Steel plate pitting corrosion simulation method and system based on cellular automaton model

Abstract

The invention relates to the technical field of computer aided design, in particular to a steel plate pitting corrosion simulation method and system based on a cellular automaton model.The method comprises the steps that all slope curve segments corresponding to all corrosion pit morphology curves are obtained, and the directional variation index of each depth position is determined according to the depth position of the slope curve segments; acquiring a plurality of slope areas in the corrosion pit under different corrosion time with different solution concentrations to analyze the abnormal corrosion condition, and determining the local corrosion progressive coefficient of each depth position by combining the expansion coefficient; obtaining the number of intersections between the depth positions of each type of slopes and each type of local erosion progressive coefficients, and generating a dynamic rule of evolution of the cells to the corresponding type of slopes in combination with a directional variation index; on the basis of a dynamic rule, the cellular automaton model with the optimal goodness of fit is obtained by training the LSTM network, the problem of directional evolution under a finite element rule is avoided, and powerful help is provided for researching the real pitting process of the steel plate.

Description

Technical Field

[0001] The present invention relates to the technical field of computer-aided design, and particularly relates to a method and system for simulating pitting corrosion of steel plates based on a cellular automata model. Background Art

[0002] Metal corrosion damage is a major focus in the durability research of steel-concrete composite structures in civil engineering, and pitting corrosion is the main form of steel plate corrosion. Pitting corrosion has the characteristics of concealment and suddenness. Especially in industries such as construction, bridges, and nuclear power, pitting corrosion is likely to cause perforation of steel plates, thereby leading to stress concentration, a sharp decline in bearing capacity, and even catastrophic accidents such as collapse. Therefore, it is necessary to simulate and predict the pitting corrosion situation. Traditional experimental methods are costly and time-consuming, and it is difficult to dynamically observe the evolution process of pitting corrosion. Therefore, in recent years, due to its discretization and rule-driven characteristics, cellular automata (CA) have been widely used to simulate the mesoscopic dynamic behavior of metal corrosion.

[0003] The simulation of pitting corrosion of steel plates based on the cellular automata model focuses on simulation and the mesoscopic scale. The metal and the corrosion environment are divided into multiple "cells", and each cell updates its state according to local rules (such as electrochemical reactions and stress distributions), which can respectively simulate the development processes on the pitted surface and in the thickness direction of the steel, and are used to predict the in-service life of the structure.

[0004] During the evolution of pitting corrosion of steel plates by cellular automata, there may be dynamic competition between complex rules for each cell. Because the initiation and propagation of pitting corrosion involve multiple process competitions such as electrochemical dissolution, stress concentration, and passivation film rupture, multiple rule evolutions may exist simultaneously in a single cell. Under the finite element rules, it is insufficient to show the random influence of various factors during the corrosion process, resulting in a fixed evolution direction and being unable to describe the irregular evolution of pitting corrosion in time and space, which is not conducive to the simulation of pitting corrosion of steel plates. Summary of the Invention

[0005] In order to solve the technical problem that the evolution direction of the pitted corrosion pit in the existing pitting corrosion simulation is a fixed mode and cannot describe its irregular evolution in time and space, which is not conducive to the simulation of pitting corrosion of steel plates, the purpose of the present invention is to provide a method and system for simulating pitting corrosion of steel plates based on a cellular automata model. The specific technical solutions adopted are as follows: An embodiment of the present invention provides a method for simulating pitting corrosion of steel plates based on a cellular automata model, and the method includes the following steps: Obtain, from a CA-FEM coupling model, each inclined curve segment corresponding to the pitted surface morphology curves at several corrosion times in a stress-free environment and a tensile stress environment generated by the CA model, where the CA rules are driven by the tensile stress field calculated by FEM; Analyze the morphological anomaly conditions of the inclined surface curve segments at different depth positions and the rule constraint conditions at different depth positions according to the depth positions of the respective inclined surface curve segments, and determine the directional mutation index for each depth position; Determine the expansion coefficient corresponding to the corrosion pit surface at each two consecutive corrosion times for the same solution concentration; analyze the corrosion anomaly conditions based on all inclined surface regions inside the corrosion pits at different corrosion times for several different solution concentrations, and combine the expansion coefficient to determine the local erosion progression coefficient for each depth position under different solution concentrations; Obtain the intersection quantity between each type of inclined surface and the depth positions of each type of the local erosion progression coefficient, and generate the dynamic rule for each cell to evolve towards the corresponding type of inclined surface in combination with the directional mutation index for each depth position; Based on the dynamic rule, obtain the cellular automaton model with the best fit for realizing the pitting corrosion simulation of the steel plate by training the LSTM network.

[0006] Further, the step of analyzing the morphological anomaly conditions of the inclined surface curve segments at different depth positions and the rule constraint conditions at different depth positions according to the depth positions of the respective inclined surface curve segments, and determining the directional mutation index for each depth position includes: Sort all the inclined surface curve segments in a preset order of average depth to obtain a set of inclined surface curve segments; in the set of inclined surface curve segments, calculate the difference between the slope of each inclined surface curve segment except the first one and the slope of its previous inclined surface curve segment to obtain the morphological anomaly degree of the inclined surface curve segment at the corresponding depth position; Obtain the variation difference situation of the horizontal width difference between the same depth positions in the corrosion pit morphology curves of the cell at each two consecutive corrosion times in the stress-free environment and the tensile stress environment in the set rules, and determine the rule constraint coefficient for each depth position; Adjust the morphological anomaly degree of each depth position by using the rule constraint coefficient to obtain the directional mutation index for each depth position; wherein, both the morphological anomaly degree and the rule constraint coefficient are positively correlated with the directional mutation index.

[0007] Further, the step of determining the rule constraint coefficient for each depth position includes: Take any depth position as the target depth position, calculate the variance of the horizontal width difference between the target depth positions in the corrosion pit morphology curves of the cell at each two consecutive corrosion times in the tensile stress environment, and denote it as the first mutation index of the pitting corrosion pit at the target depth position under the set natural rules in the tensile stress environment; Wherein, the horizontal width difference is the difference between the horizontal widths at the target depth positions in the pit morphology curves at two consecutive corrosion times, and the corrosion time is the time interval corresponding to the corrosion start time point to the corrosion end time point; Calculate the variance of the horizontal width differences between the target depth positions in the pit morphology curves at every two consecutive corrosion times of the cell in a stress-free environment, and denote it as the second mutation index of the pitting pits at the target depth position under the set natural rules in the stress-free environment; Calculate the difference value between the first mutation index and the second mutation index, perform normalization processing on the difference value to obtain a normalized value, and use the normalized value as the rule constraint coefficient at the target depth position.

[0008] Furthermore, the determining the expansion coefficient corresponding to the pit surfaces at every two consecutive corrosion times with the same solution concentration includes: For the pit surfaces at any two consecutive corrosion times, perform grid segmentation using the same specification of triangular meshes; Extract the curvature, torsion, and coordinates at each vertex position on the grid, calculate the Euclidean norms of the coordinates, curvatures, and torsions between different positions in the two pit surfaces, and use the average Euclidean norm as the matching result at the corresponding position; Select the minimum matching result from the matching results at all positions, and use the minimum matching result as the expansion coefficient corresponding to the pit surfaces at two consecutive corrosion times.

[0009] Furthermore, the analyzing the corrosion anomaly situation based on all the inclined plane regions inside the pits at different corrosion times with several different solution concentrations, and combining with the expansion coefficient to determine the local erosion progression coefficient at each depth position under different solution concentrations includes: Obtain the three-dimensional morphology data of the pits at each corrosion time with several different solution concentrations, and segment the corresponding inclined plane regions inside the pits according to the three-dimensional morphology data of the pits; Along the sequence from top to bottom of the depth positions, obtain all the normal angles between each inclined plane region and the inclined plane regions adjacent to all its edges, and combine with the expansion coefficient to determine the dynamic rule competition index of each inclined plane region; Project the dynamic rule competition index onto a two-dimensional plane, and obtain the relationship curve between the dynamic rule competition index and the depth position through curve fitting; According to the relationship curve, calculate the average derivative value in time series of all the dynamic rule competition indices at each depth position with each solution concentration, and determine the local erosion progression coefficient at each depth position under different solution concentrations.

[0010] Furthermore, the obtaining all the normal angles between each inclined plane region and the inclined plane regions adjacent to all its edges includes: Taking the vertical direction as the Y-axis and the horizontal line where the centroid of the target inclined plane area is located as the X-axis, a coordinate system is created, and the normal line passing through the centroid of each inclined plane area is obtained in the coordinate system; According to the normal lines passing through the centroid of each inclined plane area, all the normal angles between each inclined plane area and the inclined plane areas adjacent to all its edges are obtained.

[0011] Further, the determining of the dynamic rule competition index of each inclined plane area includes: Taking any inclined plane area as the target inclined plane area, and taking the inclined plane areas adjacent to all the edges of the target inclined plane area as the neighborhood inclined plane areas; According to the cosine value of the normal angle between the target inclined plane area and each neighborhood inclined plane area, the Z-axis coordinate value of the centroid of each neighborhood inclined plane area, and the expansion coefficient, the dynamic rule competition index of the target inclined plane area is determined.

[0012] Further, the determining of the dynamic rule competition index of the target inclined plane area according to the cosine value of the normal angle between the target inclined plane area and each neighborhood inclined plane area, the Z-axis coordinate value of the centroid of each neighborhood inclined plane area, and the expansion coefficient includes: Calculating the product of the cosine value of the normal angle between the target inclined plane area and each neighborhood inclined plane area and the Z-axis coordinate value of the centroid of the corresponding neighborhood inclined plane area, and taking the average value of all the products as the first dynamic rule competition factor of the target inclined plane area; Exponentially amplifying the expansion coefficient, and taking the amplified value as the second dynamic rule competition factor of the target inclined plane area; Fusing the first dynamic rule competition factor and the second dynamic rule competition factor of the target inclined plane area to obtain the dynamic rule competition index of the target inclined plane area.

[0013] Further, the generating of the dynamic rule for each cell to evolve towards the corresponding type of inclined plane includes: Calculating the product of the intersection number between each type of inclined plane and the depth position of each type of the local erosion progression coefficient and the directional mutation index of the corresponding depth position, to obtain the membership degree of each type of inclined plane belonging to each type of the local erosion progression coefficient; For any cell, determining the membership degree of each type of inclined plane belonging to the weighted average value of the local erosion progression coefficients of all the cells above the cell, and taking the type of inclined plane corresponding to the maximum membership degree as the evolution direction of the cell, so that the cell generates a dynamic rule to evolve towards the corresponding type of inclined plane.

[0014] Another embodiment of the present invention also provides a pitting corrosion simulation system for steel plates based on a cellular automaton model, including a processor and a memory, and the processor is used to process the instructions stored in the memory to implement the pitting corrosion simulation method for steel plates based on the cellular automaton model.

[0015] The present invention has the following beneficial effects: The present invention provides a method and system for simulating pitting corrosion of steel plates based on a cellular automaton model. This method takes into account the influence of tensile stress and helps to predict the corrosion fatigue life under complex loads. On this basis, in view of the problem that during the evolution of pitting corrosion of steel plates by cellular automata, there may be dynamic competition between complex rules for each cell, and the finite element rules are insufficient to show the random influence of various factors during the corrosion process, resulting in a fixed evolution direction, the present invention proposes to use a combination of long short-term memory network and local erosion progression to optimize the cellular evolution rules. Specifically, it calculates the directional mutation index at different depth positions under the set finite element rules, and calculates the local erosion progression coefficient at different concentrations and depth positions under the accelerated corrosion experiment. For each type of inclined plane, it calculates its membership degree for different local erosion progression coefficients, and then combines the long short-term memory network to dynamically match the cells in the CA model evolving step by step in time series with the inclined plane, increasing the possibility of its evolution in different directions, avoiding the problem of directional evolution under finite element rules, being conducive to accurately describing its irregular evolution in time-space, and providing strong help for studying the actual pitting corrosion process of steel plates. Description of the Drawings

[0016] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0017] Figure 1 Schematic diagram of the CA model for pitting corrosion of steel plates; Figure 2 Flowchart of the steps of a method for simulating pitting corrosion of steel plates based on a cellular automaton model provided by an embodiment of the present invention; Figure 3 Schematic diagram of the morphology curve of the side profile of corrosion pits with different simulated corrosion times under the same stress simulated by the finite element model in the embodiment of the present invention; Figure 4 Schematic diagram of the morphology curve of the side profile of corrosion pits with different stress levels under the same simulated corrosion time simulated by the finite element model in the embodiment of the present invention; Figure 5 Example of pit morphology Figure 1 ; Figure 6 Example of pit morphology Figure 2 。 Detailed Embodiments

[0018] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following will, in conjunction with the accompanying drawings and preferred embodiments, detail the specific implementation manners, structures, features and their effects of the technical solutions proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0020] The application scenarios targeted by the present invention can be: If the pitting corrosion simulation of steel plates is only based on ideal discrete causal relationships (local rules) and corrosion propagation rates (corrosion concentrations), it is impossible to describe the irregular evolution of corrosion pits in time and space, which is not conducive to effective pitting corrosion simulation of steel plates. This is because pitting corrosion involves multiple competing processes such as electrochemical dissolution, stress concentration, and passive film rupture from initiation to propagation. Multiple rules may coexist in a single cell, and the dynamic competition simulation between rules is currently a research difficulty.

[0021] In order to overcome the defects existing in the existing pitting corrosion simulation of steel plates, an embodiment of the present invention provides a method for pitting corrosion simulation of steel plates based on a cellular automaton model.

[0022] First, preparations before creating the cellular automaton.

[0023] A cellular automaton is a discrete dynamic system composed of regularly arranged basic units (cells), and the cell states evolve synchronously in discrete time and space through local rules.

[0024] A cellular automaton consists of four elements: cells, cell space, neighbors, and rules.

[0025] Cells, the basic units, have finite discrete states (such as 0 / 1, colors, etc.) and are distributed on a regular grid; the cell space, the grid where the cells are distributed, can be one-dimensional, two-dimensional, or multi-dimensional; neighbors, the set of neighboring cells that determine the state of a cell at the next moment. For example, the left and right neighbors of a one-dimensional cell (radius r = 1), and the Moore neighborhood (8-neighborhood) of a two-dimensional cell; rules, local transition functions, which calculate the state at the next moment based on the states of the current cell and its neighbors.

[0026] A cellular automaton emerges global complexity through the iteration of simple local rules. As Figure 1 shown in the schematic diagram of the CA model of pitting corrosion of steel plates, it is composed of several two-dimensional cells. In Figure 1In the figure, H represents hydrogen ions, M represents metal matrix, F represents oxide film layer, R represents iron ions, and P represents iron hydroxide.

[0027] According to the application direction and field of steel plates, simulate the external environmental factors that cause pitting of steel plates, including the salinity, temperature, humidity, light, pH value, etc. of the application environment, and determine a corrosion development rate in advance, including the lateral expansion rate and the longitudinal corrosion rate. In addition, clarify the electrochemical reaction equation to provide a mechanism basis for the deduction of cellular automata.

[0028] It should be noted that the metal hydroxide is attached to the substrate in the form of precipitation to form a passive protective film. Except for the metal cells, all other types of cells in the model can move randomly.

[0029] Second, define the initialization cell settings and distribution.

[0030] In the 2D drawing software that comes with Windows, the various elements involved in the pitting process are converted into each cell in the cellular space to simulate the pitting process of metal materials in a corrosive environment. The molecules, ions or atoms represented by the actual chemical reactions are represented by cells. A total of 10 types of cells are set: W represents water, H represents hydrogen ions; M1 represents an oxide film layer; R1 represents iron ions; P1 represents iron hydroxide generated by the reaction; M2 represents a base steel plate; R2 represents ferrous ions; P2 represents ferrous hydroxide; O represents oxygen; and C represents chloride ions.

[0031] It is worth mentioning that the oxide film that appears in the CA simulation will change the simulation from simple corrosion expansion to passivation to activation of a dynamic competition system, significantly improving the realism.

[0032] Then import the PNG image (Portable Network Graphics), and process the grayscale to get a 0, 1 matrix. The initially imported image does not include the oxide layer and the solution layer, so the structural layer and the solution layer are required.

[0033] Specifically, two loop assignment statements are used. Because the lines with one layer of thickness are dim, three layers of thickness are used. ( ) is used to represent the solution layer thickness of 150 layers, represented by ( ) to indicate. Define the initial sentinel to record the corrosion area, with an initial value of 0. Define the upper and lower limits of vertical corrosion expansion, which are 40 and 350 respectively. Define the center position of horizontal corrosion expansion ( ), define the initial expansion range ( ), and the number of expansion time steps is 5.

[0034] Third, set the cellular chemical corrosion reaction and diffusion parameters.

[0035] Among them, the parameters involved in the cellular reaction rules include: when there are at least six acidic sites H near the metal site M1, the reaction occurs with a corrosion probability of P1_corr; the M1 site will be replaced by the substance R1.

[0036] When at least three neighbors of the active metal R1 are neutral sites W, hydrolyzes to , with a hydrolysis probability of P2_Hyd2, and the product is replaced by P1; the above are the laws of the electrochemical reaction of the oxide film.

[0037] When there are at least two acidic sites H near the metal site M2, the forward reaction of Fe oxidation to generate occurs, with a corrosion probability of P2_corr; the M2 site will be replaced by the substance R2.

[0038] When there is at least one acidic site H near the active metal site R2, is oxidized to , with an oxidation probability of P2_ox; the R2 site will be replaced by the substance D. When at least two neighbors of the active metal R2 are neutral sites W, hydrolyzes to , with a hydrolysis probability of P1_Hyd1; R2 and two W sites are respectively replaced by P2 and two H substances.

[0039] The final reaction is hydrolyzes to , with a hydrolysis probability of P2_Hyd2. When at least three of the adjacent sites of the D site are neutral sites W, D and three W sites will be respectively replaced by P1 and three H substances.

[0040] As an exemplary implementation, the setting of the diffusion parameters includes: The corrosion generation probabilities of 7 substances (water, hydrogen ion, oxide film layer, iron hydroxide, ferrous hydroxide, oxygen, chloride ion) are initially set (0.1433, 0.2218, 0.183, 0.1943, 0.096, 0.1252, 0.1252 respectively); 5 diffusion coefficients are set (0.5, 0.3, 0.08, 0.06, 0.01 respectively); the stress factor is set to 0.2, then the tensile coefficient is 1.2; the sedimentation factor is 2, then the sedimentation coefficient is 1.2; two hydrolysis probabilities are defined as 0.25 and 0.3; the analysis step is defined as 30.

[0041] Fourth, establish a bridge between the cellular automaton and the finite element model.

[0042] First, in the cellular automaton model before applying stress, record the positions of each cell to facilitate one-to-one correspondence with the elements in the finite element model; Secondly, the finite element method calculates the stress and strain of the cell, so Create a finite element model, create materials, assign elastic-plastic constitutive properties to components, and define assemblies; Then, create the component mesh in advance. The mesh is A small square grid corresponds to each cell; Next, create two analysis steps. The first analysis step is set to define the cellular corrosion model, and the second analysis step is to define the loading and perform finite element calculation of stress and strain. Finally, the interactions are defined and, during the initial analysis, a cellular automaton model of the corrosion morphology that occurs during electrochemical corrosion formation is created. The component becomes the shape of a pit, and the model is imported into the finite element model. After the mesh is divided, the finite element pit is created.

[0043] The transition of steel plate pitting from metastable state to stable state is random. The cells at the edge of the pit may be dissolved preferentially due to high stress, and the diffusion of surrounding corrosive substances will slow down the corrosion and re-form the metal oxide film. The cellular automaton rule considers the random walk of a single cell. After introducing the dissolution and passivation probabilities, the walk direction of the cell in three-dimensional space can be affected, that is, each direction has a weight; and in fact, only under the finite element rule, each cell has a high probability of directional mutation, which is not conducive to the simulation of the trend of steel plate pitting. Among them, the metastable state refers to possible repair, while the stable state refers to continuous development.

[0044] In order to overcome the defects of the above-mentioned cellular evolution rules, an embodiment of the present invention uses a combination of long short-term memory network and local corrosion progression to simulate the dynamic rule competition that may occur at different cell positions, and constructs a cellular automaton model with the best fit for steel plate pitting simulation based on this evolution rule, such as Figure 2 As shown, the following steps are included: S1, from the CA-FEM coupling model, obtain the various slope curve segments corresponding to the pit morphology curves of several corrosion times under the environment without tensile stress and under the environment with tensile stress generated by the CA model, where the CA rule is driven by the tensile stress field calculated by FEM.

[0045] It should be noted that in the presence of tensile stress, tensile stress can promote the increase of corrosion rate and the expansion of corrosion morphology, which is mainly reflected in the expansion of vertical corrosion depth, that is, the change of depth position.

[0046] As an exemplary implementation, obtaining the inclined curve segment includes: The FEM (Finite Element Method) calculates the tensile stress field, maps the tensile stress field to the CA grid, the CA simulates the evolution of the corrosion pit, extracts the morphology curves at different corrosion times during the evolution process, and through the inclined plane segmentation algorithm, statistically analyzes the characteristics of the inclined plane curve segments to obtain each inclined plane curve segment corresponding to the morphology curve of each corrosion pit. Of course, the morphology curve can also be obtained through the steps of establishing the bridge between the cellular automaton and the finite element model mentioned above.

[0047] Among them, the corrosion pit inclined plane needs to be clearly defined through normal vector definition or curvature calculation, such as the area where the angle between the normal direction and the vertical direction is greater than 15 degrees; for the inclined plane curve segment, obtain each slope in the corrosion pit morphology curve, and use the data points corresponding to the slopes as the segmentation points to segment the corrosion pit morphology curve to obtain each curve segment in the curve. Since the morphology curve is the inclined plane curve of the corrosion pit, the curve segment can be used as the inclined plane curve segment. The corrosion time is the time interval corresponding to the start time point to the end time point of corrosion, with the unit of day (d), such as 1 to 5 (5d), 1 to 10 (10d), 1 to 15 (15d), etc. The number of corrosion times can be set by the implementer according to the specific actual situation without specific limitation.

[0048] The tensile stress is default downward. The schematic diagrams of the morphology curves of the side profiles of corrosion pits at different simulated corrosion times under the same stress simulated by the finite element model are as Figure 3 shown. The schematic diagrams of the morphology curves of the side profiles of corrosion pits at different stress levels under the same simulated corrosion time simulated by the finite element model are as Figure 4 shown. In Figure 3 and Figure 4 , F represents the stress magnitude, and T represents the simulated corrosion time. The horizontal axis represents the width position, and the vertical axis represents the depth position.

[0049] Different inclined planes may represent the non-linear pitting corrosion trends occurring at local time periods and positions. Therefore, the erosion inclined plane can be used as the basic bearing unit of the dynamic rules. Each inclined plane curve segment has only one slope, so the inclined plane segment can be segmented by whether the slope value changes. By analyzing the evolution of the inclined plane curve segment, the anisotropy degree of stress corrosion can be quantified, providing a basis for life prediction.

[0050] So far, this embodiment has obtained several inclined plane curve segments for analyzing the directional variation conditions at different depth positions.

[0051] S2. According to the depth positions of each inclined plane curve segment, analyze the morphology abnormality of the inclined plane curve segments at different depth positions and the rule constraint conditions at different depth positions, and determine the directional variation index for each depth position.

[0052] To overcome the influence of the directional mutation of cells on the simulation of the pitting trend of steel plates, based on the depth position characteristics of the inclined plane curve segments, the directional mutation index at different depth positions is quantified. The directional mutation index can characterize the missing evolution possibility of cells at different corrosion depth positions in the pre-set regular cell model. The larger the directional mutation index, the more necessary it is to assign a competition factor to the cells at the corresponding depth position to endow them with richer dynamic competition randomness.

[0053] As an exemplary implementation, the above step S2 can be realized through steps S21 to S23: S21, sort all the inclined plane curve segments in the preset order of average depth to obtain a set of inclined plane curve segments; in the set of inclined plane curve segments, calculate the difference between the slopes of each inclined plane curve segment except the first one and the slope of its previous inclined plane curve segment to obtain the morphological anomaly degree of the inclined plane curve segment at the corresponding depth position.

[0054] Here, the morphological anomaly degree refers to the anomaly degree of the corrosion situation of the corrosion pit corresponding to the inclined plane curve segment.

[0055] In the cellular automaton, a variety of rules including environmental parameters such as dissolution probability, solution concentration, and temperature have been set, and there may also be several uncaught dynamic rules. Dynamic rules generally concentrate in the initial stage of corrosion. As the depth increases, the competition between dynamic rules becomes less. Therefore, it is necessary to sort and analyze all the inclined plane curve segments according to the average depth. Generally, in a two-dimensional finite element model, the formation of the lower inclined plane is caused by the dynamic rule competition in the upper corrosion process.

[0056] In this embodiment, all the inclined plane curve segments are sorted in descending order according to the average depth from top to bottom to obtain a set of inclined plane curve segments. Each data point in an inclined plane curve segment has its corresponding depth. The average value of the depths of all data points in the same inclined plane curve segment (average depth) is used as the depth of the corresponding inclined plane curve segment. The difference between two slopes generally refers to the absolute value of the difference between the two slopes. The slope difference can reveal the local dynamic characteristics of the evolution of the corrosion pit morphology. The larger the slope difference, the more significant the expansion speed or direction mutation of the corrosion pit in the local area, reflecting the non-uniformity of the stress field distribution, and the more abnormal the morphological characteristics of the inclined plane curve segment at the depth position.

[0057] It should be noted that each depth position in the etch pit morphology curve has its corresponding morphology abnormality degree. For the morphology abnormality degree at the same depth position, the average value of the morphology abnormality degree can be used as the morphology abnormality degree at the corresponding depth position, and the morphology abnormality degrees at each depth position corresponding to the same inclined plane curve segment are the same; the morphology abnormality degree is obtained by analyzing the slope difference between the inclined plane curve segment and the previous adjacent inclined plane curve segment. Since there is no previous inclined plane curve segment for the first inclined plane curve segment in the set of inclined plane curve segments, the morphology abnormality degree analysis is not performed on the first inclined plane curve segment in the sequence.

[0058] S22. Obtain the change difference of the horizontal width difference between the same depth positions in the etch pit morphology curves of each two consecutive corrosion times of the cell in the non-tensile stress environment and the tensile stress environment in the set rules, and determine the rule constraint coefficient for each depth position.

[0059] Here, the rule constraint coefficient is used to quantify the degree to which the inclined plane morphology abnormality is driven by the non-dynamic competition rule and is constrained by the finite evolution rule at the depth position; the depth position when analyzing the horizontal width has a one-to-one correspondence with the depth position when analyzing the morphology abnormality degree, that is, each depth position can correspond to a morphology abnormality degree and a rule constraint coefficient.

[0060] As an exemplary implementation manner, the above step S22 can be implemented through steps S221 to S223: S221. Take any depth position as the target depth position, calculate the variance of the horizontal width difference between the target depth positions in the etch pit morphology curves of each two consecutive corrosion times of the cell in the tensile stress environment, and record it as the first mutation index of the pitting corrosion pit at the target depth position under the set natural rule in the tensile stress environment.

[0061] Here, any depth position is any depth position on the etch pit morphology curve; the horizontal width difference can characterize the evolution difference of each depth position in the curve of the continuous corrosion time of the cell in the tensile stress environment in the set rules; the first mutation index can represent the variability of different depth positions of the pitting corrosion pit under the interference of the tensile stress on the set natural rule.

[0062] In this embodiment, generally, the horizontal width of a larger corrosion time is larger, and the horizontal width of a smaller corrosion time is smaller. Therefore, the horizontal width difference can be the difference between the horizontal widths between the target depth positions in the etch pit morphology curves of two consecutive corrosion times, or the absolute value of the difference between the two horizontal widths.

[0063] S222. Calculate the variance of the horizontal width differences between the target depth positions in the pit morphology curves for every two consecutive corrosion times of the cell in a stress-free environment, which is denoted as the second mutation index of the pitting pit at the target depth position under the set natural rules in the stress-free environment.

[0064] Here, the horizontal width can characterize the degree of horizontal corrosion expansion, and the horizontal width difference can characterize the evolution differences of the cell at each depth position under continuous tensile stress in the stress-free environment according to the set rules; the second mutation index represents the variability of the pitting pit at different depth positions under the interference of the stress-free environment with the set natural rules.

[0065] S223. Calculate the difference value between the first mutation index and the second mutation index, perform normalization processing on the difference value to obtain the normalized value, and use the normalized value as the rule constraint coefficient at the target depth position.

[0066] In this embodiment, calculate the absolute value of the difference between the two mutation indices as the difference value; use the exp(-) function to perform inverse normalization processing on the difference value, and use the value after normalization processing as the rule constraint coefficient at the corresponding target depth position.

[0067] During the calculation process of the rule constraint coefficient, the smaller the difference between the two mutation indices, the more it indicates that the set rules will still produce a fixed evolution direction even in the presence of tensile stress. Then, the morphological abnormality of the lower slope at the target depth position is not generated driven by the dynamic competition rules, but is restricted by the limited evolution rules.

[0068] Referring to the calculation process of the rule constraint coefficient at the target depth position, the rule constraint coefficient of each depth position can be obtained, and each depth position in the pit morphology curve has its corresponding rule constraint coefficient.

[0069] S23. Use the rule constraint coefficient to adjust the degree of morphological abnormality at each depth position to obtain the directional mutation index at each depth position; among them, the degree of morphological abnormality and the rule constraint coefficient are both positively correlated with the directional mutation index.

[0070] In this embodiment, the product of the rule constraint coefficient at each depth position and the degree of morphological abnormality of the corresponding slope segment can be used as the directional mutation index at each depth position under the set rules.

[0071] In the finite element model, the slope curve segment with more abnormal morphology and greater rule binding force represents that the probability of the set rules undergoing directional mutation at the corresponding depth position is greater under the finite element conditions. The greater the probability of directional mutation at a certain depth position, the more necessary it is to endow the cells at that depth position with a competition factor to make them carry richer dynamic competition randomness.

[0072] So far, the present embodiment has obtained the directional mutation index at each depth position in the CA model.

[0073] S3. Determine the expansion coefficient corresponding to the pit surface at every two consecutive corrosion times for the same solution concentration; analyze the corrosion anomaly situation based on all the inclined plane areas inside the pits at different corrosion times for several different solution concentrations, and combine with the expansion coefficient to determine the local erosion progression coefficient at each depth position under different solution concentrations.

[0074] Here, the expansion coefficient refers to the expansion state of the pit morphology at one corrosion time relative to the pit morphology at its previous corrosion time under the same solution concentration. The larger the expansion coefficient, the greater the overall corrosion process; the local erosion progression coefficient refers to the local corrosion progression relationship at different depth positions under different corrosion rates.

[0075] In the present embodiment, when conducting the accelerated corrosion experiment, corrosion solutions with different concentrations are dropped on the surfaces of different steel samples. The corresponding corrosion rates can be obtained in advance according to the concentrations of the corrosion solutions. Then, the experimental samples are placed in a non-laboratory environment, that is, the environment in the field of application of steel. When the corrosion time reaches 5d, 10d, 15d, 20d, 25d, 30d..., a batch is taken out each time for observing the pit morphology. The magnitude and number of the corrosion time can be consistent with the magnitude and number of the corrosion time recorded in the above step S1. Among them, the example of the pit morphology Figure 1 Such as Figure 5 shown, the example of the pit morphology Figure 2 Such as Figure 6 shown. It should be noted that Figure 5 and Figure 6 are essentially schematic diagrams, only used to show the pit morphology characteristics at different corrosion times.

[0076] From Figure 5 and Figure 6 it can be seen that most of the pits formed on the sample in the initial stage of corrosion are approximately circular. As the corrosion time prolongs, a single pit gradually develops towards the surrounding and depth directions, presenting a hemispherical shape, and finally will continue to develop deeper at the bottom of the hemisphere. Therefore, theoretically, the direction of the tensile stress at each cell position should be divergent in the early stage and concentrated downward in the later stage. Based on this, it can be known that: assuming that there are tensile stresses in several directions on each cell, the number and included angle of the cell tensile stress directions should be a function relationship with an inverse proportional non-linear decay with the corrosion time or corrosion depth. Among them, the pit morphology data can be obtained by laser scanning inside the pits of the experimental samples.

[0077] The above step S3 can be implemented through steps S31 to S32: S31. Determine the expansion coefficient corresponding to the pit surface at every two consecutive corrosion times for the same solution concentration.

[0078] Before determining the expansion coefficient, first obtain the pit surfaces of each sample at any concentration at different corrosion times. The pit surfaces can be obtained by laser scanning the topography inside the pits. By analyzing the matching situation between the pit surface at each corrosion time and the pit surface at the previous corrosion time, the expansion coefficient corresponding to the pit surface at each corrosion time under the same solution concentration is quantified. The worse the matching situation between the two pit surfaces, the greater the degree of expansion and the larger the expansion coefficient.

[0079] As an exemplary implementation, the above step S31 can be implemented through steps S311 to S313: S311, for the pit surfaces at any two consecutive corrosion times, perform grid segmentation using the same specification triangular grid.

[0080] In this embodiment, the implementation process of grid segmentation is prior art and not within the protection scope of the present invention, so it will not be elaborated here.

[0081] S312, extract the curvature, torsion, and coordinates for each vertex position on the grid, calculate the Euclidean norms of the coordinates, curvatures, and torsions between different positions in the two pit surfaces, and take the average Euclidean norm as the matching result for the corresponding position.

[0082] In this embodiment, the average Euclidean norm refers to the average of the Euclidean norms in three dimensions at the same position. Analyzing the Euclidean norm is also analyzing the matching situation of the three-dimensional morphologies at the previous and subsequent corrosion times.

[0083] S313, select the minimum matching result from the matching results at all positions, and take the minimum matching result as the expansion coefficient corresponding to the pit surfaces at two consecutive corrosion times.

[0084] In this embodiment, the minimum matching result can best represent the overall matching situation of the two pit surfaces. The larger the minimum matching result, the less matching the two pit surfaces are, the larger the expansion coefficient, and the higher the degree of increase in the pit corrosion expansion situation. By performing morphological matching analysis between each corrosion time and the previous corrosion time, the corresponding expansion coefficient can be obtained. The expansion coefficient of the pit surface at the t-th corrosion time is denoted as .

[0085] S32, analyze the corrosion anomaly situation based on all inclined plane regions inside the pits at different corrosion times for several different solution concentrations, and combine the expansion coefficient to determine the local erosion progression coefficient at each depth position under different solution concentrations.

[0086] As an exemplary implementation, the above step S32 can be implemented through steps S321 to S324: S321. Obtain the three-dimensional morphology data of the corrosion pits at each corrosion time for several different solution concentrations, and segment each corresponding inclined plane area inside the corrosion pits according to the three-dimensional morphology data of the corrosion pits.

[0087] In this embodiment, to segment the inclined plane area from the three-dimensional morphology data of the corrosion pits, it needs to be realized by combining feature extraction and region growing algorithms. Specifically, the inclined plane area can be realized through normal vector analysis, curvature filtering and region growing of the three-dimensional morphology data of the corrosion pits. The method for obtaining the inclined plane area is a prior art and not within the protection scope of the present invention, so it will not be elaborated in detail here.

[0088] S322. Along the sequence from top to bottom of the depth position, obtain all the normal angles between each inclined plane area and the inclined plane areas adjacent to all its edges, and combine the expansion coefficient to determine the dynamic rule competition index of each inclined plane area.

[0089] Here, for the dynamic rule competition index, it is determined by analyzing the matching situation between the corrosion progress of the inclined plane area and the overall corrosion progress; the more mismatched the corrosion progress of any inclined plane area is with the overall corrosion progress, the greater the probability that there is dynamic rule competition in this inclined plane area.

[0090] In this embodiment, the normal angle has positive and negative values. The larger the sum of the absolute values of the normal angles, the greater the difference between the inclined plane area and its surrounding inclined plane areas, and the steeper it is; on the contrary, the smaller the sum of the absolute values of the normal angles, the closer the normal angles are to being parallel, the smaller the difference between the inclined plane area and its surrounding inclined plane areas, and the flatter it is. The inclined plane area closer to the top of the depth position is steeper, indicating that the divergence of its tensile stress is insufficient and there may be dynamic competition of corrosion rules; the inclined plane area closer to the bottom of the depth position is flatter, indicating that its tensile stress is not concentrated enough, and at this time the flatter inclined plane area is more likely to have entered the late stage of evolution.

[0091] Therefore, for the corrosion pit morphology data at each corrosion time under each solution concentration, along the sequence from top to bottom of the depth position, calculate the dynamic rule competition index of each inclined plane area in turn.

[0092] As an exemplary implementation manner, the above step S322 can be realized through steps S3221 to S3222: S3221. Take any inclined plane area as the target inclined plane area, and take the inclined plane areas adjacent to all the edges of the target inclined plane area as the neighborhood inclined plane areas.

[0093] S3222. Determine the dynamic rule competition index of the target inclined plane area according to the cosine value of the normal angle between the target inclined plane area and each neighborhood inclined plane area, the Z-axis coordinate value of the centroid of each neighborhood inclined plane area, and the expansion coefficient.

[0094] First, calculate the product of the cosine value of the normal angle between the target inclined plane region and each neighboring inclined plane region and the Z-axis coordinate value of the centroid of the corresponding neighboring inclined plane region, and take the average value of all products as the first dynamic rule competition factor of the target inclined plane region; Secondly, exponentially amplify the expansion coefficient and take the amplified value as the second dynamic rule competition factor of the target inclined plane region; Finally, fuse the first dynamic rule competition factor and the second dynamic rule competition factor of the target inclined plane region to obtain the dynamic rule competition index of the target inclined plane region.

[0095] As an example, the calculation formula for the dynamic rule competition index of the o-th inclined plane region can be: ; where, represents the dynamic rule competition index of the o-th inclined plane region (target inclined plane region), e represents the natural constant, represents the expansion coefficient corresponding to the erosion pit surface at the t-th erosion time, represents the number of neighboring inclined plane regions of the o-th inclined plane region at the t-th erosion time, represents the Z-axis coordinate value of the centroid of the i-th neighboring inclined plane region of the o-th inclined plane region, represents the cosine value of the normal angle between the o-th inclined plane region and the i-th neighboring inclined plane region, represents the overall erosion progress at the t-th erosion time, represents the erosion progress at the location of the o-th inclined plane region at the t-th erosion time.

[0096] In the calculation formula of the dynamic rule competition index, the larger it is, the greater the overall erosion process; the larger it is, the more abnormal the erosion at the location of the o-th inclined plane region, and the slow erosion process essentially represents the slow erosion progress; the larger the dynamic rule competition index, the greater the probability that there may be dynamic rule competition at the location of the o-th inclined plane region.

[0097] Furthermore, the steps for obtaining the normal angle between the target inclined plane region and each neighboring inclined plane region include: taking the vertical direction as the Y-axis and the horizontal line where the centroid of the target inclined plane region is located as the X-axis to create a coordinate system, and obtaining the normal line passing through the centroid of each inclined plane region in the coordinate system; according to the normal lines passing through the centroids of each inclined plane region, obtain all the normal angles between each inclined plane region and all the inclined plane regions adjacent to its edges.

[0098] Referring to the dynamic rule competition index of the above-mentioned target inclined plane region, obtain the dynamic competition indexes of all inclined plane regions in the depth sequence from top to bottom.

[0099] S323. Project the dynamic rule competition index onto a two-dimensional plane, and obtain the relationship curve between the dynamic rule competition index and the depth position through curve fitting.

[0100] In this embodiment, the dynamic rule competition index is projected onto a two-dimensional plane, and the relationship curve between the dynamic rule competition index and the depth position is obtained through least squares fitting. Furthermore, the relationship curve between the dynamic competition probability and the depth position in the accelerated corrosion experiment pits for each corrosion time and its previous adjacent corrosion time under all solution concentration conditions is obtained. Among them, the implementation process of the least squares method is prior art and not within the protection scope of the present invention, so it will not be elaborated in detail here.

[0101] S324. According to the relationship curve, calculate the average derivative value of all dynamic rule competition indexes at each depth position of each solution concentration in time series, and determine the local erosion progression coefficient at each depth position under different solution concentrations.

[0102] In this embodiment, each solution concentration can represent different corrosion rates, and each corrosion time can represent the corrosion progress. Then, the changes in multiple dynamic rule competition indexes at each depth position under continuous corrosion times reflect the local corrosion progression relationship at different depth positions under different corrosion rates.

[0103] Specifically, by calculating the average derivative value of all dynamic rule competition indexes at each depth position under each solution concentration in time series, the local erosion progression coefficient at each depth position under different solution concentrations is obtained.

[0104] It should be noted that the depth refers to the longitudinal corrosion progress during the corrosion process. The deeper the corrosion, the more backward the longitudinal corrosion progress. Here, the depth position is the average value of all depth positions in a single inclined plane area; the inclined plane is different morphological results formed during the corrosion process, which can characterize the surface condition of the steel plate after corrosion; dynamic rule competition refers to the evolution results of some rule-limiting cells under different corrosion progressions. For example, when two rules such as the reformation of the metal surface oxide film and the dilution of the corrosion concentration coexist, and when more rules coexist, they may limit the deviation of the final corrosion result.

[0105] So far, this embodiment has determined the local erosion progression coefficient at each depth position under different solution concentrations.

[0106] S4. Obtain the intersection number between the depth positions of each type of inclined plane and each type of local erosion progression coefficient, and combine the directional mutation index of each depth position to generate the dynamic rules for each cell to evolve into the corresponding type of inclined plane.

[0107] Here, the directional mutation index used to represent the degree of sudden change in the corrosion direction at the depth position determines the missing evolution possibilities of the cells at different corrosion depth positions in the pre-set regular cellular model. The local erosion progression coefficient used to represent the relative speed of the corrosion rate at the depth represents the change in the dynamic competition probability at different corrosion depth positions with different corrosion speeds in the actual accelerated corrosion experiment.

[0108] Then, for the cells with partially missing evolution possibilities, increase their probabilities of evolving into different types of inclined planes. The evolution into different types of inclined planes is determined by the local erosion progression relationship of the upper cell positions, which can, to a certain extent, restore the dynamic rule competition of the upper cell positions during the pitting process in the cellular automaton and analyze the non-linear and random evolution effects on the lower cell positions.

[0109] Among them, in order to facilitate the subsequent calculation of the membership degree, the depth positions corresponding to the directional mutation index and the local erosion progression coefficient can be made to have the same size and number through numerical fitting.

[0110] In this embodiment, each type of inclined plane is a sample set obtained through an accelerated corrosion experiment and is obtained by classifying all inclined planes according to slope, area, and depth range. The similarity of the slope, area, and depth range of the same type of inclined plane is relatively high, and each type of inclined plane may be a direction evolved by the dynamic competition rule; the local erosion progression coefficients with the same numerical value are one type of local erosion progression coefficient, and there are cases where the local erosion progression coefficients corresponding to some different depth positions are the same; an inclined plane is between the depth positions at one end and the other end of the inclined plane on the Z-axis, so all the depth positions included in each type of inclined plane can be obtained. For the depth position set of one type of local erosion progression coefficient and the depth position set of one type of inclined plane, the number of intersections (identical) of the depth positions in the two depth position sets is used as the intersection number. Among them, the classification of the inclined planes can be realized by using the dynamic time rule to measure the similarity between the vectors of the inclined planes. The vector is composed of elements in three dimensions, namely slope, area, and depth range.

[0111] As an exemplary implementation manner, the above step S4 can be realized through steps S41 to S42: S41, calculate the product of the intersection number between the depth positions of each type of inclined plane and each type of local erosion progression coefficient and the directional mutation index of the corresponding depth position to obtain the membership degree of each type of inclined plane belonging to each type of local erosion progression coefficient.

[0112] The larger the number of depth positions in the intersection between all the depth positions where a certain type of inclined plane has occurred and all the depth positions where each type of local erosion progression coefficient has occurred, the greater the probability of the coincidence of depth positions; the larger the directional mutation index of a certain depth position, the more the dynamic rule competition at this depth position needs to be increased, that is, the membership degree of the inclined plane to the local erosion progression coefficient at this depth position is expanded. Expanding the membership degree can match more possible inclined planes for this depth position, and vice versa, the membership degree needs to be reduced.

[0113] As an example, the calculation formula for the membership degree of the r-th type of inclined plane belonging to the local erosion progression coefficient at the z-th position of the etching pit under the s-th solution concentration can be: ; In the formula, represents the membership degree of the r-th type of inclined plane belonging to the -th type of local erosion progression coefficient, represents the local erosion progression coefficient at the z-th depth position of the etching pit under the s-th solution concentration, represents the directional mutation index of the z-th depth position, represents the intersection number between the r-th type of inclined plane and the depth positions of the -th type of local erosion progression coefficient.

[0114] Referring to the above calculation process of the membership degree of the r-th type of inclined plane belonging to the -th type of local erosion progression coefficient, the membership degree of each type of inclined plane belonging to each type of local erosion progression coefficient can be obtained.

[0115] So far, in this embodiment, the membership degree of each type of inclined plane belonging to each type of local erosion progression coefficient has been obtained.

[0116] Furthermore, a long short-term memory network is introduced.

[0117] LSTM (Long Short-Term Memory) is good at processing time series data and long-term dependence relationships. In pitting evolution, the current state of each cell depends not only on the immediate state of neighboring cells, but also on the historical evolution paths of itself and its surrounding areas. For example, the erosion history of the upper cell position (such as the accumulation of corrosion products, the frequency of passive film rupture) will dynamically adjust the evolution probability of the lower cell through the gating mechanism (input gate, forget gate, output gate) of LSTM, so as to simulate local rule competition.

[0118] The input features of each cell include: Current state: such as corrosion probability, hydrolysis probability, sedimentation coefficient, sedimentation factor, diffusion probability, stress factor, tensile coefficient, reaction threshold, corrosion products, etc. Neighborhood state: Capturing the states of neighboring cells at the upper position of each cell. Time series: The state changes in the past several time steps (such as corrosion depth, solution concentration gradient, temperature change).

[0119] In the time sequence, a cell position may correspond to different corrosion rates, or there may be multiple local erosion progression coefficients of different degrees; according to the depth position of the cell and the current state index, an estimated corrosion rate corresponding to it is generated, and the local erosion progression coefficient at the concentration closest to this depth position is matched according to the corrosion rate; similarly, according to the depth positions and states of all cells in the neighborhood above this cell, the local erosion progression coefficients of all cells in the neighborhood are obtained; the distance Gaussian function is used to perform weighted averaging on the local erosion progression coefficients of all cells above this cell.

[0120] Each cell is associated with an LSTM unit, and the output is the probability of the evolution direction at the next time step (the probability of expanding to different inclined planes). Specifically: S42. For any cell, determine the membership degree of each type of inclined plane to the weighted average of the local erosion progression coefficients of all cells above this cell, and take the type of inclined plane with the largest membership degree as the evolution direction of this cell, so that this cell generates a dynamic rule for evolving to the corresponding type of inclined plane.

[0121] So far, referring to the determination process of generating the dynamic rule for evolving to the corresponding type of inclined plane for the above single cell, obtain the dynamic rule for each cell to generate the evolution to the corresponding type of inclined plane.

[0122] S5. On the basis of the dynamic rule, by training the LSTM network, obtain a cellular automaton model with the best fit for realizing the pitting corrosion simulation of the steel plate.

[0123] In this embodiment, on the basis of the dynamic rule for the cell to evolve to the corresponding type of inclined plane, train the LSTM network, and predict the inclined plane that the lower cell may evolve to according to the state of the upper cell. Then, the cross-entropy loss function can be used to minimize the difference between the evolution direction predicted by the LSTM and the real simulation or experimental results; set a drawing function to refresh at all times, and the morphology change of the pitting corrosion pit within the time step can be observed. The corrosion area, width, depth, depth-to-diameter ratio, etc. of the pitting corrosion pit are recorded through a sentry; finally, compare the simulation results of the LSTM-CA model and the pure probability-type CA, evaluate the fit degree of the pit morphology (such as equivalent radius, depth) and randomness distribution (such as pitting density) with the experimental data, and obtain a cellular automaton model with the best fit for realizing the pitting corrosion simulation of the steel plate.

[0124] Among them, the forgetting gate of the LSTM can control the influence of historical rules, such as whether to retain the hindering effect of early corrosion products, and the input gate determines the influence weight of the current environmental parameters on the evolution direction. The current environmental parameters are such as solution concentration and temperature. The actual training process of the LSTM network is prior art and not within the protection scope of the present invention, so it will not be elaborated in detail here.

[0125] It should be noted that by combining the long short-term memory network with local erosion progression, the limitations of the multi-dimensional regular finite element setting can be overcome, dynamically adapting to the coupling effect of multiple factors. At the same time, the path dependence in real corrosion can be reproduced, such as the early corrosion pits guiding the subsequent expansion direction, improving the accuracy of pitting corrosion simulation of steel plates.

[0126] An embodiment of the present invention also provides a pitting corrosion simulation system of a steel plate based on a cellular automaton model, including a processor and a memory. The processor is configured to process instructions stored in the memory to implement the pitting corrosion simulation method of the steel plate based on the cellular automaton model.

[0127] The above-described embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention and should all be included in the protection scope of the present invention.

Claims

1. A method for simulating steel plate pitting corrosion based on a cellular automaton model, characterized in that: The following steps are involved: From the CA-FEM coupling model, the slope curve segments corresponding to the pit morphology curves at several corrosion times under the environment without tensile stress and under the environment with tensile stress generated by the CA model are obtained, wherein the CA rule is driven by the tensile stress field calculated by FEM; According to the depth positions of the various inclined curve segments, the morphological anomalies of the inclined curve segments at different depth positions and the rule constraints at different depth positions are analyzed to determine the directional variation index at each depth position; Determine the expansion coefficient corresponding to the pit surface at every two consecutive corrosion times for the same solution concentration; analyze the corrosion anomaly based on all the inclined surface areas inside the pit at different corrosion times for several different solution concentrations, and determine the local erosion progression coefficient at each depth position at different solution concentrations in combination with the expansion coefficient; Obtain the number of intersections between each type of slope and each type of depth position of the local erosion progression coefficient, and generate a dynamic rule for each cell to evolve toward the corresponding type of slope by combining the directional variation index of each depth position; On the basis of the dynamic rules, the cellular automaton model with the best fit for realizing the simulation of steel plate pitting corrosion is obtained by training the LSTM network.

2. The method for simulating steel plate pitting corrosion based on a cellular automaton model according to claim 1, characterized in that: According to the depth position of each inclined curve segment, analyzing the morphological anomalies of the inclined curve segments at different depth positions and the rule constraints at different depth positions, and determining the directional variation index of each depth position, includes: All the inclined surface curve segments are sorted according to a preset order of average depth to obtain an inclined surface curve segment set; in the inclined surface curve segment set, the difference between the slope of each inclined surface curve segment except the first inclined surface curve segment and the slope of the previous inclined surface curve segment is calculated to obtain the morphological abnormality degree of the inclined surface curve segment at the corresponding depth position; Obtain the change difference of the horizontal width difference between the same depth positions in the pit morphology curves of the cells in the set rules under the environment without tensile stress and under the environment with tensile stress, and determine the rule constraint coefficient at each depth position; The degree of morphological anomaly at each depth position is adjusted using the rule constraint coefficient to obtain a directional variation index at each depth position; wherein the degree of morphological anomaly and the rule constraint coefficient are both positively correlated with the directional variation index.

3. The method for simulating steel plate pitting corrosion based on a cellular automaton model according to claim 2, characterized in that: The determining of the rule constraint coefficient for each depth position includes: Taking any depth position as the target depth position, the variance of the horizontal width difference between the target depth positions in the pit morphology curve of each two consecutive corrosion times of the cell under the tensile stress environment is calculated, and recorded as the first abnormal index of the pitting pit at the target depth position under the set natural rule under the tensile stress environment; The horizontal width difference is the difference between the horizontal widths of the target depth positions in the pit morphology curves of two consecutive corrosion times, and the corrosion time is the time interval corresponding to the corrosion start time point to the corrosion end time point; The variance of the horizontal width difference between the target depth positions in the pit morphology curve of each two consecutive corrosion times of the cell in the non-tensile stress environment is calculated, and recorded as the second abnormal index of the pitting pit at the target depth position under the natural rule set in the non-tensile stress environment; A difference value between the first anomaly index and the second anomaly index is calculated, the difference value is normalized to obtain a normalized value, and the normalized value is used as a rule constraint coefficient of the target depth position.

4. The method for simulating steel plate pitting corrosion based on a cellular automaton model according to claim 1, characterized in that: The step of determining the expansion coefficient corresponding to the pit surface at every two consecutive etching times under the same solution concentration includes: For any two consecutive corrosion times, the pit surface is segmented by using triangular meshes of the same specification. The curvature, torsion and coordinates of each vertex position on the grid are extracted, the Euclidean norms of the coordinates, curvature and torsion between different positions in the two pit surfaces are calculated, and the average Euclidean norm is used as the matching result of the corresponding position; The minimum matching result is selected from the matching results of all positions, and the minimum matching result is used as the expansion coefficient corresponding to the etch pit surface under two consecutive etching times.

5. The method for simulating steel plate pitting corrosion based on a cellular automaton model according to claim 1, characterized in that: The analysis of the corrosion abnormality of all the slope areas inside the etch pit at different corrosion times according to several different solution concentrations, combined with the expansion coefficient, determines the local corrosion progression coefficient at each depth position at different solution concentrations, including: Acquire three-dimensional morphological data of the etch pits at each etching time under several different solution concentrations, and segment the etch pits according to the three-dimensional morphological data to obtain the corresponding inclined surface areas inside the etch pits; Acquire all normal angles between each slope region and all slope regions adjacent to its edges along the depth position sequence from top to bottom, and determine the dynamic rule competition index of each slope region in combination with the expansion coefficient; Projecting the dynamic rule competition index in a two-dimensional plane, and obtaining a relationship curve between the dynamic rule competition index and the depth position by curve fitting; According to the relationship curve, the average derivative of all dynamic rule competition indicators at each depth position of each solution concentration in time series is calculated to determine the local erosion progression coefficient of each depth position at different solution concentrations.

6. The method for simulating steel plate pitting corrosion based on a cellular automaton model according to claim 5, characterized in that: The step of obtaining all normal angles between each bevel region and all bevel regions adjacent to its edges includes: A coordinate system is created with the vertical direction as the Y axis and the horizontal line where the centroid of the target slope area is located as the X axis, and a normal line passing through the centroid of each slope area is obtained in the coordinate system; According to the normal of each slope region passing through the centroid, all normal angles between each slope region and all the slope regions adjacent to its edges are obtained.

7. The method for simulating steel plate pitting corrosion based on a cellular automaton model according to claim 6, characterized in that: Determining the dynamic rule competition index of each slope area includes: Take any slope area as the target slope area, and take all slope areas adjacent to the edges of the target slope area as the neighboring slope areas; The dynamic rule competition index of the target slope area is determined according to the cosine value of the normal angle between the target slope area and each neighboring slope area, the Z-axis coordinate value of the centroid of each neighboring slope area, and the expansion coefficient.

8. The method for simulating steel plate pitting corrosion based on a cellular automaton model according to claim 7, characterized in that: The step of determining the dynamic rule competition index of the target slope region according to the cosine value of the normal angle between the target slope region and each neighboring slope region, the Z-axis coordinate value of the centroid of each neighboring slope region, and the expansion coefficient includes: Calculate the product of the cosine value of the normal angle between the target slope area and each neighboring slope area and the Z-axis coordinate value of the centroid of the corresponding neighboring slope area, and use the average value of all products as the first dynamic rule competition factor of the target slope area; exponentially amplifying the expansion coefficient, and using the amplified value as a second dynamic rule competition factor of the target slope area; The first dynamic rule competition factor and the second dynamic rule competition factor of the target slope area are integrated to obtain a dynamic rule competition index of the target slope area.

9. The method for simulating steel plate pitting corrosion based on a cellular automaton model according to claim 1, characterized in that: The generating of dynamic rules for each cell to evolve toward the corresponding class slope includes: Calculate the product of the number of intersections between each type of slope and each type of depth position of the local erosion progression coefficient and the directional variation index of the corresponding depth position to obtain the degree of membership of each type of slope to each type of local erosion progression coefficient; For any cell, determine the membership of each type of slope to the weighted mean of the local erosion progressive coefficients of all cells above the cell, and take the type of slope corresponding to the largest membership as the evolution direction of the cell, so that the cell generates dynamic rules for evolving toward the corresponding type of slope.

10. A steel plate pitting simulation system based on a cellular automaton model, characterized in that: It comprises a processor and a memory, wherein the processor is used to process instructions stored in the memory to implement the steel plate pitting simulation method based on a cellular automaton model as described in any one of claims 1 to 9.

Images