Cable life dynamic evaluation system based on multi-physics field coupling

Abstract

The invention discloses a cable life dynamic evaluation system based on multi-physics field coupling, and particularly relates to the field of industrial automation and control systems, which comprises a multi-physics field sensing module, a coupling analysis engine module, a dynamic life evaluation module, a digital twin interaction module and an environmental interference suppression module, through a distributed optical fiber temperature sensor, a capacitive electric field sensor and a magnetostrictive stress sensor, temperature, electric field, magnetic field and mechanical stress data of a cable are collected in real time, multi-physical field characteristics and a cable defect database are matched in real time by using a cross-scale dynamic association algorithm, a damage state is evaluated, and the cable defect detection accuracy is improved. A time sequence neural network architecture is adopted to predict the remaining life, model self-correction is achieved through digital twin comparison, interference is suppressed in combination with an environment-physical field coupling compensation matrix, sensing and evaluation of the health state of the cable, life prediction and continuous optimization of the model are achieved, and the efficiency of cable life evaluation is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial automation and control systems, more particularly, the present application relates to a cable life dynamic evaluation system based on multi-physical field coupling. BACKGROUND

[0002] With the continuous expansion of the power system scale and the increase of cable service life, the aging and failure of cables are increasingly prominent, which threatens the safe and stable operation of the power grid. As a key equipment for power transmission, the monitoring and life evaluation of the running state of the cable is of great significance to ensure the reliability of power supply.

[0003] The prior art mainly uses a single physical field monitoring combined with a statistical model or a simple algorithm to evaluate the cable life. For example, by laying temperature sensors to collect cable conductor temperature, combining a thermal aging model to calculate the remaining life, detecting partial discharge signals of the insulation layer, and evaluating the cable health status according to the correlation between the discharge quantity and the defect degree to evaluate the cable life However, in actual use, there are still some shortcomings, such as relying on single physical field information, which is difficult to fully reflect the complex failure mechanism of the multi-field coupling of "heat-electricity-magnetism-force" of the cable, which easily leads to inaccurate evaluation, secondly, the fixed threshold lacks dynamic adaptability to the cable operating conditions and environmental interference, the evaluation result accuracy is poor, the simple linear life model cannot depict the nonlinear time-varying characteristics of defect evolution, the prediction accuracy is limited, the system lacks self-learning and optimization ability, the model parameters cannot be updated and corrected according to real-time data, and it is difficult to realize accurate dynamic evaluation. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a cable life dynamic evaluation system based on multi-physical field coupling, which solves the problems raised in the background art by the following scheme.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical scheme: a cable life dynamic evaluation system based on multi-physical field coupling, comprising a multi-physical field perception module, the cable to be evaluated is recorded as a target cable, real-time collection of temperature field, electric field, magnetic field and mechanical stress field data of the target cable during operation; a coupling analysis engine module, receiving the data of the perception module and matching the multi-physical field characteristics with the pre-established cable defect database in real time through a cross-scale dynamic correlation algorithm, and outputting the target cable damage state evaluation result to a dynamic life evaluation module; a dynamic life evaluation module, based on the damage evaluation result output by the coupling analysis engine, using a multi-physical field coupled time series neural network architecture for life prediction, and synchronizing the prediction result to a digital twin interaction module; The digital twin interaction module receives actual measurement data and life prediction results, compares the simulation data of the digital twin in real time, triggers a self-correction signal when the difference exceeds a dynamic threshold, and feeds the signal back to the coupling analysis engine module for model parameter optimization. The environmental interference suppression module is interconnected with the multi-physical field perception module, and dynamically corrects the original measurement data through an environment-physical field coupling compensation matrix. The corrected data is input to the coupling analysis engine module.

[0006] Preferably, the multi-physical field perception module includes a distributed optical fiber temperature sensor array, a capacitive electric field sensor group, and a magnetostrictive stress sensor. The distributed optical fiber temperature sensor is arranged in a spiral winding manner on the surface of the cable conductor, the capacitive electric field sensor is embedded in the cable insulation shielding layer in the form of an orthogonal array, and the magnetostrictive stress sensor is arranged at equal intervals along the axial direction of the cable.

[0007] Preferably, the temperature field data includes the surface temperature of the target cable conductor, the temperature at different depths of the insulation layer, and the surface temperature of the shielding layer. The electric field data includes the instantaneous field strength of the target cable terminal, the voltage value between the conductor and the shielding layer. The magnetic field data includes the magnetic induction intensity of the center point of the target cable conductor surface, the load current in the cable line, the induced current of the metal shielding layer, and the ground current. The mechanical stress field data includes the instantaneous axial / radial tension of the target cable and the instantaneous bending strain force.

[0008] Preferably, the cross-scale dynamic correlation algorithm uses cross-scale feature extraction technology to perform feature dimension reduction and parameter processing on the multi-physical field data collected by the perception module, to obtain a feature vector that reflects the overall operation state of the cable. Meanwhile, the algorithm calls a pre-established cable defect database to perform real-time calculation on the correlation between the feature vector and the cable defects.

[0009] Preferably, the cable defect database includes defect types and corresponding feature vector data, corresponding multi-physical field response characteristics, and defect evolution rules. The cable defect database is dynamically updated based on experimental data and actual operation cable defect detection case data.

[0010] Preferably, the damage state assessment result includes the damage type, damage location, and damage degree corresponding to the defect. The result is output to the dynamic life assessment module using a standardized data interface and a hierarchical identifier.

[0011] Preferably, the input end of the time series neural network architecture adopts a multi-channel parallel structure, embeds a coupling layer, and accesses a bidirectional long-term memory network layer to calculate the evolution function and time-dependent coefficient of the damage state over time. The architecture is trained in combination with historical damage data of the cable and the life attenuation curve of similar cables, and outputs the residual life prediction value of the target cable.

[0012] Preferably, the model parameter optimization is directed to the difference between the digital twin simulation data and the actual measurement data, and the optimization locates the multi-physical field coupling coefficients and the feature matching threshold in the cross-scale correlation algorithm, and the optimization process adopts an adaptive particle swarm algorithm combined with historical optimization records and parameter calibration experience of similar cables for dynamic adjustment.

[0013] Preferably, the environment-physical field coupling compensation matrix is a mathematical model for environment interference suppression, which quantifies the coupling influence relationship between environmental factors and each physical field in the form of a multi-dimensional matrix, the row dimension of the matrix corresponds to each data of the temperature field, the electric field, the magnetic field and the mechanical stress field, the column dimension corresponds to the environmental interference factors including temperature, humidity, air pressure and wind speed, and the matrix element is a coupling coefficient calibrated by experimental and measured data, wherein the matrix coefficient is periodically updated according to the cable laying scene and seasonal change.

[0014] Technical effects and advantages of the present application: The present application realizes multi-dimensional perception and comprehensive analysis of the cable operation state through the multi-physical field coupling technology, compared with the traditional single parameter monitoring method, simultaneously collects temperature field, electric field, magnetic field and mechanical stress field data, and comprehensively reflects the operation health status of the cable through multi-dimensional information fusion, effectively improving the integrity and accuracy of the state monitoring; The present application significantly improves the real-time and adaptability of the evaluation by introducing a dynamic life prediction and self-optimization mechanism, the system adopts a time sequence neural network architecture for life prediction, and dynamically corrects the model parameters by combining with the digital twin technology, continuously optimizes the evaluation model according to the actual operation data, adapts to complex working conditions and environmental changes, and greatly improves the accuracy and reliability of the life prediction; The present application compares the actual data with the simulation data in real time, automatically triggers parameter optimization when the difference exceeds the threshold, adjusts the coupling coefficients and the matching threshold by combining the adaptive particle swarm algorithm and historical experience, effectively reduces the influence of environmental interference and model error, and ensures that the system always maintains high-precision evaluation ability in complex and changeable operating environment. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 It is a schematic diagram of the overall structure of the present application; Figure 2 It is a schematic diagram of the target cable defect determination of the present application; Figure 3 It is a schematic diagram of the target cable life prediction of the present application. DETAILED DESCRIPTION

[0016] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of the present application.

[0017] As shown in FIGS. 1, 2, 3 and 4, the cable life dynamic evaluation system based on multi-physical field coupling comprises a multi-physical field sensing module. Figure 1 Figure 2 As shown in FIGS. 1, 2, 3 and 4, the cable life dynamic evaluation system based on multi-physical field coupling comprises a multi-physical field sensing module. Figure 3 The multi-physical field sensing module comprises a distributed optical fiber temperature sensor array, a capacitive electric field sensor group and a magnetostrictive stress sensor.

[0018] It needs to be particularly pointed out that the multi-physical field sensing module comprises a distributed optical fiber temperature sensor array, a capacitive electric field sensor group and a magnetostrictive stress sensor.

[0019] The temperature field data comprises the surface temperature of the target cable conductor, the temperature at different depths of the insulation layer, and the surface temperature of the shielding layer.

[0020] It needs to be further pointed out that the reason why the sensors are arranged in different ways comprises: The distributed optical fiber temperature sensor is arranged on the surface of the conductor in a spiral winding manner, which maximizes the temperature sensing range of the conductor in the axial and circumferential directions through the spiral path, and the elastic laying manner of the spiral winding does not hinder the normal thermal expansion and contraction of the cable.

[0021] The capacitive electric field sensor is embedded in the insulation shielding layer in an orthogonal array form.

[0022] ​The magnetostrictive stress sensors are arranged at equal intervals in the axial direction, and the mechanical stress of the cable in operation is distributed in the axial direction in a gradient manner, the equal interval arrangement is used to judge the stress concentration area through the stress difference of adjacent sensors, and the mechanical deformation is directly converted into a magnetic permeability change signal, so that high sensitivity monitoring of a small stress fluctuation is realized.

[0023] It needs to be further explained that the reason for selecting the temperature field, electric field, magnetic field and mechanical stress field as the multi-physical field sensing object includes: The temperature field, electric field, magnetic field and mechanical stress field jointly constitute the core physical process that is mutually related and influenced in the operation of the cable, and comprehensively reflects the working state and health degree of the cable. The cable is essentially an electric energy transmission carrier. The temperature field is the result of energy loss in electric energy transmission and external environment heat dissipation balance. Temperature abnormalities are often early signals of faults. The electric field is the basis for the existence and transmission of electric energy, and the distribution state directly determines the safety of the insulation system. The magnetic field is generated by the current and is closely related to the load size, conductor structure and shielding method. The mechanical stress field is derived from installation, thermal expansion and contraction and external environmental mechanical action, and directly affects the mechanical integrity and structural stability of the cable. The temperature field, electric field, magnetic field and mechanical stress field interact with each other. For example, overloading will increase the current, leading to an increase in the magnetic field and an increase in temperature, which in turn causes thermal expansion to generate additional mechanical stress. At the same time, high temperature will accelerate the aging of the insulation and change the electric field distribution. The multi-physical field sensing of the target cable "thermal-electric-magnetic-force" is helpful to the integrity of the target cable fault damage state evaluation and life prediction.

[0024] The coupling analysis engine module receives the sensing module data and performs real-time matching of the multi-physical field characteristics with a pre-established cable defect database through a cross-scale dynamic correlation algorithm, and outputs the target cable damage state evaluation result to the dynamic life evaluation module.

[0025] It needs to be specifically explained that the cross-scale dynamic correlation algorithm performs feature dimension reduction and parameter processing on the multi-physical field data collected by the sensing module through a cross-scale feature extraction technology to obtain a feature vector that can reflect the overall operation state of the cable, and simultaneously calls the pre-established cable defect database to perform real-time calculation of the correlation between the feature vector and the cable defect.

[0026] The cable defect database includes defect types and corresponding feature vector data, corresponding multi-physical field response characteristics and defect evolution rules. The cable defect database is dynamically updated based on experimental data and actual operation cable defect detection case data.

[0027] The damage state evaluation result includes the damage type, damage location and damage degree corresponding to the defect, and is output to the dynamic life evaluation module in a standardized data interface and hierarchical identification.

[0028] It needs to be further explained that the multi-physical field data collected by the perception module is subjected to feature dimension reduction and parameter processing, and the specific analysis methods include: The arithmetic mean of the temperature of multiple sites on the surface of the conductor is calculated, denoted as the average temperature of the conductor, the ratio of the temperature difference at different depths in the radial direction of the insulation layer to the radial distance is calculated, denoted as the temperature gradient of the insulation layer, the arithmetic mean of the temperature of multiple sites on the surface of the shielding layer is calculated, denoted as the average temperature of the shielding layer, the ratio of the voltage between the conductor and the shielding layer to the thickness of the insulation layer is calculated, denoted as the average electric field strength, the ratio of the instantaneous field strength of the target cable terminal to the average electric field strength is calculated, denoted as the electric field distortion coefficient, the vector sum of the magnetic induction intensity in x, y, and z directions of the center point on the surface of the target cable conductor is calculated, denoted as the resultant magnetic induction intensity, the ratio of the difference and the average value of the maximum and minimum of the load current in the target cable line within a fixed time period is calculated, denoted as the load current fluctuation coefficient, the ratio of the induced current of the metal shielding layer to the conductor current is calculated, denoted as the shielding current ratio, the effective value of the current on the grounding wire is measured, denoted as the grounding current value, the covariance of the instantaneous axial / radial tension of the target cable is calculated, denoted as the axial / radial tension coupling coefficient, and the ratio of the difference in instantaneous bending strain force at different axial positions of the target cable at the same time to the axial distance is calculated, denoted as the bending strain spatial distribution gradient.

[0029] The above-mentioned average temperature of the conductor, temperature gradient of the insulation layer, average temperature of the shielding layer, average electric field strength, electric field distortion coefficient, resultant magnetic induction intensity, load current fluctuation coefficient, shielding current ratio, grounding current value, axial / radial tension coupling coefficient, and bending strain spatial distribution gradient are denoted as the characteristic vector reflecting the overall operation state of the cable.

[0030] The real-time calculation of the correlation between the characteristic vector and the cable defect includes: real-time extraction and calculation of the characteristic vector of the target cable, using the K-nearest neighbor algorithm, retrieving the top 20 historical cases similar to the characteristic vector from the cable defect database, the similarity is measured by the Euclidean distance, and the distance wherein, is the feature i, is the feature i of the kth historical case, and the similarity S threshold is set to 0.85. If there is a historical case with a similarity greater than S, it is determined that the target cable has the defect type corresponding to the case, and if the similarity of all cases is lower than S, the evaluation result of 'no high-confidence defect identified' is output, and the set of characteristic vectors is marked as a sample to be analyzed, which is used for subsequent database expansion and model optimization.

[0031] The specific steps of dynamically updating the cable defect database include: synchronously collecting standardized multi-physical field data of laboratory simulated defects and operation data of actual defect cases in the field, statistically analyzing and expanding the feature vector interval of known defect types, using cluster analysis to extract unique features and define identification for new defect types, optimizing single-field physical features through correlation analysis, supplementing multi-field coupling rules, using regression models to iteratively update defect evolution parameters based on time series data, correcting biased features through a feedback mechanism after double verification of experimental samples and field cases, and completing dynamic updating of the cable defect database once every quarter.

[0032] The damage type includes common defects and other defects, the damage location is positioned based on the target cable axial radial coordinate axis, and the damage degree includes slight damage, moderate damage, and severe damage.

[0033] The feature vector of the conductor average temperature, the insulation layer temperature gradient, the shielding layer average temperature, the ground loop value, the axial radial tension coupling coefficient, and the bending strain spatial distribution gradient is used to determine whether there are common defects such as conductor overheating, insulation aging, shielding layer failure, mechanical damage, and abnormal ground system. The average electric field intensity, the electric field distortion coefficient, the synthetic magnetic induction intensity, the load current fluctuation coefficient, and the shielding loop current ratio are used to determine whether there are other defects in which a single feature vector deviates from the specified threshold.

[0034] Based on the system, the difference degree is the deviation degree of the actual measurement normal use threshold value, the defect degree determination is based on the dynamic threshold value, the slight damage refers to the difference degree exceeding the dynamic threshold value and being lower than the dynamic threshold value + 2%, the moderate damage refers to the difference degree exceeding the dynamic threshold value + 2% and being lower than the dynamic threshold value + 5%, and the severe damage refers to the difference degree exceeding the dynamic threshold value + 5%. The dynamic threshold value is adjusted according to the cable operation years, which is 3% for less than 2 years, 5% for 2-5 years, and 8% for more than 5 years.

[0035] The dynamic life assessment module, based on the damage assessment results output by the coupling analysis engine module, uses a multi-physical field coupled time series neural network architecture for life prediction, and synchronously outputs the prediction results to the digital twin interaction module.

[0036] It should be specifically noted that the input end of the time series neural network architecture adopts a multi-channel parallel structure, embeds a coupling layer, accesses a bidirectional long-term memory network layer, calculates the evolution function of the damage state over time and the time-dependent coefficient, combines the historical damage data of the cable with the life attenuation curve of similar cables for training, and outputs the residual life prediction value of the target cable.

[0037] It should be further noted that the specific analysis method of the evolution function of the damage state over time and the time-dependent coefficient is as follows: wherein correspond to the normalized damage degree of the five common defects respectively, and the value range is , 1 represents complete failure, and the evolution function is defined as: wherein, is the damage degree at the initial time , which is obtained from historical data or the first detection value, is the basic evolution rate of the defect at the time , which is output by a bidirectional LSTM layer, is a multi-physical field coupling coefficient, which is calculated by a coupling layer and reflects the acceleration / slowing effect of temperature, electric field and other parameters on evolution.

[0038] The time-dependent coefficient is defined as: wherein, is the current time, is the historical time, is a time decay factor, which is optimized by neural network training and takes 30 days, that is, the damage weight of the recent 30 days is higher.

[0039] In the bidirectional LSTM layer, the time-dependent coefficient acts on the historical data through the attention mechanism, and the formula is: wherein, is the weighted real-time evolution rate, so that the model pays more attention to the evolution trend of recent damage.

[0040] The historical data of the same type of cable are collected, including multi-physical field monitoring values, damage degree labels and actual failure time in the whole life cycle.

[0041] The digital same type of cable life attenuation curve is collected, and the reference life model is fitted: wherein, is the average damage degree, and a, b and c are curve parameters.

[0042] The historical data is used to train the time series neural network, the multi-physical field data is input, the damage evolution rule is learned by bidirectional LSTM, and the damage degree is output, and the loss function is: The life prediction head is connected, the same type of curve is fused, history, wherein, is the prediction value based on the historical data, is the weight obtained by training, and the life prediction error is used to optimize the model, the real-time monitoring data of the target cable is input, the current average damage degree and the evolution trend are calculated, and is obtained by substituting the reference model, and the final prediction value is output after being corrected in combination with the historical trend. Finally history.

[0043] The digital twin interaction module receives actual measurement data and life prediction results, compares the simulation data of the digital twin in real time, triggers a self-correction signal when the difference exceeds a dynamic threshold, and feeds the signal back to the coupled analysis engine module for model parameter optimization.

[0044] Need to be specific, model parameter optimization aims at the difference between digital twin simulation data and actual measurement data, locates the optimization of multi-physical field coupling coefficient and the feature matching threshold in cross-scale correlation algorithm, and the optimization process uses adaptive particle swarm algorithm combined with historical optimization records and parameter calibration experience of similar cables for dynamic adjustment.

[0045] Need to be further explained, the digital twin is a simulation model based on multi-physical field coupling theory, the input is the design parameters of the cable, the real-time running condition and the environmental boundary condition corrected by the environmental interference suppression module, the model parameter optimization of the digital twin interaction module is implemented according to the process of "difference positioning-adaptive optimization-experience feedback", through residual analysis and sensitivity evaluation, the multi-physical field coupling coefficient and the feature matching threshold in cross-scale dynamic correlation algorithm are locked, the residual sum of squares is taken as the target, the adaptive particle swarm algorithm is used for optimization, the initial solution space is based on the calibration experience of similar cables, the inertia weight is dynamically adjusted in iteration, the search step of stable adjustment parameter is reduced combined with historical optimization records, the optimized parameters are fed back to the coupled analysis engine, if the difference still exceeds the threshold, repeat the iteration, if the difference still exceeds the threshold for three times, trigger the self-correction signal.

[0046] The environmental interference suppression module is interconnected with the multi-physical field perception module, and the original measurement data is dynamically corrected by the environmental-physical field coupling compensation matrix, and the corrected data is input into the coupled analysis engine module.

[0047] Need to be specific, the environmental-physical field coupling compensation matrix is a mathematical model for environmental interference suppression, which quantifies the coupling influence relationship between environmental factors and each physical field in the form of a multi-dimensional matrix. The row dimension of the matrix corresponds to the data of temperature field, electric field, magnetic field and mechanical stress field, and the column dimension corresponds to environmental interference factors including temperature, humidity, air pressure and wind speed. The matrix elements are coupling coefficients calibrated by experimental and measured data, wherein the matrix coefficients are updated periodically according to the cable laying scene and seasonal changes.

[0048] It needs to be further explained that the dynamic modification of the original measurement data by the environment-physical field coupling compensation matrix is implemented as follows: first, a 12-row 4-column compensation matrix M is constructed, wherein the row vectors correspond to 12 core data of temperature field (target cable conductor surface temperature, temperature at different depths of insulation layer in radial direction, shielding layer surface temperature), electric field (target cable terminal instantaneous field strength, voltage value between conductor and shielding layer), magnetic field (magnetic induction intensity of target cable conductor surface center point, load current in cable line, induced current of metal shielding layer, ground loop current), mechanical stress field (target cable instantaneous axial / radial tension, instantaneous bending strain force) in turn, and the column vectors correspond to environmental factors temperature T, humidity H, air pressure P, and wind speed V, and the matrix elements for the first for the first for the first for the first for the first for the first for the first

[0049] Secondly, only the structures involved in the disclosed embodiments are involved in the drawings of the disclosed embodiments, other structures can be referred to the usual design, and the same embodiments and different embodiments of the present application can be combined with each other under the condition of no conflict. Finally, the above only describes the preferred embodiments of the present application, and is not used to limit the present application, and any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A cable life dynamic assessment system based on multi-physical field coupling, characterized in that, The method comprises the following steps: A multi-physical field perception module, which takes the cable to be evaluated as a target cable, collects temperature field, electric field, magnetic field and mechanical stress field data of the target cable in real time; A coupling analysis engine module, which receives the data from the perception module and matches the multi-physical field characteristics with a pre-established cable defect database in real time through a cross-scale dynamic correlation algorithm, and outputs the target cable damage state evaluation result to a dynamic life assessment module; A dynamic life assessment module, which, based on the damage evaluation result output by the coupling analysis engine, adopts a multi-physical field coupled time series neural network architecture for life prediction, and synchronizes the prediction result to a digital twin interaction module; A digital twin interaction module, which receives actual measurement data and life prediction results, compares the simulation data of the digital twin body in real time, triggers a self-correction signal when the difference exceeds a dynamic threshold, and feeds back the signal to the coupling analysis engine module for model parameter optimization; An environmental interference suppression module, which is interconnected with the multi-physical field perception module data, dynamically modifies the original measurement data through an environment-physical field coupling compensation matrix, and inputs the modified data to the coupling analysis engine module.

2. The multi-physics coupling based cable life dynamic assessment system according to claim 1, wherein: The multi-physical field perception module comprises a distributed optical fiber temperature sensor array, a capacitive electric field sensor group and a magnetostrictive stress sensor, wherein the distributed optical fiber temperature sensor is arranged in a spiral winding manner on the surface of the cable conductor, the capacitive electric field sensor is embedded in the cable insulation shielding layer in the form of an orthogonal array, and the magnetostrictive stress sensor is arranged at equal intervals along the axial direction of the cable.

3. The multi-physics coupling based cable life dynamic assessment system according to claim 1, wherein: The temperature field data includes the surface temperature of the target cable conductor, the temperature at different depths of the insulation layer, and the surface temperature of the shielding layer. The electric field data includes the instantaneous field strength of the target cable terminal, the voltage value between the conductor and the shielding layer. The magnetic field data includes the magnetic induction intensity of the center point of the target cable conductor surface, the load current in the cable line, the induced current of the metal shielding layer, and the ground circulating current. The mechanical stress field data includes the instantaneous axial / radial tension of the target cable and the instantaneous bending strain force.

4. The multi-physics coupling based cable life dynamic assessment system according to claim 1, wherein: The cross-scale dynamic correlation algorithm uses cross-scale feature extraction technology to reduce the dimensionality of the multi-physical field data collected by the perception module and process the parameters to obtain a feature vector that reflects the overall operating state of the cable. At the same time, the pre-established cable defect database is called to calculate the correlation between the feature vector and the cable defects in real time.

5. The multi-physics coupling based cable life dynamic assessment system according to claim 1, wherein: The cable defect database includes defect types and corresponding feature vector data, corresponding multi-physical field response characteristics and defect evolution rules, and is dynamically updated based on experimental data and actual operating cable defect detection case data.

6. The multi-physics coupling based cable life dynamic assessment system according to claim 1, wherein: The damage state evaluation result includes the damage type, damage location and damage degree corresponding to the defect, and is output to the dynamic life assessment module in a standardized data interface and hierarchical identification.

7. The multi-physics coupling based cable life dynamic assessment system according to claim 1, wherein: The input end of the time series neural network architecture adopts a multi-channel parallel structure, embeds a coupling layer, accesses a bidirectional long-term memory network layer, calculates the evolution function and time-dependent coefficient of the damage state over time, and combines the historical damage data of the cable with the life attenuation curve of the same type of cable to train and output the remaining life prediction value of the target cable.

8. The multi-physics coupling based cable life dynamic assessment system of claim 1, wherein: The model parameter optimization is aimed at the difference between the digital twin simulation data and the actual measurement data, and is used for locating the optimization of the multi-physical field coupling coefficient and the feature matching threshold in the cross-scale correlation algorithm, and the optimization process adopts an adaptive particle swarm algorithm and dynamically adjusts the optimization process by combining historical optimization records and parameter calibration experience of similar cables.

9. The multi-physics coupling based cable life dynamic assessment system of claim 1, wherein: The environment-physical field coupling compensation matrix is a mathematical model for environment interference suppression, quantifies the coupling influence relationship between environmental factors and various physical fields in the form of a multi-dimensional matrix, the row dimension of the matrix corresponds to various data of the temperature field, the electric field, the magnetic field and the mechanical stress field, the column dimension corresponds to environmental interference factors including temperature, humidity, air pressure and wind speed, and the matrix elements are coupling coefficients calibrated through experiments and measured data, wherein the matrix coefficients are periodically updated according to the cable laying scene and seasonal changes.

Images