static strain index representing quasi-static tensile strain are calculated respectively.

[0016] Time domain reflectometry is used to monitor the real-time impedance of the optical cable loop in real time, and the sheath damage index is obtained based on the relative change rate between the real-time impedance and the healthy state reference impedance.

[0017] The polarization state rotation rate along the sensing optical fiber inside the optical cable is measured using polarization time domain analysis technology. The cumulative torsion angle is obtained by integrating and accumulating the polarization state rotation rate. The torsional stress index is calculated based on the cumulative torsion angle and the rated safe torsion angle limit of the optical cable.

[0018] Furthermore, generating the axial eigenvector, the sheath eigenvector, and the torsional eigenvector comprises the following steps:

[0019] The dynamic impact index, static strain index, sheath damage index and torsional stress index are timestamp aligned and each index is normalized. The normalized dynamic impact index and static strain index are fused by weighted average to obtain the axial eigenvector and construct a two-dimensional vector matrix. The first component of the two-dimensional vector matrix is ​​the sheath damage index or torsional stress index at the current moment, and the second component is the difference between the sheath damage index or torsional stress index at the current moment and the previous moment.

[0020] Furthermore, obtaining the laying risk index includes the following steps:

[0021] Obtain historical installation data, obtain axial characteristic vectors, sheath characteristic vectors and torsional characteristic vectors based on the historical installation data, perform statistical analysis on the axial characteristic vectors, sheath characteristic vectors and torsional characteristic vectors, obtain the contribution rate of each vector to the damage of the optical cable, determine the contribution weight of each vector based on the contribution rate, perform weighted summation of the axial characteristic vectors, sheath characteristic vectors and torsional characteristic vectors based on the contribution weights, and generate a installation risk index.

[0022] Furthermore, closed-loop control of the traction equipment of the cable laying engineering vehicle based on the analysis results includes the following steps:

[0023] The obtained laying risk index is compared with the preset first risk threshold and second risk threshold. When the laying risk index is lower than the first risk threshold, the traction speed of the cable laying engineering vehicle is controlled to smoothly recover to the set value. When the laying risk index is between the first risk threshold and the second risk threshold, the laying speed of the cable laying engineering vehicle is dynamically adjusted according to the actual value of the laying risk index. When the laying risk index is not less than the second risk threshold, the movement of the cable laying engineering vehicle is stopped.

[0024] Furthermore, before performing analysis based on the laying risk index, the traction status data of the traction equipment on the laying engineering vehicle is obtained in real time, and the traction deviation factor representing the stability of the traction system is calculated based on the traction status data, and a first default value and a second default value are set. The first default value and the second default value are adjusted based on the traction deviation factor to obtain a first risk threshold and a second risk threshold.

[0025] The present invention also provides a system for identifying and processing abnormalities in optical cable laying, the system being used to execute the method for identifying and processing abnormalities in optical cable laying, comprising:

[0026] A sensor module, wherein during the installation of the optical cable, the sensor module acquires multimodal sensing data of the optical cable at predetermined time intervals, the multimodal sensing data including vibration data, internal axial strain data, external sheath integrity data, and internal torsional stress data;

[0027] A preprocessing module extracts a static strain index, a dynamic impact index, a sheath damage index, and a torsional stress index based on the multimodal sensing data, and generates an axial eigenvector, a sheath eigenvector, and a torsional eigenvector based on the static strain index, the dynamic impact index, the sheath damage index, and the torsional stress index;

[0028] A calculation module obtains the contribution weights of the axial eigenvector, sheath eigenvector, and torsional eigenvector to the damage of the optical cable, performs a weighted summation of the eigenvectors based on the contribution weights, and obtains a laying risk index. The laying risk index is used to quantify the damage risk of the optical cable.

[0029] The analysis module performs analysis based on the laying risk index and obtains analysis results, and performs closed-loop control of the traction equipment of the cable laying engineering vehicle based on the analysis results.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] The present invention deploys a monitoring system on the laying engineering vehicle to synchronously acquire multimodal sensing data such as the internal axial strain, vibration, external sheath integrity and internal torsional stress of the optical cable in real time. It can accurately capture instantaneous and dynamic damage events caused by passing through bends, encountering foreign objects or improper traction, and achieve full-process, high-frequency seamless monitoring of the health status of the optical cable. Secondly, the present invention generates a single, quantitative laying risk index, which comprehensively reflects the coupling effects of multiple damage factors such as stretching, torsion, and scratching. Compared with the traditional method of only monitoring a single physical quantity, it can more comprehensively and accurately quantify the real risks faced by the optical cable. Finally, the present invention directly uses the quantitative laying risk index for closed-loop control of the traction equipment, thereby avoiding the occurrence and accumulation of potential damage to the greatest extent, and significantly improving the laying quality, long-term reliability and operational safety of the optical cable. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 Schematic diagram of the overall method flow of the present invention;

[0033] Figure 2 This is a block diagram of the system modules of the present invention. DETAILED DESCRIPTION

[0034] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0035] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0036] like Figure 1 As shown, a method for identifying and processing abnormalities in optical cable laying includes the following specific steps:

[0037] Step S1: During the optical cable laying process, multimodal sensing data of the optical cable is obtained at predetermined time intervals. The multimodal sensing data includes vibration data, internal axial strain data, external sheath integrity data, and internal torsional stress data. Static strain index, dynamic impact index, sheath damage index, and torsional stress index are extracted based on the multimodal sensing data.

[0038] Step S2: generating an axial eigenvector, a sheath eigenvector, and a torsional eigenvector based on the static strain index, the dynamic impact index, the sheath damage index, and the torsional stress index.

[0039] Step S3: Obtain the contribution weights of the axial eigenvector, sheath eigenvector, and torsional eigenvector to the damage of the optical cable, perform weighted summation on each eigenvector based on the contribution weights, and obtain a laying risk index. The laying risk index is used to quantify the damage risk of the optical cable.

[0040] Step S4: Analyze the laying risk index and obtain analysis results, and perform closed-loop control on the traction equipment of the cable laying engineering vehicle based on the analysis results.

[0041] In this embodiment, the cable is laid by an optical cable laying vehicle, and multimodal sensing data of the optical cable is obtained at predetermined time intervals, such as 1 minute. The laying risk index is obtained by analyzing the multimodal sensing data. The larger the laying risk index, the greater the possibility of damage to the optical cable. When the laying risk index is large, the cable laying engineering vehicle sends an alarm to on-site personnel, who can check whether the optical cable is damaged in time. The traditional method is to conduct a visual inspection along the cable laying path after the cable laying is completed. Therefore, the present invention can speed up work efficiency and make timely repairs when the cable is abnormal.

[0042] In this embodiment, before laying the optical cable, remote sensing images of the laid line at multiple time points are obtained, and the growth starting point is calibrated in the optical cable laying area of ​​each remote sensing image. The growth starting point is extended based on the region growing algorithm to divide the line area where the optical cable is laid in each remote sensing image. The line area where the optical cable is laid is gridded and divided along the line area where the optical cable is laid to obtain multiple regional grids. The regional grids are divided into a first type and a second type based on the backscattering intensity, wherein the first type of regional grid has no water accumulation and the second type of regional grid has water accumulation. A line area of ​​a remote sensing image is selected, and the second type of regional grid is located therein. The first type of regional grid at the same position is located in other remote sensing images, and the second type of regional grid is restored based on the first type of regional grid to obtain a complete line area.

[0043] Conduct risk analysis on the line area, identify high-risk points in each regional grid, obtain the geographic coordinates of the high-risk points, and acquire multimodal sensing data if the predetermined time interval is not reached when the optical cable is laid through the high-risk points.

[0044] First, high-resolution remote sensing images of the area where the optical cable is laid are obtained at multiple time points. For example, three remote sensing images are obtained, with the remote sensing images of adjacent time points separated by one week. These remote sensing images are then subjected to radiometric correction, atmospheric correction, and geometric correction to ensure spatial and spectral consistency between the remote sensing images. Then, in the remote sensing images of each time phase, one or more clear growth starting points are manually selected as the starting points of the region growing algorithm. The region growing threshold is set based on spectral, texture, and other characteristics, and the region mask covering the line is gradually expanded to form. The region growing algorithm is an existing technology and will not be described in detail here. For the line area formed in each time phase, the extension direction of the line area is divided into multiple equal-sized regional grids, with the size of the regional grid being, for example, 100 meters by 100 meters.

[0045] Then, using microwave remote sensing backscatter intensity data from the corresponding time phase, grids are divided into two types: Type 1 (no water accumulation) and Type 2 (water accumulation). The specific judgment rule is based on the difference in backscatter intensity between waterlogged and non-waterlogged ground. If a grid area is identified as Type 2, some parts of it may be obscured by water, making it impossible to obtain its specific line information. In this scenario, this embodiment obtains the corresponding Type 1 grid area at the same spatial location, that is, the non-waterlogged grid area. Image replacement is used to restore the image of the waterlogged area to the non-waterlogged image, thereby obtaining a complete and continuous line area.

[0046] After obtaining a complete route area, high-risk points are located by combining the degree of curvature and terrain height. For example, if a certain area has large terrain fluctuations, it will be designated as a high-risk point. High-risk points can also be designated manually. When the optical cable line passes through a high-risk point, the multimodal sensing system automatically activates to monitor the cable laying process, even if the pre-set multimodal sensing data collection interval has not been reached.

[0047] In this embodiment, extracting the static strain index, dynamic impact index, sheath damage index, and torsional stress index based on multimodal sensing data includes the following steps:

[0048] Phase-sensitive optical time-domain reflectometry is used to collect the phase change information of the backscattered Rayleigh light along the internal sensing optical cable. The phase change information is processed by short-time Fourier transform and low-pass filtering, and the dynamic impact index characterizing high-frequency impact events and the static strain index characterizing quasi-static tensile strain are calculated respectively.

[0049] The real-time impedance of the optical cable loop is monitored using time domain reflectometry technology, and the sheath damage index is obtained based on the relative change rate between the real-time impedance and the healthy reference impedance.

[0050] The polarization state rotation rate along the sensing optical fiber inside the optical cable is measured using polarization time domain analysis technology. The cumulative torsion angle is obtained by integrating and accumulating the polarization state rotation rate. The torsional stress index is calculated based on the cumulative torsion angle and the rated safe torsion angle limit of the optical cable.

[0051] During the installation process, a monitoring system deployed on a cable-laying vehicle synchronously acquires multimodal sensor data along the optical cable. Specifically, when acquiring internal axial strain and vibration data, phase-sensitive optical time-domain reflectometry (PSTAR) technology is used to collect real-time phase change information of backscattered Rayleigh light from the optical cable's internal sensing area. This phase change information is then subjected to a short-time Fourier transform (SFT) to obtain a phase change representing high-frequency shock. Specifically, a short time window is defined at predetermined intervals, and the phase change representing high-frequency shock within the short time window is acquired. The first difference between this phase change and the shock value of the normal vibration level is calculated. This first difference reflects the degree to which the current vibration within the short time window deviates from the normal range. The first difference is then normalized to a value between 0 and 1, using, for example, a sigmoid function, to convert it into a dimensionless value. After mapping, if the dynamic shock index approaches 0, it indicates a smooth installation process with low vibration and a very low risk of dynamic damage. When the dynamic impact index approaches 1, it indicates that the optical cable has suffered severe impact or severe friction, and the risk of dynamic damage is extremely high, requiring immediate intervention.

[0052] The static strain index is used to quantify the quasi-static, continuous tensile strain experienced by optical cables during installation. The static strain index is calculated by first low-pass filtering the phase change data within a short time window to obtain the low-frequency component of the signal. This is then linearly transformed to obtain the real-time static strain. The real-time static strain is then normalized to the static strain index using a three-stage mapping method. For example, if the real-time static strain is less than or equal to the warning strain threshold, the static strain index is mapped to 0, indicating that the cable is within the safe range. If the real-time static strain is greater than or equal to the ultimate strain threshold, the static strain index is mapped to 1, indicating that the cable is within the dangerous range. If the real-time static strain is between the warning strain threshold and the ultimate strain threshold, the ultimate strain threshold is used as the maximum value and the warning strain threshold as the minimum value. Using the maximum-minimum normalization method, the real-time static strain is mapped to a range between 0 and 1. The warning strain threshold and ultimate strain threshold are set based on experience; generally, the warning strain threshold is 0.1 times the maximum acceptable strain risk, and the ultimate strain threshold is 0.7 times the maximum acceptable strain value.

[0053] To calculate the sheath damage index, the real-time impedance representing the insulation performance of the optical cable sheath is first obtained. The ratio between this impedance and a preset baseline impedance is then calculated. The impedance degradation ratio is calculated by subtracting this ratio from 1. The baseline impedance represents the impedance value of the optical cable measured in a dry environment before installation, representing its healthy state. Similarly, the impedance degradation ratio is normalized to a value between 0 and 1. The core function of an optical cable sheath is electrical insulation, characterized by high impedance. When a damaged sheath comes into contact with an external conductive environment (such as moist soil or metal channels), its equivalent impedance inevitably decreases. This method directly uses the change in this core physical quantity as input, ensuring the fundamental validity of the monitoring results. By using the impedance degradation ratio rather than the absolute value of the real-time impedance, this ratio calculation effectively offsets inherent, non-fault-related impedance fluctuations along the optical cable, allowing the calculated results to focus more on the relative changes caused by the damage variable, thereby improving the signal-to-noise ratio and accuracy of the detection. When the sheath damage index approaches 0, it means that the optical cable sheath is intact and the insulation performance is good. When it approaches 1, it means that the optical cable sheath is severely damaged at that position and a low-impedance path has been formed with the external environment.

[0054] When calculating internal torsional stress data, polarization time-domain analysis (PDO) technology is used to determine the polarization rotation rate, which represents the local torsion rate of the optical cable. This is then integrated along the cable length to obtain the cumulative torsion angle, representing the total degree of torsion. It should be noted that real-time internal torsional stress data along the cable is acquired by measuring the local birefringence distribution of the sensing fiber within the optical cable. The ratio of the cumulative torsion angle to the preset safety limit is calculated as the torsion ratio. The absolute value of the torsion ratio is then normalized to a value between 0 and 1. By taking the absolute value, the risk is accurately focused on the magnitude of the torsional stress, which is consistent with the mechanism of physical damage. When the internal torsional stress data approaches 0, the cumulative torsional stress at that location is low, indicating better torsion control during installation. When it approaches 1, the cumulative torsion angle at that location has reached or far exceeded the safety limit, indicating that the cable has developed microcracks or permanent deformation.

[0055] In particular, if the detected sensor fails, is disconnected, or is not powered on, the index is output as -1, making it easier for relevant personnel to distinguish whether it is a sensor failure or a cable failure.

[0056] Generating the axial eigenvectors, sheath eigenvectors, and torsional eigenvectors involves the following steps:

[0057] The dynamic impact index, static strain index, sheath damage index and torsional stress index are timestamp aligned and each index is normalized. The normalized dynamic impact index and static strain index are fused by weighted average to obtain the axial eigenvector and construct a two-dimensional vector matrix. The first component of the two-dimensional vector matrix is ​​the sheath damage index or torsional stress index at the current moment, and the second component is the difference between the sheath damage index or torsional stress index at the current moment and the previous moment.

[0058] Specifically, each dynamic impact index, static strain index, sheath damage index, and torsional stress index has a corresponding generation time, which is used as a timestamp for alignment. After alignment, the indices are normalized to further eliminate dimensional differences between the different indices. When calculating the axial eigenvector, the synchronized dynamic impact index and static strain index are combined using a weighted average. Since both indices have already been normalized, the result obtained after weighted averaging will be between 0 and 1. The weights of the dynamic impact index and the static strain index can be preset empirically, for example, the weight of the dynamic impact index can be set to 0.6 and the weight of the static strain index to 0.4. Alternatively, a dynamic weighting approach can be used, where the weights of the dynamic impact index and the static strain index increase with the value of the other. The closer the dynamic impact index is to 1, the greater the weight of the static strain index. Similarly, the closer the static impact index is to 1, the greater the weight of the dynamic strain index. In other words, the risk of one index increases as the risk of the other increases.

[0059] The sheath eigenvector generation method is described using the example of a sheath eigenvector. The torsion eigenvector generation method is the same as the sheath eigenvector generation method. First, the sheath damage index at the previous moment is directly used as the first component of the two-dimensional vector. The second component is then constructed. The second component is the difference between the sheath damage index at the current moment and the sheath damage index at the previous moment. The two-dimensional vector containing the first and second components is used as the sheath damage index.

[0060] In this embodiment, obtaining the laying risk index includes the following steps:

[0061] Obtain historical installation data, obtain axial characteristic vectors, sheath characteristic vectors and torsional characteristic vectors based on the historical installation data, perform statistical analysis on the axial characteristic vectors, sheath characteristic vectors and torsional characteristic vectors, obtain the contribution rate of each vector to the damage of the optical cable, determine the contribution weight of each vector based on the contribution rate, perform weighted summation of the axial characteristic vectors, sheath characteristic vectors and torsional characteristic vectors based on the contribution weights, and generate a installation risk index.

[0062] Specifically, historical sample data is first obtained. The historical sample data includes axial eigenvectors, sheath eigenvectors, torsional eigenvectors, and the degree of cable damage. The degree of cable damage, for example, includes a scale of 1 to 10. Based on the historical sample data, a correlation analysis is performed between the axial eigenvectors, sheath eigenvectors, torsional eigenvectors, and the degree of cable damage, thereby obtaining correlation coefficients between the various eigenvectors and the degree of cable damage. The correlation coefficients are, for example, Pearson correlation coefficients. The contribution rate of each vector is determined based on the Pearson correlation coefficients. The greater the contribution rate, the greater the weight of the corresponding vector in affecting the cable damage. For example, if the correlation coefficients of the axial eigenvectors, sheath eigenvectors, and torsional eigenvectors are 0.5, 0.4, 0.6, 0.7, and 0.8, respectively, and the total value is 3, then the corresponding contribution weights are 0.16 = (0.5 / 3), 0.13 = (0.4 / 3), 0.2 = (0.6 / 3), 0.23 = (0.7 / 3), and 0.26 = (0.8 / 3), respectively.

[0063] In other embodiments, a decision tree model may be established based on historical sample data, and SHAP value analysis may be performed on each eigenvector of the established decision tree model to obtain the SHAP value of each eigenvector. The larger the SHAP value, the greater the influence of the eigenvector on the result.

[0064] In particular, based on the previous introduction, the sheath eigenvector and torsional eigenvector include two components, so there are five contribution rates. Finally, the axial eigenvector, sheath eigenvector, and torsional eigenvector are weighted summed based on the contribution weights to obtain the installation risk index.

[0065] The closed-loop control of the traction equipment of the cable laying engineering vehicle based on the analysis results includes the following steps:

[0066] The obtained laying risk index is compared with the preset first risk threshold and second risk threshold. When the laying risk index is lower than the first risk threshold, the traction speed of the cable laying engineering vehicle is controlled to smoothly recover to the set value. When the laying risk index is between the first risk threshold and the second risk threshold, the laying speed of the cable laying engineering vehicle is dynamically adjusted according to the actual value of the laying risk index. When the laying risk index is not less than the second risk threshold, the movement of the cable laying engineering vehicle is stopped.

[0067] Before performing analysis based on the laying risk index, the traction status data of the traction equipment on the laying engineering vehicle is obtained in real time, and the traction deviation factor representing the stability of the traction system is calculated based on the traction status data, and a first default value and a second default value are set. The first default value and the second default value are adjusted based on the traction deviation factor to obtain a first risk threshold and a second risk threshold.

[0068] The first risk threshold is lower than the second risk threshold. In this embodiment, the first risk threshold and the second risk threshold can be fixed values ​​or dynamic values. In this embodiment, they are dynamic values, and their initial values ​​are 0.6 and 0.85, respectively. When the traction machine is controlled by a PLC, when the laying risk index is lower than the first risk threshold, the PLC enters a speed optimization mode, which smoothly restores the current speed to a preset speed setting value instead of an instantaneous jump. When the laying risk index is between the first risk threshold and the second risk threshold, the PLC controller enters a proportional speed reduction mode. Specifically, a comparison table of traction speeds and risk indices can be set. In the comparison table, each traction speed corresponds to a range of risk indices. The PLC determines the corresponding traction speed based on the received risk index and the comparison table, and reduces the current speed to the determined traction speed. This proportional adjustment allows the traction speed to be accurately matched to the risk level. The higher the risk, the slower the speed.

[0069] When the installation risk index is no less than the adaptive second risk threshold, a specific data frame with the highest bus priority is immediately sent to the PLC module, triggering the PLC's hardware interrupt service routine. The PLC then immediately sends an STO signal to the traction motor inverter via its safety output. Upon receiving this signal, the inverter hardware-basedly cuts off the IGBT gate drive pulses to the motor, causing the motor to immediately enter free-coast mode, with torque dropping to zero within milliseconds. Simultaneously, the PLC outputs a brake activation signal, such as a 24VDC signal, to the relay or solenoid valve controlling the electromagnetic brake via another high-speed output. This causes the mechanical brake (normally closed) to immediately lock the drive shaft of the traction wheel, achieving fast and reliable physical braking.

[0070] The traction deviation factor represents the degree to which the working state of the traction system deviates from the ideal. The larger the traction deviation factor, the more unstable the traction system is, which means that it is more likely to cause laying damage. As for the calculation method of the traction deviation factor, this embodiment first divides the short time window into multiple test windows, obtains the traction force of the motor on the cable within the test window, and calculates the first average value of the traction force of the tractor in the short time window based on this, determines the second average value of the ideal traction force, calculates the first difference between the first average value and the second average value, and uses the first average value as the traction deviation factor. Then, a discount coefficient comparison table is set. In this comparison table, traction deviation factors of different size ranges correspond to different discount coefficients. For example, if the traction deviation factor is 300N, within the range of 300N-400N, its corresponding discount coefficient is 0.9. The discount coefficient is multiplied by the first default value and the second default value to obtain the first risk threshold and the second risk threshold.

[0071] See also Figure 2A system for identifying and processing abnormalities in optical cable laying is provided, wherein the system is used to execute the method for identifying and processing abnormalities in optical cable laying, comprising:

[0072] The sensor module acquires multimodal sensing data of the optical cable at predetermined time intervals during the optical cable laying process. The multimodal sensing data includes vibration data, internal axial strain data, external sheath integrity data, and internal torsional stress data.

[0073] The preprocessing module extracts the static strain index, dynamic impact index, sheath damage index and torsional stress index based on the multimodal sensing data, and generates the axial eigenvector, sheath eigenvector and torsional eigenvector based on the static strain index, dynamic impact index, sheath damage index and torsional stress index.

[0074] The calculation module obtains the contribution weights of the axial eigenvector, sheath eigenvector and torsional eigenvector to the damage of the optical cable, performs weighted summation on each eigenvector based on the contribution weights, and obtains the laying risk index, which is used to quantify the damage risk of the optical cable.

[0075] The analysis module performs analysis based on the laying risk index and obtains analysis results, and performs closed-loop control of the traction equipment of the cable laying engineering vehicle based on the analysis results.

[0076] It should be noted that all calculation formulas in this application document use regression analysis including but not limited to machine learning algorithms to deeply analyze the relevant parameters collected and identify their natural trends and interrelationships. Use professional software, such as Python's Scikit-learn library or R language, to automatically generate mathematical models that match the data. Then, objectively evaluate the performance of the model through methods such as cross-validation, and combine continuous feedback and optimization to ensure that the created formula truly reflects the inherent laws of the data, thereby ensuring its effectiveness and accuracy. In all calculation formulas in this application, the parameters in each formula are dimensionally non-scaled within a consistent range to ensure that different physical quantities are compared on the same scale; dimensionless technical means include but are not limited to Min-Max-Normalization and Z-Score standardization.

[0077] The technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random-access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0078] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0079] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for identifying and processing abnormalities in optical cable laying, characterized in that: The specific steps include: During the optical cable laying process, multimodal sensing data of the optical cable is acquired at predetermined time intervals. The multimodal sensing data includes vibration data, internal axial strain data, external sheath integrity data, and internal torsional stress data. Static strain index, dynamic impact index, sheath damage index, and torsional stress index are extracted based on the multimodal sensing data. generating an axial eigenvector, a sheath eigenvector, and a torsional eigenvector based on a static strain index, a dynamic impact index, a sheath damage index, and a torsional stress index; Obtain the contribution weights of the axial eigenvector, sheath eigenvector, and torsional eigenvector to the optical cable damage, perform weighted summation of each eigenvector based on the contribution weights, and obtain the laying risk index, which is used to quantify the damage risk of the optical cable. Analyze the laying risk index and obtain analysis results, and perform closed-loop control on the traction equipment of the cable laying engineering vehicle based on the analysis results; Before laying the optical cable, remote sensing images of the laying line at multiple time points are obtained, a growth starting point is calibrated in the optical cable laying area of ​​each remote sensing image, the growth starting point is extended based on a region growing algorithm to divide the line area where the optical cable is laid in each remote sensing image, the line area where the optical cable is laid is gridded along the line area where the optical cable is laid to obtain multiple regional grids, the regional grids are divided into a first type and a second type based on backscattering intensity, wherein the first type of regional grid has no accumulated water and the second type of regional grid has accumulated water, a line area of ​​a remote sensing image is selected, and the second type of regional grid is located therein, the first type of regional grid at the same position is located in other remote sensing images, the second type of regional grid is restored based on the first type of regional grid to obtain a complete line area; Conduct risk analysis on the line area, identify high-risk points in each regional grid, obtain the geographic coordinates of the high-risk points, and acquire multimodal sensor data if the predetermined time interval is not reached when the optical cable is laid through the high-risk points; Extracting static strain index, dynamic impact index, sheath damage index, and torsional stress index based on multimodal sensing data includes the following steps: Phase-sensitive optical time-domain reflectometry is used to collect the phase change information of the backscattered Rayleigh light along the internal sensing optical cable. The phase change information is processed by short-time Fourier transform and low-pass filtering, and the dynamic impact index representing high-frequency impact events and the static strain index representing quasi-static tensile strain are calculated respectively. Time domain reflectometry is used to monitor the real-time impedance of the optical cable loop in real time, and the sheath damage index is obtained based on the relative change rate between the real-time impedance and the healthy state reference impedance. The polarization state rotation rate along the sensing optical fiber inside the optical cable is measured using polarization time domain analysis technology. The cumulative torsion angle is obtained by integrating and accumulating the polarization state rotation rate. The torsional stress index is calculated based on the cumulative torsion angle and the rated safe torsion angle limit of the optical cable.

2. The method for identifying and processing abnormalities in optical cable laying according to claim 1, characterized in that: Generating the axial eigenvectors, sheath eigenvectors, and torsional eigenvectors involves the following steps: The dynamic impact index, static strain index, sheath damage index and torsional stress index are timestamp aligned and each index is normalized. The normalized dynamic impact index and static strain index are fused by weighted average to obtain the axial eigenvector and construct a two-dimensional vector matrix. The first component of the two-dimensional vector matrix is ​​the sheath damage index or torsional stress index at the current moment, and the second component is the difference between the sheath damage index or torsional stress index at the current moment and the previous moment.

3. The method for identifying and processing abnormalities in optical cable laying according to claim 2, wherein: Obtaining the laying risk index involves the following steps: Obtain historical installation data, obtain axial characteristic vectors, sheath characteristic vectors and torsional characteristic vectors based on the historical installation data, perform statistical analysis on the axial characteristic vectors, sheath characteristic vectors and torsional characteristic vectors, obtain the contribution rate of each vector to the damage of the optical cable, determine the contribution weight of each vector based on the contribution rate, perform weighted summation of the axial characteristic vectors, sheath characteristic vectors and torsional characteristic vectors based on the contribution weights, and generate a installation risk index.

4. The method for identifying and processing abnormalities in optical cable laying according to claim 3, wherein: The closed-loop control of the traction equipment of the cable laying engineering vehicle based on the analysis results includes the following steps: The obtained laying risk index is compared with the preset first risk threshold and second risk threshold. When the laying risk index is lower than the first risk threshold, the traction speed of the cable laying engineering vehicle is controlled to smoothly recover to the set value. When the laying risk index is between the first risk threshold and the second risk threshold, the laying speed of the cable laying engineering vehicle is dynamically adjusted according to the actual value of the laying risk index. When the laying risk index is not less than the second risk threshold, the movement of the cable laying engineering vehicle is stopped.

5. The method for identifying and processing abnormalities in optical cable laying according to claim 4, characterized in that: Before performing analysis based on the laying risk index, the traction status data of the traction equipment on the laying engineering vehicle is obtained in real time, and the traction deviation factor representing the stability of the traction system is calculated based on the traction status data, and a first default value and a second default value are set. The first default value and the second default value are adjusted based on the traction deviation factor to obtain a first risk threshold and a second risk threshold.

6. A system for identifying and processing abnormalities in optical cable laying, characterized by: The system is used to execute the optical cable laying abnormality identification and processing method according to any one of claims 1 to 5, comprising: A sensor module, wherein during the installation of the optical cable, the sensor module acquires multimodal sensing data of the optical cable at predetermined time intervals, the multimodal sensing data including vibration data, internal axial strain data, external sheath integrity data, and internal torsional stress data; A preprocessing module extracts a static strain index, a dynamic impact index, a sheath damage index, and a torsional stress index based on the multimodal sensing data, and generates an axial eigenvector, a sheath eigenvector, and a torsional eigenvector based on the static strain index, the dynamic impact index, the sheath damage index, and the torsional stress index; A calculation module obtains the contribution weights of the axial eigenvector, sheath eigenvector, and torsional eigenvector to the damage of the optical cable, performs a weighted summation of the eigenvectors based on the contribution weights, and obtains a laying risk index. The laying risk index is used to quantify the damage risk of the optical cable. The analysis module performs analysis based on the laying risk index and obtains analysis results, and performs closed-loop control of the traction equipment of the cable laying engineering vehicle based on the analysis results.

Images