Corrosion evolution and performance degradation simulation method for three-dimensional adaptive cellular automaton

Abstract

A three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method belongs to the field of corrosion evolution simulation, and comprises the following steps: S1, determining a corrosion environment parameter and a corrosion evolution parameter rule; s2, generating m groups of corrosion environment parameters based on the corrosion environment; s3, carrying out n times of cellular automaton simulation of the current time step by using each group of corrosion environment parameters based on the result of the previous time step, and calculating the accuracy of corrosion evolution parameter simulation under each group of parameters; s4, extracting a% m groups with high accuracy from the m groups of corrosion environment parameters, exchanging adjacent array sub-parameters with a b% probability, changing the sub-parameters with a c% probability, generating (1-a%) m groups of parameters, and merging the (1-a%) m groups of parameters into the extracted a% m groups to regenerate m groups of parameters; s5, repeating the steps S3-S4 until the accuracy reaches a target value; and S6, if k times of repetition are still not reached, performing 3-to-2 fission on each cell, and re-executing the steps S2-S5 and the like. According to the method, corrosion parameters can be adaptively optimized to improve the precision.

Description

Technical Field

[0001] The invention relates to the field of corrosion evolution simulation, and in particular to a three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method. Background Art

[0002] In the context of the rapid development of modern industry, the corrosion problem of metal materials has become particularly prominent, especially in extreme environments such as the ocean, industrial environment and urban atmosphere. Metal corrosion not only reduces the performance of the material, but may also cause structural damage, leading to serious economic and safety problems. Although experimental studies can provide direct corrosion data, these methods are usually time-consuming and costly, and it is difficult to reproduce the corrosion evolution of the entire process of long service time. In recent years, cellular automaton model technology has been gradually applied to the field of material performance evolution research, but some existing metal corrosion models mostly use two-dimensional cellular automaton simulation, or use cellular evolution under fixed parameters, which cannot effectively obtain the changes in metal corrosion environment parameters in actual service, and it is difficult to obtain simulation results that meet the actual corrosion evolution parameters. Summary of the invention

[0003] To solve the above problems, the present invention provides a three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method, which is used to solve the problems of cumbersome three-dimensional cellular automaton parameter adjustment and difficult corrosion model conversion into finite element in the whole process of metal corrosion.

[0004] The technical solution adopted by the present invention to solve the technical problem is: a three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method, comprising the following steps:

[0005] S1: Determine the corrosion environment parameters and corrosion evolution parameter rules, and construct a cellular automaton model;

[0006] S2: Generate m groups of corrosion environment parameters based on the corrosion environment;

[0007] S3: Based on the results of the previous time step, the n-fold cellular automaton simulation of the current time step is carried out with each set of corrosion environment parameters, and the accuracy of the simulation of the corrosion evolution parameters under each set of parameters is calculated;

[0008] S4: extracting a%m groups with high accuracy from the m groups of corrosion environment parameters, exchanging adjacent array sub-parameters with a b% probability, changing sub-parameters with a c% probability, generating (1-a%)m groups of parameters, and merging them into the extracted a%m groups to regenerate m groups of parameters;

[0009] S5: Repeat S3-S4 until the accuracy reaches the target value;

[0010] S6: If the target is not reached after k repetitions, each cell undergoes 3-way binary fission and S2-S5 are executed again;

[0011] S7: Complete the cellular automaton calculation of the current time step, convert it into finite elements using the point cloud mesh generation method, and carry out performance degradation simulation.

[0012] Furthermore, in step S1, the corrosion environment parameters include the concentration of the corrosive medium w c , corrosion probability P c , passivation probability P p and the probability of passive film ablation P b , where corrosion and passivation are based on the premise that the corrosive medium cell contacts the metal cell. The two events are mutually exclusive and P c +P p <1, when neither event occurs, the probability that the metal cell remains in its original state is 1-(P c +P p ); the passivation rupture is based on the premise that the corrosive medium cell contacts the passivation cell. When this event does not occur, the probability that the metal cell remains in its original state is 1-P b .

[0013] Furthermore, in step S1, constructing a cellular automaton model includes the following steps:

[0014] (1) The various elements involved in the corrosion process are converted into cellular units, including four types of cells:

[0015] Metal cell M: When a metal cell is adjacent to a corrosive medium unit, corrosion may occur and the cell may be transformed into a passivated cell or a state where no chemical reaction occurs. The cell position remains unchanged during the simulation process.

[0016] Passivation cell DM: It refers to the metal in the passivation state. It will not corrode or repeat the passivation reaction, but the passivation film may be ablated or no chemical reaction may occur. It is located in the outermost layer of the metal and has a fixed position.

[0017] Corrosive medium cell C: Corrosive medium cells and water or inert environmental factor cells together constitute a corrosive environment. Corrosive medium cells are corrosive and can react with metal cells to cause the metal cells to corrode and disappear. The cell position can move freely in three-dimensional directions.

[0018] Water or inert environmental factor cell W: Water or inert environmental factor that does not react with metal cells, and the cell position can move freely in three-dimensional directions;

[0019] (2) Establishing the three-dimensional cellular space and calculation area for metal corrosion simulation:

[0020] A three-dimensional cellular space composed of I×J×K cube cells is constructed. The upper and lower sides of the space are defined as fixed boundaries to prohibit cells from moving outside the upper and lower boundaries. The surrounding areas of the space are defined as periodic boundaries to ensure that cells will return to the three-dimensional cellular space from the other side when they move outside the boundaries. The von Neumann neighborhood is used to determine the contact relationship between cells. The selected cell may only react with its 26 adjacent cells.

[0021] Considering the actual corrosion reaction, the constructed three-dimensional cellular space is divided into the upper corrosion environment area, the activation reaction area and the lower stable metal area:

[0022] Corrosion environment area: the upper middle of the initially generated cellular space The regional cells are the corrosive environment areas. When the cellular automaton is iteratively calculated, the area occupied by the upper cells C and W only contains cells C and W, and the number of the two types of cells is determined by the concentration of the corrosive medium w. c Decide:

[0023] Activation reaction area: The cells DM and M that have von Neumann contact relationship with the corrosion environment area constitute the activation reaction area, which means the metal cells and passivation cells that are in direct contact with the corrosion environment may react with the corrosive medium to cause corrosion disappearance, conversion into passivation cells or passive film ablation reaction;

[0024] Stable metal region: All cells below the activation reaction region are stable metal regions, which are not in contact with the corrosive medium, so no reaction occurs in each step of the calculation;

[0025] (3) Clarify the cellular computing method and evolution rules:

[0026] At the beginning of each iteration, the corrosive environment area is calculated according to the concentration of the corrosive medium w c Randomly generate cells C or W, close the stable metal region in the previous iterative calculation results, and the activation reaction region in the previous calculation results contains metal cells M and passivation cells DM;

[0027] In the current iteration, the cell DM in the activation reaction area is searched with probability P b The transformation of the cell DM into the cell M represents the process of the passive film ablation and transformation into the corrodible metal cell M;

[0028] Each corrosive medium cell C in the corrosive environment area randomly selects one direction from the 26 adjacent cells to move; different evolution modes occur according to the type of contact cells in the moving direction of cell C: when cell C selects cell W or cell C to contact in the moving direction, the two position cells are exchanged; when cell C selects cell M to contact in the moving direction, cell M moves in the direction of P. cThe probability is converted into a cell W, which represents the corrosion disappearance process, or P b The probability is converted into a passivation cell DM, which represents the metal passivation process; when the cell C selects the cell DM in the moving direction, the two position cells remain unchanged;

[0029] Based on the cell state when the current iterative calculation is completed, the cells DM and M that have direct von Neumann contact with the corrosion environment area are searched to update the activation reaction area flag.

[0030] Furthermore, the corrosion evolution parameters include mass loss rate μ, average corrosion depth D ave , Maximum corrosion depth D m and non-uniform corrosion depth D nave ;

[0031] The calculation method for the measured values ​​of the above parameters in the metal corrosion test is as follows:

[0032] In the metal corrosion test, the initial mass of the specimen is assumed to be G 0 , after x time steps the remaining mass is G x , the expanded plane area of ​​the corrosion surface is A, then the mass loss rate μ is calculated as Measured average corrosion depth The maximum corrosion depth measured is D′ m , the measured non-uniform corrosion depth is D n ' ave =D m -D ave ;

[0033] The calculation method of the above parameter simulation values ​​in cellular automaton simulation is as follows:

[0034] In the cellular automaton simulation, each cell is identified by the three-dimensional cell number (i, j, k). Assuming that the side length of a single cell unit is l, the metal height at the plane position (i, j) is: d(i, j) = lmink(i, j) where

[0035] ,mink(i,j) is the z-direction number of the cell in the active state at the plane position (i,j);

[0036] Then the average corrosion depth D in the cellular automaton simulation is ave The calculation method is: The accuracy of the average corrosion depth is:

[0037] Maximum corrosion depth D in cellular automaton simulation m The calculation method is: D m =H 0 -min(d(i,j)), where H 0is the height of the metal layer before corrosion, and the accuracy of the maximum corrosion depth is:

[0038] Non-uniform corrosion depth D in cellular automaton simulation nave The calculation method is: D nave =D ave -min(d(i,j)), the accuracy of non-uniform corrosion depth is: The accuracy of the corrosion evolution parameters is close to 1, indicating that the simulated parameters are close to the measured parameters.

[0039] Furthermore, in step S2, the generated corrosion environment parameters include the concentration of the corrosive medium w c , corrosion probability P c , passivation probability P p and the probability of passive film ablation P b 4-factor array, each sub-parameter is randomly generated based on the actual corrosion environment parameter variation range; the concentration of the corrosive medium w c The variation interval is (0, 1), the corrosion probability P c The variation interval is (0, 0.75), the passivation probability P p The variation interval is (P c , 0.25) and the probability of passivation film ablation P b The variation interval is (0, 1); the generated m groups of corrosion environment parameters are expressed as:

[0040] Furthermore, the specific steps of step S3 are as follows:

[0041] (1) The result of the previous time step is the cellular state and relative position of the activation area and the corrosion environment area after the cellular automation calculation of the previous time step;

[0042] (2) For each set of corrosion environment parameters, n cellular automaton operations are performed. The process of each cellular automaton operation is as follows: based on the selected set of corrosion environment parameters [w c (y),P c (y),P p (y),P b (y)] (y=1,2,…m) is the concentration of corrosive medium w c (y) Update the proportion of corrosive cells in the corrosive environment area based on P c (y),P p (y),P b (y) Carry out the cellular automaton evolution operation of the current time step, and calculate the average corrosion depth accuracy R(D) of the current operation step after each evolution operation. ave ), if the average corrosion depth accuracy R(D ave)<0.98, the cellular automaton evolution operation is continued until the average corrosion depth accuracy R(D ave )≧0.98, the cellular automaton operation is completed; the average corrosion depth accuracy R(D ave ) z , maximum corrosion depth accuracy R(D m ) z and non-uniform corrosion depth accuracy R(D nave ) z , calculate the comprehensive accuracy of the cellular automaton operation e r (z): Where z is the number of cellular automaton operations currently being performed, z = 1, 2, ... n;

[0043] (3) For the selected yth group of corrosion environment parameters, the n-th cellular automation result is used to calculate the overall accuracy under the yth group of corrosion environment parameters.

[0044] (4) Record the minimum comprehensive accuracy e in n-fold cellular automaton operations r (z) The corresponding cellular automaton activation area and corrosion environment area cellular state and relative position calculation results;

[0045] (5) This step carries out mn cellular automaton operations, and each set of corrosion environment parameters can obtain an overall accuracy E r , and the cellular states and relative positions of the cellular automaton activation area and corrosion environment area corresponding to the minimum comprehensive accuracy.

[0046] Furthermore, the specific steps of step S4 are as follows:

[0047] (1) extracting ceil(a%m) of the corrosion environment parameters from m groups of corrosion environment parameters in order of overall accuracy from high to low, where ceil(a%m) is the smallest integer greater than or equal to a%m;

[0048] (2) Randomly select two groups from the ceil (a% m) groups of corrosion environment parameters as adjacent arrays, extract ceil (a% m) times, and obtain ceil (a% m) columns of adjacent arrays. Interchange the corrosive medium concentrations w in the adjacent arrays one by one with a probability of b%. c , corrosion probability P c , passivation probability P p and the probability of passive film ablation P b Parameters, generate ceil (a% m) group to update corrosion environment parameters;

[0049] (3) Update the corrosion environment parameters of the ceil (a% m) group and randomly vary the concentration of the corrosive medium w with a probability of c%c , corrosion probability P c , passivation probability P p and the probability of passive film ablation P b Parameters, the variation range must conform to the actual variation range of each corrosion environment parameter, and the ceil (a% m) group is obtained to update the variation corrosion environment parameters;

[0050] (4) Execute step S3 to obtain the overall accuracy E corresponding to the ceil (a% m) group of updated variant corrosion environment parameters. r , and the cellular states and relative positions of the cellular automaton activation area and the corrosion environment area corresponding to the minimum comprehensive accuracy;

[0051] (5) Extract the m-ceil (a% m) group of parameters with the highest comprehensive accuracy from the ceil (a% m) group of updated variant corrosion environment parameters, merge them into the initially extracted ceil (a% m) group of corrosion environment parameters, and re-obtain m groups of corrosion environment parameters and the overall accuracy E corresponding to each group of parameters r The cellular states and relative positions of the cellular automaton activation area and corrosion environment area corresponding to the minimum comprehensive accuracy.

[0052] Further, in step S5, the method for determining whether the accuracy reaches the target value is as follows: selecting |E r -1|The smallest parameter group, such as min(|E r (y)-1|)<0.05, the accuracy reaches the target value, corresponding to min(|E r The cellular states and relative positions of the cellular automaton activation area and the corrosion environment area corresponding to the minimum comprehensive accuracy under the corrosion environment parameters of (y)-1|) are the cellular automaton calculation results when the calculation of the current time step is completed.

[0053] Furthermore, in step S6, the method of the three-way binary fission of each cell is as follows: each cell in the activation area and the corrosion environment area after the cellular automaton calculation in the previous time step is evenly divided into two along three directions, and the cell properties and parameters remain unchanged, and each cell fissions into 8 cells of the same type with the same size and half the side length l.

[0054] Furthermore, in step S7, the grid generation method is:

[0055] (1) Obtaining the position coordinate data of the cell M and the cell DM in the activation reaction area in the cellular automaton, and calculating the coordinates of the center points of each cell in the activation reaction area according to the cell unit side length l;

[0056] (2) Generate a surface model of the corrosion morphology surface based on the center points of each cell in the activation reaction area;

[0057] (3) Using triangular or quadrilateral meshes to mesh the surface model, based on the surface mesh and controlling the size of the volume unit, volume units are generated for the stable metal area;

[0058] (4) Import the generated solid unit model into the finite element analysis software to give the metal material mechanical characteristics.

[0059] Compared with the prior art, the beneficial effects of the present invention are: the corrosion environment parameters and cell size of the cellular automaton can be adaptively adjusted within a reasonable range according to the corrosion evolution parameters, and the effective corrosion calculation area can be updated and judged in real time, so as to achieve a more efficient and more actual corrosion characteristic full-process corrosion evolution prediction, and evaluate the metal corrosion degradation performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a flow chart of the three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method;

[0061] Figure 2 is a schematic diagram of the cell space;

[0062] Figure 3 It is a schematic diagram of the von Neumann contact relationship of a three-dimensional cell;

[0063] Figure 4 It is a schematic diagram of three-dimensional cell types and cell evolution rules;

[0064] Figure 5 It is a schematic diagram of the adaptive change process of corrosion environment parameters;

[0065] Figure 6 It is a schematic diagram of three-dimensional cell fission;

[0066] Figure 7 It is a comparison chart of the corrosion morphology evolution parameters between cellular automaton simulation and experiment;

[0067] Figure 8 It is a schematic diagram of the process of converting cellular automaton simulation into finite element model. DETAILED DESCRIPTION

[0068] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0069] like Figure 1 As shown, a three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method includes the following steps:

[0070] S1: Determine the corrosion environment parameters and corrosion evolution parameter rules, and construct a cellular automaton model;

[0071] S2: Generate m groups of corrosion environment parameters based on the corrosion environment;

[0072] S3: Based on the results of the previous time step, the n-fold cellular automaton simulation of the current time step is carried out with each set of corrosion environment parameters, and the accuracy of the simulation of the corrosion evolution parameters under each set of parameters is calculated;

[0073] S4: extracting a%m groups with high accuracy from the m groups of corrosion environment parameters, exchanging adjacent array sub-parameters with a b% probability, changing sub-parameters with a c% probability, generating (1-a%)m groups of parameters, and merging them into the extracted a%m groups to regenerate m groups of parameters;

[0074] S5: Repeat S3-S4 until the accuracy reaches the target value;

[0075] S6: If the target is not reached after k repetitions, each cell undergoes 3-way binary fission and S2-S5 are executed again;

[0076] S7: Complete the cellular automaton calculation of the current time step, convert it into finite elements using the point cloud mesh generation method, and carry out performance degradation simulation.

[0077] The corrosion environment parameters include the concentration of the corrosive medium w c , corrosion probability P c , passivation probability P p and the probability of passive film ablation P b , where corrosion and passivation are based on the premise that the corrosive medium cell contacts the metal cell. The two events are mutually exclusive and P c +P p <1, when neither event occurs, the probability that the metal cell remains in its original state is 1-(P c +P p ); the passivation rupture is based on the premise that the corrosive medium cell contacts the passivation cell. When this event does not occur, the probability that the metal cell remains in its original state is 1-P b .

[0078] like Figure 2-Figure 4 As shown, the cellular automaton includes 4 types of cells:

[0079] Metal cell M: When a metal cell is adjacent to a corrosive medium unit, corrosion may occur and the cell may disappear, transform into a passivated cell, or no chemical reaction may occur. The cell position remains unchanged during the simulation.

[0080] Passivation cell DM: refers to the metal in the passivation state. It will not corrode or repeat the passivation reaction, but the passivation film may be ablated or no chemical reaction may occur. It is located in the outermost layer of the metal and has a fixed position.

[0081] Corrosive medium cell C: Corrosive medium cells and water or inert environmental factor cells together constitute a corrosive environment. Corrosive medium cells are corrosive and can react with metal cells to cause the metal cells to corrode and disappear. The cell position can move freely in three dimensions.

[0082] Water or inert environmental factor cell W: Water or inert environmental factor that does not react with metal cells, and the cell position can move freely in three-dimensional directions.

[0083] During the initial calculation, a three-dimensional cellular space consisting of I×J×K cube cells is constructed. The upper and lower sides of the space are defined as fixed boundaries to prohibit cells from moving outside the upper and lower boundaries. The surrounding areas of the space are defined as periodic boundaries to ensure that when cells move outside the boundaries, they will return to the three-dimensional cellular space from the other side. The von Neumann neighborhood is used to determine the contact relationship between cells. The selected cell may only react with its 26 adjacent cells, such as Figure 3 shown.

[0084] like Figure 2 As shown, the constructed three-dimensional cellular space is divided into the upper corrosion environment area, the activation reaction area and the lower stable metal area:

[0085] Corrosion environment area: the upper middle of the initially generated cellular space The regional cells are the corrosive environment areas. When the cellular automaton is iteratively calculated, the area occupied by the upper cells C and W only contains cells C and W, and the number of the two types of cells is determined by the concentration of the corrosive medium w. c Decide:

[0086] Activation reaction area: The cells DM and M that have von Neumann contact relationship with the corrosion environment area constitute the activation reaction area, which means that the metal cells and passivation cells that are in direct contact with the corrosion environment may react with the corrosive medium to cause corrosion disappearance, conversion into passivation cells or passive film ablation reaction.

[0087] Stable metal region: All cells below the activation reaction region are stable metal regions that are not in contact with the corrosive medium, so no reaction occurs in each step of the calculation.

[0088] Figure 4 The following shows the cellular computing method and evolution rules:

[0089] At the beginning of each iterative calculation, the corrosion environment area randomly generates cells C or cells W according to the concentration of the corrosive medium wc, closes the stable metal area in the previous iterative calculation result, and the activation reaction area in the previous calculation result contains metal cells M and passivation cells DM.

[0090] In the current iteration, the cell DM in the activation reaction area is searched with probability P b The transformation of cell DM into cell M represents the process of the passive film being ablated and transformed into the corrodible metal cell M.

[0091] Each corrosive medium cell C in the corrosive environment area randomly selects one direction from the 26 adjacent cells to move; different evolution modes occur according to the type of contact cells in the moving direction of cell C: when cell C selects cell W or cell C to contact in the moving direction, the two position cells are exchanged; when cell C selects cell M to contact in the moving direction, cell M moves in the direction of P. c The probability is converted into a cell W, which represents the corrosion disappearance process, or P b The probability is converted into a passivation cell DM, which represents the metal passivation process; when the cell C selects the cell DM to be in contact in the moving direction, the two position cells remain unchanged.

[0092] Based on the cell state when the current iterative calculation is completed, the cells DM and M that have direct von Neumann contact with the corrosion environment area are searched to update the activation reaction area flag.

[0093] The corrosion evolution parameters include mass loss rate μ, average corrosion depth D ave , Maximum corrosion depth D m and non-uniform corrosion depth D nave ;

[0094] The calculation method for the measured values ​​of the above parameters in the metal corrosion test is as follows:

[0095] In the metal corrosion test, the initial mass of the specimen is assumed to be G 0 , after x time steps the remaining mass is G x , the expanded plane area of ​​the corrosion surface is A, then the mass loss rate μ is calculated as Measured average corrosion depth The maximum measured corrosion depth is D′ m , the measured non-uniform corrosion depth is D n ' ave =D m -D ave .

[0096] The calculation method of the above parameter simulation values ​​in cellular automaton simulation is as follows:

[0097] In the cellular automaton simulation, each cell is identified by the three-dimensional cell number (i, j, k). Assuming that the side length of a single cell unit is l, the metal height at the plane position (i, j) is: d(i, j) = lmink(i, j) where

[0098] ,mink(i,j) is the z-direction number of the cell in the active state at the plane position (i,j).

[0099] Then the average corrosion depth D in the cellular automaton simulation is ave The calculation method is: The accuracy of the average corrosion depth is:

[0100] Maximum corrosion depth D in cellular automaton simulation m The calculation method is: D m =H 0 -min(d(i,j)), where H 0 is the height of the metal layer before corrosion, and the accuracy of the maximum corrosion depth is:

[0101] Non-uniform corrosion depth D in cellular automaton simulation nave The calculation method is: D nave =D ave -min(d(i,j)), the accuracy of non-uniform corrosion depth is: The accuracy of the corrosion evolution parameters is close to 1, indicating that the simulated parameters are close to the measured parameters.

[0102] like Figure 5 As shown, the generated corrosion environment parameters include the concentration of the corrosive medium w c , corrosion probability P c , passivation probability P p and the probability of passive film ablation P b 4-factor array, each sub-parameter is randomly generated based on the actual corrosion environment parameter variation range. c The variation interval is (0, 1), the corrosion probability P c The variation interval is (0, 0.75), the passivation probability P p The variation interval is (P c , 0.25) and the probability of passivation film ablation P b The variation interval is (0, 1); the generated m groups of corrosion environment parameters are expressed as:

[0103]

[0104] In the cellular automaton calculation of each time step, (1) the result of the previous time step is the state and relative position of the cells in the activation area and the corrosion environment area after the cellular automaton calculation of the previous time step; (2) for each set of corrosion environment parameters, n cellular automaton operations are performed, and each cellular automaton operation process is: based on the selected set of corrosion environment parameters [w c (y),P c (y),P p (y),P b (y)] (y=1,2,…m) is the concentration of corrosive medium w c (y) Update the proportion of corrosive cells in the corrosive environment area based on P c (y),P p(y),P b (y) Carry out the cellular automaton evolution operation of the current time step, and calculate the average corrosion depth accuracy R(D) of the current operation step after each evolution operation. ave ), if the average corrosion depth accuracy R(D ave )<0.98, the cellular automaton evolution operation is continued until the average corrosion depth accuracy R(D ave )≧0.98, the cellular automaton operation is completed; the average corrosion depth accuracy R(D ave ) z , maximum corrosion depth accuracy R(D m ) z and non-uniform corrosion depth accuracy R(D nave ) z , calculate the comprehensive accuracy of the cellular automaton operation e r (z): Where z is the number of cellular automata operations currently being performed, z = 1, 2, ... n; (3) For the nth cellular automata results under the selected yth group of corrosion environment parameters, the overall accuracy under the yth group of corrosion environment parameters is calculated. (4) Record the minimum comprehensive accuracy e in n-fold cellular automaton operations r (z) The corresponding cellular automaton activation area and corrosion environment area cellular state and relative position calculation results; (5) This step carries out mn times of cellular automaton calculations, and each set of corrosion environment parameters can obtain an overall accuracy E r , and the cellular states and relative positions of the cellular automaton activation area and corrosion environment area corresponding to the minimum comprehensive accuracy.

[0105] like Figure 5 As shown, the steps for adaptive adjustment of corrosion environment parameters are as follows:

[0106] (1) From m groups of corrosion environment parameters, extract the number of groups of corrosion environment parameters that is ceil(a%m) (a%≥50%) in order of overall accuracy from high to low, where ceil(a%m) is the smallest integer greater than or equal to a%m;

[0107] (2) Randomly select two groups from the ceil (a% m) groups of corrosion environment parameters as adjacent arrays, extract ceil (a% m) times, and obtain ceil (a% m) columns of adjacent arrays. Interchange the corrosive medium concentrations w in the adjacent arrays one by one with a probability of b%. c , corrosion probability P c , passivation probability P p and the probability of passive film ablation P b Parameters, generate ceil (a% m) group to update corrosion environment parameters;

[0108] (3) Update the corrosion environment parameters of the ceil (a% m) group and randomly vary the concentration of the corrosive medium w with a probability of c% c , corrosion probability P c , passivation probability P p and the probability of passive film ablation P b Parameters, the variation range must conform to the actual variation range of each corrosion environment parameter, and the ceil (a% m) group is obtained to update the variation corrosion environment parameters;

[0109] (4) Execute step S3 to obtain the overall accuracy E corresponding to the ceil (a% m) group of updated variant corrosion environment parameters. r , and the cellular states and relative positions of the cellular automaton activation area and the corrosion environment area corresponding to the minimum comprehensive accuracy;

[0110] (5) Extract the m-ceil (a% m) group of parameters with the highest comprehensive accuracy from the ceil (a% m) group of updated variant corrosion environment parameters, merge them into the initially extracted ceil (a% m) group of corrosion environment parameters, and re-obtain m groups of corrosion environment parameters and the overall accuracy E corresponding to each group of parameters r The cellular states and relative positions of the cellular automaton activation area and corrosion environment area corresponding to the minimum comprehensive accuracy.

[0111] The method for judging whether the accuracy reaches the target value is as follows: select |E r -1|The smallest parameter group, such as min(|E r (y)-1|)<0.05, the accuracy reaches the target value, corresponding to min(|E r The cellular states and relative positions of the cellular automaton activation area and the corrosion environment area corresponding to the minimum comprehensive accuracy under the corrosion environment parameters of (y)-1|) are the cellular automaton calculation results when the calculation of the current time step is completed.

[0112] like Figure 6 As shown, the three-way binary fission method of each cell is as follows: after the cellular automaton calculation in the previous time step, each cell in the activation area and the corrosion environment area is evenly divided into two along three directions, and the cell properties and parameters remain unchanged. Each cell fissions into 8 cells of the same type with the same size and half the side length l.

[0113] Figure 7 In order to adopt the practical effect of the proposed invention method, the corrosion environment parameters and cell size of the cellular automaton can be adaptively adjusted within a reasonable range according to the corrosion evolution parameters to obtain high-precision predictions of the mass loss rate, average corrosion depth and maximum corrosion depth.

[0114] The proposed mesh generation method process is as follows Figure 8As shown, the method mainly includes the following steps: (1) obtaining the position coordinate data of the cell M and the cell DM in the activation reaction area in the cellular automaton, and calculating the coordinates of the center points of each cell in the activation reaction area according to the cell unit side length l. (2) generating a surface model of the corrosion morphology surface based on the center points of each cell in the activation reaction area. (3) using triangular or quadrilateral meshes to mesh the surface model, and generating volume units for the stable metal area based on the surface mesh and controlling the volume unit size. (4) importing the generated volume unit model into the finite element analysis software to give the metal material mechanical characteristics.

[0115] Although the preferred embodiments of the present invention have been described above in conjunction with the accompanying drawings, the present invention is not limited to the above-mentioned specific embodiments, which are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which fall within the scope of protection of the present invention.

Claims

1. A three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method, characterized in that: The following steps are involved: S1: Determine the corrosion environment parameters and corrosion evolution parameter rules, and construct a cellular automaton model; S2: Generate m groups of corrosion environment parameters based on the corrosion environment; S3: Based on the results of the previous time step, the n-fold cellular automaton simulation of the current time step is carried out with each set of corrosion environment parameters, and the accuracy of the simulation of the corrosion evolution parameters under each set of parameters is calculated; S4: extracting a%m groups with high accuracy from the m groups of corrosion environment parameters, exchanging adjacent array sub-parameters with a b% probability, changing sub-parameters with a c% probability, generating (1-a%)m groups of parameters, and merging them into the extracted a%m groups to regenerate m groups of parameters; S5: Repeat S3-S4 until the accuracy reaches the target value; S6: If the target is not reached after k repetitions, each cell undergoes 3-way binary fission and S2-S5 are executed again; S7: Complete the cellular automaton calculation of the current time step, convert it into finite elements using the point cloud mesh generation method, and carry out performance degradation simulation.

2. The three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method according to claim 1, characterized in that: In step S1, the corrosion environment parameters include the concentration of the corrosive medium w c , corrosion probability P c , passivation probability P p and the probability of passive film ablation P b , where corrosion and passivation are based on the premise that the corrosive medium cell contacts the metal cell. The two events are mutually exclusive and P c +P p <1, when neither event occurs, the probability that the metal cell remains in its original state is 1-(P c +P p ); the passivation rupture is based on the premise that the corrosive medium cell contacts the passivation cell. When this event does not occur, the probability that the metal cell remains in its original state is 1-P b .

3. The three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method according to claim 1, characterized in that: In step S1, constructing a cellular automaton model includes the following steps: (1) The various elements involved in the corrosion process are converted into cellular units, including four types of cells: Metal cell M: When a metal cell is adjacent to a corrosive medium unit, corrosion may occur and the cell may be transformed into a passivated cell or a state where no chemical reaction occurs. The cell position remains unchanged during the simulation process. Passivation cell DM: It refers to the metal in the passivation state. It will not corrode or repeat the passivation reaction, but the passivation film may be ablated or no chemical reaction may occur. It is located in the outermost layer of the metal and has a fixed position. Corrosive medium cell C: Corrosive medium cells and water or inert environmental factor cells together constitute a corrosive environment. Corrosive medium cells are corrosive and can react with metal cells to cause the metal cells to corrode and disappear. The cell position can move freely in three-dimensional directions. Water or inert environmental factor cell W: Water or inert environmental factor that does not react with metal cells, and the cell position can move freely in three-dimensional directions; (2) Establishing the three-dimensional cellular space and calculation area for metal corrosion simulation: A three-dimensional cellular space composed of I×J×K cube cells is constructed. The upper and lower sides of the space are defined as fixed boundaries to prohibit cells from moving outside the upper and lower boundaries. The surrounding areas of the space are defined as periodic boundaries to ensure that cells will return to the three-dimensional cellular space from the other side when they move outside the boundaries. The von Neumann neighborhood is used to determine the contact relationship between cells. The selected cell may only react with its 26 adjacent cells. Considering the actual corrosion reaction, the constructed three-dimensional cellular space is divided into the upper corrosion environment area, the activation reaction area and the lower stable metal area: Corrosion environment area: the upper middle of the initially generated cellular space The regional cells are the corrosive environment areas. When the cellular automaton is iteratively calculated, the area occupied by the upper cells C and W only contains cells C and W, and the number of the two types of cells is determined by the concentration of the corrosive medium w. c Decide: Activation reaction area: The cells DM and M that have von Neumann contact relationship with the corrosion environment area constitute the activation reaction area, which means the metal cells and passivation cells that are in direct contact with the corrosion environment may react with the corrosive medium to cause corrosion disappearance, conversion into passivation cells or passive film ablation reaction; Stable metal region: All cells below the activation reaction region are stable metal regions, which are not in contact with the corrosive medium, so no reaction occurs in each step of the calculation; (3) Clarify the cellular computing method and evolution rules: At the beginning of each iterative calculation, the corrosion environment area randomly generates cells C or W according to the concentration of the corrosive medium wc, closes the stable metal area in the previous iterative calculation result, and the activation reaction area in the previous calculation result contains metal cells M and passivation cells DM; In the current iteration, the cell DM in the activation reaction area is searched with probability P b The transformation of the cell DM into the cell M represents the process of the passive film ablation and transformation into the corrodible metal cell M; Each corrosive medium cell C in the corrosive environment area randomly selects one direction from the 26 adjacent cells to move; different evolution modes occur according to the type of contact cells in the moving direction of cell C: when cell C selects cell W or cell C to contact in the moving direction, the two position cells are exchanged; when cell C selects cell M to contact in the moving direction, cell M moves in the direction of P. c The probability is converted into a cell W, which represents the corrosion disappearance process, or P b The probability is converted into a passivation cell DM, which represents the metal passivation process; when the cell C selects the cell DM in the moving direction, the two position cells remain unchanged; Based on the cell state when the current iterative calculation is completed, the cells DM and M that have direct von Neumann contact with the corrosion environment area are searched to update the activation reaction area flag.

4. The three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method according to claim 1, characterized in that: The corrosion evolution parameters include mass loss rate μ, average corrosion depth D ave , Maximum corrosion depth D m and non-uniform corrosion depth D nave ; The calculation method for the measured values ​​of the above parameters in the metal corrosion test is as follows: In the metal corrosion test, it is assumed that the initial mass of the specimen is G0, and the remaining mass after x time steps is G x , the expanded plane area of ​​the corrosion surface is A, then the mass loss rate μ is calculated as Measured average corrosion depth The maximum corrosion depth measured is D′ m , the measured non-uniform corrosion depth is D n ' ave =D m -D ave ; The calculation method of the above parameter simulation values ​​in cellular automaton simulation is as follows: In the cellular automaton simulation, each cell is identified by the three-dimensional cell number (i, j, k). Assuming that the side length of a single cell unit is l, the metal height at the plane position (i, j) is: d(i, j) = lmink(i, j) where, mink(i,j) is the z-direction number of the cell in the active state at the plane position (i,j); Then the average corrosion depth D in the cellular automaton simulation is ave The calculation method is: The accuracy of the average corrosion depth is: Maximum corrosion depth D in cellular automaton simulation m The calculation method is: D m =H0-min(d(i,j)), where H0 is the height of the metal layer before corrosion, and the accuracy of the maximum corrosion depth is: Non-uniform corrosion depth D in cellular automaton simulation nave The calculation method is: D nave =D ave -min(d(i,j)), the accuracy of non-uniform corrosion depth is: The accuracy of the corrosion evolution parameters is close to 1, indicating that the simulated parameters are close to the measured parameters.

5. The three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method according to claim 1, characterized in that: In step S2, the generated corrosion environment parameters include the concentration of the corrosive medium w c , corrosion probability P c , passivation probability P p and the probability of passive film ablation P b 4-factor array, each sub-parameter is randomly generated based on the actual corrosion environment parameter variation range; the concentration of the corrosive medium w c The variation interval is (0, 1), the corrosion probability P c The variation interval is (0, 0.75), the passivation probability P p The variation interval is (P c , 0.25) and the probability of passivation film ablation P b The variation interval is (0, 1); the generated m groups of corrosion environment parameters are expressed as:

6. The three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method according to claim 1, characterized in that: The specific steps of step S3 are as follows: (1) The result of the previous time step is the cellular state and relative position of the activation area and the corrosion environment area after the cellular automation calculation of the previous time step; (2) For each set of corrosion environment parameters, n cellular automaton operations are performed. Each cellular automaton operation process is as follows: based on the selected set of corrosion environment parameters [w c (y),P c (y),P p (y),P b (y)] (y=1,2,…m) is the concentration of corrosive medium w c (y) Update the proportion of corrosive cells in the corrosive environment area based on P c (y),P p (y),P b (y) Carry out the cellular automaton evolution operation of the current time step, and calculate the average corrosion depth accuracy R(D) of the current operation step after each evolution operation. ave ), if the average corrosion depth accuracy R(D ave )<0.98, the cellular automaton evolution operation is continued until the average corrosion depth accuracy R(D ave )≧0.98, the cellular automaton operation is completed; the average corrosion depth accuracy R(D ave ) z , maximum corrosion depth accuracy R(D m ) z and non-uniform corrosion depth accuracy R(D nave ) z , calculate the comprehensive accuracy of the cellular automaton operation e r (z): Where z is the number of cellular automaton operations currently being performed, z = 1, 2, ... n; (3) For the selected yth group of corrosion environment parameters, the n-th cellular automation result is used to calculate the overall accuracy under the yth group of corrosion environment parameters. (4) Record the minimum comprehensive accuracy e in n-fold cellular automaton operations r (z) The corresponding cellular automaton activation area and corrosion environment area cellular state and relative position calculation results; (5) This step carries out mn cellular automaton operations, and each set of corrosion environment parameters can obtain an overall accuracy E r , and the cellular states and relative positions of the cellular automaton activation area and corrosion environment area corresponding to the minimum comprehensive accuracy.

7. The three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method according to claim 1, characterized in that: The specific steps of step S4 are as follows: (1) extracting ceil(a%m) of the corrosion environment parameters from m groups of corrosion environment parameters in order of overall accuracy from high to low, where ceil(a%m) is the smallest integer greater than or equal to a%m; (2) Randomly select two groups from the ceil (a% m) groups of corrosion environment parameters as adjacent arrays, extract ceil (a% m) times, and obtain ceil (a% m) columns of adjacent arrays. Interchange the corrosive medium concentrations w in the adjacent arrays one by one with a probability of b%. c , corrosion probability P c , passivation probability P p and the probability of passive film ablation P b Parameters, generate ceil (a% m) group to update corrosion environment parameters; (3) Update the corrosion environment parameters of the ceil (a% m) group and randomly vary the concentration of the corrosive medium w with a probability of c% c , corrosion probability P c , passivation probability P p and the probability of passive film ablation P b Parameters, the variation range must conform to the actual variation range of each corrosion environment parameter, and the ceil (a% m) group is obtained to update the variation corrosion environment parameters; (4) Execute step S3 to obtain the overall accuracy E corresponding to the ceil (a% m) group of updated variant corrosion environment parameters. r , and the cellular states and relative positions of the cellular automaton activation area and the corrosion environment area corresponding to the minimum comprehensive accuracy; (5) Extract the m-ceil (a% m) group of parameters with the highest comprehensive accuracy from the ceil (a% m) group of updated variant corrosion environment parameters, merge them into the initially extracted ceil (a% m) group of corrosion environment parameters, and re-obtain m groups of corrosion environment parameters and the overall accuracy E corresponding to each group of parameters r The cellular states and relative positions of the cellular automaton activation area and corrosion environment area corresponding to the minimum comprehensive accuracy.

8. The three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method according to claim 1, characterized in that: In step S5, the method for determining whether the accuracy reaches the target value is as follows: select |E r -1|The smallest parameter group, such as min(|E r (y)-1|)<0.05, the accuracy reaches the target value, corresponding to min(|E r The cellular states and relative positions of the cellular automaton activation area and the corrosion environment area corresponding to the minimum comprehensive accuracy under the corrosion environment parameters of (y)-1|) are the cellular automaton calculation results when the calculation of the current time step is completed.

9. The three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method according to claim 1, characterized in that: In step S6, the method of three-way binary fission of each cell is as follows: each cell in the activation area and the corrosion environment area after the cellular automaton calculation in the previous time step is evenly divided into two along three directions, and the cell properties and parameters remain unchanged. Each cell fissions into 8 cells of the same type with the same size and half the side length l.

10. The three-dimensional adaptive cellular automaton corrosion evolution and performance degradation simulation method according to claim 1, characterized in that: In step S7, the grid generation method is: (1) Obtaining the position coordinate data of the cells M and DM in the activation reaction area in the cellular automaton, and calculating the coordinates of the center points of each cell in the activation reaction area according to the side length l of the cell unit; (2) Generate a surface model of the corrosion morphology surface based on the center points of each cell in the activation reaction area; (3) Using triangular or quadrilateral meshes to mesh the surface model, based on the surface mesh and controlling the size of the volume unit, volume units are generated for the stable metal area; (4) Import the generated solid unit model into the finite element analysis software to give the metal material mechanical characteristics.

Images