Suspension bridge main cable corrosion evolution prediction method and system

Abstract

The invention provides a method and system for predicting corrosion evolution of a main cable of a suspension bridge, and the method comprises the steps: dividing the main cable into a plurality of cable segments, collecting the non-real-time physical information of the cable segments, and constructing a cable segment-level data structure model; a multi-dimensional input vector is generated by utilizing the characteristics of structural stress, environmental exposure, bridge age, maintenance records and the like, and corrosion risk scoring is realized in combination with the trained XGBoost model; further constructing a multi-state corrosion evolution model based on a Markov chain, dividing the corrosion state into pitting, expansion, precursor and critical stages, and establishing a time evolution path; risk distribution is visually displayed in a corrosion thermodynamic diagram mode, and residual life prediction and early warning information is output in combination with a threshold strategy; according to the method, the limitation of dependence on a real-time sensor is broken through, active prediction can be completed based on historical and structural parameters under a passive condition, and the corrosion identification accuracy and early warning timeliness are effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of bridge main cable monitoring, in particular to a suspension bridge main cable corrosion evolution prediction method and system. BACKGROUND

[0002] As the largest bridge structure with the strongest crossing capacity, the main cable of the suspension bridge is a key load-bearing component, and the structural integrity thereof is directly related to the overall safety and service life of the bridge. During the service process, the main cable is often subjected to multiple corrosion environments, such as marine salt spray, hot and humid climate, air pollution, etc., and is prone to internal steel wire corrosion, fracture and other damage problems.

[0003] At present, the monitoring of the corrosion of the main cable mainly relies on periodic cable inspection, surface observation or strain / humidity monitoring by laying sensors, but there are still the following technical defects in practice:

[0004] The existing corrosion monitoring method of the main cable of the suspension bridge mainly relies on long-term collection of environmental parameters such as temperature and humidity to infer the corrosion risk level inside the main cable, but there is a serious data disconnection problem between the internal microenvironment of the main cable and the actual corrosion state, which makes it difficult to truly reflect the electrochemical corrosion process and local risk state of the steel wire, thereby leading to lagging corrosion evaluation results or even misjudgment;

[0005] Therefore, a suspension bridge main cable corrosion evolution prediction method and system are provided. SUMMARY

[0006] Therefore, the present application provides a suspension bridge main cable corrosion evolution prediction method and system to solve or alleviate the technical problems in the prior art, and at least provides a beneficial choice.

[0007] The technical scheme of the present application is implemented as follows: a suspension bridge main cable corrosion evolution prediction method, comprising the following steps:

[0008] S1. Constructing a main cable data structure model, dividing the target main cable of the suspension bridge at the cable segment level, dividing the main cable into multiple spatial discrete units in the length direction, collecting and coding non-real-time physical information for each cable segment;

[0009] S2. Establishing a structure-environment factor fusion model, using a feature vector to represent the structure stress characteristics E(x) of the cable segment, the environmental exposure characteristics S(x), the bridge age influence function H(x) and the maintenance operation record function M(x), forming a multi-dimensional input vector set, and constructing a corrosion risk score function R(x) based on the multi-dimensional input vector set: c (x)=f(E(x),S(x),H(x),M(x));

[0010] S3. Constructing a corrosion evolution path model, dividing the main cable corrosion process into a pitting initial stage, a corrosion expansion stage, a failure precursor stage, and a critical damage stage, and training and constructing a four-state Markov state transition matrix based on historical corrosion data to model the time evolution trend of the cable segment corrosion state;

[0011] S4. Constructing a spatial corrosion thermal map model, mapping the corrosion risk score function of each cable segment to the corresponding four-state transition result to the whole bridge main cable spatial model to generate a corrosion thermal map evolving over time, which is used to reflect the high-risk area and its evolution path;

[0012] S5. Outputting the prediction result and warning suggestion, performing risk threshold judgment on the corrosion thermal map, identifying the high-risk cable segment whose corrosion level reaches the preset safety threshold, and outputting the residual life prediction value of the cable segment.

[0013] Further preferably, the non-real-time physical information collected in step S1 includes:

[0014] Main cable structure parameters, including main cable cross-section structure form, anti-corrosion coating type, and cable arrangement mode;

[0015] Environmental statistical parameters, including historical meteorological data at the bridge site, sea salt concentration, wind speed and wind frequency distribution, and annual rainfall;

[0016] Structure design and service parameters, including the connection form of the cable segment and the main tower or anchor, the main cable oil sealing time, and the maintenance period of the corresponding cable segment.

[0017] Further preferably, the main cable division method in step S1 is equal-interval division, and each cable segment is assigned a unique number, which is used as an index in the main cable data structure model calculation for positioning and tracking the corrosion state of the main cable.

[0018] Further preferably, the corrosion risk score function constructed in step S2 is a nonlinear regression prediction model based on supervised learning, which is constructed using the XGBoost gradient boosting tree algorithm, including the following training process:

[0019] The structure stress feature E(x), the environmental exposure feature S(x), the bridge age influence function H(x), and the maintenance operation record function M(x) of each cable segment collected are combined to form an input feature vector;

[0020] The corrosion grade or cross-section reduction ratio obtained by manual cable disassembly detection in actual engineering is used as the label variable in supervised learning;

[0021] The XGBoost algorithm is used to train the nonlinear mapping relationship between the input feature vector and the label variable, and a regression tree cluster is constructed to minimize the loss function;

[0022] The corrosion risk score function outputs a corrosion risk score value for each cable segment after training, which is a continuous variable between [0, 1], and the greater the score value, the higher the corrosion risk.

[0023] Further preferably, the corrosion evolution path model constructed in step S3 is modeled using a four-state Markov chain, and the four states include:

[0024] State S1: Initial stage of pitting corrosion, with scattered micro-corrosion spots formed on the surface of the cable segment;

[0025] State S2: Corrosion propagation stage, with pitting corrosion connected to form a network of corrosion zones, and the cross-section weakening ratio <10%;

[0026] State S3: Precursor stage of failure, with signs of partial wire erosion or fracture appearing, accompanied by tension decay;

[0027] State S4: Critical damage stage, reaching the structural safety limit and requiring immediate repair;

[0028] The four-state model evolves in discrete time steps, and the state transition matrix P is represented as: P = [p ij ]; p ij represents the probability of the cable segment transitioning from state S4 to state S j at a certain time step.

[0029] Further preferably, the corrosion thermodynamic map generated in step S4 is a two-dimensional matrix structure composed of cable segment numbers and predicted times, with cable segment numbers as the horizontal coordinate and predicted times as the vertical coordinate. The corrosion risk score and evolution state of each cable segment at each future time point are visually mapped.

[0030] Further preferably, the risk threshold judgment in step S5 includes multi-level corrosion grade judgment and residual life assessment, including:

[0031] Set the corrosion grade judgment threshold, with a risk score greater than 0.6 defined as a high-risk area and a risk score greater than 0.85 defined as a critical risk area;

[0032] When the risk score of any cable segment at any future time point exceeds the threshold, output the corresponding risk warning;

[0033] If there is a continuous score rise and a threshold crossing trend, trigger the "residual life estimation" module;

[0034] "Residual life estimation" estimates the time interval for the cable segment to reach state S4 based on the current risk state and state transition model, and gives an output prompt of "estimated remaining safe life less than 2.5 years".

[0035] Further preferably, the remaining life prediction value output in the step S5 comprises a quantitative estimation of the remaining safe life of each cable segment, which is based on a Markov state transition matrix, predicts the expected time required for the current cable segment to migrate from the existing corrosion state to the critical damage state S4, and outputs the time point or the remaining life limit in years when the state S4 is expected to be reached.

[0036] The application also provides a system for predicting corrosion evolution of a main cable of a suspension bridge, comprising:

[0037] A cable segment data modeling module is configured to divide the main cable of the suspension bridge into a plurality of equally spaced spatial discrete cable segments, collect and encode non-real-time physical information of each cable segment;

[0038] A feature vector construction and scoring module is configured to construct the structural stress feature E(x), the environmental exposure feature S(x), the bridge age influence function H(x) and the maintenance operation record function M(x) of each cable segment into a multi-dimensional input vector, and calculate the corrosion risk score function of each cable segment based on the trained XGBoost;

[0039] An corrosion evolution modeling module is configured to model the corrosion process of the main cable into a pitting initial stage, a corrosion expansion stage, a failure precursor stage and a critical damage stage based on a Markov state transition model, and calculate the evolution path of the cable segment state over time based on the state transition probability;

[0040] A corrosion thermodynamic map generation module is configured to map the corrosion score and evolution state of the cable segment to the spatial structure of the main cable;

[0041] A result output and early warning module is configured to determine the state of the cable segment according to the set corrosion level threshold.

[0042] The embodiments of the application have the following advantages due to the use of the above technical solutions:

[0043] The application adopts cable segment level data modeling, constructs a structure-environment factor fusion model and a corrosion risk score function, quantitatively evaluates the corrosion sensitivity of each cable segment, establishes a time evolution process of the corrosion state through a multi-state Markov chain, realizes modeling of the complete corrosion path of the cable segment from the initial pitting to the critical damage, and outputs a dynamic risk thermodynamic map and a remaining life estimation.

[0044] The above summary is intended to illustrate the application and is not intended to be limiting thereof. Further aspects, embodiments and features of the application will become apparent from the detailed description referred to in conjunction with the accompanying drawings and the following detailed description. BRIEF DESCRIPTION OF DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0046] Figure 1 The step flow chart of the present application;

[0047] Figure 2 The system module architecture diagram of the present application. DETAILED DESCRIPTION

[0048] In the following, only some exemplary embodiments are simply described. As those skilled in the art can recognize, the described embodiments can be modified in various different ways without departing from the spirit or scope of the present application. Therefore, the drawings and the description are considered to be exemplary in nature rather than limiting.

[0049] The embodiments of the present application will be described in detail below with reference to the drawings.

[0050] As Figure 1 shown, the embodiment of the present application provides a method for predicting corrosion evolution of main cable of suspension bridge, comprising:

[0051] Step S1: constructing a main cable data structure model

[0052] According to the bridge design drawings and the field structure arrangement information, the main cable is divided into a plurality of cable segments at equal intervals in the length direction, each cable segment is provided with a unique number, and a discrete structure set of cable segment level is formed;

[0053] A multi-source data item record structure is established for each cable segment, and the following three types of non-real-time physical information are collected:

[0054] (1) main cable structure parameters: including the cross-section form of the position of each cable segment, i.e. solid / hollow, multi-strand winding structure, anti-corrosion coating type, i.e. epoxy coating, hot-dip galvanizing, paraffin oil sealing, and cable arrangement mode, i.e. staggered / uniform winding of inner and outer layers;

[0055] (2) Environmental statistical parameters: including historical meteorological data at the bridge site, i.e. annual average temperature, annual rainfall, wind speed and frequency distribution, surrounding sea aerosol concentration (expressed by sea salt Cl- mass concentration), regional relative humidity variation range, etc.

[0056] (3) Structure design and service parameters: including the connection relationship between the cable segment and the main tower or anchor (suspension type, anchoring type), the first oil sealing time of the main cable, the maintenance period and repair record of the cable segment;

[0057] The collected data is converted into structural stress characteristics E(x) by normalization mapping, the environmental exposure characteristics S(x) are normalized by statistical distribution to form environmental factors, and the service parameters are converted into bridge age influence function H(x) and maintenance operation record function M(x). All types of information are indexed by the same cable segment number to build a main cable structure database X s , which is used for subsequent model unified call.

[0058] Step S2: Establishing a structure-environment factor fusion model and a corrosion risk scoring function

[0059] The goal of this step is to build a multi-dimensional score representing the corrosion risk of the cable segment. For each cable segment, the following standardized feature variables are extracted to form a vector:

[0060] Structural stress characteristics E(x), environmental exposure characteristics S(x), bridge age influence function H(x) and maintenance operation record function M(x) are combined into input vector X(x) = [E(x), S(x), H(x), M(x)];

[0061] Based on the supervised learning method, the corrosion grade (classified as 0 to 4) obtained by artificial detection of the historical bridge or the section weakening ratio of the cable segment is taken as the label variable Y(x), the XGBoost gradient boosting tree algorithm is selected as the modeling core, the nonlinear mapping relationship between the input vector X and the corrosion index Y is trained, the five-fold cross-validation is used in the training process to ensure the robustness of the model, the mean square error MSE is selected as the loss function, the leaf node weight is updated and the optimal tree model cluster is constructed in each iteration, and the model output is the corrosion risk scoring function R c (x), the value range is [0, 1], where close to 1 represents a very high corrosion risk.

[0062] After the model training is completed, the parameterized function R c (x) is saved, which can output the corrosion risk score of any cable segment at any time state, serving as the input basis for subsequent corrosion evolution modeling.

[0063] Step S3: Building a corrosion evolution path model

[0064] The corrosion of the main cable is not a static state, but gradually evolves over time with environmental influences, so a path model of its evolution over time needs to be constructed. A multi-state Markov model is used to abstract the corrosion process of the main cable into the following four states:

[0065] State S1: Initial stage of pitting corrosion, characterized by the formation of local small rust spots on the surface of the steel wire, the macroscopic structure strength has not been damaged, usually first appears in visual inspection or coating micro-damage area, electrochemical activity is enhanced but no continuous corrosion path is formed;

[0066] State S2: Corrosion expansion stage, the pitting area expands outward, multiple pitting areas merge to form a sheet or network of rust areas, the coating begins to fall off, the effective cross-section of the steel wire is gradually weakened, and the loss ratio is <10%;

[0067] State S3: Pre-failure stage, multiple steel wires break, plastic erosion, cable tension decreases, and other signs appear, the structure deformation risk increases, the strength of the steel wire decreases significantly, usually accompanied by uneven tension or abnormal vibration response;

[0068] State S4: Critical damage stage, the structure reaches or approaches the limit state, such as the serious consequences of main cable slip, fracture, and cable clamp damage, structural reinforcement or replacement measures need to be taken immediately, this state is an irreversible absorbing state.

[0069] The transition probability relationship between the above states is established using a discrete-time Markov chain, and a state transition matrix P = [p ij ] is constructed, where:

[0070] p ij represents the probability of the cable segment transitioning from state S4 to state S j at a certain time step;

[0071] represents the total probability conservation of state transition;

[0072] p 44 = 1, state S4 is an absorbing state, once entered, it no longer transitions.

[0073] State transition is only allowed between adjacent states (e.g., S1→S2, S2→S3), and state rollback (i.e., no S3→S2 path) is not supported, to comply with the irreversibility and time directionality of corrosion.

[0074] In actual engineering, the values of matrix P are obtained through the following methods:

[0075] (1) For bridges with continuous annual cable inspection data, the state transition frequency between states is estimated by statistical analysis of the state sequence of each cable segment in different years, and the transition probability is obtained after normalization;

[0076] (2) Based on the existing statistical prior, the new detection data is accessed in real time during operation to dynamically modify the matrix, improving the robustness and adaptability of the model;

[0077] (3) In the absence of a large amount of measured data, the initial transition probability can be set by an expert team through factors such as structural design life, coating type, and environmental grade;

[0078] According to the current state and score R c (x), the system can use the state transition matrix to calculate the expected number of steps T e from the current state to the critical state S4

[0079] Where π i is the probability of the cable segment currently being in state S i , t i is the expected number of steps from S i to S4, which can be solved by absorbing Markov chain, and finally converted to "remaining corrosion life" in years.

[0080] Step S4: Constructing a spatial corrosion thermodynamic map model

[0081] (1) Construct a thermodynamic map matrix, with cable segment number as horizontal coordinate, prediction time point as vertical coordinate, and each cell filled with risk score value R c (x, t) and state label (S1-S4) of the corresponding cable segment and time;

[0082] (2) Risk level is presented in a color gradient mapping manner: green (low risk) - yellow (medium) - orange (high risk) - red (critical risk);

[0083] (3) The state is displayed with a digital number or icon marker overlay for easy identification of evolution trends.

[0084] (4) The thermodynamic map results are embedded into the bridge BIM platform or digital twin system for real-time updating, interactive filtering, and engineering visualization management.

[0085] Step S5: Output prediction results and early warning

[0086] This step is based on the corrosion score results, corrosion evolution state, and state transition matrix to quantitatively judge and estimate the future risk development trend of each cable segment;

[0087] (1) Risk threshold judgment mechanism

[0088] The system presets multiple score threshold values, combined with the output value of the corrosion risk score function R c (x), the risk level of each cable segment at any prediction time is divided into:

[0089] Safety zone (0≤R c <0.3): Very slight corrosion, can be included in the routine cycle inspection;

[0090] Attention zone (0.3≤R c <0.6): There are signs of mild corrosion, it is recommended to increase the frequency of inspection;

[0091] High-risk zone (0.6≤R c <0.85): Corrosion is obviously expanded, it is recommended to carry out local maintenance or coating reinforcement;

[0092] Critical zone (R c ≥0.85): May enter the failure precursor or critical state, immediate inspection or structure evaluation is required.

[0093] By traversing the main cable section, the score of each cable section at all predicted time points is compared with the threshold, all over-limit cable sections are marked, and the "risk cable section list" is output, and highlighted in the corrosion heat map.

[0094] If the score rises at several consecutive time points, or the state prediction shows a monotonic transition trend of S1→S2→S3, it is determined as a deterioration path.

[0095] As shown in Figure 2 The present application also provides a system for predicting corrosion evolution of a main cable of a suspension bridge, comprising:

[0096] A cable section data modeling module is configured to divide the main cable of the suspension bridge into a plurality of equally spaced spatial discrete cable sections, collect and encode non-real-time physical information of each cable section;

[0097] A feature vector construction and scoring module is configured to construct the structural stress feature E(x), the environmental exposure feature S(x), the bridge age influence function H(x) and the maintenance operation record function M(x) of each cable section into a multi-dimensional input vector, and calculate the corrosion risk score function of each cable section based on the trained XGBoost;

[0098] A corrosion evolution modeling module is configured to model the corrosion process of the main cable into a pitting initial stage, a corrosion expansion stage, a failure precursor stage and a critical damage stage based on a Markov state transition model, and calculate the evolution path of the cable section state over time based on the state transition probability;

[0099] A corrosion heat map generation module is configured to map the corrosion score and evolution state of the cable section to the spatial structure of the main cable;

[0100] A result output and early warning module is configured to determine the state of the cable section according to the set corrosion level threshold.

[0101] The above merely illustrates the specific embodiments 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 various changes or replacements within the technical range disclosed by 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 predicting the corrosion evolution of a suspension bridge main cable, characterized by: The following steps are involved: S1. Divide the main cable of the target suspension bridge into multiple discrete spatial units along its length, and collect and encode non-real-time physical information for each cable segment. S2. Use the characteristic vector method to represent the structural stress characteristics E(x), environmental exposure characteristics S(x), bridge age impact function H(x), and maintenance operation record function M(x) corresponding to the cable segment to form a multidimensional input vector set. Based on the multidimensional input vector set, a corrosion risk scoring function is constructed: R c (x)=f(E(x),S(x),H(x),M(x)); S3. The main cable corrosion process is divided into the initial pitting corrosion stage, the corrosion expansion stage, the failure precursor stage, and the critical damage stage. A four-state Markov state transition matrix is ​​constructed based on historical corrosion data training to build a corrosion evolution path model. S4. Map the corrosion risk scoring function of each cable segment and the corresponding four-state migration results to the spatial model of the main cables of the entire bridge, generating a corrosion heat map that evolves over time to reflect high-risk areas and their evolution paths; S5. Perform risk threshold judgment on the corrosion thermodynamic map, identify high-risk cable segments whose corrosion levels reach a preset safety threshold, and output the remaining life prediction value of the cable segment.

2. The method for predicting the corrosion evolution of a suspension bridge main cable according to claim 1, characterized in that: The non-real-time physical information collected in step S1 includes: Main cable structural parameters, including main cable cross-section structure, anti-corrosion coating type, and cable arrangement; Environmental statistical parameters, including historical meteorological data, sea salt concentration, wind speed and frequency distribution, and annual rainfall at the bridge location; Structural design and service parameters, including the connection between the cable section and the main tower or anchor, the oil sealing time of the main cable and the maintenance life of the corresponding cable section.

3. The method for predicting the corrosion evolution of a suspension bridge main cable according to claim 1, characterized in that: The main cable segmentation in step S1 is performed at equal intervals, and each cable segment is assigned a unique number. The number is used as an index in the calculation of the main cable data structure model to locate and track the corrosion status of the main cable.

4. The method for predicting corrosion evolution of a suspension bridge main cable according to claim 1, characterized in that: The corrosion risk scoring function constructed in step S2 is a nonlinear regression prediction model based on supervised learning. The nonlinear regression prediction model is constructed using the XGBoost gradient boosting tree algorithm and includes the following training process: The collected structural stress characteristics E(x), environmental exposure characteristics S(x), bridge age impact function H(x) and maintenance operation record function M(x) of each cable segment are used to form the input feature vector; The corrosion grade or cross-section weakening ratio obtained from manual cable removal inspection in actual projects is used as the label variable in supervised learning; The XGBoost algorithm is used to train the nonlinear mapping relationship between the input feature vector and the label variable, and a regression tree cluster is constructed to minimize the loss function; The corrosion risk scoring function obtained after training is completed outputs a corrosion risk score value for each cable segment. The corrosion risk score value is a continuous variable between [0, 1], and a larger score value indicates a higher corrosion risk.

5. The method for predicting corrosion evolution of a suspension bridge main cable according to claim 1, characterized in that: The corrosion evolution path model constructed in step S3 is modeled using a four-state Markov chain, where the four states include: State S1: Initial stage of pitting corrosion, scattered micro-corrosion spots form on the cable surface; State S2: Corrosion expansion stage, pitting is connected to form a network corrosion area, and the cross-section weakening ratio is <10%; State S3: Pre-failure stage, some steel wires show signs of erosion or breakage, accompanied by tension decay; State S4: critical damage stage, reaching the structural safety limit, requiring immediate repair; The four-state model evolution is modeled in discrete time steps, and the state transition matrix P is expressed as: P = [p ij ];p ij Indicates that the cable segment transitions from state S4 to state S at a certain time step j probability.

6. The method for predicting corrosion evolution of a suspension bridge main cable according to claim 1, characterized in that: The corrosion thermodynamic map generated in step S4 is a two-dimensional matrix structure consisting of cable segment numbers and predicted times, with the cable segment number as the horizontal coordinate and the predicted time as the vertical coordinate, visually mapping the corrosion risk score and evolution state of each cable segment at each future time point.

7. The method for predicting corrosion evolution of a suspension bridge main cable according to claim 1, characterized in that: The risk threshold determination in step S5 includes multi-level corrosion level determination and remaining life assessment, including: Set the threshold for corrosion level determination. A risk score greater than 0.6 is defined as a high-risk area, and a risk score greater than 0.85 is defined as a critical risk area. When the risk score of any cable segment exceeds the threshold at any time in the future, a corresponding risk alarm will be output; If there is a trend of continuous score increase and crossing the threshold, the "Remaining Life Assessment" will be triggered, and the qualitative recommendation of "Expected Remaining Safe Life is less than 2.5 years" will be output.

8. The method for predicting corrosion evolution of a suspension bridge main cable according to claim 1, characterized in that: The remaining life prediction value output in step S5 includes a quantitative estimation of the remaining safe life of each cable segment. The quantitative estimation is based on the Markov state transition matrix, predicting the expected time required for the current cable segment to migrate from the existing corrosion state to the critical damage state S4, and outputting the expected time point to reach state S4 or the remaining life in years.

9. A system for predicting the corrosion evolution of a main cable of a suspension bridge according to any one of claims 1 to 8, characterized in that: include: The cable segment data modeling module is used to divide the main cable of the suspension bridge into multiple equally spaced spatially discrete cable segments, and collect and encode the non-real-time physical information of each cable segment; The feature vector construction and scoring module is used to construct the structural stress characteristics E(x), environmental exposure characteristics S(x), bridge age impact function H(x), and maintenance operation record function M(x) of each cable segment into a multidimensional input vector, and calculate the corrosion risk scoring function of each cable segment based on the trained XGBoost; The corrosion evolution modeling module is used to model the main cable corrosion process into the initial pitting corrosion stage, corrosion expansion stage, failure precursor stage and critical damage stage based on the Markov state transition model, and calculate the evolution path of the cable segment state over time based on the state transition probability; A corrosion heat map generation module is used to map the corrosion score and evolution state of the cable segment to the main cable spatial structure; The result output and early warning module is used to determine the status of the cable section according to the set corrosion level threshold.

Images