Method for quantifying corrosion sensitivity and synergistic corrosion effect of steel anchor box plate

Abstract

The invention provides a steel anchor box plate corrosion sensitivity and synergistic corrosion effect quantification method. The method comprises the steps that a cellular automaton model is adopted for simulation, and a geometric model of a corrosion plate is obtained; a random traffic flow load spectrum is generated based on MATLAB platform programming, and cable force time history data of the stay cable are obtained through an influence line dynamic loading technology; cable force time history data are imported into finite element software ABAQUS, fatigue life calculation and analysis of the steel anchor box are carried out in combination with fatigue analysis software FE-SAFE, and then the influence of plate corrosion on the overall fatigue performance of the anchor box is quantitatively evaluated. According to the invention, the simulation of a corrosion-load-structure response full-random process can be realized, and the nonlinear accelerated degradation effect of the multi-plate synergistic corrosion of the steel anchor box on the fatigue life is accurately quantified; the structural fatigue behavior under the combined action of corrosion damage and complex random load in an actual service environment can be simulated more truly, and the engineering credibility and predictive capacity of an evaluation result are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of bridge structure durability technology, specifically to a method for evaluating the corrosion effect of a steel anchor box-type cable-stayed beam anchorage structure. Background Technology

[0002] As the core force-transmitting component of a cable-stayed bridge structure, the cable-stayed girder anchorage structure plays a crucial role in transferring the static and dynamic loads from the bridge deck to the cables. Its mechanical behavior and fatigue characteristics under cyclic loading directly affect the bridge's service safety and performance. Anchor box structures, as the current mainstream cable-stayed girder anchorage method, are widely used due to their advantages such as clear force transmission paths, convenient construction, high anchorage accuracy, and strong overall aesthetics. However, throughout the anchor box's lifespan, it faces multiple corrosion threats: the corrosion and wear of the cables have a transmission effect on the cable-stayed girder anchorage area; and when the external protective layer of the cables ages or is damaged, chloride-containing media easily penetrate along the cable body and remain inside the anchor box, forming a high-concentration corrosive environment. Such harsh conditions not only induce continuous electrochemical corrosion of metal components but also significantly weaken the load-bearing capacity and fatigue life of the cable-stayed girder anchorage structure. Therefore, conducting research on the fatigue performance of anchor boxes under corrosive environments has significant engineering value.

[0003] Current research largely focuses on the impact of corrosion on individual plate components (such as upper / lower web plates and bottom plates) on the fatigue performance of anchor boxes. However, with increasing service life, anchor box protection systems face multiple degradation risks: after the anti-corrosion coating fails due to aging or mechanical damage, environmental factors such as salt spray penetration and condensate accumulation will exacerbate the overall corrosion process. Of particular concern is the lack of clarity regarding the differences in corrosion sensitivity of key load-bearing plates (such as anchor plates and bearing plates), the quantitative correlation mechanism between multi-plate synergistic corrosion and the overall fatigue life of the structure. Research in this area will help determine anchor box maintenance priorities and clarify the attenuation law of corrosion on anchor box fatigue life. Furthermore, most studies use the uniform thickness reduction method or pitting corrosion to simulate bridge corrosion, which fails to reflect the randomness of corrosion. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method for quantifying the corrosion sensitivity and synergistic corrosion effect of the main load-bearing plates of steel anchor boxes, in view of the above-mentioned problems in the prior art.

[0005] The present invention solves its technical problem by adopting the following technical solution: The present invention provides a method for quantifying the corrosion sensitivity and synergistic corrosion effect of steel anchor box plates. Specifically, it employs a cellular automata model to dynamically simulate the electrochemical corrosion process, generating random corrosion morphology by defining state transition rules between corrosion medium cells, metal cells, and oxide film cells; designs systematic corrosion conditions based on multi-level mass loss rates, covering single-factor independent corrosion and multi-factor synergistic corrosion combinations of key load-bearing plates; integrates Monte Carlo random traffic flow generation and influence line dynamic loading technology, generates random traffic flow load spectra through MATLAB programming, and extracts cable force time history data that conforms to actual traffic parameters through influence line dynamic loading technology; constructs a corrosion-fatigue coupled evaluation framework, imports the geometric model of the corroded plate and the cable force load spectrum into ABAQUS to establish a refined finite element model, and calculates the fatigue life of the steel anchor box using the Brown-Miller multiaxial fatigue criterion and the Morrow mean stress correction algorithm in conjunction with the FE-SAFE platform; and quantitatively outputs the corrosion sensitivity classification conclusions and synergistic corrosion accelerated deterioration laws of the plates through multi-condition comparison.

[0006] The geometric model of the corroded plate is used to subsequently establish the overall finite element model of the steel anchor box with corrosion morphology. This geometric model is a solid model generated by reverse engineering surface reconstruction and geometric smoothing based on the three-dimensional discrete point coordinate point cloud output by cellular automata simulation. Its surface has randomly distributed corrosion pit morphology features.

[0007] The present invention can obtain the geometric model of the corroded plate using the following method: First, the electrochemical corrosion process under a preset mass loss rate is simulated based on a cellular automata model. The simulation terminates when the target mass loss rate is reached, outputting corrosion surface data including three-dimensional discrete point coordinates, maximum corrosion depth, and mass loss rate. Then, the discrete coordinates are imported into the reverse engineering software Geomagic Wrap, and a continuous corrosion morphology surface is reconstructed through surface fitting. Next, the geometric discontinuities caused by local sharp pits in the reconstructed surface are appropriately smoothed to eliminate geometric defects that may cause finite element mesh distortion while preserving the corrosion evolution characteristics. After smoothing, the processed corrosion surface is imported into the three-dimensional modeling software Unigraphics. In NX, the original uncorroded plate solid is assembled. The corroded areas in the original solid are removed using the trim body operation, and the smooth surfaces are seamlessly integrated to generate a solid model with realistic corrosion morphology. Finally, the solid model is imported into the finite element software ABAQUS. For the corrosion feature areas, C3D8R hexahedral reduced integral elements are used to implement local mesh refinement (2mm element size in the thickness direction and 10mm in the planar direction). Non-critical areas are coarsened with mesh. At the same time, the boundary constraints of the original model are inherited to complete the finite element solid modeling of integrated corrosion morphology.

[0008] The present invention can generate random traffic load spectra using the following methods: First, based on the traffic characteristics of the area where the bridge is located, six typical standard fatigue vehicle models that can reflect the characteristics of traffic flow are adopted to clarify the axle load, wheelbase and vehicle type ratio parameters of each type. Second, based on the actual engineering background, the average daily traffic volume of the bridge design (e.g., speed of 80 km / h and average daily traffic volume of 80,000 vehicles / day) is adopted, and a program written in Matlab is used to randomly generate a traffic flow sequence that conforms to the spatial distribution of multiple lanes using the Monte Carlo method.

[0009] The cable force time history data of the cable-stayed bridge described in this invention includes: a time-cable force value sequence, specifically the time point of dynamic vehicle loading, constant load, and the total cable force contribution value of random traffic flow load in each lane.

[0010] The present invention can obtain cable force time history data of stay cables using the following method: The influence line data of the six-lane cable force of the target cable was extracted using Midas Civil. Finally, the load was mapped on the random traffic flow of the two-way six lanes in a step size of 0.5m using the dynamic influence line superposition algorithm. The time-varying cable force contribution value during the entire vehicle operation process was calculated, and parallel computing technology was used to improve the data processing efficiency, thereby generating a high-fidelity random traffic flow load spectrum.

[0011] The present invention can import cable stress time history data into the finite element software ABAQUS using the following method: The cable force time history data generated by MATLAB is defined as a time-varying surface load using the amplitude curve function of ABAQUS and applied directly to the surface of the anchor plate.

[0012] The present invention can use the following method to calculate the fatigue life of steel anchor boxes: To address the localized stress concentration caused by corrosion morphology, the Brown-Miller multiaxial fatigue criterion was used to calculate the equivalent amplitude of the critical plane, and the stress gradient effect caused by corrosion pits was compensated by Morrow mean stress correction. Finally, the fatigue life under different corrosion conditions was quantified by combining the rainflow counting method and Miner's linear cumulative damage theory.

[0013] The present invention can use the following method to analyze the fatigue life calculation results of steel anchor boxes: Through comparative analysis of parametric corrosion conditions, multiple sets of single / synergistic corrosion conditions with different mass loss rates were established, fatigue life attenuation rates were quantified, and corrosion degradation patterns were revealed based on sensitivity ranking.

[0014] The present invention provides a method for quantifying the corrosion sensitivity and synergistic corrosion effect of steel anchor box plates, which is used to evaluate the corrosion effect of steel anchor box cable-stayed bridge anchorage structures and improve the service safety and performance of cable-stayed bridges.

[0015] Compared with the prior art, the present invention has the following main advantages: This invention constructs a complete end-to-end evaluation framework, integrating cellular automata corrosion simulation, 3D corrosion morphology reconstruction, refined finite element modeling, generation of random traffic flow load spectra based on existing traffic parameters, and advanced multiaxial fatigue life analysis. Through dynamic corrosion simulation coupled with random load technology, it achieves the simulation of a fully stochastic process of corrosion-load-structural response, accurately quantifying the nonlinear accelerated degradation effect of multi-plate synergistic corrosion on fatigue life in steel anchor boxes.

[0016] This paper presents an effective method for accurately quantifying the differences in corrosion sensitivity of key load-bearing plates in steel anchor boxes and the impact of synergistic corrosion effects of multiple plates on the overall fatigue life of the anchor box. This integrated approach overcomes the limitations of traditional segmented studies, enabling a more realistic simulation of structural fatigue behavior under the combined effects of corrosion damage and complex random loads in actual service environments. This significantly improves the engineering credibility and predictive ability of the evaluation results. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the finite element model of the entire bridge.

[0018] Figure 2 This is a schematic diagram of cell distribution.

[0019] Figure 3 Figure 1 is a schematic diagram of the cell transformation rules, where: Figure (a) is a schematic diagram of the rule of the diffusion of corrosive media to vacancies, Figure (b) is a schematic diagram of the rule of multiple corrosive media competing for the same target site, Figure (c) is a schematic diagram of the rule of the spatial repulsion effect between corrosive media, Figure (d) is a schematic diagram of the rule of the metal dissolution reaction, Figure (e) is a schematic diagram of the rule of the passivation film damage-corrosion diffusion, and Figure (f) is a schematic diagram of the rule of the passivation film barrier function.

[0020] Figure 4 This is a schematic diagram of the three-dimensional corrosion coordinate points.

[0021] Figure 5 A schematic diagram of the modeling process for a corroded entity is shown, where: Figure (a) is the reconstruction of the corroded surface, and Figure (b) is the corroded entity after smoothing.

[0022] Figure 6 A schematic diagram of mesh generation for corroded solids.

[0023] Figure 7 A schematic diagram is provided for setting the boundary conditions of each component of the anchor box and the box girder segments.

[0024] Figure 8The following are schematic diagrams of fatigue vehicle models: Figure (a) is a 2-axle Class I vehicle, Figure (b) is a 2-axle Class II vehicle, Figure (c) is a 3-axle vehicle, Figure (d) is a 4-axle vehicle, Figure (e) is a 5-axle vehicle, and Figure (f) is a 6-axle vehicle.

[0025] Figure 9 This is a schematic diagram of the axle load distribution of random traffic flow on lane 1.

[0026] Figure 10 This is a schematic diagram of the fatigue life distribution under the baseline undamaged condition.

[0027] Figure 11 Figure 1 shows the fatigue life distribution of the lower support plate under various mass loss rates, where: Figure (a) is at a mass loss rate of 5%, Figure (b) is at a mass loss rate of 10%, Figure (c) is at a mass loss rate of 15%, and Figure (d) is at a mass loss rate of 20%.

[0028] Figure 12 Figure 1 shows the fatigue life distribution of the anchor plate under various mass loss rates, where: Figure (a) is at a mass loss rate of 5%, Figure (b) is at a mass loss rate of 10%, Figure (c) is at a mass loss rate of 15%, and Figure (d) is at a mass loss rate of 20%.

[0029] Figure 13 The following is a schematic diagram of the fatigue life distribution of all major load-bearing plates under synergistic corrosion at various mass loss rates: Figure (a) is at a mass loss rate of 5%, Figure (b) is at a mass loss rate of 10%, Figure (c) is at a mass loss rate of 15%, and Figure (d) is at a mass loss rate of 20%.

[0030] Figure 14 Figure 1 shows the calculation results of corrosion conditions for a single main load-bearing plate. Figure 2 shows the relationship between mass loss rate and fatigue life, and Figure 3 shows the relationship between mass loss rate and fatigue life reduction rate.

[0031] Figure 15 Figure 1 shows the calculation results of the main load-bearing plate under the synergistic corrosion condition. Figure 2 shows the relationship between mass loss rate and fatigue life, and Figure 3 shows the relationship between mass loss rate and fatigue life reduction rate. Detailed Implementation

[0032] The method for quantifying the corrosion sensitivity and synergistic corrosion effect of steel anchor box plates provided by this invention is as follows: a cellular automata model is used to simulate and obtain the geometric model of the corroded plate; a random traffic load spectrum is generated based on the MATLAB platform; and the cable force time history data of the stay cable is obtained through influence line dynamic loading technology. The cable force time history data is imported into the finite element software ABAQUS and combined with the fatigue analysis software FE-SAFE to calculate and analyze the fatigue life of the steel anchor box, thereby quantitatively evaluating the impact of plate corrosion on the overall fatigue performance of the anchor box.

[0033] The method for quantifying the corrosion sensitivity and synergistic corrosion effect of steel anchor box plates provided by this invention includes the following steps: corrosion morphology generation based on cellular automata, random traffic load spectrum construction, corrosion-fatigue coupled life assessment, and sensitivity and synergistic effect quantification. The details are as follows: 1. Corrosion morphology generation based on cellular automata model: A three-dimensional cellular space is established, and the transformation rules of the corrosive medium cell C, metal cell M, and oxide film cell P are defined. Random pit morphology is generated with the mass loss rate as the termination condition. Three-dimensional point cloud data is output, and the subsequent finite element model is established.

[0034] 2. Construction of random traffic flow load spectrum: Based on a typical fatigue vehicle model, a multi-lane random traffic flow sequence is generated using the Monte Carlo method, and cable force time history data is obtained through dynamic loading of influence lines.

[0035] 3. Corrosion-fatigue coupled life assessment: The cable stress time history was imported into the fatigue life calculation software FE-SAFE, and the fatigue life of the anchor box under single / synergistic corrosion conditions was calculated by combining the Brown-Miller multiaxial fatigue criterion and the Morrow mean stress correction.

[0036] 4. Sensitivity and synergistic effect quantification: The results were analyzed to determine the mapping relationship between the mass loss rate and fatigue life of each plate component, identify highly sensitive plates (anchor plates, bearing plates, and lower support plates), and reveal the pattern of synergistic corrosion of multiple plates.

[0037] The present invention will be further described below with reference to specific examples and accompanying drawings, but this description does not limit the scope of the invention.

[0038] 1. For example Figure 1 As shown, a full-bridge model was established using the finite element analysis software Midas Civil. In this model, the structural self-weight, secondary dead load, cable forces under the completed bridge condition, vehicle live load, and temperature changes were applied sequentially to conduct an overall structural stress performance analysis of the bridge. The finite element calculation results showed that the cable force under the most unfavorable load combination was 4435.15 kN. The steel anchor box corresponding to this cable will be the subject of subsequent fatigue performance analysis.

[0039] 2. A cellular automata model was used to obtain the corrosion morphology of key load-bearing plates (including the upper support plate, lower support plate, two end plates, bearing plate, and anchor plate). The size of a single cell was 1mm × 1mm × 1mm. The cellular space was set as a three-dimensional cuboid structure based on the geometry of the anchor box plate, with its planar dimensions consistent with the actual length and width of the plate, and set to twice the thickness of the plate along the thickness direction to accommodate the corrosive environment layer. This cellular space was divided into three regions: corrosion cell region C, metal cell region M, and oxide film cell region P. Region C is distributed within the corrosive environment layer and is used to simulate corrosive media such as salt spray or water vapor; region M constitutes the main body of the metal component; and region P covers the metal surface as a passivation protective layer. The spatial distribution of the three regions is shown in [see figure]. Figure 2 To simplify the CA evolution, it is assumed that the simulation ignores the effects of trace elements, ambient temperature, and other factors, and does not consider the impact of corrosion product formation on subsequent simulations. In each time step of the corrosion simulation, the corrosion cell C randomly selects neighboring cells and migrates its state according to a transition rule. For example... Figure 3 As shown in (a), if the adjacent position of cell C is empty, cell C is moved to that position and the original position is cleared; Figure 3 As shown in (b), when multiple cells C compete for the same target (vacancy or cell M), all competing cells C remain in their original positions, reflecting the random competition of local medium diffusion; as Figure 3 As shown in (c), if the adjacent cells are other cells C, and cell C is stationary, it represents spatial repulsion between the media; as Figure 3 As shown in (d), if the adjacent cell is M, cell C moves to that position and dissolves cell M, simulating the electrochemical anodic reaction; Figure 3 As shown in (e), if the adjacent cells are P and cell C is located inside the metal or at the interface, cell C occupies the P site (cell P state to cell C), simulating the corrosion diffusion caused by passivation film damage; Figure 3 As shown in (f), if the adjacent cell is cell P and cell C is outside the metal, cell C is stationary, reflecting the barrier effect of the passivation film on the medium. The mass loss rate is used as a quantitative indicator of corrosion degree, defined as the percentage of dissolved metal cells to the initial total number of metal cells. The simulation terminates when the mass loss rate reaches preset thresholds of 5%, 10%, 15%, and 20%, and the output is as follows. Figure 4 The key parameters shown, such as the three-dimensional discrete point coordinates of the corroded surface, the maximum corrosion depth, and the mass loss rate, can be used to dynamically verify the corrosion process.

[0040] 3. The 3D reverse engineering software Geomagic Wrap is used to fit the eroded 3D coordinate points into a continuous surface, such as... Figure 5 As shown in (a), appropriate geometric smoothing is applied to sharp local pits to avoid convergence problems caused by mesh distortion in finite element analysis. This process strictly controls the smoothing degree to suppress excessive distortion and ensure the authenticity of the geometric features and evolution laws of corrosion morphology.

[0041] 4. After importing the geometrically smoothed corroded surface into Unigraphics NX 3D modeling software and assembling it with the original uncorroded plate solid, a smoothed corroded morphology solid model is constructed through the trimming volume operation, such as... Figure 5 As shown in (b), the model was imported into the finite element software ABAQUS and integrated into the overall finite element model of the steel anchor box for fatigue analysis. Corrosion-prone areas were locally meshed using C3D8R hexahedral reduced integral elements, with the mesh element size set to 2mm in the thickness direction and 10mm in the planar direction. Non-critical areas used coarser meshes to save computational resources. The mesh generation results are shown in [reference needed]. Figure 6 .

[0042] 5. Using the obtained solid model of the corroded plate, a refined model of the steel anchor box girder segment of the cable-stayed bridge was implemented in ABAQUS finite element software. Since the cross-section of the girder segment is symmetrical along the bridge centerline, a half-section modeling strategy was adopted to improve computational efficiency. In the modeling of the steel anchor box structure, when the thickness-to-width ratio of the plate exceeds 1:10, shell element simulation is prone to distortion due to the lack of stress gradient in the thickness direction and cannot accurately represent the complex stress state of the weld. Therefore, solid element modeling was used for the steel anchor box structure, while shell element modeling was used for the remaining plates of the girder segment, balancing model accuracy and computational load control. Fixed constraints were set on both sides of the girder segment, and symmetrical constraints were applied to the center surface. The boundary settings for the anchor box components and box girder segments are detailed in [link to documentation]. Figure 7 N1-N9 correspond to the upper and lower support plates, pressure plate, stiffening plate, a-side end plate, b-side end plate, anchor plate, side stiffening plate, and cable anchor pipe, respectively; where the a-side end plate corresponds to the far side of the box girder web, and the b-side end plate corresponds to the near side of the web. Welding between the plates is simulated through constraint binding. Q345C steel is used, with an elastic modulus of [missing value]. MPa, Poisson's ratio is taken as 0.3, and tensile strength is 550MPa.

[0043] 6. Based on a cellular automata model, corrosion morphology entities of key load-bearing plates are generated, covering the upper support plate, lower support plate, bearing plate, two end plates, and anchor plates. A multi-condition corrosion damage finite element model is constructed through parameter-driven assembly, and a set of comparative working conditions is systematically established using single-factor and multi-factor combined design methods. Details are as follows: First, a baseline non-destructive model was constructed to simulate the condition of an intact anchor box. Second, independent corrosion models were set up for six types of key load-bearing plates, each with four levels of mass loss rate (5%, 10%, 15%, and 20%), resulting in 24 single-factor working conditions. Subsequently, a multi-plate collaborative corrosion model was further established, including three modes: support plate combination, double-sided end plate combination, and simultaneous corrosion of all plates. Each mode also had four levels of mass loss rate, resulting in 12 multi-factor working conditions. Finally, a total of 37 comparative working conditions were constructed, and equivalent boundary constraints were applied to all models to ensure comparability of mechanical responses.

[0044] 7. A MATLAB program was developed to generate random traffic flow and dynamically load influence lines, acquiring time-varying load data on the stay cables throughout the vehicle passage process, providing a load input basis for fatigue damage assessment. Influence line data for six lanes was extracted using Midas Civil, combined with... Figure 8 The data shown reflects the traffic flow characteristics of the existing highway cable-stayed bridge using six standard fatigue vehicle model data, covering axle load and wheelbase. The simulation assumes uniform vehicle speed, using the bridge's design speed of 80 km / h, and a daily average traffic flow of 80,000 vehicles / day (based on reports from local authorities). The simulation step size is set to 0.5 meters. Parallel computing technology is applied to significantly improve data processing efficiency. A single-day multi-lane spatially distributed random traffic flow sequence is generated using the Monte Carlo method, such as... Figure 9 The generated random traffic flow redistribution on lane 1 for a single day is shown. The influence line superposition algorithm was used to accurately calculate the time-varying cable force contribution of the vehicle load on the target cable in both directions of the six lanes, with a total calculation step of 1,472,961 steps. This bidirectional loading mechanism effectively characterizes the spatial distribution characteristics of actual traffic loads, ensuring the accuracy of bridge dynamic response simulation. The cable force time history was treated as a time-varying load using the finite element software ABAQUS and applied as a surface load to the anchor plate.

[0045] This invention integrates 36 sets of finite element models of steel anchor box girder segments with corrosion damage and daily random traffic flow cable load spectra, and utilizes the compatibility of the fatigue analysis software FE-SAFE and ABAQUS to calculate the fatigue life of the steel anchor box. Basic mechanical parameters of Q345C steel, including Poisson's ratio, tensile strength, and elastic modulus, are defined on the FE-SAFE platform, and the material SN curve is automatically generated. The Brown-Miller multiaxial fatigue criterion is used to calculate the equivalent amplitude, and the Morrow criterion is used to correct the mean stress. The critical load-bearing plate with the shortest fatigue life in the steel anchor box system, i.e., the component with significant stress concentration and the highest risk of crack initiation, is used as the fatigue life assessment standard for the overall steel anchor box structure. This evaluation method obtains the fatigue life distribution patterns under all 37 working conditions, thereby quantifying the differences in corrosion sensitivity of plates and the synergistic effect mechanism.

[0046] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described below in conjunction with the accompanying drawings.

[0047] Compared to simplified corrosion simulation methods that rely on uniform thickness reduction or pre-defined geometric pitting, this invention employs a cellular automata model based on the fundamental physical mechanisms of electrochemical corrosion. It defines the corrosive medium cell C, the metal substrate cell M, and the passivation film cell P, as well as the state transition rules between cells (see details). Figure 3This model dynamically and self-organizes the evolution of corrosion damage with highly spatial heterogeneity and complex three-dimensional morphology through local, stochastic cellular interaction processes (such as dielectric diffusion, metal dissolution, and passivation film destruction). Figure 4 The three-dimensional discrete point cloud shown and Figure 5 The simulated entity reconstruction results are shown. This simulation method fundamentally overcomes the morphological distortion problem caused by artificially pre-setting corrosion morphology in traditional methods, and can more realistically reproduce the non-uniform distribution of corrosion pits, random pit sizes, and irregular pit contours formed by the combined effects of local microenvironment differences, material inhomogeneity, and random factors during the actual corrosion process. Therefore, the corrosion geometry generated by the CA model (such as...) Figure 5 (b) This significantly improves the geometric fidelity of the subsequently constructed finite element model of the corroded plate, providing a geometric basis that is closer to the actual damage state in engineering and reflects the characteristics of random corrosion for fatigue life analysis on the ABAQUS and FE-SAFE joint platform. This fundamentally improves the physical rationality and predictive reliability of the corrosion-fatigue coupled life assessment results.

[0048] The fatigue life distribution of steel anchor boxes under the benchmark non-destructive working condition is as follows: Figure 10 As can be seen, the lifespan result output by LOGLife represents the fatigue life. N The logarithmic form of 1, lgN, gives the fatigue life of the steel anchor box under this condition as: , The study of fatigue life distribution of steel anchor boxes under 37 working conditions shows that the key fatigue weak areas in all working conditions are concentrated in the weld ends of the upper support plate, the lower support plate, and the box girder, as well as the area around the anchor holes of the anchor plate. Other plates have a relatively low risk of fatigue failure. The following are examples of four working conditions for the lower support plate at various mass loss rates: Figure 11It is evident that, except for the support plate and anchor plate, the other plates did not exhibit fatigue failure risk. The weld end connecting the lower support plate to the box girder, bearing the stress caused by the enormous alternating cable force, is the most susceptible area to fatigue failure; therefore, the fatigue life of this area characterizes the overall fatigue life of the anchor box structure. The area around the anchor holes of the anchor plate, due to geometrical abrupt changes leading to significant stress concentration and the fact that the anchor plate is under triaxial stress during actual loading, is also a location prone to fatigue failure. The fatigue lives of the steel anchor box with mass loss rates of 5%, 10%, 15%, and 20% were 114.47 years, 98.33 years, 87.04 years, and 74.59 years, respectively. Compared to the non-corrosion condition, the fatigue life of the steel anchor box decreased by 11.69%, 24.15%, 32.86%, and 42.46%, respectively. With increasing mass loss rate, the fatigue life of the steel anchor box exhibits a non-linear degradation effect. The mass loss rate of the lower support plate exhibits a significant, nonlinear, negative correlation with the overall fatigue life of the anchor box. This result is visually presented using a LOGLife logarithmic life cloud map, which identifies highly corrosion-sensitive areas and reveals the degradation mechanism of corrosion damage on the anchor box's fatigue life, providing crucial information for anchor box protection design.

[0049] Figure 12 The fatigue life distribution of the anchor plate under various mass loss rates (5%, 10%, 15%, and 20%) is shown. The results indicate that the corrosion sensitivity of the anchor plate is higher than that of the lower support plate. The fatigue life of the steel anchor box at mass loss rates of 5%, 10%, 15%, and 20% are 107.82 years, 85.65 years, 72.39 years, and 61.33 years, respectively. Compared with the non-corrosion condition, the fatigue life of the steel anchor box decreases by 16.82%, 33.93%, 44.16%, and 52.69%, respectively, indicating that the corrosion of the anchor plate has a stronger impact on the fatigue life.

[0050] Figure 13The fatigue life distribution of all major load-bearing plates under synchronous corrosion at the same mass loss rate is presented. The analysis results show that the synergistic corrosion of the major load-bearing plates significantly damages the fatigue performance of the steel anchor box. When the mass loss rate is 5%, 10%, 15%, and 20%, the fatigue life of the steel anchor box is 89.48 years, 73.02 years, 51.17 years, and 23.91 years, respectively. Compared with the uncorroded condition, the fatigue life reduction rates are 30.97%, 43.67%, 60.53%, and 81.56%, respectively. This reveals the effect of multi-plate synergistic corrosion leading to a nonlinear accelerated deterioration of fatigue life. For example, at a 20% mass loss rate, the fatigue life reduction rate caused by synergistic corrosion is as high as 81.56%, significantly higher than the 52.69% reduction rate when the most sensitive plate, the anchor pad, is corroded alone. This comparison indicates a significant mutual aggravation effect between the corrosion damage of different plates, and its combined effect far exceeds the damage effect of a single plate. This method effectively quantifies this key synergistic effect.

[0051] like Figure 14 As shown, the results indicate significant differences in corrosion susceptibility among different plates. The corrosion susceptibility of the main load-bearing plates, ranked from highest to lowest, is as follows: anchor plate, lower support plate, bearing plate, upper support plate, and end plate. Anchor plate corrosion has the most significant deterioration effect on the fatigue life of the steel anchor box compared to other main load-bearing plates. When the mass loss rate increases from 5% to 20%, the decrease rate increases from 16.82% to 52.69%, revealing the significant destructive impact of anchor plate corrosion on fatigue performance. Furthermore, corrosion of the lower support plate and bearing plate also induces significant life attenuation, with degradation rates of 11.69%–42.46% and 9.64%–33.47%, respectively, and their impact cannot be ignored. In comparison, the corrosion of the upper support plate and the a and b side end plates has a relatively small impact on fatigue life. When the mass loss rate increases from 5% to 20%, the fatigue life of the anchor box decreases by 5.70%~19.05%, 1.68%~11.75%, and 3.50%~14.01% respectively compared to the uncorroded condition. This difference was quantified through parameter sensitivity analysis. Based on the above patterns, it is recommended that the durability protection design of steel anchor boxes prioritize prevention and control strategies for the anchor plate, bearing plate, and lower support plate. Optimizing the drainage structure at the bottom of the anchor box is crucial to eliminate electrochemical corrosion inducing factors, i.e., the bottom water accumulation area, which corresponds to the corrosion-sensitive plates of the lower support plate, bearing plate, and anchor plate. This will help optimize the allocation of maintenance resources and improve the safety and economy of the anchor box throughout its entire life cycle.

[0052] like Figure 15Analysis of the multi-plate synergistic corrosion conditions shows that when the mass loss rate of the end plates increases from 5% to 20%, the fatigue life of the steel anchor box decreases from 119.79 years to 93.42 years, with the rate of decrease increasing from 7.59% to 27.93%, further verifying that the corrosion of the end plates has a relatively small impact on the fatigue life of the anchor box. However, when the upper and lower support plates corrode simultaneously at a 5% mass loss rate, the fatigue life of the steel anchor box is 98.95 years, but this decreases significantly to 57.84 years when the mass loss rate is 20%, indicating that the corrosion of the support plates has a strong destructive effect on the overall structural fatigue performance of the steel anchor box. When the main load-bearing plates of the anchor box corrode simultaneously at a 20% mass loss rate, their fatigue life is 23.91 years, with a rate of decrease of 81.56%. At this point, the overall structure of the anchor box no longer possesses safety. Based on the above analysis, it can be concluded that the synergistic corrosion of the main load-bearing plates of the steel anchor box significantly accelerates the structural failure process.

[0053] Tables 1-4 provide detailed data for each corrosion condition. Table 1 shows the relationship between fatigue life and mass loss rate of steel anchor boxes under single plate corrosion conditions; Table 2 shows the relationship between fatigue life reduction rate and mass loss rate of steel anchor boxes under single plate corrosion conditions; Table 3 shows the relationship between fatigue life and mass loss rate of steel anchor boxes under synergistic corrosion conditions; and Table 4 shows the relationship between fatigue life reduction rate and mass loss rate of steel anchor boxes under synergistic corrosion conditions.

[0054] The method for quantifying the corrosion sensitivity and synergistic corrosion effect of the main load-bearing plates of steel anchor boxes provided by this invention has the following technical innovations: 1. Systematically quantify the accelerated degradation mechanism of fatigue life caused by multi-plate synergistic corrosion. By designing a multi-level working condition comparison covering independent corrosion of single plates and combined corrosion of multiple plates, it is revealed that the attenuation intensity of synergistic corrosion on the overall fatigue performance of the anchor box is much greater than that of single plate corrosion, and the anchor plate, lower support plate and bearing plate are identified as highly corrosion-sensitive plates. 2. By integrating cellular automata corrosion modeling, Monte Carlo random traffic flow generation, influence line dynamic loading cable force technology, and life analysis based on multi-axis fatigue criteria, the system achieves multi-dimensional precise coupling of corrosion morphology randomness, traffic load randomness, and structural fatigue response. 3. Output a corrosion sensitivity classification system for key load-bearing plates and locate fatigue-sensitive areas, establish the conclusion that the anchor plate is the most sensitive plate, accurately identify the weld ends and the area around the anchor holes as the core area for fatigue crack initiation, and provide a direct theoretical basis for the protective design of steel anchor boxes.

[0055] Table 1. Relationship between fatigue life and mass loss rate of steel anchor boxes under single-plate corrosion conditions.

[0056] Table 2. Relationship between fatigue life reduction rate and mass loss rate of steel anchor boxes under single-plate corrosion conditions.

[0057] Table 3. Relationship between fatigue life and mass loss rate of steel anchor boxes under synergistic corrosion conditions.

[0058] Table 4. Relationship between fatigue life reduction rate and mass loss rate of steel anchor boxes under synergistic corrosion conditions.

Claims

1. A method for quantifying the corrosion sensitivity and synergistic corrosion effect of steel anchor box plates, characterized in that, A cellular automata model was used to dynamically simulate the electrochemical corrosion process. Random corrosion morphology was generated by defining state transition rules between corrosion medium cells, metal cells, and oxide film cells. Systematic corrosion conditions were designed based on multi-level mass loss rates, covering single-factor independent corrosion and multi-factor synergistic corrosion combinations of key load-bearing plates. Monte Carlo random traffic flow generation and influence line dynamic loading technology were integrated. Random traffic flow load spectra were generated through programming on the MATLAB platform, and cable force time history data that conformed to actual traffic parameters were extracted through influence line dynamic loading technology. A corrosion-fatigue coupled evaluation framework was constructed. The geometric model of the corroded plate and the cable force load spectrum were imported into ABAQUS to establish a refined finite element model. The fatigue life of the steel anchor box was calculated by combining the FE-SAFE platform with the Brown-Miller multiaxial fatigue criterion and the Morrow mean stress correction algorithm. The corrosion sensitivity classification conclusion and synergistic corrosion accelerated deterioration law of the plate were quantitatively output through multi-condition comparison.

2. The method for quantifying the corrosion sensitivity and synergistic corrosion effect of steel anchor box plates according to claim 1, characterized in that, The geometric model of the corroded plate is a solid model generated by reverse engineering surface reconstruction and geometric smoothing based on the three-dimensional discrete point coordinate point cloud output by cellular automata simulation. Its surface has randomly distributed corrosion pit morphology features.

3. The method for quantifying the corrosion sensitivity and synergistic corrosion effect of steel anchor box plates according to claim 2, characterized in that, The geometric model of the corroded plate was obtained using the following method: First, the electrochemical corrosion process under a preset mass loss rate is simulated based on a cellular automata model. The simulation terminates when the target mass loss rate is reached, and the output includes corrosion surface data containing three-dimensional discrete point coordinates, maximum corrosion depth, and mass loss rate. Then, the discrete coordinate points are imported into the reverse engineering software Geomagic Wrap, and the continuous corrosion morphology surface is reconstructed through surface fitting. Next, the geometric discontinuities caused by local sharp pits in the reconstructed surface are appropriately smoothed to eliminate geometric defects that may cause finite element mesh distortion while preserving the corrosion evolution characteristics. After smoothing, the processed eroded surface is imported into the 3D modeling software Unigraphics NX and assembled with the original uncorroded plate solid. The eroded area in the original solid is removed by trimming the body operation and the smoothed surface is seamlessly integrated to generate a solid model with a realistic erosion morphology. Finally, the solid model is imported into the finite element software ABAQUS. Local mesh refinement is implemented for the eroded feature areas using C3D8R hexahedral reduced integral elements, and the non-critical areas are coarsened. At the same time, the boundary constraints of the original model are inherited to complete the finite element solid modeling of the integrated erosion morphology.

4. The method for quantifying the corrosion sensitivity and synergistic corrosion effect of steel anchor box plates according to claim 1, characterized in that, The random traffic load spectrum is generated using the following method: First, based on the traffic characteristics of the area where the bridge is located, six typical standard fatigue vehicle models that can reflect the characteristics of traffic flow are adopted to clarify the axle load, wheelbase and vehicle type ratio parameters of each vehicle type. Second, based on the actual engineering background, the average daily traffic volume of the bridge design is used to randomly generate traffic flow sequences that conform to the spatial distribution of multiple lanes by writing a program in Matlab and combining the Monte Carlo method.

5. The method for quantifying the corrosion sensitivity and synergistic corrosion effect of steel anchor box plates according to claim 1, characterized in that, The cable force time history data of the cable-stayed bridge includes: time-cable force value sequence, specifically the time point of vehicle dynamic loading, constant load, and the total cable force contribution value of random traffic load in each lane.

6. The method for quantifying the corrosion sensitivity and synergistic corrosion effect of steel anchor box plates according to claim 5, characterized in that, The following method was used to obtain the cable force time history data of the stay cables: The influence line data of the six-lane cable force of the target cable was extracted using Midas Civil. Finally, the load was mapped on the random traffic flow of the two-way six lanes with a step size of 0.5m using the dynamic influence line superposition algorithm. The time-varying cable force contribution value during the entire vehicle operation process was calculated, and parallel computing technology was used to improve the data processing efficiency, thereby generating a high-fidelity random traffic flow load spectrum.

7. The method for quantifying the corrosion sensitivity and synergistic corrosion effect of steel anchor box plates according to claim 6, characterized in that, The cable stress time history data were imported into the finite element software ABAQUS using the following method: The cable force time history data generated by MATLAB is defined as a time-varying surface load using the amplitude curve function of ABAQUS and applied directly to the surface of the anchor plate.

8. The method for quantifying the corrosion sensitivity and synergistic corrosion effect of steel anchor box plates according to claim 1, characterized in that, The fatigue life of the steel anchor box is calculated using the following method: To address the localized stress concentration caused by corrosion morphology, the Brown-Miller multiaxial fatigue criterion was used to calculate the equivalent amplitude of the critical plane, and the stress gradient effect caused by corrosion pits was compensated by Morrow mean stress correction. Finally, the fatigue life under different corrosion conditions was quantified by combining the rainflow counting method and Miner's linear cumulative damage theory.

9. The method for quantifying the corrosion sensitivity and synergistic corrosion effect of steel anchor box plates according to claim 8, characterized in that, The fatigue life calculation results of the steel anchor box were analyzed using the following method: Through comparative analysis of parametric corrosion conditions, multiple sets of single / synergistic corrosion conditions with different mass loss rates were established, fatigue life attenuation rates were quantified, and corrosion degradation patterns were revealed based on sensitivity ranking.

10. The method for quantifying the corrosion sensitivity and synergistic corrosion effect of steel anchor box plates according to any one of claims 1 to 9, characterized in that, Used to evaluate the corrosion effects of steel anchor box girder anchorage structures, thereby improving the service safety and performance of cable-stayed bridges.

Images