In-situ monitoring method and system for monitoring corrosion behavior of weathering resistant steel welded joint

Abstract

The invention provides an in-situ monitoring method and system for monitoring the corrosion behavior of a weathering resistant steel welded joint, and relates to the technical field of corrosion monitoring. An electrochemical impedance measurement value of a monitoring area is collected and serves as a boundary condition to be input into a pre-established electric field distribution model for solving, and the corrosion behavior of the weathering resistant steel welded joint is monitored. Therefore, a two-dimensional conductivity distribution is obtained in each monitoring period and a conductivity map is generated. A first conductivity map sequence is formed by repeating the process, and key components with the conductivity changing along with time are analyzed and extracted. And reconstructing a conductivity background in an uncorroded state based on the key components, and deducting the background from the first sequence to obtain a second conductivity map sequence representing conductivity change caused by local corrosion. And when the conductivity variable quantity in the second sequence exceeds a preset threshold value, judging that the monitoring area is corroded, and determining the corrosion position, outline and stage. According to the method, accurate positioning, form recognition and stage judgment of the corrosion behavior of the weathering resistant steel welded joint can be achieved.

Description

Technical Field

[0001] This application relates to the field of corrosion monitoring technology, and in particular to an in-situ monitoring method and system for monitoring the corrosion behavior of weathering steel welded joints. Background Technology

[0002] In current industrial and infrastructure construction, metal structures exposed to harsh environments for extended periods, especially their critical connections, face a significant technical challenge in early and accurate corrosion detection and on-site monitoring to ensure structural safety and extend service life. Traditional inspection methods struggle to detect minute changes in the early stages of corrosion and cannot pinpoint its location in real time, posing considerable difficulties in preventing sudden structural damage and ensuring effective maintenance.

[0003] To address these issues, a method using ultrasonic guided wave technology to monitor corrosion has been developed. This method involves emitting and receiving ultrasonic waves at the material's surface, and analyzing changes in these waves to determine the extent of corrosion within the material or on its surface. This is because corrosion causes materials to thin or soften, affecting the propagation of ultrasonic waves, thus allowing for indirect assessment of corrosion.

[0004] However, this ultrasonic guided wave technology also has some drawbacks in practical applications. Its biggest problem is its insensitivity to small, localized corrosion, especially in the early stages of corrosion. The material changes it causes may not significantly alter the propagation path or intensity of the ultrasonic waves, making early corrosion difficult to detect. Furthermore, this technology typically only provides an overall corrosion profile, making it difficult to pinpoint the exact location of corrosion or assess uneven corrosion distribution. For complex structures requiring meticulous maintenance, it cannot meet the demands for high-precision localized corrosion monitoring. Summary of the Invention

[0005] The purpose of this application is to provide an in-situ monitoring method and system for monitoring the corrosion behavior of weathering steel welded joints, so as to solve the problem of insufficient accurate identification of the location and development level of local corrosion in weathering steel welded joints in the prior art.

[0006] To address the aforementioned technical problems, in a first aspect, this application provides an in-situ monitoring method for monitoring the corrosion behavior of weathering steel welded joints, comprising: Electrochemical impedance measurements were collected for multiple monitoring areas, including the surface of the weathering steel welded joint and a preset range area of ​​the joint. A model is pre-established to describe the conductivity distribution corresponding to multiple monitoring areas. The model is solved by taking the electrochemical impedance measurement value corresponding to each monitoring area as the boundary condition input, and the conductivity distribution of the corresponding monitoring area is output by the model to construct a conductivity map. The acquisition and solution process is repeated at preset time intervals to form a first conductivity map sequence. The first conductivity map sequence is then analyzed to extract key components that characterize the change of conductivity over time. Based on the key changing components, the conductivity background characterizing the uncorroded state is reconstructed, and background subtraction is performed on each conductivity map in the first conductivity map sequence based on the conductivity background to obtain a second conductivity map sequence. The second conductivity map sequence contains multiple conductivity maps whose conductivity changes due to the corroded state. When the change in conductivity corresponding to the conductivity map in the second conductivity map sequence exceeds a preset threshold, it is determined that corrosion has occurred in the monitoring area corresponding to the conductivity map, and the corrosion result is determined.

[0007] Optionally, the pre-established model for describing the electric field distribution in the monitoring area includes: Based on the geometry, material properties, and layout of the electrode array of the monitoring area, a mathematical expression is constructed that can reflect the law of charge transport within the monitoring area. The mathematical expression is discretized to form a set of equations containing the conductivity parameter to be determined, wherein each unknown parameter corresponds to a key location within the monitoring area; By preprocessing the equation set, a mapping relationship between the measured electrochemical impedance values ​​and the conductivity distribution within the monitoring area is established to form a model for describing the electric field distribution in the monitoring area.

[0008] Optionally, the step of solving the model by using the electrochemical impedance measurement value corresponding to each monitoring area as a boundary condition input, and outputting the conductivity distribution of the corresponding monitoring area through the model to construct a conductivity map includes: In each monitoring cycle, the measured electrochemical impedance value is used as a boundary condition for the mathematical expression; Using an iterative optimization method, under the premise of satisfying the boundary conditions, the conductivity parameter to be determined in the equation system is adjusted until the equation system reaches a convergent state; The converged conductivity parameters to be determined are arranged according to their positions within the monitoring area to form a two-dimensional data matrix, and the two-dimensional data matrix is ​​used as the conductivity map of the corresponding monitoring area.

[0009] Optionally, the acquisition and solution process is repeated at preset time intervals to form a first conductivity map sequence, and the first conductivity map sequence is analyzed to extract key components characterizing the change in conductivity over time, including: Within a preset monitoring time interval, the process of acquiring electrochemical impedance measurements and constructing conductivity maps is repeated to obtain a series of conductivity maps arranged in chronological order, forming the first conductivity map sequence. Data processing is performed on each conductivity map in the first conductivity map sequence to identify and quantify the time-varying characteristic components in the conductivity map that are indicative of corrosion behavior. These characteristic components are key components characterizing the change in conductivity over time.

[0010] Optionally, the step of reconstructing the conductivity background characterizing the uncorroded state based on the key changing components, and performing background subtraction on each conductivity map in the first conductivity map sequence based on the conductivity background to obtain the second conductivity map sequence includes: Using the aforementioned key variable components, a reference image representing the conductivity distribution of the monitored area under non-corrosion conditions is constructed as a conductivity background through data modeling methods. Perform a pixel-by-pixel difference operation between each conductivity map in the first conductivity map sequence and the conductivity background to obtain the difference operation result; The difference calculation results are used to form a second conductivity map sequence. Each conductivity map in the second conductivity map sequence reflects the amount of conductivity change caused by local corrosion in the monitoring area.

[0011] Optionally, the corrosion results include the corrosion location, corrosion profile, and corrosion stage. When the change in conductivity corresponding to a conductivity map in the second conductivity map sequence exceeds a preset threshold, it is determined that corrosion has occurred in the monitoring area corresponding to the conductivity map, and the corrosion result is determined, including: For each conductivity map in the second conductivity map sequence, the conductivity change corresponding to the conductivity map is compared point by point with a preset corrosion discrimination threshold. When the conductivity change at any monitoring point in the monitoring area exceeds the corrosion discrimination threshold, the monitoring point is marked as a corrosion occurrence point. By analyzing the set of all marked corrosion sites, corrosion locations, corrosion profiles, and corrosion stages determined based on the degree of change in conductivity are identified to form corrosion results.

[0012] Optionally, by analyzing the set of all marked corrosion sites, corrosion locations, corrosion profiles, and corrosion stages determined based on the degree of change in conductivity, corrosion results are formed, including: Spatial clustering analysis is performed on the set of points marked as corrosion occurrences to determine continuous regions where corrosion occurs. The centroid or geometric center of the continuous region is taken as the corrosion location, and the outer boundary of the continuous region forms the corrosion profile. The change in conductivity within the continuous region is statistically analyzed and compared with a preset corrosion stage classification standard to determine the current corrosion stage. The identified corrosion locations, corrosion profiles, and determined corrosion stages are integrated to form the corrosion result.

[0013] Secondly, this application provides an in-situ monitoring system for monitoring the corrosion behavior of weathering steel welded joints, comprising: The acquisition module is used to acquire electrochemical impedance measurements corresponding to multiple monitoring areas, including the surface of the weathering steel welded joint and a preset range area of ​​the joint; The module is used to pre-build a model to describe the conductivity distribution corresponding to multiple monitoring areas. The model is solved by taking the electrochemical impedance measurement value corresponding to each monitoring area as the boundary condition input, and the conductivity distribution of the corresponding monitoring area is output by the model to construct a conductivity map. The extraction module is used to repeatedly perform the acquisition and solution process at preset time intervals to form a first conductivity map sequence, and to analyze the first conductivity map sequence to extract key change components that characterize the change of conductivity over time. A generation module is used to reconstruct the conductivity background characterizing the uncorroded state based on the key changing components, and to perform background subtraction on each conductivity map in the first conductivity map sequence based on the conductivity background to obtain a second conductivity map sequence. The second conductivity map sequence contains multiple conductivity maps whose conductivity changes due to the corroded state. The determination module is used to determine that corrosion has occurred in the monitoring area corresponding to the conductivity map when the change in conductivity corresponding to the conductivity map in the second conductivity map sequence exceeds a preset threshold, and to determine the corrosion result.

[0014] Thirdly, this application provides an electronic device, comprising: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of an in-situ monitoring method for monitoring the corrosion behavior of weathering steel welded joints as described in the first aspect above.

[0015] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of an in-situ monitoring method for monitoring the corrosion behavior of weathering steel welded joints as described in the first aspect above.

[0016] This application acquires electrochemical data for key areas of welded joints by collecting electrochemical impedance spectroscopy (EIS) measurements from multiple monitoring areas, providing a basis for corrosion analysis. By using these EIS measurements as boundary conditions and solving the model, the model outputs a two-dimensional conductivity distribution for the corresponding monitoring areas, which is recorded as a conductivity map. This transforms the EIS data into a two-dimensional conductivity map, visually displaying the charge distribution and providing a basis for corrosion localization. By repeating the acquisition and solution process at preset time intervals, a first conductivity map sequence is formed. This sequence is then analyzed to extract key components characterizing conductivity changes over time, dynamically generating a new conductivity map sequence and extracting information reflecting corrosion. The changing characteristic information supports early warning and trend analysis; by reconstructing the conductivity background representing the uncorroded state based on the key changing components, and performing background subtraction on each conductivity map in the first conductivity map sequence based on the conductivity background, a second conductivity map sequence is obtained. By subtracting the non-corrosion background, the conductivity changes caused by local corrosion are highlighted and quantified, improving detection sensitivity and accuracy; when the conductivity change corresponding to the conductivity map in the second conductivity map sequence exceeds a preset threshold, it is determined that corrosion has occurred in the monitoring area corresponding to the conductivity map, and the corrosion result is determined. Based on the conductivity change threshold, corrosion is automatically determined to have occurred, and the corrosion location, outline, and stage are provided, providing a basis for protection decisions. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A schematic flowchart illustrating an in-situ monitoring method for monitoring the corrosion behavior of weathering steel welded joints, provided in an embodiment of this application. Figure 2 A schematic flowchart illustrating another in-situ monitoring method for monitoring the corrosion behavior of weathering steel welded joints, provided in an embodiment of this application; Figure 3 A schematic flowchart illustrating another in-situ monitoring method for monitoring the corrosion behavior of weathering steel welded joints, provided in an embodiment of this application; Figure 4 This is a schematic diagram of an in-situ monitoring system for monitoring the corrosion behavior of weathering steel welded joints, provided in an embodiment of this application. Detailed Implementation

[0019] In the current industrial and infrastructure construction sectors, achieving early, accurate, and effective corrosion monitoring of metal structures, especially their critical connection points, exposed to complex environments for extended periods, to ensure structural integrity and extend service life, has become an urgent technological requirement. Traditional macroscopic detection methods often fail to capture the microscopic details of corrosion initiation and cannot provide real-time, spatially resolved corrosion status information, posing a significant challenge to preventing sudden structural failures and optimizing maintenance strategies. Even advanced technologies such as ultrasonic guided waves often fail to meet the demand for high-precision localized corrosion monitoring due to insufficient sensitivity to minute localized corrosion and the inability to accurately provide information on the specific location and spatial distribution of corrosion.

[0020] To address the aforementioned technical deficiencies, this application proposes an in-situ monitoring method for monitoring the corrosion behavior of weathering steel welded joints. This method generates and analyzes a two-dimensional conductivity map sequence by collecting electrochemical impedance data and combining it with model calculations, enabling dynamic tracking of minute changes in the charge transport characteristics within the material. Through precise background subtraction techniques, this application can effectively separate and quantify conductivity anomalies caused by localized corrosion, thereby achieving early detection, precise location, morphological identification, and stage assessment of corrosion. This significantly improves the sensitivity and spatial resolution of corrosion monitoring, overcoming the shortcomings of existing technologies in early corrosion identification and refined assessment of localized corrosion, and providing more reliable technical support for the safe operation and maintenance of metal structures.

[0021] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] The core of this application is to provide an in-situ monitoring method for monitoring the corrosion behavior of weathering steel welded joints, and a flowchart of one specific implementation is shown below. Figure 1 As shown, the method includes: 101. Collect electrochemical impedance measurements corresponding to multiple monitoring areas, wherein the multiple monitoring areas include the surface of the weathering steel welded joint and the preset range area of ​​the joint.

[0023] Electrochemical impedance spectroscopy (EIS) measurements, obtained through electrochemical impedance spectroscopy, reflect the impedance response of a material under the influence of alternating current signals at different frequencies. These data include the material's electrochemical properties such as resistance and capacitance, and can reveal information related to the corrosion process occurring on the material's surface.

[0024] The monitoring area refers to the specific physical region selected in and around the weathering steel welded joint for electrochemical impedance spectroscopy measurements. These areas are divided into multiple discrete measurement points or small blocks to obtain spatially resolved electrochemical information.

[0025] In this embodiment, the monitoring object and scope are first determined, that is, the weathering steel welded joint to be monitored and the surrounding preset range area are clearly defined. For example, in a large bridge structure, the weathering steel welded joint of the key load-bearing part is selected as the monitoring object, and its weld area and the area extending 5 cm on both sides of the weld are defined as the monitoring range.

[0026] Secondly, by arranging an electrode array within the monitoring area, multiple working electrodes, reference electrodes, and counter electrodes are fixed to the surface of the monitoring area in a predetermined geometric arrangement. For example, within the defined monitoring range, an array consisting of 16 miniature working electrodes, 1 reference electrode, and 1 counter electrode is uniformly arranged to ensure that each electrode can effectively cover a small monitoring sub-area.

[0027] Next, by applying alternating current signals and acquiring impedance responses, a series of small alternating current signals of different frequencies were applied to the electrode array using an electrochemical workstation, and the voltage and current responses between each working electrode and the reference electrode were measured simultaneously. For example, the electrochemical workstation applied an alternating voltage perturbation of 5 mV to each working electrode in a frequency range from 100 kHz to 0.01 Hz and recorded the corresponding current response, thereby obtaining electrochemical impedance spectral data for each monitoring sub-region.

[0028] Finally, by converting the collected electrochemical impedance spectroscopy data into electrochemical impedance measurements, the impedance spectra of each monitoring sub-region are processed to extract key parameters such as the impedance magnitude and phase angle at specific frequencies, which are then used as the electrochemical impedance measurements for that sub-region. For example, the impedance magnitude and phase angle at 1 Hz are extracted from the impedance spectra of each electrode, and these values ​​will serve as input data for model solving in subsequent steps.

[0029] 102. A model is pre-established to describe the conductivity distribution corresponding to multiple monitoring areas. The model is solved by taking the electrochemical impedance measurement value corresponding to each monitoring area as the boundary condition input, and the conductivity distribution of the corresponding monitoring area is output by the model to construct a conductivity map.

[0030] The conductivity distribution model is a mathematical framework that establishes a link between externally measured electrochemical signals (such as electrochemical impedance spectroscopy) and the conductivity distribution within a material. This model considers the physical shape of the monitoring area, the electrical properties of the material itself, and the placement of the electrodes used for measurement, aiming to deduce the conductivity values ​​at different locations within the material through calculation.

[0031] Conductivity maps are a visual representation, usually presented as two-dimensional images, where different colors or gray levels represent the conductivity at different locations within the monitored area, thus intuitively showing how easily current flows through the material.

[0032] Specifically, the process of "pre-establishing a model to describe the electric field distribution in the monitoring area" in step 102 includes: constructing a mathematical expression that reflects the charge transport law in the monitoring area based on the geometry, material properties, and electrode array layout of the monitoring area; discretizing the mathematical expression to form a set of equations containing the conductivity parameters to be determined, wherein each unknown parameter corresponds to a key location in the monitoring area; and establishing a mapping relationship between the electrochemical impedance measurement value and the conductivity distribution inside the monitoring area by preprocessing the set of equations, so as to form a model to describe the electric field distribution in the monitoring area.

[0033] First, based on the geometry, material properties, and electrode array layout of the monitoring area, a mathematical expression reflecting the charge transport pattern within the monitoring area is constructed. This process forms the basis of the model, which utilizes physical laws (such as Ohm's law and the law of conservation of charge) to describe how current flows within the monitoring area. For example, on a two-dimensional monitoring plane, the charge transport pattern can be represented by a mathematical expression that states the divergence of the product of conductivity and the rate of change of potential space is zero. Specifically, it takes the following form: div(σ·grad(V))=0 Where σ represents the electrical conductivity of the material, which may vary at different locations; V represents the electric potential, which also varies with location; div is the divergence operation, representing the degree of divergence of the vector field; and grad is the gradient operation, representing the direction of the maximum rate of change of the scalar field. This expression illustrates the distribution of current within the material in the absence of additional charge generation or loss.

[0034] Secondly, by discretizing the mathematical expression, a system of equations is formed containing the conductivity parameters to be determined, where each unknown parameter corresponds to a key location within the monitoring area. Since computers cannot directly process continuous differential equations, they need to be transformed into a series of algebraic equations that can be solved at discrete points. This is typically achieved using numerical techniques such as the finite element method, which divides the monitoring area into many small, interconnected units (e.g., small cubes or triangles). Within each unit, conductivity is considered an unknown parameter to be determined; that is, the unknown parameter corresponds to each unit (key location) within the monitoring area. In this way, a continuous physical problem is transformed into a large system of linear or nonlinear algebraic equations, for example, of the form Ax = b, where x is a vector containing the conductivity values ​​of all these unknown units, A is a coefficient matrix, and b is a known vector, both determined by the geometry and electrode layout of the monitoring area.

[0035] Finally, by preprocessing the equations, a mapping relationship is established between the electrochemical impedance measurements and the conductivity distribution within the monitoring region, forming a model to describe the electric field distribution in the monitoring region. Preprocessing aims to improve the efficiency and stability of solving the equations, for example, by optimizing the coefficient matrix. This mapping relationship is a key function of the model, defining how to infer the conductivity distribution within the material from electrochemical impedance data measured at external electrodes (as input to the model). For example, by analyzing the sensitivity of each electrode measurement to conductivity changes in different internal regions, the model can establish a reverse-engineering mechanism, allowing the internal conductivity distribution to be inferred from the model given the impedance measurement at each electrode.

[0036] Furthermore, step 102, the process of "solving the model by using the electrochemical impedance measurement value corresponding to each monitoring area as a boundary condition input, and outputting the conductivity distribution of the corresponding monitoring area through the model to construct a conductivity map," includes: 1021. In each monitoring cycle, the measured electrochemical impedance value is used as the boundary condition of the mathematical expression.

[0037] 1022. Using an iterative optimization method, under the premise of satisfying the boundary conditions, adjust the conductivity parameter to be determined in the equation system until the equation system reaches a convergent state.

[0038] 1023. Arrange the converged conductivity parameters to be determined according to their positions in the monitoring area to form a two-dimensional data matrix, and use the two-dimensional data matrix as the conductivity map of the corresponding monitoring area.

[0039] In this embodiment, firstly, in each monitoring cycle, the electrochemical impedance measurement value is used as the boundary condition of the mathematical expression. The electrochemical impedance data collected at the electrode location in step 101 is converted into the voltage or current values ​​required for model solving, and these are input as known information into the discretized equation set. For example, if the electrochemical impedance measurement value is obtained at a specific frequency and the model is based on a steady-state electric field, then the impedance value needs to be converted into an equivalent resistance or conductance, and combined with the applied current or measured voltage, as the input to the model at the electrode location. Specifically, if the applied current I_k and the measured voltage V_k on the electrode are known, then these values ​​will be directly used as the boundary conditions of the equation set.

[0040] Secondly, using an iterative optimization method, under the premise of satisfying the boundary conditions, the conductivity parameters to be determined in the equation system are adjusted until the equation system reaches a convergent state. The iterative optimization method is a computer algorithm that starts with an initial guess of the conductivity distribution and repeatedly adjusts these conductivity parameters to minimize the difference between the electrode response calculated by the model and the actual measured electrochemical impedance value. For example, the algorithm starts with an initial guess of the conductivity distribution and calculates the voltage or current value predicted by the model at the electrode location under that conductivity distribution. Then, these predicted values ​​are compared with the actual measured electrochemical impedance value, and the error is calculated. Based on the magnitude and direction of the error, the algorithm adjusts the internal conductivity parameters to be determined and repeats this process until the error is sufficiently small, or the change in the conductivity parameters is below a preset threshold, i.e., convergence is achieved. For example, convergence can be considered achieved when the root mean square error between the model-predicted electrode voltage and the measured voltage is less than 0.01 mV.

[0041] Finally, by arranging the converged conductivity parameters to be determined according to their positions within the monitoring area, a two-dimensional data matrix is ​​formed, and this two-dimensional data matrix serves as the conductivity map for the corresponding monitoring area. Once the iterative optimization process is completed and convergence is achieved, the conductivity value of each discrete unit within the monitoring area is obtained. These values ​​are organized into a two-dimensional grid or matrix according to their actual spatial positions within the monitoring area. For example, if the monitoring area is divided into a 10x10 grid, then the 100 converged conductivity parameters will fill this 10x10 matrix. This two-dimensional data matrix is ​​the digital form of the conductivity map, which can be further processed and visualized, for example, by color-coding different conductivity values ​​to different colors, thereby visually displaying the conductivity distribution within the monitoring area.

[0042] This application, by establishing an accurate electric field model and combining iterative optimization algorithms, can transform discrete electrochemical impedance measurements into continuous, high-resolution two-dimensional conductivity distribution images. This not only overcomes the limitation of traditional point measurements in providing spatial information, but also intuitively and accurately reflects the charge transport characteristics within the material.

[0043] 103. Repeat the acquisition and solution process at preset time intervals to form a first conductivity map sequence, and analyze the first conductivity map sequence to extract key change components that characterize the change of conductivity over time.

[0044] The first conductivity map sequence refers to a collection of two-dimensional conductivity maps arranged in chronological order, where each conductivity map represents the spatial distribution of conductivity within a monitoring area at a specific time point.

[0045] Key changing components refer to specific patterns or features in the conductivity diagram sequence that show significant changes over time. These changes are directly related to the occurrence or development of corrosion. By extracting these components, the dynamic characteristics of corrosion can be highlighted and quantified.

[0046] Optionally, such as Figure 2 As shown, step 103 may specifically include: 1031. Within a preset monitoring time interval, the process of acquiring electrochemical impedance measurements and constructing conductivity maps is repeated to obtain a series of conductivity maps arranged in chronological order, forming the first conductivity map sequence.

[0047] The preset monitoring time interval is the pre-defined time length between two consecutive measurements and the construction of the conductivity map. It determines the frequency of assessing the corrosion state of the material, thereby enabling the capture of corrosion dynamics that change over time.

[0048] For example, suppose a weathering steel structure named "Structure A" has welded joints as the primary monitoring target. The monitoring system is configured to perform measurements every 24 hours (a preset monitoring interval). At 08:00 AM on Day 1, the system initiates its process: it collects electrochemical impedance data from 100 designated monitoring points on the specific welded joints. This data is then input into a conductivity distribution model, generating the conductivity map for Day 1. At 08:00 AM on Day 2, this process is repeated to generate the conductivity map for Day 2. This process continues for 30 days, ultimately forming a sequence of 30 conductivity maps, each corresponding to a specific day, thus constituting the "First Conductivity Map Sequence".

[0049] 1032. Perform data processing on each conductivity map in the first conductivity map sequence to identify and quantify the time-varying characteristic components in the conductivity map that are indicative of corrosion behavior, wherein the characteristic components are key change components characterizing the change of conductivity over time.

[0050] Among them, the characteristic components that change over time and are indicative of corrosion behavior refer to specific changes in the conductivity pattern over time. These specific changes can serve as reliable indicators of corrosion initiation, development, or severity, i.e., key change components.

[0051] In step 1032, key information related to corrosion is extracted by analyzing the "first conductivity map sequence." First, each conductivity map in the sequence undergoes data processing. This typically involves techniques such as image processing, statistical analysis, or machine learning algorithms. For example, a common method is differential analysis. In this method, a baseline conductivity map (e.g., the first map in the sequence, representing the initial state) is subtracted from subsequent conductivity maps. The resulting differential map highlights areas where conductivity changes.

[0052] The formula for difference analysis can be expressed as: baseline in, This represents the conductivity change matrix over time t. This represents the conductivity matrix at time t; baseline represents the baseline conductivity matrix.

[0053] Alternatively, Principal Component Analysis (PCA) or Independent Component Analysis (ICA) can be applied to the entire sequence. These techniques decompose multidimensional time-series data (conductivity map sequences) into a set of orthogonal components. Components that capture the most significant variance over time, particularly those associated with localized decreases in conductivity (indicating material degradation due to corrosion), are identified as "key variation components." For example, PCA aims to find a set of orthogonal vectors (principal components) that explain the largest variance in the data. The first few principal components typically represent the most significant time variations.

[0054] The process then involves quantifying these identified characteristic components. This might mean calculating the magnitude of conductivity changes within a specific region, tracking the growth rate of low-conductivity areas, or assigning a corrosion severity index based on the extracted characteristics. The output of this sub-step is a set of quantified “key change components” that directly characterize how conductivity changes over time, providing a clear indicator of corrosion behavior.

[0055] For example, continuing with "Structure A," after obtaining the 30-day "first conductivity map sequence," the system performs data processing. It uses differential analysis to compare the daily conductivity map with the initial map from day 1. For example, the conductivity map for day 10 is subtracted from the map for day 1. Regions with significantly negative conductivity differences (indicating a decrease in conductivity) are marked. The system then quantifies these changes. For example, it might calculate the average decrease in conductivity for a specific region over 30 days. If a specific region on the weld joint (e.g., "Region X") consistently shows a decreasing trend in conductivity in the differential map, that trend is identified as a "key component of change." The rate of decrease and the magnitude of "Region X" are then quantified.

[0056] 104. Reconstruct the conductivity background characterizing the uncorroded state based on the key changing components, and perform background subtraction on each conductivity map in the first conductivity map sequence based on the conductivity background to obtain the second conductivity map sequence.

[0057] The conductivity background refers to a reference image constructed using data modeling methods that represents the ideal conductivity distribution of the monitored area under non-corrosion conditions. It serves as a benchmark for distinguishing between the inherent conductivity of the material and conductivity changes caused by corrosion in subsequent analyses.

[0058] The second conductivity map sequence refers to a series of conductivity maps arranged in chronological order after background subtraction, where each map specifically reflects the amount of conductivity change caused by local corrosion in the monitored area.

[0059] Optionally, such as Figure 3 As shown, step 104 may specifically include: 1041. Using the aforementioned key variation components, a reference image representing the conductivity distribution of the monitored area under non-corrosion conditions is constructed as a conductivity background through data modeling methods. 1042. Perform a pixel-by-pixel difference operation between each conductivity map in the first conductivity map sequence and the conductivity background to obtain the difference operation result; By performing pixel-by-pixel difference calculations between each conductivity map in the first conductivity map sequence and the conductivity background, the inherent conductivity distribution in the uncorroded state is eliminated, thereby highlighting the conductivity changes caused by corrosion.

[0060] 1043. The difference calculation results are used to form a second conductivity map sequence, where each conductivity map in the second conductivity map sequence reflects the amount of conductivity change caused by local corrosion in the monitoring area.

[0061] The second conductivity map sequence contains multiple conductivity maps whose conductivity changes due to corrosion.

[0062] In this embodiment of the application, firstly, by using the key variable components in step 1041, a reference image representing the conductivity distribution of the monitoring area under non-corrosion conditions is constructed as a conductivity background through a data modeling method.

[0063] Data modeling methods refer to methods that use techniques such as statistics or machine learning to build mathematical models from existing data in order to infer or predict a certain state.

[0064] To accurately identify corrosion, a pristine "uncorroded" state is needed as a reference. Even in uncorroded materials, the conductivity distribution may exhibit subtle inherent differences. This step utilizes the "key variation components" (i.e., corrosion-related conductivity variation patterns) identified in step 1032 to infer and construct an ideal "conductivity background" using data modeling methods. This can be achieved in several ways, such as: 1. Statistical Averaging Method: If the "key variation components" are local and sparse, then in the "first conductivity map sequence," most regions may still be in an uncorroded state in the early stages. The system can statistically average the conductivity values ​​of these stable regions across multiple early conductivity maps (e.g., calculate the median or weighted average) to construct a conductivity background representing the uncorroded state.

[0065] 2. Machine Learning Method: Train a machine learning model (e.g., an autoencoder or anomaly detection model) to learn the conductivity patterns of healthy or stable regions in the "first conductivity map sequence". When applied to the entire monitoring area, the output of this model is the "conductivity background" after removing the effects of corrosion.

[0066] Using these methods, the system can generate a two-dimensional conductivity map that accurately depicts the conductivity distribution at various locations within the monitoring area when no corrosion occurs, serving as a benchmark for subsequent corrosion detection.

[0067] For example, suppose that after the monitoring system for "Structure A" obtains a 30-day "first conductivity map sequence," analysis reveals that in the first few days of the sequence, except for a few tiny areas, the conductivity changes in most areas are very small, and no "key change components" as defined in step 1032 appear. The system selects these early, stable conductivity maps (e.g., the maps from the first 5 days) and performs pixel-level averaging on the areas unaffected by the "key change components." Through this statistical averaging method, the system constructs a "conductivity background" map representing the welded joint of "Structure A" in an ideal, uncorroded state. This background map smoothly reflects the inherent conductivity distribution of the material itself, eliminating interference that corrosion may cause.

[0068] Secondly, in step 1042, a pixel-by-pixel difference operation is performed between each conductivity map in the first conductivity map sequence and the conductivity background to obtain the difference operation result. By performing a pixel-by-pixel difference operation between each conductivity map in the first conductivity map sequence and the conductivity background, the inherent conductivity distribution in the uncorroded state is eliminated, thereby highlighting the conductivity change caused by corrosion.

[0069] Pixel-by-pixel interpolation is an image processing technique that generates a new image by subtracting the value of the corresponding pixel in another image from the value of each pixel in one image.

[0070] The core of the process involves performing a simple arithmetic operation: "background subtraction." For each conductivity map in the "first conductivity map sequence," the system performs a pixel-by-pixel subtraction operation with the "conductivity background" constructed in step 1041. Specifically, for each pixel in the conductivity map... Its current conductivity value The conductivity value at the corresponding position in the conductivity background Perform difference operations.

[0071] The formula for this operation is defined as: in, Indicates time t and position The change in electrical conductivity caused by corrosion. Indicates time t and position The conductivity value is derived from the "first conductivity sequence". Indicates the location The conductivity background value is derived from the "conductivity background" constructed in step 1041.

[0072] The purpose of this subtraction operation is to eliminate the inherent conductivity distribution of the material in its uncorroded state, so that any remaining changes in conductivity can be directly attributed to the occurrence or development of corrosion. In this way, the weak signals caused by corrosion can be effectively amplified and highlighted.

[0073] For example, continuing with "Structure A," after obtaining the "conductivity background," the system iterates through each conductivity map in the 30-day "first conductivity map sequence." For the conductivity map of day 15, the system performs a pixel-by-pixel subtraction operation with the "conductivity background." Suppose that at a specific pixel, the background conductivity is 120 units, while the corresponding pixel in the day 15 conductivity map has a conductivity of 110 units. The difference for that pixel is 110 - 120 = -10 units. This negative value directly indicates a decrease in conductivity at that location relative to the uncorroded state, potentially foreshadowing corrosion.

[0074] Finally, the difference calculation results are used in step 1043 to form a second conductivity map sequence. Each conductivity map in the second conductivity map sequence reflects the amount of conductivity change caused by local corrosion in the monitoring area.

[0075] This step organizes each "difference calculation result" obtained in step 1042 into a new time series. Each "difference calculation result" is itself a two-dimensional image, representing the change in conductivity relative to the "conductivity background" at each location within the monitoring area at a specific time point. The system collects these time-series-arranged difference images to form the "second conductivity map sequence." Each conductivity map in this sequence directly quantifies the conductivity change caused by localized corrosion, making the identification of corrosion areas and the assessment of corrosion severity more intuitive and accurate.

[0076] For example, continuing with "Structure A", the system arranges the "difference calculation results" obtained each day for 30 days in chronological order, forming a "second conductivity map sequence" containing 30 difference maps. In this new sequence, if a difference map shows that the pixel value of a certain area is continuously negative and the value gradually increases (absolute value), it indicates that the conductivity of that area is continuously decreasing and corrosion is developing.

[0077] This application effectively eliminates the interference of the inherent conductivity distribution of the material on the corrosion signal through this step, greatly improving the sensitivity and accuracy of corrosion detection.

[0078] 105. When the change in conductivity corresponding to the conductivity map in the second conductivity map sequence exceeds a preset threshold, it is determined that corrosion has occurred in the monitoring area corresponding to the conductivity map, and the corrosion result is determined.

[0079] The corrosion results include the corrosion location, corrosion profile, and corrosion stage. The corrosion location refers to the center coordinates of the corrosion area. The corrosion profile refers to the outer boundary of the corrosion area. The corrosion stage refers to the classification based on the severity or development state of the corrosion.

[0080] The corrosion detection threshold is a pre-defined critical value for the change in conductivity, used to determine whether corrosion has occurred at a monitoring point. When the change in conductivity at a monitoring point (usually a negative value, indicating a decrease) falls below this threshold, corrosion is considered to have occurred at that point. A corrosion point is a monitoring point in the conductivity graph where the change in conductivity exceeds the corrosion detection threshold.

[0081] Specifically, step 105 may include: 1051. For each conductivity map in the second conductivity map sequence, the conductivity change corresponding to the conductivity map is compared point by point with a preset corrosion discrimination threshold. When the conductivity change at any monitoring point in the monitoring area exceeds the corrosion discrimination threshold, the monitoring point is marked as a corrosion occurrence point. 1053. By analyzing the set of all marked corrosion sites, the corrosion location, corrosion profile, and corrosion stage determined based on the degree of change in conductivity are identified to form corrosion results.

[0082] Specifically, step 1053 first involves performing spatial clustering analysis on the set of points marked as corrosion sites to determine continuous regions where corrosion occurs. The centroid or geometric center of these continuous regions is taken as the corrosion location, and the outer boundary of these continuous regions is used to form a corrosion profile. Second, the changes in conductivity within these continuous regions are statistically analyzed and compared with a preset corrosion stage classification standard to determine the current corrosion stage. Finally, the identified corrosion locations, corrosion profiles, and determined corrosion stages are integrated to form the corrosion result.

[0083] In this embodiment of the application, firstly, in step 1051, for each conductivity map in the second conductivity map sequence, the conductivity change corresponding to the conductivity map is compared with a preset corrosion discrimination threshold point by point. When the conductivity change of any monitoring point in the monitoring area exceeds the corrosion discrimination threshold, the monitoring point is marked as a corrosion occurrence point.

[0084] The preset corrosion detection threshold is a negative value determined based on material properties, environmental conditions, and empirical data. For example, -5 units means that a decrease in conductivity of 5 units is considered a sign of corrosion.

[0085] Specifically, the system will iterate through each conductivity map in the "second conductivity map sequence" and process all pixels in the map. Conduct a thorough inspection of each monitoring point. The change in its conductivity corrosion It will be compared with the preset corrosion discrimination threshold, Thresholdcorrosion.

[0086] The mathematical expression for this discrimination process is defined as: If corrosion Thresholdcorrosion, thenMark asCorrosionPoint in, corrosion Indicates time t and position The change in conductivity caused by corrosion (derived from step 1043). Thresholdcorrosion represents the preset corrosion discrimination threshold (a negative value). Mark asCorrosionPoint indicates that monitoring points that meet the criteria are marked as corrosion sites.

[0087] If the change in conductivity at a monitoring point is below (i.e., a larger negative value) this threshold, then that monitoring point will be marked as a "corrosion site". This process generates a binary image for each conductivity map, where the marked points represent potential corrosion areas.

[0088] For example, continuing with "Structure A," assuming the preset corrosion detection threshold is -5 units, the system will check each conductivity map in the "Second Conductivity Map Sequence." In the conductivity map of day 20, at a certain monitoring point... The change in conductivity is -7 units. Since -7 < -5, at this point... This was marked as a corrosion site. Another monitoring point... The conductivity change is -3 units. Since -3 is not less than -5, this point is not marked. By comparing the entire conductivity map point by point, the system can identify all monitoring points where the conductivity change meets the corrosion discrimination criteria.

[0089] Secondly, in step 1053, by analyzing the set of all points marked as corrosion sites, the corrosion location, corrosion profile, and corrosion stage determined based on the degree of change in conductivity are identified to form a corrosion result.

[0090] Spatial clustering analysis is a data analysis technique used to group spatially adjacent data points into clusters or regions with similar characteristics. The corrosion stage classification standard is a pre-defined rule that divides the corrosion process into different severity levels based on the varying degrees of change in conductivity.

[0091] The system performs spatial clustering analysis on all sets of points marked as "corrosion occurrence points" in step 1051. This process aims to group spatially close corrosion occurrence points together, thereby identifying contiguous regions composed of adjacent corrosion occurrence points. Commonly used algorithms include density-based clustering algorithms (such as DBSCAN) or connected component analysis. For example, connected component analysis scans all marked points and groups any points that are interconnected (e.g., by sharing edges or corners) into a connected region.

[0092] Once these "continuous regions" are identified, the system will further analyze each region to determine detailed corrosion information: 1. Corrosion Location: For each identified contiguous region, the system calculates its centroid or geometric center. This center point is determined as the precise location of the corroded region.

[0093] 2. Erosion Contour: The outer boundary of the continuous region is extracted, forming the geometric contour of the erosion. This can be achieved using edge detection algorithms in image processing, such as identifying the pixels that constitute the outermost layer of the region.

[0094] Secondly, the system performs statistical analysis on the conductivity changes at all corrosion sites within each continuous region (e.g., calculating average, minimum, or maximum values). These statistical results are then compared to preset corrosion stage classification criteria. For example, if the average conductivity decreases between -5 and -10 units, it may be classified as "early corrosion"; if the decrease is between -10 and -20 units, it is considered "intermediate corrosion"; and if the decrease exceeds -20 units, it is classified as "severe corrosion." These criteria are pre-set based on experimental data and engineering experience.

[0095] Finally, the system integrates the identified corrosion locations, corrosion profiles, and determined corrosion stages to form the final "corrosion result." This result can be presented in a structured data format (such as JSON or XML) or as a visual report, clearly indicating the location, extent, and severity of the corrosion.

[0096] For example, continuing with "Structure A," the system identified a set of corrosion sites in the conductivity graph on day 20. Through spatial clustering analysis, the system found that these sites formed two independent continuous regions within a specific area of ​​the welded joint, named "Region 1" and "Region 2," respectively. For "Region 1," the system calculated its geometric center as (X_A, Y_A) and extracted its outer boundary as the corrosion profile. Simultaneously, the system statistically analyzed the conductivity changes of all corrosion sites within "Region 1," finding an average decrease of -12 units. Based on the preset corrosion stage classification criteria (e.g., -10 to -20 units for intermediate corrosion), "Region 1" was determined to be in the "intermediate corrosion stage." Similarly, the location, profile, and corrosion stage of "Region 2" were determined. Finally, this information was integrated into a detailed corrosion results report.

[0097] This application achieves automated, precise identification, location, and classification of corrosion through this step. By setting discrimination thresholds and performing spatial cluster analysis, the system can accurately extract corrosion areas from complex conductivity variation data and quantify their severity. This greatly improves the efficiency and reliability of corrosion monitoring, providing crucial and actionable information for structural safety assessments and maintenance decisions, thereby effectively avoiding potential structural failure risks.

[0098] Figure 4 This application provides a schematic diagram of a specific embodiment of an in-situ monitoring system for monitoring the corrosion behavior of weathering steel welded joints, with reference to... Figure 4 The system may include: The acquisition module 41 is used to acquire electrochemical impedance measurement values ​​corresponding to multiple monitoring areas, including the surface of the weathering steel welded joint and a preset range area of ​​the joint. The construction module 42 is used to pre-establish a model for describing the conductivity distribution corresponding to multiple monitoring areas. The model is solved by taking the electrochemical impedance measurement value corresponding to each monitoring area as the boundary condition input, and the conductivity distribution of the corresponding monitoring area is output through the model to construct a conductivity map. The extraction module 43 is used to repeatedly perform the acquisition and solution process at preset time intervals to form a first conductivity map sequence, and to analyze the first conductivity map sequence to extract key change components that characterize the change of conductivity over time. The generation module 44 is used to reconstruct the conductivity background characterizing the uncorroded state based on the key change components, and to perform background subtraction on each conductivity map in the first conductivity map sequence based on the conductivity background to obtain a second conductivity map sequence. The second conductivity map sequence contains multiple conductivity maps whose conductivity changes due to the corrosion state. The determination module 45 is used to determine that corrosion has occurred in the monitoring area corresponding to the conductivity map when the change in conductivity corresponding to the conductivity map in the second conductivity map sequence exceeds a preset threshold, and to determine the corrosion result.

[0099] The in-situ monitoring system for monitoring the corrosion behavior of weathering steel welded joints in this application embodiment is used to implement the aforementioned in-situ monitoring method for monitoring the corrosion behavior of weathering steel welded joints. Therefore, the specific implementation of the in-situ monitoring system for monitoring the corrosion behavior of weathering steel welded joints can be found in the embodiment section of the in-situ monitoring method for monitoring the corrosion behavior of weathering steel welded joints mentioned above. The specific implementation can be referred to the description of the corresponding embodiments, and will not be repeated here.

[0100] This application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of the in-situ monitoring method for monitoring the corrosion behavior of weathering steel welded joints as described above.

[0101] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the in-situ monitoring method for monitoring the corrosion behavior of weathering steel welded joints as described above.

[0102] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, random access memory, portable hard drives, magnetic disks, or optical disks.

[0103] Embodiments of the present invention also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above embodiments of the in-situ monitoring method for monitoring the corrosion behavior of weathering steel welded joints.

[0104] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0105] The above provides a detailed description of an in-situ monitoring method and system for monitoring the corrosion behavior of weathering steel welded joints. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. An in-situ monitoring method for monitoring corrosion behavior of a weathering steel welded joint, characterized in that, The method comprises the following steps: collecting electrochemical impedance measurement values corresponding to a plurality of monitoring areas, the plurality of monitoring areas including a surface of a weathering-resistant steel welded joint and a preset range area of the joint; pre-establishing a model for describing the conductivity distribution of the plurality of monitoring areas, inputting the electrochemical impedance measurement value corresponding to each monitoring area as a boundary condition to solve the model, and outputting the conductivity distribution of the corresponding monitoring area through the model to construct a conductivity map; repeating the collection and solving process at a preset time interval to form a first conductivity map sequence, and analyzing the first conductivity map sequence to extract a key change component representing the change of conductivity over time; reconstructing a conductivity background representing a non-corrosion state based on the key change component, and performing background subtraction on each conductivity map in the first conductivity map sequence based on the conductivity background to obtain a second conductivity map sequence, the second conductivity map sequence including a plurality of conductivity maps in which the conductivity change amount is changed due to the corrosion state; when the conductivity change amount corresponding to the conductivity map in the second conductivity map sequence exceeds a preset threshold, it is determined that the monitoring area corresponding to the conductivity map has corrosion, and the corrosion result is determined.

2. The method of claim 1, wherein, The pre-established model for describing the electric field distribution of the monitoring area comprises: based on the geometric shape, material properties and layout of the electrode array of the monitoring area, a mathematical expression is constructed, which can reflect the transmission law of electric charge in the monitoring area; discretize the mathematical expression to form an equation group containing to-be-determined conductivity parameters, wherein each to-be-determined conductivity parameter corresponds to a key position in the monitoring area; by preprocessing the equation group, a mapping relationship between the electrochemical impedance measurement value and the internal conductivity distribution of the monitoring area is established to form a model for describing the electric field distribution of the monitoring area.

3. The method of claim 2, wherein, The method comprises the following steps: in each monitoring period, the electrochemical impedance measurement value is used as the boundary condition of the mathematical expression; using an iterative optimization method, the to-be-determined conductivity parameters in the equation group are adjusted under the premise of meeting the boundary condition until the equation group reaches a convergence state; arrange the converged to-be-determined conductivity parameters according to their positions in the monitoring area to form a two-dimensional data matrix, and use the two-dimensional data matrix as the conductivity map of the corresponding monitoring area.

4. The method of claim 1, wherein, The method comprises the following steps: within a preset monitoring time interval, the collection of electrochemical impedance measurement values and the construction of conductivity maps are repeatedly performed to obtain a series of conductivity maps arranged in time sequence to form a first conductivity map sequence; data processing is performed on each of the first sequence of conductivity maps to identify and quantify a characteristic component in the conductivity map that is indicative of corrosion behavior over time, the characteristic component being a key variation component representing conductivity variation over time.

5. The method of claim 1, wherein, The conductivity background representing the non-corrosion state is reconstructed based on the key variation component, and background subtraction is performed on each of the first sequence of conductivity maps based on the conductivity background to obtain a second sequence of conductivity maps, including: a reference image representing the conductivity distribution of the monitoring area under the non-corrosion condition is constructed as the conductivity background by a data modeling method using the key variation component; a pixel-by-pixel difference operation is performed between each of the first sequence of conductivity maps and the conductivity background to obtain a difference operation result; the difference operation result forms a second sequence of conductivity maps, and each of the second sequence of conductivity maps reflects the amount of conductivity variation caused by local corrosion in the monitoring area.

6. The method of claim 1, wherein, The corrosion result includes the corrosion position, the corrosion profile, and the corrosion stage; when the conductivity variation amount corresponding to the conductivity map in the second sequence of conductivity maps exceeds the preset threshold, it is determined that corrosion occurs in the monitoring area corresponding to the conductivity map, and the corrosion result is determined, including: For each of the second sequence of conductivity maps, the conductivity variation amount corresponding to the conductivity map is compared with the preset corrosion discrimination threshold point by point, and when the conductivity variation amount of any monitoring point in the monitoring area exceeds the corrosion discrimination threshold, the monitoring point is marked as a corrosion occurrence point; By analyzing all the sets of marked corrosion occurrence points, the corrosion position, the corrosion profile, and the corrosion stage determined according to the degree of conductivity variation amount are identified to form the corrosion result.

7. The method of claim 6, wherein, By analyzing all the sets of marked corrosion occurrence points, the corrosion position, the corrosion profile, and the corrosion stage determined according to the degree of conductivity variation amount are identified to form the corrosion result, including: spatial clustering analysis is performed on the sets of marked corrosion occurrence points to determine a continuous area where corrosion occurs, and the centroid or geometric center of the continuous area is taken as the corrosion position, and the outer boundary of the continuous area forms the corrosion profile; the conductivity variation amount in the continuous area is counted and compared with the preset corrosion stage division standard to determine the corrosion stage; the identified corrosion position, corrosion profile, and determined corrosion stage are integrated to form the corrosion result.

8. An in-situ monitoring system for monitoring corrosion behavior of a weathering steel welded joint, characterized in that, including: a collection module for collecting electrochemical impedance measurement values corresponding to a plurality of monitoring areas, the plurality of monitoring areas including the surface of a weathering steel welded joint and a preset range area of the joint; a construction module for pre-establishing a model for describing the conductivity distribution of the plurality of monitoring areas, inputting the electrochemical impedance measurement value corresponding to each monitoring area as a boundary condition to solve the model, and outputting the conductivity distribution of the corresponding monitoring area through the model to construct a conductivity map; The extraction module is configured to repeatedly implement the acquisition and solution process at preset time intervals to form a first conductivity map sequence, and analyze the first conductivity map sequence to extract a key change component representing the change of conductivity over time. The generation module is configured to reconstruct a conductivity background representing the uncorroded state based on the key change component, and perform background subtraction on each conductivity map in the first conductivity map sequence based on the conductivity background to obtain a second conductivity map sequence, wherein the second conductivity map sequence includes a plurality of conductivity maps in which the amount of conductivity change is changed due to the corroded state. The determination module is configured to determine that a monitoring area corresponding to a conductivity map in the second conductivity map sequence is corroded when the amount of conductivity change corresponding to the conductivity map exceeds a preset threshold, and determine a corrosion result.

9. An electronic device, comprising: The computer program is stored in the computer readable storage medium and is executed by the processor to implement the steps of the in-situ monitoring method for monitoring the corrosion behavior of the weathering steel welded joint according to any one of claims 1 to 7. The computer program is stored in the computer readable storage medium and is executed by the processor to implement the steps of the in-situ monitoring method for monitoring the corrosion behavior of the weathering steel welded joint according to any one of claims 1 to 7. ​ 10. A computer-readable storage medium, characterized in that, ​

Images