A method for predicting the life of stainless steel in a polluted urban atmosphere environment

Abstract

The present application belongs to the technical field of metal material corrosion evaluation, and particularly relates to a method for predicting the service life of stainless steel under the polluted urban atmospheric environment, and the specific steps are as follows: S1: exposure test; S2: corrosion parameter determination; S3: grey correlation analysis; S4: model establishment; and S5: model verification. The method introduces the grey correlation analysis into the multi-parameter modeling of stainless steel corrosion under the polluted urban atmospheric environment, determines the corrosion parameter most related to the maximum pitting depth based on the long-term exposure data of small samples, has high model precision, and can predict the change of the maximum pitting depth of stainless steel in the polluted urban atmospheric environment with time, or calculate the time required to reach a certain maximum pitting depth according to the corrosion weight loss rate, so that the model has high goodness-of-fit, and provides a basis for material selection, structure life evaluation and maintenance cycle formulation.

Description

Technical Field

[0001] This invention belongs to the field of metal material corrosion assessment technology, and in particular relates to a method for predicting the service life of stainless steel in polluted urban environments. It is especially applicable to the long-term service performance assessment and selection of typical stainless steel materials such as 304, 316, and 430 in industrially polluted urban environments. Background Technology

[0002] Stainless steel, due to its excellent corrosion resistance and good mechanical properties, is widely used in urban infrastructure, building curtain walls, transportation facilities, and other applications exposed to the urban atmosphere. However, with industrial development, urban atmospheres often contain high concentrations of SO2 and NO. X Pollutants such as particulate matter create a corrosive urban atmospheric environment. This environment easily induces pitting corrosion and other localized corrosion in stainless steel, significantly shortening its service life. Therefore, studying the corrosion behavior of stainless steel in polluted urban environments and establishing a scientifically reliable life prediction model has significant engineering and economic value for material selection, safety assessment, and maintenance decisions for urban infrastructure.

[0003] Currently, scholars at home and abroad have conducted extensive research on the corrosion behavior of stainless steel in the atmospheric environment. However, most of the research focuses on ordinary atmospheric environments, while there are relatively few studies on the atmospheric environment of polluted cities. Moreover, existing life prediction models mostly rely on extrapolation of short-term accelerated test data, which has low accuracy and poor engineering applicability. Summary of the Invention

[0004] The purpose of this invention is to provide a method for predicting the lifespan of stainless steel in polluted urban environments, which solves the problem that existing lifespan prediction models rely heavily on extrapolation of short-term accelerated test data, resulting in low accuracy and poor engineering applicability.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A method for predicting the lifespan of stainless steel in polluted urban environments, comprising the following steps: S1: Exposure test Stainless steel samples of different grades with a surface condition of 2B were selected and placed in a polluted urban atmospheric environment for exposure tests for at least 1 year, 2 years and 4 years, in accordance with GB / T 14165-2008 "General Requirements for Field Tests of Atmospheric Corrosion Testing of Metals and Alloys". S2: Corrosion parameter measurement S3: Grey Relational Analysis Using the maximum pitting depth as the reference sequence X0, and other corrosion parameters such as corrosion weight loss rate X1, average pitting depth X2, and red rust coverage X3 as comparison sequences, grey relational analysis was performed. The specific steps are as follows: (1) Initialize the original data: ; ; Where k=1, 2, 3, corresponding to exposure times of 1 year, 2 years, and 4 years, and i=0, 1, 2, 3; (2) Calculate the absolute difference sequence: ; (3) Calculate the correlation coefficient: ; Where the resolution coefficient is 0 < ρ < 1; (4) Calculate the grey relational degree: n=3; (5) Select the parameter with the highest correlation to the maximum pitting depth. γ>0.6 indicates good correlation, and the larger the value, the better the correlation between sequences. S4: Model Building S5: Model Validation The model's predicted values ​​are compared with the measured values ​​to verify the model's accuracy.

[0006] Preferably, in step S1, the test surface of the sample faces south and is placed at a 45° angle to the ground.

[0007] Preferably, the specific details of the corrosion parameter determination in step S2 are as follows: After the exposure test, the corrosion products of the samples were removed (according to GB / T 16545-2015), and the following corrosion parameters were measured: (1) Maximum pitting depth (P) max ): The depth of at least 10 pits on each sample was measured using an ultra-depth-of-field microscope (such as Keyence VHX-2000), and the maximum value was taken; (2) Corrosion weight loss rate (W): The weight loss per unit area (g·cm³) is calculated by weighing. -2 ·a -1 or mm·a -1 ); (3) Average pitting depth (P) avg ): The arithmetic mean of all measured pitting depths; (4) Red rust coverage (R): The percentage of the area occupied by the red rust area on the sample surface is calculated using image analysis software.

[0008] Preferably, the specific content of the model establishment in step S4 is as follows: Based on the corrosion weight loss rate, the parameter with the highest correlation, a corrosion lifetime prediction model is established using a power function form: P max = A⋅t BOr t = C⋅W D Where t is the exposure time (years), W is the corrosion weight loss rate, and A, B, C, and D are material-related fitting constants. The model was obtained by fitting the experimental data using the least squares method, and the goodness of fit R0 was calculated. 2 A value greater than 0.9 indicates that the model is reliable.

[0009] Preferably, the resolution coefficient ρ in step S3 is 0.5.

[0010] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: (1) For the first time, grey relational analysis was introduced into the multi-parameter modeling of stainless steel corrosion in the atmospheric environment of polluted cities. Based on long-term exposure data of small samples, the corrosion parameters most related to the maximum pitting depth were determined. The model has high accuracy and strong applicability. (2) A corrosion life prediction method with the maximum pitting depth as the core indicator is proposed. The maximum pitting depth is a key parameter for measuring the local corrosion hazard and is of greater engineering significance. (3) Predicting long-term corrosion behavior based on short-term exposure data can significantly shorten the test cycle and reduce R&D costs; (4) This method predicts the change of the maximum pitting depth of stainless steel in polluted urban atmospheric environment over time, or calculates the time required to reach a certain maximum pitting depth based on the corrosion weight loss rate. The model has a high goodness of fit and provides a basis for material selection, structural life assessment and maintenance cycle formulation. Detailed Implementation

[0011] The technical solution of the present invention will be described in detail below with reference to the embodiments.

[0012] Three typical stainless steels, 304, 316, and 430, were selected. The sample size was 75 mm × 150 mm, the surface condition was 2B, and the test surface of the sample faced south, placed at a 45° angle to the ground. The method for predicting the lifespan of stainless steel in polluted urban atmospheric environments according to this invention was adopted, and the specific steps are as follows: Example 1

[0013] S1: Exposure test 304 stainless steel samples (chemical composition: C 0.064%, Si 0.34%, Mn 1.14%, Cr 18.03%, Ni 8.17%, balance Fe) were placed at the Qingdao Polluted City Atmospheric Test Station for 1 year, 2 years and 4 years, with 3 parallel samples in each group.

[0014] S2: Corrosion parameter measurement After the exposure test, the corrosion products were removed in accordance with GB / T 16545-2015, and the corrosion parameters were measured. The results are shown in Table 1.

[0015] Table 1 Corrosion parameters of 304 stainless steel after exposure

[0016] S3: Grey Relational Analysis The maximum pitting depth was taken as the reference sequence X0, and the corrosion weight loss rate X1, average pitting depth X2, and red rust coverage X3 were taken as the comparison sequences. The initial values ​​are shown in Table 2.

[0017] Table 2 Initial values ​​of corrosion parameters for 304 stainless steel

[0018] The absolute difference sequences are shown in Table 3.

[0019] Table 3 Absolute difference sequences

[0020] Table 3 shows that Δmax = 1.0310, Δmin = 0, and ρ = 0.5. The correlation coefficient and degree of correlation are calculated: Correlation degree with corrosion weight loss rate: γ 01 =0.78 Correlation between γ and average pitting depth 02 =0.92 and correlation with red rust coverage: γ 03 =0.65.

[0021] The results showed that the maximum pitting depth of 304 stainless steel had the highest correlation with the average pitting depth (0.92), but the corrosion weight loss rate also had a high correlation (0.78). Since the corrosion weight loss rate is a macroscopically measurable parameter and is convenient for engineering applications, this embodiment selected the corrosion weight loss rate as the modeling parameter.

[0022] S4: Model Building Corrosion weight loss rate W (unit: ×10) -4 g·cm -2 ·a -1 () is the independent variable, and the maximum pitting depth P is the independent variable. max (Unit: μm) is the dependent variable. A power function model is fitted, and the data is subjected to nonlinear fitting to obtain: P max =7.0585t 0.6507 Or it can be expressed as the relationship between time and weightlessness: t=0.4863W 0.5628 , Model fit R 2 =0.9983, indicating that the model is reliable.

[0023] S5: Model Validation The model's predicted maximum pitting depth over 4 years (17.47 μm) was compared with the measured value (16.95 μm), and the error was less than 3%, indicating high prediction accuracy.

[0024] Example 2 S1: Exposure test 316 stainless steel samples (chemical composition: C 0.0069%, Si 0.45%, Mn 1.14%, Cr 16.55%, Ni 10.19%, Mo 2.21%, balance Fe) were placed at the Qingdao Polluted City Atmosphere Test Station for 1 year, 2 years, and 4 years, with 3 parallel samples in each group.

[0025] S2: Corrosion parameter measurement After the exposure test, the corrosion products were removed in accordance with GB / T 16545-2015, and the corrosion parameters were measured as shown in Table 4.

[0026] Table 4 Corrosion parameters of 316 stainless steel after exposure

[0027] S3: Grey Relational Analysis The maximum pitting depth was taken as the reference sequence X0, and the corrosion weight loss rate X1, average pitting depth X2, and red rust coverage X3 were taken as the comparison sequences. The initial values ​​are shown in Table 5.

[0028] Table 5 Initial values ​​of corrosion parameters for 316 stainless steel

[0029] The absolute difference sequences are shown in Table 6.

[0030] Table 6 Absolute Difference Sequences

[0031] Table 6 shows that Δmax = 2.02284, Δmin = 0, and ρ = 0.5. The correlation coefficient and correlation degree are calculated: correlation degree with corrosion weight loss rate: γ 01 =0.71 and the correlation γ with the average pitting depth 02 =0.65 and correlation with red rust coverage: γ 03 =0.66.

[0032] The results show that the maximum pitting depth of 316 stainless steel is most correlated with the corrosion weight loss rate. Therefore, corrosion weight loss rate is selected as the modeling parameter in this embodiment.

[0033] S4: Model Building The fitting yielded: Pmax = 3.6378t 0.9285 Or it can be expressed as the relationship between time and weightlessness: t = 0.5652W 0.6363 , Goodness of fit R 2 =0.9943.

[0034] S5: Model Validation The model's predicted maximum pitting depth over 4 years (13.08 μm) was compared with the measured value (13.55 μm), and the error was less than 4%, indicating high prediction accuracy.

[0035] Example 3 S1: Exposure test 430 stainless steel samples (chemical composition: C 0.039%, Si 0.26%, Mn 0.41%, Cr 16.38%, balance Fe) were placed at the Qingdao Polluted City Atmospheric Test Station for 1 year, 2 years and 4 years, with 3 parallel samples in each group.

[0036] S2: Corrosion parameter measurement After the exposure test, the corrosion products were removed in accordance with GB / T 16545-2015, and the corrosion parameters were measured as shown in Table 7.

[0037] Table 7 Corrosion parameters of 430 stainless steel after exposure

[0038] S3: Grey Relational Analysis The maximum pitting depth was taken as the reference sequence X0, and the corrosion weight loss rate X1, average pitting depth X2, and red rust coverage X3 were taken as the comparison sequences. The initial values ​​are shown in Table 8.

[0039] Table 8 Initial values ​​of corrosion parameters for 430 stainless steel

[0040] The absolute difference sequences are shown in Table 9.

[0041] Table 9 Absolute Difference Sequences

[0042] Table 9 shows that Δmax = 1.92528, Δmin = 0, and ρ = 0.5. The correlation coefficient and correlation degree are calculated: correlation degree with corrosion weight loss rate: γ 01 =0.88 and the correlation between γ and the average pitting depth 02 =0.73 and correlation with red rust coverage: γ 03 =0.65.

[0043] The results show that the maximum pitting depth of 430 stainless steel is most correlated with the corrosion weight loss rate. Therefore, corrosion weight loss rate is selected as the modeling parameter in this embodiment.

[0044] S4: Model Building The fitting yielded: P max =9.8427t0.9027 Or it can be expressed as the relationship between time and weightlessness: t = 1.3721 W 0.2482 , Goodness of fit R 2 =0.9957.

[0045] S5: Model Validation The model's predicted maximum pitting depth over 4 years (34.62 μm) was compared with the measured value (33.64 μm), and the error was less than 3%, indicating high prediction accuracy.

[0046] The above embodiments demonstrate that the method of the present invention is applicable to the prediction of corrosion life of different types of stainless steel in polluted urban atmospheric environments, with high model accuracy and good universality.

Claims

1. A method for predicting the lifespan of stainless steel in polluted urban environments, characterized in that, The specific steps are as follows: S1: Exposure test Stainless steel samples of different grades with a surface condition of 2B were selected and placed in a polluted urban atmospheric environment for exposure tests for at least 1 year, 2 years and 4 years, in accordance with GB / T 14165-2008 "General Requirements for Field Tests of Atmospheric Corrosion Testing of Metals and Alloys". S2: Corrosion parameter measurement S3: Grey Relational Analysis Using the maximum pitting depth as the reference sequence X0, and other corrosion parameters such as corrosion weight loss rate X1, average pitting depth X2, and red rust coverage X3 as comparison sequences, grey relational analysis was performed. The specific steps are as follows: (1) Initialize the original data: ； ； Where k=1, 2, 3, corresponding to exposure times of 1 year, 2 years, and 4 years, and i=0, 1, 2, 3; (2) Calculate the absolute difference sequence: ; (3) Calculate the correlation coefficient: ; Where the resolution coefficient is 0 < ρ < 1; (4) Calculate the grey relational degree: n=3; (5) Select the parameter with the highest correlation to the maximum pitting depth. γ>0.6 indicates good correlation, and the larger the value, the better the correlation between sequences. S4: Model Building S5: Model Validation The model's predicted values ​​are compared with the measured values ​​to verify the model's accuracy.

2. The method for predicting the lifespan of stainless steel in a polluted urban atmosphere according to claim 1, characterized in that, In step S1, the test surface of the sample faces south and is placed at a 45° angle to the ground.

3. The method for predicting the lifespan of stainless steel in a polluted urban atmosphere according to claim 1, characterized in that, The specific details of the corrosion parameter determination in step S2 are as follows: After the exposure test, the corrosion products of the samples were removed (according to GB / T 16545-2015), and the following corrosion parameters were measured: (1) Maximum pitting depth (P) max ): The depth of at least 10 pits on each sample was measured using an ultra-depth-of-field microscope (such as Keyence VHX-2000), and the maximum value was taken; (2) Corrosion weight loss rate (W): The weight loss per unit area (g·cm³) is calculated by weighing. -2 ·a -1 or mm·a -1 ); (3) Average pitting depth (P) avg ): The arithmetic mean of all measured pitting depths; (4) Red rust coverage (R): The percentage of the area occupied by the red rust area on the sample surface is calculated using image analysis software.

4. The method for predicting the lifespan of stainless steel in a polluted urban atmosphere according to claim 1, characterized in that, The specific details of model establishment in step S4 are as follows: Based on the corrosion weight loss rate, the parameter with the highest correlation, a corrosion lifetime prediction model is established using a power function: P max = A⋅t B Or t = C⋅W D Where t is the exposure time (years), W is the corrosion weight loss rate, and A, B, C, and D are material-related fitting constants. The model was obtained by fitting the experimental data using the least squares method, and the goodness of fit R0 was calculated. 2 A value greater than 0.9 indicates that the model is reliable.

5. The method for predicting the lifespan of stainless steel in a polluted urban atmosphere according to claim 1, characterized in that, In step S3, the resolution coefficient ρ is set to 0.5.

Images