Cable health state prediction method and system based on dynamic risk assessment

Abstract

The invention relates to the technical field of cable health state prediction, in particular to a cable health state prediction method and system based on dynamic risk assessment, and the method and system comprehensively quantify the risk demand of a target cable and scientifically mark the risk state prediction cable by combining the aging, operation load, external environment and scene adaptation characteristics of the cable. The system preferentially focuses on a high-risk cable, sorts node association of each state dimension, compares risk conductivity with a corresponding threshold value, screens an effective risk state guide edge, forms a health state influence link according to one-way conduction characteristics in series, visualizes risk transmission logic, and solves the problem that risk roots are difficult to trace in traditional assessment. And a target end point node is locked to predict the health state by quantifying the link risk intensity. And in combination with the node health index and the fault probability model, the prediction result is scientific and instructive, a refined decision basis is provided for cable operation and maintenance, and the health management efficiency and accuracy are effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cable health state prediction, more particularly, it relates to a cable health state prediction method and system based on dynamic risk assessment. BACKGROUND

[0002] In the field of cable health state prediction, the traditional evaluation method often has the limitation of single risk consideration dimension, and only focuses on a single state indicator or influencing factor of the cable, which is difficult to comprehensively capture the complex composition and dynamic changes of the cable risk, resulting in inaccurate quantification of the risk demand of the target cable, and unable to provide a reliable preliminary basis for subsequent health state prediction. The existing technology has obvious shortcomings in the analysis of cable risk transmission logic, and lacks systematic construction and analysis of the correlation between nodes of each state dimension of the cable. The traditional method is difficult to identify the transmission path of the risk from the source node to the deterioration node, and cannot convert the abstract risk transmission process into a structured analysis object, which further leads to difficulty in tracing the root cause of the risk, and the subsequent health evaluation and intervention measures lack clear targeting, making it difficult to accurately solve the potential core risk problem of the cable.

[0003] The traditional cable health state prediction mechanism lacks scientific priority ranking and comprehensive dynamic evaluation logic, either failing to highlight the disposal priority of high-risk cables, resulting in unreasonable resource allocation, or ignoring the long-term cumulative risk of low-risk cables, which is easy to cause evaluation omission. At the same time, the prediction results are mainly single state judgments, which lack systematization and guidance closely combined with operation and maintenance practice, and are difficult to directly provide fine basis for cable operation and maintenance decision-making, affecting the overall efficiency and accuracy of cable health management. SUMMARY

[0004] In view of the deficiencies of the prior art, the purpose of the present application is to provide a cable health state prediction method and system based on dynamic risk assessment.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0006] The cable health state prediction method based on dynamic risk assessment, the method steps are as follows:

[0007] Step 1: Determine all target cables in the scene, periodically generate dynamic risk demand indexes of each target cable, and mark the corresponding risk state prediction cable based on the comparison result of the dynamic risk demand index and the dynamic risk demand threshold;

[0008] Step two: generate a health state prediction demand ranking queue, select the risk state prediction cable ranked first in the health state prediction demand ranking queue, generate a state node topology network model of the risk state prediction cable, obtain the risk conduction degree of each relationship conduction edge in the cable target topology network model, mark the corresponding risk state guide edge based on the comparison result of the risk conduction degree and the risk conduction degree threshold, and form multiple cable health state influence links in turn according to the direction of the risk state guide edge.

[0009] Step three: obtain the risk intensity of each cable health state influence link, select the target terminal node of each cable health state influence link, and then perform cable health state prediction.

[0010] Further, the dynamic risk demand index of the target cable is generated in the following manner: select a target cable, obtain the aging acceleration factor , the load stress coefficient , the environmental coupling coefficient and the scenario correction coefficient of the target cable, calculate the dynamic risk demand index of the target cable through the formula .

[0011] Further, the scenario correction coefficient of the target cable is the voltage level correction coefficient; the functional importance coefficient.

[0012] Further, the generation steps of the health state prediction demand ranking queue are as follows: all risk state prediction cables are sorted in descending order of the value of the dynamic risk demand index to generate a health state prediction demand ranking queue.

[0013] Further, the generation method of the state node topology network model of the risk state prediction cable is as follows: obtain each state dimension node of the risk state prediction cable, obtain the risk conduction degree of each relationship conduction edge, and form a state node topology network model of the risk state prediction cable according to the state dimension node and the relationship conduction edge.

[0014] Further, the cable health state influence link risk intensity acquisition manner: selecting a cable health state influence link, acquiring the risk conduction degree of each risk state guide edge in the cable health state influence link, synchronously acquiring the relationship type of each risk state guide edge, determining the type weight of each risk state guide edge based on the relationship type, performing product calculation on the risk conduction degree of each risk state guide edge and the corresponding type weight, obtaining the influence intensity of each risk state guide edge, and performing sum mean calculation on the influence intensity of each risk state guide edge to obtain the risk intensity of the cable health state influence link.

[0015] Further, the relationship type includes a heat conduction relationship, an electrical conduction relationship, and an environmental conduction relationship. When the relationship type is the heat conduction relationship, the corresponding risk conduction degree is ; is the actual value of the source state dimension node in the relationship conduction edge; is the rated reference value of the source state dimension node in the relationship conduction edge; is the heat conduction efficiency coefficient.

[0016] Further, when the relationship type is the electrical conduction relationship, the corresponding risk conduction degree is ; is the electrical safety threshold of the source state dimension node in the relationship conduction edge; is the electrical conduction sensitivity coefficient.

[0017] Further, when the relationship type is the environmental conduction relationship, the corresponding risk conduction degree is ; is the non-erosion reference value of the source state dimension node; is the environmental conduction efficiency coefficient.

[0018] Further, the cable health state prediction system based on dynamic risk assessment includes a dynamic risk assessment module, a health influence link generation module, a health influence link quantification module, and a health state prediction module.

[0019] The dynamic risk assessment module is used to determine the risk state prediction cable in the scene.

[0020] The health influence link generation module is used to form a plurality of cable health state influence links according to the risk state prediction cable.

[0021] The health influence link quantification module is used to acquire the risk intensity of each cable health state influence link and select a target terminal node of each cable health state influence link.

[0022] The health state prediction module is used to perform cable health state prediction.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] The system and method of this invention comprehensively and accurately quantifies the risk requirements of target cables by considering their own aging state, operating load pressure, external environmental influences, and scenario adaptability. This allows for the scientific labeling of cables with risk status predictions. This invention prioritizes cables with the highest risk levels, systematically analyzes the relationships between nodes in each state dimension of the cable, and precisely identifies effective risk status guiding edges by comparing and filtering risk transmission degree with corresponding type thresholds. Based on the unidirectional transmission characteristics of the guiding edges, a cable health status influence link is formed, transforming the abstract risk transmission process into a structured and visualized path presentation. This clearly reveals the transmission logic of risk from the source node to the deterioration node, breaking through the bottleneck of traditional health assessments that make it difficult to trace the root cause of risks, and providing a clear path for subsequent precise intervention.

[0025] By quantifying the risk intensity of each cable's health status affecting the link, the target endpoint node is identified for health status prediction. Subsequently, cables with predicted risk status in the sorting queue are evaluated in turn, forming a dynamic prediction mechanism that prioritizes key areas and provides comprehensive coverage. This mechanism highlights the priority of handling high-risk cables while avoiding omissions in the evaluation of low-risk cables. Combining the node health status index with the fault probability prediction model, the prediction results are both scientific and instructive, providing a systematic and refined decision-making basis for cable operation and maintenance, and effectively improving the efficiency and accuracy of cable health management. Attached Figure Description

[0026] Fig. 1 The flowchart shows the cable health status prediction method based on dynamic risk assessment.

[0027] Fig. 2 Flowchart for generating the state node topology network model of a cable for risk state prediction;

[0028] Fig. 3 A flowchart for predicting cable health status. Detailed Implementation

[0029] Example 1: Refer to Figs. 1 to 3 A method and system for predicting cable health status based on dynamic risk assessment are presented, and the method steps are as follows:

[0030] Step one: determine all target cables in the scene (for example: in a certain city 10kV distribution network, covering 12 power transmission cables and 5 distribution cables in 3 urban areas, these cables are all target cables), periodically generate the dynamic risk demand index of each target cable (periodically corresponding to the time interval, refer to the historical data of the target cable's past dynamic risk demand index in the scene, according to the risk state prediction cable's occurrence frequency and cumulative duration, comprehensive setting), when the dynamic risk demand index of the target cable is higher than the dynamic risk demand threshold (the dynamic risk demand threshold is based on the historical data of the dynamic risk demand index of the target cable in the scene, combined with the risk state prediction cable's occurrence frequency, comprehensive setting), the corresponding target cable is marked as a risk state prediction cable.

[0031] The dynamic risk demand index of the target cable is generated as follows: select a target cable, obtain the aging acceleration factor , load stress coefficient , environmental coupling coefficient and scenario correction coefficient , calculate the dynamic risk demand index of the target cable by formula . .

[0032] The aging acceleration factor of the target cable ; is the average temperature of the conductor in the target cable in the last period (calculated according to the optical fiber grating temperature sensor acquisition), is the critical temperature for safe operation of the cable (factory setting, different materials of the cable are not the same) ; is the insulation aging degree , wherein is the current value of the main insulation resistance of the target cable, collected by the insulation resistance monitor; is the initial value of the main insulation resistance of the target cable, factory setting) ; is the actual contact resistance of the target cable joint (collected by the wireless joint monitoring terminal) ; is the standard contact resistance of the target cable joint (factory setting) ; The target cable's sheath damage rate (i.e., the ratio of the damaged area to the total monitored area) is calculated as follows: Sheath damage detectors (common types: electromagnetic induction, ultrasonic, infrared imaging) are installed on key monitoring sections of the cable sheath surface (such as near joints or underground directly buried sections prone to corrosion), with a preset total coverage area. The detectors capture sheath damage signals using specific principles (e.g., electromagnetic induction detects changes in the electric field caused by insulation integrity failure, ultrasonic detects abnormal sound wave reflection at the damage site). After confirming damage, a multi-sensor array or scanning path is used to locate the outline boundary of the damaged area (e.g., identifying signal abrupt changes at the damage edge to delineate the approximate shape of irregular damage). Damage area quantification: Based on the located damage outline, the area is quantified using a preset algorithm: For regular shapes (circular, rectangular): key dimensions of the outline (such as diameter, length × width) are directly measured and substituted into geometric formulas for calculation; the "pixel segmentation method" (for infrared imaging detectors) or "segmentation approximation method" is used to break down the damaged area into multiple simple geometric shapes. The total area is obtained by summing the data of the various graphs. Data correction: detector errors (such as false signals caused by environmental interference) are eliminated, and the final true value of the damaged area is output. Data transmission and synchronization: the collected damaged area data is transmitted to the system platform wirelessly / wired, and associated with the total monitored area of ​​the cable to automatically calculate the sheath damage rate, providing parameters for subsequent calculation of aging acceleration factors. Among them, k1, k2, and k3 are weighting coefficients, k1+k2+k3=1. Since thermal aging is the core driving factor of cable body deterioration, the continuous high conductor temperature will directly accelerate the thermal-oxidative aging of the insulation material and reduce the mechanical properties of the material, which is one of the most common causes of cable failure. Therefore, it is given the highest weight. Electrical aging (insulation deterioration) is related to the electrical safety of the cable, and mechanical aging (poor joint contact, sheath damage) is related to the structural integrity of the cable. The contribution of the two to cable aging is similar, and the degree of influence is weaker than that of thermal aging. Therefore, they are assigned equal second-highest weights. Therefore, the value of k1 can be 0.4, the value of k2 can be 0.3, and the value of k3 can be 0.3.

[0033] target cable load stress coefficient ; The maximum value of the target cable's operating load in the previous cycle (obtained through the power system SCADA system). The rated load of the target cable (i.e., the maximum allowable load for safe operation of the target cable, set at the factory). The standard deviation of the target cable's operating load in the previous cycle; This refers to the number of times overvoltage occurred in the target cable during the previous cycle (overvoltage is defined as a situation where the actual operating voltage exceeds 1.1 times the cable's rated voltage, collected and statistically analyzed in real time using an overvoltage recorder); when When the load is low, take 1.0 (to avoid the coefficient being too small at low loads). is the ratio of the standard deviation of load fluctuation to the rated load, reflecting the degree of instability of the load; multiplied by 2 is the stress contribution weight of amplifying the load fluctuation, frequent load fluctuation will cause the cable to repeatedly bear the mechanical stress of thermal expansion and cold contraction, accelerating the aging of insulation and joints, so the coefficient 2 is used to strengthen this influence; is the number of overvoltages, overvoltage will exert additional electrical stress on the cable insulation, multiplied by 0.1 is the stress contribution weight of single overvoltage, the more the number of overvoltages, the stronger the cumulative electrical stress, through the coefficient of 0.1 to realize the quantization logic of "the more the number, the greater the stress coefficient", which is consistent with the cumulative degradation effect of overvoltage.

[0034] The environmental coupling coefficient of the target cable ; is the average relative humidity of the environment in which the target cable is located in the last period (obtained by periodically collecting and calculating the temperature and humidity sensor), is the soil corrosion concentration (acquisition method: bury the soil corrosion sensor (common ion selective electrode sensor) in the soil around the cable sheath (close to the cable laying position, to ensure that the collected data is the soil data of the actual environment of the cable), the sensor is in direct contact with the soil medium and senses the concentration of target corrosion ions in real time. The single corrosion ion concentration acquisition sensor is aimed at the core medium in the soil that has the strongest corrosion effect on the cable sheath (metal armor, corrosion protection layer) (mainly chloride ions and sulfate ions, the real-time concentration values of which are collected, for example: the collected chloride ion concentration = 0.5 mg / cm 3 ; the collected sulfate ion concentration = 0.3 mg / cm 3 . Due to the difference in corrosion activity of different ions (such as the corrosion rate of chloride ions is much higher than that of sulfate ions), it is necessary to assign a corrosion activity weight coefficient to each ion (because the chloride ion has a small radius and strong penetration, it can easily penetrate into the micropores of the corrosion protection layer of the cable sheath and react with the metal armor (such as steel armor) to form a pitting (local corrosion pit) - this type of corrosion develops fast and is destructive, easily penetrates the sheath, and causes water and impurities to invade the cable body, while the corrosion rate of sulfate ions is slow and tends to form a stable corrosion product (such as a sulfate layer) with the metal, which is much weaker than that of chloride ions in terms of damage degree and diffusion speed, usually the weight of chloride ions , the weight of sulfate ions : ); is the environmental temperature fluctuation value of the target cable in the last period (the difference between the maximum and minimum values of the environmental temperature of the target cable in the last period, which is calculated after collecting data by the temperature and humidity sensor); The laying environment correction coefficient (the coupling correction coefficient adjusted according to the difference of cable laying scenes, reflecting the superposition effect of environmental erosion in different scenes (such as more complex corrosion medium in chemical industry park, higher correction coefficient), the value standard example: underground pipe network = 1.0, chemical industry park = 1.3, high-altitude erection = 0.9, tunnel = 1.1); Coefficients 0.008, 0.2, 0.01: the influence weight of humidity, corrosion concentration, and temperature fluctuation (obtained by fitting through multi-scene environmental erosion experiment); Corrosion concentration is the core dominant factor of environmental erosion - chloride ions and sulfate ions directly cause irreversible chemical corrosion with cable sheath and metal armor, destroy the integrity of the protection structure, and the corrosion rate increases linearly with the increase of concentration (experiments show: Every increase of 1 mg / cm³, the cable corrosion rate increases about 20%). Therefore, the highest weight 0.2 is given to highlight its "decisive contribution" to the comprehensive erosion. Humidity is a corrosion accelerating factor rather than a direct erosion factor. In dry environment, corrosion almost stops, and in humid environment, corrosion ions migrate and electrochemical reactions occur, but the influence of humidity is "indirect and moderate": even if the humidity reaches 90% RH, the contribution is only 0.72 after multiplying by 0.008, which is much lower than the influence of corrosion concentration (such as 0.1 contribution), which is consistent with the logic that "humidity assists corrosion, not dominant erosion" in reality. Temperature fluctuation indirectly affects through physical fatigue, thermal expansion and contraction cause sheath cracking and joint seal failure, but this influence is a long-term cumulative physical fatigue, and the erosion contribution of single fluctuation is weak (experiments show: every increase of 10℃ of temperature fluctuation, the cable sheath cracking risk only increases about 10%). Therefore, the minimum weight 0.01 is given to quantify its "slow accumulation, auxiliary erosion" characteristics and avoid overestimation.

[0035] Scenario correction coefficient of target cable ; Voltage level correction coefficient (different voltage level correction coefficients correspond to different voltage levels of target cable, if the voltage level of target cable is 10kV, the voltage level correction coefficient is 1.0, 35kV = 1.1, 110kV = 1.2, 220kV = 1.3); Function importance coefficient (different function importance coefficients correspond to different function importance of target cable, for example: main cable = 1.2, branch cable = 1.0, backup cable = 0.8).

[0036] Step two: sort all risk state prediction cables in descending order according to the value of dynamic risk demand index, generate a health state prediction demand sorting queue, select the risk state prediction cable at the top of the health state prediction demand sorting queue, generate the state node topology network model of the risk state prediction cable, obtain the risk conduction degree of each relationship conduction edge in the cable target topology network model, when the risk conduction degree is higher than the risk conduction degree threshold value (the risk conduction degree threshold value of the relationship conduction edge of different relationship types is not the same, the risk conduction degree threshold value of the relationship conduction edge is set based on the actual conduction situation in history, for example, when the conduction degree is 0.5, the probability of temperature exceeding standard failure caused by heat conduction is 70%, the risk conduction degree threshold value of the heat conduction relationship can be set to 0.5), mark the corresponding relationship conduction edge as a risk state guide edge (the length of the risk state guide edge is set according to the difference ΔD between the risk conduction degree and the risk conduction degree threshold value, for example: the basic rule is that the reference length is L0 (such as the reference length in the topology graph is 1 cm), the length of the length risk state guide edge=L0*(1+k*ΔD) (k is the proportionality coefficient, which is uniformly taken as 2 to ensure that the length change is controllable)), mark the corresponding relationship conduction edge as a risk state guide edge, and sequentially connect according to the direction of the risk state guide edge to form multiple cable health state influence links.

[0037] According to the direction of the risk state guide edge, multiple cable health state influence links are sequentially connected, for example: the scene contains the following risk state guide edges: 1, load peak value→average temperature; 2, overvoltage times→insulation aging degree; 3, contact resistance→insulation aging degree; 4, soil corrosion concentration→sheath damage rate; 5, average relative humidity→sheath damage rate; 6, sheath damage rate→insulation aging degree; 7, average temperature→insulation aging degree; then 3 cable health state influence links are generated, link 1: soil corrosion concentration→sheath damage rate→insulation aging degree; link 2: average relative humidity→sheath damage rate→insulation aging degree; link 3: load peak value→average temperature→insulation aging degree.

[0038] The state node topology network model of the risk state prediction cable is generated in the following manner: obtaining each state dimension node (the state dimension nodes include the following: average temperature node, insulation aging degree node, contact resistance node (corresponding to the actual contact resistance of the target cable joint), sheath damage rate node, load peak node (corresponding to the maximum value of the operating load of the target cable in the last cycle), overvoltage frequency node (corresponding to the number of overvoltage occurrences of the target cable in the last cycle), average relative humidity node (corresponding to the average relative humidity of the environment in which the target cable is located in the last cycle), and soil corrosion concentration node) of the risk state prediction cable, taking the association relationship between each state dimension node as a relationship transmission edge, and explicitly defining the direction of each relationship transmission edge (the direction of the relationship transmission edge is unidirectional; for example, the risk source of the heat conduction relationship is the active heat generation node, the target node is the passive heat absorption node, the heat transfer has a clear direction from the heat generation end to the heat receiving end, and it is irreversible; for example, the risk source of the electrical conduction relationship is the electrical stress application node, and the target node is the electrical deterioration bearing node. The deterioration process caused by electrical stress is irreversible, and there is no reverse influence; the risk source of the environmental conduction relationship is the external environmental factor node, and the target node is the cable state node. The environment factor is independent of the cable, and the cable state will not change the external environment in the opposite direction), obtaining the risk transmission degree of each relationship transmission edge, and forming the state node topology network model of the risk state prediction cable according to the state dimension node and the relationship transmission edge.

[0039] The risk transmission degree of the relationship transmission edge is obtained in the following manner: determining the relationship type of the relationship transmission edge, the relationship type including a heat conduction relationship, an electrical conduction relationship, and an environmental conduction relationship; when the relationship type is a heat conduction relationship, the corresponding risk transmission degree ; is the actual value of the source state dimension node in the relationship transmission edge (the source state dimension node is the state dimension node at the starting end of the relationship transmission arrow, and the actual value is, for example, the contact resistance , load peak ); is the rated reference value of the source state dimension node in the relationship transmission edge (for example, the contact resistance standard value , rated load ); is the heat conduction efficiency coefficient (determined according to the core heat conduction material of the target cable, if the core heat conduction material of the target cable is copper, the value of may be 0.9; if the core heat conduction material of the target cable is aluminum, the value of may be 0.7; and if the core heat conduction material of the target cable is steel, the value of may be 0.4).

[0040] When the relationship type is an electrical conduction relationship, the corresponding risk transmission degree ; risk transmission degree in the formula of may be the number of overvoltage occurrences , contact resistance ; is the electrical safety threshold of the source state dimension node in the relationship transmission edge (definition of the electrical safety threshold: the maximum allowed value of the source state dimension node to ensure its own safe operation and not to cause the target node (such as the insulation aging degree node, ) to produce electrical degradation within a preset time period (consistent with the dynamic risk assessment period); for example, the electrical safety threshold corresponding to the number of overvoltage occurrences can be set to 3 times; the electrical safety threshold corresponding to the contact resistance can be set to 60 mΩ); is the electrical transmission sensitivity coefficient (the formula is as follows: , is the basic sensitivity value, which is taken from the source-target node electrical degradation coupling strength table , such as overvoltage → insulation aging 0.9; the index steps of the source-target node electrical degradation coupling strength table are as follows: screening effective “source-target” combinations to only keep node pairs with electrical conduction relationship (excluding thermal / environmental conduction combinations), and combing all combinations actually existing electrical degradation coupling (such as “overvoltage number → insulation aging degree”, “contact resistance → insulation aging degree”, etc., excluding combinations without electrical association such as “load peak value → soil corrosion concentration”). Step 2: defining the coupling strength grading standard according to “directness / impact degree of source node anomaly on target node degradation”, which is divided into 3-4 levels, corresponding to value interval ([0.7, 0.95] range): 1, coupling strength level: extremely strong, core determination basis: source node anomaly directly causes irreversible degradation of target node, 0.90~0.95; 2, coupling strength level: strong; core determination basis: source node anomaly indirectly leads to accelerated degradation of target node; 0.80~0.89; 3, coupling strength level: medium; core determination basis: source node anomaly has an auxiliary effect on target node degradation; 0.70~0.79; according to cable aging experiment data and engineering failure statistics, each “source-target” combination is assigned a value: experimental data support: through accelerated aging experiments, the correlation between source node parameter changes and target node degradation rate is measured (the higher the correlation coefficient, the stronger the coupling); engineering case support: statistics of the occurrence probability of “source node anomaly → target node failure” in historical failures (the higher the probability, the stronger the coupling), and finally, the source-target node electrical degradation coupling strength table is prepared); the meaning of 0.3 in the formula: The weighting coefficient for (voltage level) represents that "the voltage level has a higher priority in influencing electrical conduction sensitivity"—the voltage level directly determines the electric field strength that the cable withstands, and is a core driving factor for electrical degradation coupling (e.g., the conduction effect of source node anomalies on target nodes is more significant under high voltage). Therefore, it is assigned a higher weight (0.3) to strengthen its influence on electrical conduction sensitivity. The correction strength. The meaning of 0.2 in the formula: The weighting coefficient of (laying environment) represents that "the impact of the laying environment on electrical conductivity sensitivity is of secondary priority" - the environment (such as chemical industrial parks and underground pipe networks) indirectly affects the insulation performance of cables through corrosion and humidity, and thus affects electrical conductivity efficiency. However, it is an auxiliary influencing factor (not a core driver), so it is assigned a low weight (0.2) to avoid over-amplifying the indirect effect of the environment.

[0041] When the relationship type is an environment transmission relationship, the corresponding risk transmission degree is... Risk transmission degree In the formula It can be the average relative humidity Soil corrosion concentration ; The non-erosion baseline value for the source state dimension node (definition of non-erosion baseline value: the lowest / benchmark parameter value where the source state dimension node has no erosive effect on the cable state dimension node and does not cause cable degradation; national standards / industry consensus should be given priority, and experimental data should be used when there is no standard; if the average relative humidity node is the source state dimension node) The value is taken as 40%RH, based on the national standard GB / T 2423.4, which defines 40%RH as the critical humidity for cable corrosion resistance. If the average relative humidity node is the soil corrosion concentration node, The value is set at 0 mg / cm³, based on the principle that there is no corrosion when there are no corrosive ions, serving as an absolute benchmark. The environmental conduction efficiency coefficient (defined as a quantitative indicator of the efficiency of erosion transfer from source state nodes (environmental factors, such as humidity and corrosion concentration) to target nodes (cable conditions, such as sheath damage rate and insulation aging)) is defined as follows: Effective environmental conduction combinations are selected by retaining only unidirectional combinations of "environmental factors → cable conditions" (e.g., "soil corrosion concentration → sheath damage rate"), excluding combinations without erosion correlation (e.g., "peak load → soil corrosion concentration"). A baseline value is determined. The lookup table method classifies the corrosion of the source node on the target node according to the "directness / degree of impact of the source node corrosion on the target node deterioration". Based on national standards, cable corrosion test data, and engineering fault statistics, a standardized comparison table is formed: 1. Coupling strength level: High, judgment basis: environmental factors directly cause cable material corrosion / structural damage. Value interval 0.85~0.95; 2, coupling strength level: medium, judgment basis: environmental factors indirectly accelerate cable degradation, Value interval 0.65~0.84; 3, coupling strength level: low, judgment basis: environmental factors only play a slight auxiliary role in cable degradation, Value interval 0.50~0.64; the micro-adjusted reference value defined in the technical scheme The formula is unified as follows: Weight 0.2: the influence of environmental scene on conduction efficiency is auxiliary (lower than the core role of erosion coupling strength); min function: limit the maximum value to not more than 0.95, to avoid over amplifying the conduction efficiency).

[0042] Step three: obtain the risk strength of each cable health status influence link, select the target end node of each cable health status influence link (the definition of the target end node: in the cable health status influence link, the node at the end of the one-way conduction path without subsequent risk state guiding edges pointing to it), and then perform cable health status prediction, and then sequentially perform cable health status prediction on the subsequent risk state prediction cables in the health status prediction demand sorting queue.

[0043] The risk strength of the cable health status influence link is obtained as follows: select a cable health status influence link, obtain the risk conduction degree of each risk state guiding edge in the cable health status influence link, and simultaneously obtain the relationship type of each risk state guiding edge, determine the type weight of each risk state guiding edge based on the relationship type (if the relationship type is electrical conduction relationship, the type weight can be 1.2, and the value is assigned as follows: directly causing irreversible failure such as insulation breakdown, with the greatest harm; if the relationship type is thermal conduction relationship, the type weight can be 1.0, and the value is assigned as follows: accelerating thermal aging, with relatively direct influence; if the relationship type is environmental conduction relationship, the type weight can be 0.8, and the value is assigned as follows: requiring long-term accumulation to cause significant degradation, with the least harm), multiplying the risk conduction degree of each risk state guiding edge and the corresponding type weight to obtain the influence strength of each risk state guiding edge (influence strength=risk conduction degree*type weight), and performing sum mean calculation on the influence strength of each risk state guiding edge to obtain the risk strength of the cable health status influence link;

[0044] The steps of cable health status prediction are as follows: if multiple cable health status influence links point to the same target end node (such as all pointing to "insulation aging degree"), the comprehensive link risk of the node needs to be integrated to avoid repeated calculation; the integration rule is as follows: let the target end node be N, and there are m cable health status influence links associated with N, and the risk strength of each cable health status influence link is , , , the integrated link risk intensity of node N (maximum value, highlighting the dominant role of the most dangerous driving link); if only 1 cable health state influence link is associated, then . Example: insulation aging degree node is associated with 2 cable health state influence links, with risk intensities of 0.85 and 0.72 respectively, then . Taking the integrated link risk intensity as the core, combined with the node's own parameter abnormality, the node health state is quantified; input parameters: integrated link risk intensity ; scene correction coefficient ; HSI calculation model: ; constraints: , the closer the value is to 1, the better the node health state. Modeling method: trend extrapolation method. Input feature sequence (time step = 3 periods): core features: node HSI sequence ( , , ), integrated link risk intensity sequence ( , , ). Prediction output: next period node health level: based on HSI sequence trend prediction , mapped to the corresponding health level; short-term failure probability (logic: the lower the HSI, the higher the link risk, the greater the failure probability; 0.8 is the normalization coefficient, to ensure that HSI = 0 failure probability ≈ 1.0). Node health level division table, for example: 1, node health level: excellent, range: [0.8, 1.0]; 2, node health level: good, range: [0.6, 0.8); 3, node health level: medium, range: [0.4, 0.6); 4, node health level: poor, range: [0.2, 0.4); 5, node health level: dangerous, range: [0, 0.2).

[0045] Example two: a cable health state prediction system based on dynamic risk assessment, including a dynamic risk assessment module, a health influence link generation module, a health influence link quantification module, and a health state prediction module;

[0046] The dynamic risk assessment module is used to determine the risk state prediction cable in the scene;

[0047] The health influence link generation module is used to form multiple cable health state influence links according to the risk state prediction cable;

[0048] The health influence link quantification module is configured to obtain the risk intensity of each cable health state influence link and select a target end node of each cable health state influence link.

[0049] The health state prediction module is configured to perform cable health state prediction.

[0050] The above formulas are all dimensionless values, and the preset parameters in the formulas are set by a person skilled in the art according to actual conditions.

[0051] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through a wired or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0052] It should be understood that the size of the sequence number of each process described above in various embodiments of the present application does not mean the order of execution, and the execution order of each process should be determined by its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0053] Those skilled in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. A person 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 the present application.

[0054] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0055] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are merely schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or components shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0056] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part of the prior art or the part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0057] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for cable health state prediction based on dynamic risk assessment, characterized in that, The steps are as follows: Step 1: Identify all target cables in the scenario, periodically generate the dynamic risk demand index for each target cable, and mark the corresponding risk status prediction cable based on the comparison results between the dynamic risk demand index and the dynamic risk demand threshold. Step 2: Generate a health status prediction demand ranking queue, select the first-ranked risk status prediction cable in the health status prediction demand ranking queue, generate the state node topology network model of the risk status prediction cable, obtain the risk transmission degree of each relation transmission edge in the target cable topology network model, mark the corresponding risk status guiding edge based on the comparison result of the risk transmission degree and the risk transmission degree threshold, and connect them sequentially according to the direction of the risk status guiding edge to form multiple cable health status influence links. Step 3: Obtain the risk intensity of each cable's health status affecting the link, select the target endpoint node of each cable's health status affecting the link, and then perform cable health status prediction. Subsequently, perform cable health status prediction on the subsequent risk status prediction cables in the health status prediction demand sorting queue.

2. The cable health status prediction method based on dynamic risk assessment according to claim 1, characterized in that, The dynamic risk demand index of the target cable is generated in the following manner: selecting a target cable, obtaining the aging acceleration factor of the target cable , the load stress coefficient , the environmental coupling coefficient , and the scenario correction coefficient , and calculating the dynamic risk demand index of the target cable through the formula . ​ 3. The cable health status prediction method based on dynamic risk assessment according to claim 2, characterized in that, Scene-based correction factors for target cable ; are voltage level correction factors; are function importance factors.

4. The method for cable health condition prediction based on dynamic risk assessment according to claim 1, characterized in that, The steps for generating the health status prediction demand ranking queue are as follows: Sort all risk status prediction cables in descending order of dynamic risk demand index value to generate the health status prediction demand ranking queue.

5. The method for cable health condition prediction based on dynamic risk assessment as claimed in claim 1, wherein, The state node topology network model of the risk state prediction cable is generated as follows: obtain the state dimension nodes of the risk state prediction cable, obtain the risk transmission degree of each relation transmission edge, and form the state node topology network model of the risk state prediction cable based on the state dimension nodes and relation transmission edges.

6. The method for cable health condition prediction based on dynamic risk assessment as claimed in claim 1, wherein, Method for obtaining the risk intensity of cable health status affecting the link: Select a link affected by cable health status, obtain the risk transmission degree of each risk state guiding edge in the link affected by cable health status, simultaneously obtain the relationship type of each risk state guiding edge, determine the type weight of each risk state guiding edge based on the relationship type, multiply the risk transmission degree of each risk state guiding edge with the corresponding type weight to obtain the influence intensity of each risk state guiding edge, and sum and average the influence intensity of each risk state guiding edge to calculate the risk intensity of the link affected by the cable health status.

7. The cable health status prediction method based on dynamic risk assessment according to claim 6, characterized in that, The relationship type includes a heat conduction relationship, an electricity conduction relationship, and an environment conduction relationship. When the relationship type is the heat conduction relationship, the corresponding risk conduction degree ; is an actual value of the source state dimension node in the relationship conduction edge; is a rated reference value of the source state dimension node in the relationship conduction edge; is a heat conduction efficiency coefficient.

8. The cable health status prediction method based on dynamic risk assessment according to claim 7, characterized in that, When the relationship type is an electrical conduction relationship, the corresponding risk conduction degree ; is an electrical safety threshold for the source state dimension node in the relationship conduction edge; is an electrical conduction sensitivity coefficient.

9. The cable health status prediction method based on dynamic risk assessment according to claim 7, characterized in that, When the relationship type is an environmental conduction relationship, the corresponding risk conduction degree ; The non-erosion reference value of the source state dimension node; The environmental conduction efficiency coefficient.

10. A system for cable health condition prediction based on dynamic risk assessment, applied to the method for cable health condition prediction based on dynamic risk assessment according to any one of claims 1-9, characterized in that, It includes a dynamic risk assessment module, a health impact link generation module, a health impact link quantification module, and a health status prediction module; The dynamic risk assessment module is used to determine the risk status prediction cable within the scenario; The health impact link generation module is used to predict the health status of cables and form multiple cable health status impact links based on the risk status of the cable. The Health Impact Link Quantification Module is used to obtain the risk intensity of the impact of each cable's health status on the link and to select the target endpoint node of each cable's health status impact on the link. The health status prediction module is used to predict the health status of cables.

Images