A steel member analysis method considering the evolution method of corrosion pits

Abstract

The present application discloses a steel member analysis method considering the corrosion pit evolution method, belonging to the technical field of steel members. The technical key points are as follows: Step 1: Use a three-dimensional topography scanner to scan the three-dimensional topography of the corroded surface of the same steel member at multiple corrosion times; Step 2: Give the relationship between the expected corrosion pit depth and time and the relationship between the relevant standard deviation and time; Step 3: Give the relationship between the expected corrosion pit radius bt ‑ and the relationship between the relevant standard deviation and time; Step 4: Determine the total number Q of corrosion pits; Step 5: Establish a numerical calculation model of the steel member and then conduct analysis. By using the method of the present application, it is possible to better predict the influence of corrosion on the bearing capacity and deformation of the steel structure at a certain future moment.

Description

Technical Field

[0001] The present invention relates to the field of building structures (especially steel members), and particularly relates to an analysis method of steel members considering the evolution method of pitting corrosion. Background Art

[0002] Traditional researchers used the method of artificial pitting corrosion (for example: Reference 1: Jie, Sheng, Junwu, et al. Effect of simulated pitting corrosion on the tensile properties of steel [J]. Construction and Building Materials, 2017; Reference 2: Yuan, Yao, Yang, etc. Experimental study on generalized constitutive model of hull structural plate with multi-parameter pitting corrosion [J]. Ocean Engineering, 2018.) to simulate pitting corrosion, and conducted systematic experimental studies by changing parameters such as the distribution pattern, depth, and diameter of pitting corrosion.

[0003] However, actual pitting corrosion occurs randomly (Reference 3: Meng Z, Wang F, Shi G. A novel evolution model of pitting failure and effect on time-varying meshing stiffness of spur gears [J]. Engineering Failure Analysis, 2020, 120(8): 105068; Reference 4: Zhao Z, Zhang H, Xian L, et al. Tensile strength of Q345 steel with random pitting corrosion based on numerical analysis [J]. Thin-Walled Structures, 2019.), so the results obtained by using the ordered arrangement of artificial pitting corrosion to study the influence of pitting corrosion on the mechanical properties of steel need to be further verified.

[0004] In addition, researchers have also used finite element analysis to study the stress concentration effect caused by pitting corrosion. However, most studies have only focused on the effect of one or two pits. On the other hand, the change of pits over time has been monitored in-situ, and the depth and width of pits on the steel surface at different time periods have been statistically analyzed.

[0005] Therefore, in order to accurately describe the relationship between the degradation of the mechanical properties of steel caused by pits and the corrosion time, the following basic idea is proposed: combining the microscopic statistical results of pits at different corrosion times with the macroscopic mechanical properties, and then establishing a corresponding random pitting model based on finite element software to analyze the mechanical properties of steel. However, how the above idea can be developed into a method that can guide engineers in practical use still requires systematic research. Summary of the Invention

[0006] The purpose of the present invention is to provide an analysis method and its application for corroded steel based on pit fusion to overcome the deficiencies of the prior art.

[0007] The solution of this application is as follows:

[0008] An analysis method for steel members considering the pit evolution method, which includes the following steps:

[0009] Step 1: Use a three-dimensional topography scanner to scan the three-dimensional topography of the corroded surface at multiple corrosion times:

[0010] Record a t , σ at , b t , σ bt at different times t; a t , σ at represent the average value and standard deviation of the pit depth; b t , σ bt represent the average value and standard deviation of the pit radius;

[0011] Step 2: Give the relationship between the expected value a t- of the pit depth and t, and the relationship between σ at and t:

[0012]

[0013] Step 3: Give the relationship between the expected value b t- of the pit radius and t, and the relationship between σ bt and t:

[0014]

[0015] Step 4: The total number of pits: the total number q tn of the pits measured on the last day in Step 1The total number Q of corrosion pits at the predicted tx;

[0016] Step Five: Use ABAQUS software to establish a finite element model that reflects the random pitting distribution at the calculation time of t x and reflects the random pitting distribution:

[0017] S5.1 Calculate a x at time t tx , σ atx , b tx , σ btx :

[0018]

[0019] S5.2 Set q tn corrosion pits in a predetermined area on the surface of the steel member. The depths of the above Q corrosion pits conform to the normal distribution law with an expectation of a tx and a standard deviation of σ atx . The radii of the above Q corrosion pits conform to the normal distribution law with an expectation of b tx and a standard deviation of σ btx .

[0020] Furthermore, Step One also includes:

[0021] S1.0, at time t0:

[0022] Record the total number of corrosion pits, q t0 ;

[0023] According to the obtained sequence of a values of the q t0 corrosion pits: {a 1-t0 a 2-t0 a 3-t0 ............}, fit according to the normal distribution, and calculate the expected value a t0 of the corrosion pit depth and the standard deviation σ at0 ;

[0024] According to the obtained sequence of b values of the q t0 corrosion pits: {b 1-t0 b 2-t0 b 3-t0 ............}, fit according to the normal distribution, and calculate the expected value b t0 of the top surface radius of the corrosion pit and the standard deviation σ bt0 ;

[0025] S1.1, at time t1:

[0026] Record the total number of corrosion pits, q t1 ;

[0027] According to the obtained q t1The a-value sequence of the etch pits: {a 1-t1 a 2-t1 a 3-t1 ............}, fitting according to the normal distribution, calculate the expected value a of the etch pit depth t1 and the standard deviation σ t1 ;

[0028] According to the recorded q t1 The b-value sequence of the etch pits: {b 1-t1 b 2-t1 b 3-t1 ............}, fitting according to the normal distribution, calculate the expected value b of the top surface radius of the etch pit t1 and the standard deviation σ bt1 ;

[0029] ……

[0030] S1.n, t n Time:

[0031] Record the total number of etch pits, q tn

[0032] According to the recorded q tn The a-value sequence of the etch pits: {a 1-tn a 2-tn a 3-tn ............}, fitting according to the normal distribution, calculate the expected value a of the etch pit depth tn and the standard deviation σ tn ;

[0033] According to the recorded q tn The b-value sequence of the etch pits: {b 1-tn b 2-tn b 3-tn ............}, fitting according to the normal distribution, calculate the expected value b of the top surface radius of the etch pit tn and the standard deviation σ btn 。

[0034] Furthermore, step two is as follows

[0035] a t0 、a t1 ………a tn As the a t column, t0, t1………t n As the t column, use the following formula for fitting:

[0036] a t =At B

[0037] σat0 , σ at1 ………σ atn As the σ at column, t0, t1………t n As the t column, the following formula is used for fitting:

[0038] σ at = Ct.

[0039] Step 3: Give the relationship between the expected value of the pit radius bt-t, and the relationship between σ bt -t:

[0040] b t0 , b t1 ………b tn As the b t column, t0, t1………t n As the t column, the following formula is used for fitting:

[0041] b t = Dt E

[0042] σ bt0 , σ bt1 ………σ btn As the σ bt column, t0, t1………t n As the t column, the following formula is used for fitting:

[0043] σ bt = Ft.

[0044] Furthermore, for 5.2 in Step 5, the implementation method is as follows:

[0045] (a) Import the Python NumPy library to call the required functions: randomstate function, normal function, uniform function, randint function;

[0046] (b) Call the randomstate function to record the surface of the pit depth, position and distribution in the initial evolution stage;

[0047] (c) Call the normal function and assign it the pit depth variable and pit radius variable: input a t , σ at , b tx , σ btx , to achieve the normal distribution of the pit depth and pit radius, and by applying the randomstate function, achieve the in-situ increase of the pit depth with the increase of corrosion time;

[0048] (d) Call the uniform function and let it control the pit position variable: Define the boundary region where pits occur and apply it in combination with the randomstate function to keep the pit position unchanged during the evolution process;

[0049] (e) Determine the pit distribution surface through two random integers (0, 1) output by the randint function: If the output is 1, the pits are distributed on surface A; otherwise, they are distributed on surface B. Apply it in combination with the randomstate function to keep the pit distribution surface fixed.

[0050] A method for establishing a numerical model considering the pit evolution method

[0051] Step 1: Use a three-dimensional topography scanner to scan the three-dimensional topography of multiple corrosion time corrosion surfaces:

[0052] Record a t , σ at , b t at different times t; a t , σ at represent the average value and standard deviation of the pit depth; b t represents the average value of the pit radius;

[0053] Step 2: Give the relationship between the expected value a t- of the pit depth and t, and the relationship between σ at and t:

[0054]

[0055] Step 3: Give the relationship between the expected value b of the pit radius and t - t:

[0056] b t = Dt E

[0057] Step 4: The total number of pits: The total number q tn of the pits measured on the last day in Step 1 is the predicted total number of pits at tx;

[0058] Step 5: Use ABAQUS software to establish a finite element model that reflects the random pitting distribution at time t x :

[0059] S5.1 Calculate a x , σ tx , b atx at time t tx :

[0060]

[0061] Set q in a predetermined area on the surface of the steel member tn erosion pits, and the depth of the above Q erosion pits meets the expectation: a tx , with a standard deviation of σ atx ; the depth-to-diameter ratio of each erosion pit satisfies: a tx / b tx .

[0062] Furthermore, step one also includes:

[0063] S1.0, at time t0:

[0064] Record the total number of erosion pits, q t0 ;

[0065] According to the sequence of a values of the q t0 erosion pits obtained from the record: {a 1-t0 a 2-t0 a 3-t0 ............}, fit according to the normal distribution, and calculate the expected erosion pit depth a t0 and the standard deviation σ at0 ;

[0066] According to the sequence of b values of the q t0 erosion pits obtained from the record: {b 1-t0 b 2-t0 b 3-t0 ............}, fit according to the normal distribution, and calculate the expected top surface radius b of the erosion pit t0 ;

[0067] S1.1, at time t1:

[0068] Record the total number of erosion pits, q t1 ;

[0069] According to the sequence of a values of the q t1 erosion pits obtained from the record: {a 1-t1 a 2-t1 a 3-t1 ............}, fit according to the normal distribution, and calculate the expected erosion pit depth a t1 and the standard deviation σ t1 ;

[0070] According to the sequence of b values of the q t1 erosion pits obtained from the record: {b 1-t1 b 2-t1 b 3-t1 ............}, fit according to the normal distribution, and calculate the expected top surface radius b of the erosion pit t1 ;

[0071] ……

[0072] S1.n, t n Time:

[0073] Record the total number of corrosion pits, q tn

[0074] q obtained according to the record tn The a-value sequence of q corrosion pits: {a 1-tn a 2-tn a 3-tn ............}, Fit according to the normal distribution and calculate the expected value a of the corrosion pit depth tn And the standard deviation σ tn ;

[0075] q obtained according to the record tn The b-value sequence of q corrosion pits: {b 1-tn b 2-tn b 3-tn ............}, Fit according to the normal distribution and calculate the expected value b of the top surface radius of the corrosion pit tn .

[0076] Furthermore, step two includes:

[0077] a t0 、a t1 ………a tn As the a t column, t0, t1………t n As the t column, use the following formula for fitting (t in this application represents the time counted from the start of use):

[0078] a t = At B

[0079] σ at0 、σ at1 ………σ atn As the σ at column, t0, t1………t n As the t column, use the following formula for fitting:

[0080] σ at = Ct.

[0081] Furthermore, step three includes:

[0082] Give the relationship between the expected value b of the corrosion pit radius and t, and the relationship between σ bt -t:

[0083] b t0 、b t1 ………b tn As bt Column, t0, t1………t n As the t column, the fitting is performed using the following formula:

[0084] b t = Dt E .

[0085] Furthermore, for 5.2 in Step 5, the implementation method is as follows:

[0086] (a) Import the Python NumPy library to call the required functions: randomstate function, normal function, uniform function, randint function;

[0087] (b) Call the randomstate function to record the surface of the pit depth, position, and distribution at the initial evolution stage;

[0088] (c) Call the normal function and assign it the pit depth variable and depth-to-diameter ratio: input a t , σ at , a tx / b tx , to achieve the normal distribution of the pit depth, and by applying the randomstate function, achieve the in-situ increase of the pit depth with the corrosion time;

[0089] (d) Call the uniform function and let it control the pit position variable: define the boundary region where the pits occur, and keep the pit position unchanged during the evolution process by applying it with the randomstate function;

[0090] (e) The surface of the pit distribution is judged by two random integers (0, 1) output by the randint function: if the output is 1, the pits are distributed on the A surface, otherwise they are distributed on the B surface, and keep the pit distribution surface fixed by applying it with the randomstate function.

[0091] The beneficial effects of this application are as follows:

[0092] First, the basic concept of this application is to solve the problem of "combining the microscopic statistical results of pits at different corrosion times with the macroscopic mechanical properties".

[0093] Second, the second inventive point of this invention is to propose two methods to establish a numerical simulation model:

[0094] 2.1 Based on practice, it is found that: a t , σ at , b t , σ bt The relationship with t satisfies:

[0095]

[0096] This is the basic condition for the implementation of this application.

[0097] 2.2 In addition to the double normal distribution method, a simplified implementation method is also proposed, that is:

[0098] Step Five: Use ABAQUS software to establish a finite element model that reflects the random pitting distribution at time t x :

[0099] S5.1 Calculate a x , σ tx , b atx at time t tx :

[0100] S5.2 Set q tn pits in a predetermined area on the surface of the steel member. The depths of the above Q pits conform to the normal distribution law with an expectation of: a tx , and a standard deviation of σ atx ; the depth-to-diameter ratio of each pit satisfies: a tx / b tx . Brief Description of the Drawings

[0101] The following further describes this application in detail with reference to the embodiments in the drawings, but does not constitute any limitation to this application.

[0102] Figure 1 is a schematic diagram of pit evolution.

[0103] Figure 2 is a schematic diagram of the change in pit depth distribution with corrosion time.

[0104] Figure 3 is a schematic diagram of the expected pit depth at different corrosion times of this application.

[0105] Figure 4 is a schematic diagram of the standard deviation of pit depth at different corrosion times of this application.

[0106] Figure 5 is a schematic diagram of the expected pit diameter at different corrosion times of this application.

[0107] Figure 6 is a schematic diagram of the variance of pit diameter at different corrosion times of this application.

[0108] Figure 7 is a schematic diagram of the random pitting finite element model of this application.

[0109] Figure 8 is a schematic diagram of the pit evolution finite element model of this application.

[0110] Figure 9 It is a schematic diagram for comparing the finite element simulation results and test results of this application.

[0111] Figure 10 It is a flowchart of the method of this application. Detailed implementation manners

[0112] <Basic research: Method for pit evolution>

[0113] Key issue 1: Description of the pit shape.

[0114] For a pit, its shape can be assumed to be semi-ellipsoidal, and its shape expression is:

[0115]

[0116] wherein, the X-axis, Y-axis, and Z-axis are perpendicular to each other; the Y-axis represents the depth direction of the pit, and the X-axis represents the length direction of the pit surface;

[0117] That is, by knowing the three parameters a, b, and c, the shape of the pit can be described.

[0118] In the actual process, during the pit evolution process, it is generally considered that b = c (that is, the surface projection plane of the pit is considered to be circular). That is, for any pit, only the following need to be detected: the depth of the pit and the maximum length of the pit surface (for example: if the actual shape of the pit has irregular edges, take its maximum opening distance) to describe it.

[0119] Traditional problem: As described in the background art, for traditional pitting corrosion simulation methods, the randomness of random generation and development is ignored, that is, artificially arranged pits in an orderly manner are used, or only the stress concentration caused by one or two pits is studied. This leads to errors between the statistical results and the actual situation.

[0120] To solve the above problems, in Example 1, a three-dimensional topography scanner is used to scan the three-dimensional topography of the corroded surface, and statistical methods are used to statistically analyze the parameters:

[0121] <Example 1: A method for establishing a numerical model considering the pit evolution method>

[0122] Step 1: Use a three-dimensional topography scanner to scan the three-dimensional topography of the corroded surface at multiple corrosion times:

[0123] S1.0, at time t0:

[0124] Record the total number of pits, q t0 ;

[0125] According to the recorded q t0 a-value sequence of pits: {a 1-t0a 2-t0 a 3-t0 ............}, Fit according to the normal distribution and calculate the expected value a of the pit depth t0 and the standard deviation σ at0 ;

[0126] The sequence of b values of q pits obtained from the records: {b t0 1-t0 b 2-t0 b 3-t0 ............}, Fit according to the normal distribution and calculate the expected value b of the top radius of the pit t0 and the standard deviation σ bt0 ;

[0127] S1.1, time t1:

[0128] Record the total number of pits, q t1 ;

[0129] The sequence of a values of q pits obtained from the records: {a t1 1-t1 a 2-t1 a 3-t1 ............}, Fit according to the normal distribution and calculate the expected value a of the pit depth t1 and the standard deviation σ t1 ;

[0130] The sequence of b values of q pits obtained from the records: {b t1 1-t1 b 2-t1 b 3-t1 ............}, Fit according to the normal distribution and calculate the expected value b of the top radius of the pit t1 and the standard deviation σ bt1 ;

[0131] ...

[0132] S1.n, time t n :

[0133] Record the total number of pits, q tn

[0134] The sequence of a values of q pits obtained from the records: {a tn 1-tn a 2-tn a 3-tn ............}, Fit according to the normal distribution and calculate the expected value a of the pit depth tn and the standard deviation σ tn ;

[0135] ​​​​q obtained from the record tn b value sequences of tn pits: {b 1-tn b 2-tn b 3-tn ............}, fitting according to the normal distribution, calculating the expected value b of the top surface radius of the pits tn and the standard deviation σ btn ;

[0136] Step 2: Give the expected value a of the pit depth t- the relationship between a and t, and the relationship between σ at and -t:

[0137] a t0 、a t1 ………a tn as the a t column, t0, t1………t n as the t column, and use the following formula for fitting (t in this application represents the time counted from the start of use):

[0138] a t =At B

[0139] σ at0 、σ at1 ………σ atn as the σ at column, t0, t1………t n as the t column, and use the following formula for fitting:

[0140] σ at =Ct

[0141] Step 3: Give the relationship between the expected value bt of the pit radius and t, and the relationship between σ - and -t: bt -t:

[0142] b t0 、b t1 ………b tn as the b t column, t0, t1………t n as the t column, and use the following formula for fitting:

[0143] b t =Dt E

[0144] σ bt0 、σ bt1 ………σ btn as the σ bt column, t0, t1………t n as the t column, and use the following formula for fitting:

[0145] σ bt = Ft

[0146] Step Four: Determine the total number of corrosion pits: According to research, the number of corrosion pits remains basically unchanged within a certain period of time, especially after a large amount of corrosion begins (the solution t0 of this application is actually carried out after there are a large number of corrosion pits on the steel).

[0147] Step Five: Use ABAQUS software to establish a finite element model that reflects the random pitting distribution at time t x :

[0148] S5.1 Calculate a x , σ tx , b atx , σ tx at time t btx :

[0149]

[0150] S5.2 Set Q tn corrosion pits in a predetermined area on the surface of the steel member. The depths of the above Q corrosion pits conform to the normal distribution law with an expectation of a tx and a standard deviation of σ atx . The radii of the above Q corrosion pits conform to the normal distribution law with an expectation of b tx and a standard deviation of σ btx .

[0151] Step Five, the implementation method is as follows:

[0152] (a) Import the Python NumPy library to call the required functions: randomstate function, normal function, uniform function, randint function;

[0153] (b) Call the randomstate function to record the surface of the corrosion pit depth, position and distribution in the initial evolution stage;

[0154] (c) Call the normal function and assign it the corrosion pit depth variable and corrosion pit radius variable: input a t , σ at , b tx , σ btx to achieve the normal distribution of the corrosion pit depth and corrosion pit radius (for the mathematical method of the normal distribution with two parameters, see https: / / zhuanlan.zhihu.com / p / 38704136), and by applying the randomstate function, achieve the in-situ increase of the corrosion pit depth with the increase of corrosion time;

[0155] (d) Call the uniform function and let it control the pit position variable: Define the boundary area where pits occur and apply it in combination with the randomstate function to keep the pit position unchanged during the evolution process;

[0156] (e) The pit distribution surface is judged by two random integers (0, 1) output by the randint function: If the output is 1, the pits are distributed on surface A, otherwise they are distributed on surface B, and apply it in combination with the randomstate function to keep the pit distribution surface fixed.

[0157] Actual case:

[0158] Figure 8 For an example: such as Figure 6 shown by a steel member, based on steps 1 to 3, the following can be obtained:

[0159] a = 13.480t 1.483

[0160] σ a = 15.966t

[0161] b = 2302.3201t 0.342

[0162] σ b = 51.147t

[0163] To avoid the random error generated by the random model, establish the random pitting model 20 times (10 times is also okay), and perform 20 evolutions for each subsequent corrosion time period. Carry out monotonic tensile simulation on the established finite element model and extract the yield strength and ultimate strength of each group of finite element models, and take the mean value of the 20 repeated simulation results at each corrosion time as the strength after corrosion.

[0164] Such as Figure 8 shown, when the corrosion times are 6, 8, and 9 months, the comparison between the simulation results and the actual results is made, and the results are in good agreement.

[0165] <Example 2: A method for establishing a numerical model considering the evolution method of pits simplified>

[0166] Example 2 is a simplified method of Example 1.

[0167] Step 1: Use a three-dimensional topography scanner to scan and obtain the three-dimensional topography of the corrosion surface at multiple corrosion times:

[0168] S1.0, at time t0:

[0169] Record the total number of pits, q t0 ;

[0170] According to the recorded qt0 The a-value sequence of a number of corrosion pits: {a 1-t0 a 2-t0 a 3-t0 ............}, fit according to the normal distribution, and calculate the expected value a of the corrosion pit depth t0 and the standard deviation σ at0 ;

[0171] According to the recorded q t0 The b-value sequence of a number of corrosion pits: {b 1-t0 b 2-t0 b 3-t0 ............}, fit according to the normal distribution, and calculate the expected value b of the top surface radius of the corrosion pit t0 ;

[0172] S1.1, at time t1:

[0173] Record the total number of corrosion pits, q t1 ;

[0174] According to the recorded q t1 The a-value sequence of a number of corrosion pits: {a 1-t1 a 2-t1 a 3-t1 ............}, fit according to the normal distribution, and calculate the expected value a of the corrosion pit depth t1 and the standard deviation σ t1 ;

[0175] According to the recorded q t1 The b-value sequence of a number of corrosion pits: {b 1-t1 b 2-t1 b 3-t1 ............}, fit according to the normal distribution, and calculate the expected value b of the top surface radius of the corrosion pit t1 ;

[0176] ……

[0177] S1.n, at time t n Time:

[0178] Record the total number of corrosion pits, q tn

[0179] According to the recorded q tn The a-value sequence of a number of corrosion pits: {a 1-tn a 2-tn a 3-tn ............}, fit according to the normal distribution, and calculate the expected value a of the corrosion pit depth tn and the standard deviation σ tn ;

[0180] q obtained from the record tn b value sequences of tn corrosion pits: {b 1-tn b 2-tn b 3-tn ............}, fitting according to the normal distribution, calculating the expected value b of the top surface radius of the corrosion pits tn ;

[0181] Step 2: Give the expected value a of the corrosion pit depth t- The relationship between a and t, and the relationship between σ at -t:

[0182] a t0 、a t1 ………a tn As the a t column, t0, t1………t n As the t column, the following formula is used for fitting (t in this application represents the time counted from the start of use):

[0183] a t =At B

[0184] σ at0 、σ at1 ………σ atn As the σ at column, t0, t1………t n As the t column, the following formula is used for fitting:

[0185] σ at =Ct

[0186] Step 3: Give the relationship between the expected value bt of the corrosion pit radius and t, and the relationship between σ - and t: bt -t:

[0187] b t0 、b t1 ………b tn As the b t column, t0, t1………t n As the t column, the following formula is used for fitting:

[0188] b t =Dt E

[0189] Step 4: Determine the total number of corrosion pits: According to the research, the number of corrosion pits is basically unchanged within a certain period of time, especially after a large amount of corrosion starts (in fact, t0 in the solution of this application is also carried out after there are a large number of corrosion pits on the steel);

[0190] Step 5: Use ABAQUS software to establish a calculation with a time of t xAt this time, a finite element model reflecting the random pitting corrosion distribution:

[0191] S5.1 Calculate a at time t x at time t tx , σ atx , b tx :

[0192]

[0193] S5.2 Set q pits in a predetermined area on the surface of the steel member tn The depths of the above Q pits conform to a normal distribution law with an expectation of: a tx , and a standard deviation of σ atx ; the depth-to-diameter ratio of each pit satisfies: a tx / b tx .

[0194] Step five, the implementation method is as follows:

[0195] (a) Import the Python NumPy library to call the required functions: randomstate function, normal function, uniform function, randint function;

[0196] (b) Call the randomstate function to record the surface of the pit depth, position, and distribution at the initial evolution stage;

[0197] (c) Call the normal function and assign it the pit depth variable and depth-to-diameter ratio: input a t , σ at , a tx / b tx , to achieve the normal distribution of the pit depth, and by applying the randomstate function, the in-situ pit depth increases with the corrosion time;

[0198] (d) Call the uniform function and let it control the pit position variable: define the boundary area where the pits occur, and apply it with the randomstate function to keep the pit position unchanged during the evolution process;

[0199] (e) The pit distribution surface is judged by two random integers (0, 1) output by the randint function: if the output is 1, the pits are distributed on surface A, otherwise they are distributed on surface B, and apply it with the randomstate function to keep the pit distribution surface fixed.

[0200] The method of Example 2 only requires a single parameter to satisfy the normal distribution, and its effect is that the assignment is faster and the running speed of the system is faster (the scheme of Example 1 is for such as Figure 6The component has an assignment operation time of about half an hour; while for the solution of the second embodiment, it only takes 3 - 5 minutes).

[0201] The above - mentioned embodiments are preferred embodiments of the present invention, which are only used to conveniently illustrate the present invention and do not impose any formal restrictions on the present invention. Any person with ordinary knowledge in the technical field, within the scope of not departing from the technical features of the present invention, makes local changes or modified equivalent embodiments by using the technical content disclosed by the present invention, and without departing from the technical feature content of the present invention, still belongs to the scope of the technical features of the present invention.

Claims

1. A steel member analysis method considering the pitting evolution method, characterized in that, It includes the following steps: Step 1: Use a three-dimensional topography scanner to scan the three-dimensional topography of the corroded surface of the same steel member at multiple corrosion times; Measure and record a at different times t t , σ at , b t , σ bt ; a t , σ at represent the average value and standard deviation of the pit depth; b t , σ bt represent the average value and standard deviation of the pit radius; Step 2: Give the expected value a of the etch pit depth t- The relationship between and t, and σ at The relationship between -t a t0 、a t1……… a tn As a t column, t0, t 1……… t n As the t column, the fitting is carried out by the following formula: a t = At B σ at0 and σ at1……… σ atn As the σ at column, t0, t 1……… t n As the t column, the fitting is performed using the following formula: σ at = Ct; Step 3: Give the relationship between the expected radius of the etch pit $b_{t - t}$ and that between $\sigma$ bt -t: b t0 and b t1……… b tn as b t column, t0, t 1……… t n As the t column, the following formula is used for fitting: b t = Dt E σ bt0 and σ bt1……… σ btn As the σ bt column, t0, t 1……… t n As the t column, the following formula is used for fitting: σ bt = Ft Step 4: Determine the total number Q of corrosion pits; Step 5: Establish a numerical calculation model of the steel member and then conduct analysis; 5.

1. Calculate a when t x is calculated, tx σ atx b tx σ btx where tx represents the time point at which the steel member needs to be analyzed; The a obtained by using Step 2 and Step 3 t -t, σ at -t, b t -t, σ bt -t relationship to determine a tx , σ atx , b tx , σ btx : 5.2, set Q corrosion pits in a predetermined area on the surface of the steel member, and the depths of the above Q corrosion pits conform to a normal distribution law with an expected value of a tx and a standard deviation of σ atx , and the radii of the above Q corrosion pits conform to a normal distribution law with an expected value of b tx and a standard deviation of σ btx ; The total number q of corrosion pits measured on the last day in Step 1 tn is the total number Q of corrosion pits at the predicted tx.

2. The analysis method of a steel member considering the corrosion pit evolution method according to claim 1, wherein: Step 1 further includes: S1.0, at time t0: Record the total number of etch pits, q t0 ; q obtained according to the record t0 The sequence of a values of { 1-t0 a 2-t0 a 3-t0 ............} for individual corrosion pits, and fitting according to the normal distribution to calculate the expected value a of the corrosion pit depth t0 and the standard deviation σ at0 ; q obtained from the record t0 b-value sequences of t0 pits: {b 1-t0 b 2-t0 b 3-t0 ............}, fitting according to the normal distribution, calculate the expected value b of the top surface radius of the pits t0 and the standard deviation σ bt0 ; S1.1, at time t1: Record the total number of etch pits, q t1 ; q obtained according to the record t1 a value sequence of t1 corrosion pits: {a 1-t1 a 2-t1 a 3-t1 ............}, fitting according to the normal distribution, calculating the expected value a of the corrosion pit depth t1 and the standard deviation σ t1 ; q obtained according to the record t1 b value sequences of t1 pitting corrosion pits: {b 1-t1 b 2-t1 b 3-t1 ............}, fitting according to the normal distribution, calculating the expected value b of the top surface radius of its pitting corrosion pits t1 and the standard deviation σ bt1 ; …… S1.n, t n Time: Record the total number of etch pits, q tn q obtained according to the record tn Sequence of a values of tn pits: {a 1-tn a 2-tn a 3-tn ............}, fitted according to the normal distribution, calculate the expected value a of the pit depth tn and the standard deviation σ tn ; q obtained from the record tn b-value sequences of tn erosion pits: {b 1-tn b 2-tn b 3-tn ............}, fitting according to the normal distribution, calculate the expected value b of the top surface radius of its erosion pits tn and the standard deviation σ btn .

3. The analysis method of a steel member considering the corrosion pit evolution method according to claim 1, wherein: For 5.2 in Step 5, the implementation method is as follows: (a) Import the Python NumPy library and call the required functions: randomstate function, normal function, uniform function, randint function; (b) Call the randomstate function to record the depths, positions and distribution surfaces of the corrosion pits in the initial evolution stage; (c) Call the normal function and assign it the variables of the pit depth and the pit radius: input a t , σ at , b tx , σ btx , to achieve the normal distribution of the pit depth and the pit radius, and by applying the randomstate function, to achieve the in-situ increase of the pit depth with the increase of the corrosion time; (d) Call the uniform function and let it control the corrosion pit position variable: define the boundary region where the corrosion pits occur, and apply it in combination with the randomstate function to keep the corrosion pit positions unchanged during the evolution process; (e) The corrosion pit distribution surface is judged by two random integers (0, 1) output by the randint function: if the output is 1, the corrosion pits are distributed on surface A, otherwise they are distributed on surface B, and apply it in combination with the randomstate function to keep the corrosion pit distribution surface fixed.

4. An analysis method of a steel member considering the corrosion pit evolution method, wherein: Step 1: Use a three-dimensional topography scanner to scan the three-dimensional topography of the corroded surface at multiple corrosion times; Record a at different times t t , σ at , b t ; a t , σ at represent the average value and standard deviation of the pit depth; b t represents the average value of the pit radius; Step 2: Give the expected value a of the etch pit depth t- The relationship between t, and σ at The relationship between -t; a t0 、a t1……… a tn As a t column, t0, t 1……… t n As the t column, fitting is performed using the following formula: a t = At B σ at0 and σ at1……… σ atn As the σ at column, t0, t 1……… t n As the t column, fitting is performed using the following formula: σ at = Ct Step 3: Give the relationship between the expected corrosion pit radius bt-t; b t0 and b t1……… b tn As b t column, t0, t 1……… t n As the t column, fitting is performed using the following formula: b t = Dt E Step 4: Determine the total number Q of corrosion pits; Step 5: Use ABAQUS software to establish a finite element model that reflects the random pitting distribution at time t x : S5.1 Calculate t x for a tx , σ atx , b tx : S5.2 Set Q corrosion pits in a predetermined area on the surface of the steel member, and the depth of the above Q corrosion pits conforms to a normal distribution law with an expected value of a tx , and a standard deviation of σ atx ; the depth-to-diameter ratio of each corrosion pit satisfies: a tx / b tx ; The total number q of etch pits measured on the last day in Step 1 tn is the total number Q of etch pits at the predicted tx.

5. The analysis method of a steel member considering the corrosion pit evolution method according to claim 4, wherein: Step 1 further includes: S1.0, at time t0: Record the total number of etch pits, q t0 ; q obtained from the record t0 The sequence of a values of t0 pits: {a 1-t0 a 2-t0 a 3-t0 ............}, fitted according to the normal distribution, calculate the expected value a of the pit depth t0 and the standard deviation σ at0 ; q obtained according to the record t0 b-value sequence of t0 pits: {b 1-t0 b 2-t0 b 3-t0 ............}, fitting according to the normal distribution, calculating the expected value b of the top surface radius of the pits t0 ; S1.1, at time t1: Record the total number of etch pits, q t1 ; q obtained according to the record t1 a value sequence of t1 pits: {a 1-t1 a 2-t1 a 3-t1 ............}, fitting according to the normal distribution, calculating the expected value a of the pit depth t1 and the standard deviation σ t1 ; q obtained according to the record t1 b value sequences of t1 corrosion pits: {b 1-t1 b 2-t1 b 3-t1 ............}, fitting according to the normal distribution, calculating the expected value b of the top surface radius of the corrosion pits t1 ; …… S1.n, t n Time: Record the total number of etch pits, q tn q obtained from the record tn a value sequence of tn pits: {a 1-tn a 2-tn a 3-tn ............}, fitting according to the normal distribution, calculate the expected value a of the pit depth tn and the standard deviation σ tn ; q obtained according to the record tn b-value sequence of tn pits: {b 1-tn b 2-tn b 3-tn ............}, fitting according to the normal distribution, calculating the expected value b of the top surface radius of the pits tn .

6. The steel member analysis method considering the evolution method of corrosion pits according to claim 2 or 5, characterized in that: Step Four: Q = q tn .

7. The steel member analysis method considering the pitting evolution method according to claim 4, characterized in that Step 2 includes: For 5.2 in Step 5, the implementation method is as follows: (a) Import the Python NumPy library and call the required functions: randomstate function, normal function, uniform function, randint function; (b) Call the randomstate function to record the depths, positions and distribution surfaces of the corrosion pits in the initial evolution stage; (c) Call the normal function and assign it the variables of the pit depth and the depth-to-diameter ratio: input a t , σ at , a tx / b tx , to achieve the normal distribution of the pit depth, and by applying the randomstate function, to achieve the in-situ increase of the pit depth with the increase of the corrosion time; (d) Call the uniform function and let it control the corrosion pit position variable: define the boundary region where the corrosion pits occur, and apply it in combination with the randomstate function to keep the corrosion pit positions unchanged during the evolution process; (e) The corrosion pit distribution surface is judged by two random integers (0, 1) output by the randint function: if the output is 1, the corrosion pits are distributed on surface A, otherwise they are distributed on surface B, and apply it in combination with the randomstate function to keep the corrosion pit distribution surface fixed.

Images