A method for evaluating mechanical performance degradation reliability of corroded steel bars based on average corrosion rate

Abstract

The present application relates to a kind of based on average corrosion rate corrosion steel bar mechanical property degradation reliability evaluation method. Including the following steps: 1) take the diameter of steel bar sample before rusting, length, quality, after rusting of steel bar sample is removed rust and weighed, obtain its residual mass, calculate its average corrosion rate η s ;2) corrosion uneven coefficient R λ It meets I type extreme value distribution, based on average corrosion rate η s , steel bar diameter and length, determine its parameter position parameter μ Rλ And scale parameter σ Rλ ;3) based on the position parameter μ Rλ And scale parameter σ Rλ , according to the yield strength f y Before steel bar rusting, ultimate strength f u , elastic modulus E s , strengthening modulus k and ultimate strain ε u , calculate the normal distribution mean and standard deviation of each mechanical property parameter of corroded steel bar after degradation.The present application finds out the best parameter for measuring uneven corrosion, and establishes a kind of corrosion steel bar mechanical property reliability calculation method based on average corrosion rate using this parameter as key intermediate variable.

Description

Technical Field

[0001] This invention relates to a reliability assessment method for the degradation of mechanical properties of rusted steel bars based on average corrosion rate, belonging to the technical field of mechanical property assessment methods for rusted steel bars. Background Technology

[0002] Steel corrosion is a major durability problem in concrete structures. The degradation of the mechanical properties of corroded steel is a primary factor contributing to the loss of load-bearing capacity, shortened service life, and premature failure of corroded components. However, there is currently no mature and effective method to quantitatively analyze the degradation of the mechanical properties of corroded steel. The models proposed by the few existing researchers are rather crude, merely simple regressions of a few experimental results. Furthermore, due to differences in the accelerated corrosion techniques used, the different corrosion rate indices (such as cross-sectional corrosion rate, volumetric corrosion rate, weight loss rate, corrosion depth, etc.), and the different diameters and lengths of the steel bars, the results obtained vary greatly. The degradation mechanism and whether the modulus of elasticity degrades remain controversial.

[0003] In view of the shortcomings of the existing technology, there is an urgent need for a reliability assessment method for the degradation of mechanical properties of corroded steel bars based on the average corrosion rate that can solve the above technical problems. Summary of the Invention

[0004] This invention clarifies the mechanical property degradation mechanism of corroded steel bars and reveals that uneven corrosion is the main cause of nominal material property degradation. Based on this, the optimal parameter for measuring uneven corrosion is identified, and this parameter is used as a key intermediate variable to establish a method for calculating the reliability of the mechanical properties of corroded steel bars based on the average corrosion rate.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A reliability assessment method for the degradation of mechanical properties of corroded steel bars based on average corrosion rate, characterized by the following steps:

[0007] 1) The diameter of the uncorroded steel bar sample is D, the length is L, and the mass is m0. After the rust is removed, the steel bar sample is weighed to obtain its residual mass m1. Calculate its average corrosion rate (weight loss rate) η. s The calculation formula (1) is as follows:

[0008]

[0009] 2) Corrosion unevenness coefficient R λ It conforms to a type I extreme value distribution, based on the average corrosion rate η s Given the diameter D and length L of the reinforcing bar, determine its positional parameter μ. Rλ and scale parameter σ Rλ The calculation formula (2) is as follows:

[0010]

[0011] In the formula, c5 is a comprehensive parameter, c5 = 0.32ln(η) s +2.41;

[0012] Based on statistical analysis, σ Rλ The overall distribution is normal, with a mean of 0.038 and a standard deviation of 0.016. In practical applications, the quantile value of 0.064, which has a 95% guarantee rate, is taken.

[0013] 3) If the corroded steel bars are obtained through a laboratory accelerated corrosion method, the resulting location parameter μ Rλ and scale parameter σ Rλ The following adjustments will be made:

[0014] When the number of specimens n < 30, the position parameter μ Rλ Corrected by multiplying by a coefficient γ, the scaling parameter σ Rλ Keeping it constant, the coefficient γ is determined as follows: The R of each specimen is measured. λ Let η be the measured corrosion rate of the i-th specimen. s,i , through η s,i The theoretical value μ was calculated. Rλ,i In fact, the coefficient of corrosion non-uniformity is R. λ,i The correction coefficient γ for the i-th specimen is obtained. i =R λ，i / μ Rλ，i For all γ i The final correction coefficient is obtained by taking the average value.

[0015] The number of specimens n≥30, and the measured R of each specimen λ The position parameter μ is obtained by directly estimating the value using the maximum likelihood estimation method. Rλ and scale parameter σ Rλ ;

[0016] 4) Based on the obtained position parameter μ Rλ and scale parameter σ Rλ Based on the yield strength f of the steel bar before corrosion y Ultimate strength f u Elastic modulus E s Strengthening modulus k and ultimate strain ε u The mean and standard deviation of the normal distribution of various mechanical property parameters of the corroded steel bars after degradation were calculated.

[0017] Nominal yield strength f y The mean μ of ′ FY and standard deviation σ FYThe calculation formula (3) is as follows:

[0018]

[0019] Nominal ultimate strength f′ u mean μ FU and standard deviation μ FU The calculation formula (4) is as follows:

[0020]

[0021] Nominal elastic modulus E′ s mean μ ES and standard deviation ε ES The calculation formula (5) is as follows:

[0022]

[0023] Nominal limit strain ε′ u mean μ EU and standard deviation ε EU The calculation formula (6) is as follows:

[0024]

[0025] The present invention provides a reliability assessment method for the degradation of mechanical properties of corroded steel bars based on average corrosion rate, which has the following advantages:

[0026] (1) Laser scanning of corroded steel bar specimens was used to accurately obtain large geometric data of surface features, which formed the basis for analysis;

[0027] (2) To reveal the mechanism of nominal material property degradation of corroded steel bars, and on this basis, to find the non-uniform corrosion coefficient R that has the best correlation with mechanical property degradation. λ ;

[0028] (3) Through big data statistical analysis, R under different corrosion rates was determined. λ Probability distribution model;

[0029] (4) Established with R λ A functional model for the nominal yield strength, ultimate strength, ultimate strain, and elastic modulus of the parameters;

[0030] (5) Reliability analysis methods were used to determine the reliability models of various nominal material properties based on the average corrosion rate or the non-uniform corrosion coefficient.

[0031] In summary, this invention, through a comparative study of numerical tensile tests based on a physical model and tensile tests on actual steel reinforcement specimens, clarifies the mechanism of mechanical property degradation in corroded steel reinforcement and reveals that uneven corrosion is the primary cause of nominal material property degradation. Based on this, the optimal parameter for measuring uneven corrosion is identified, and using this parameter as a key intermediate variable, a method for calculating the reliability of mechanical properties of corroded steel reinforcement based on the average corrosion rate is established. Attached Figure Description

[0032] Figure 1 Diagram of steel reinforcement with uneven corrosion;

[0033] Figure 2 Image of unevenly corroded steel bars after mechanical processing;

[0034] Figure 3 Comparison between corroded steel bars and the actual model;

[0035] Figure 4 : A typical distribution curve of the discretized residual cross-sectional area along the longitudinal direction of the reinforcing bar;

[0036] Figure 5 : Tensile test diagram of corroded steel bar specimen;

[0037] Figure 6 : Diagram of the test section and the physical model section. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] A reliability assessment method for the degradation of mechanical properties of corroded steel bars based on average corrosion rate is proposed, and the research mechanism is as follows:

[0040] 1) Three-dimensional laser scanning to create a 3D solid model of the corroded steel bars.

[0041] The corroded steel bars were acid-washed to remove rust, then a laser scanner was used to perform a 3D scan to obtain point cloud data of the bar surface. Noise was removed, and the surface was meshed to create a curved surface. Finally, the surface was closed and filled in Pro / E software to obtain a solid model of the corroded steel bars, as shown in the attached figure. Figure 3 As shown, the resulting solid model maintains the same geometric features as the prototype of the corroded steel bar;

[0042] 2) Accurately obtain the geometric characteristic parameters of corroded steel bars through solid modeling.

[0043] Along the length of the rebar sample, starting from the end, cross-sections of the rusted rebar were cut at 1mm intervals, and their residual cross-sectional areas were measured to obtain discretized cross-sectional area data. (See attached image) Figure 4 The distribution curve of the residual cross-sectional area along the longitudinal direction of the steel bar is a typical discrete curve. Based on a large amount of data on the distribution of the residual cross-sectional area of ​​a large number of corroded steel bar samples, subsequent statistical analysis and research on corrosion non-uniformity are carried out.

[0044] 3) Obtain the coefficient of non-uniform corrosion R λ data

[0045] From the solid model of the reinforcing steel, arbitrarily cut a section of reinforcing steel specimen of length L, and determine the maximum residual cross-sectional area A within it. max Minimum residual cross-sectional area A min The non-uniform corrosion coefficient R of this sample was calculated. λ =A max / A min The residual volume was calculated using numerical integration, and the corrosion rate (weight loss rate) η was calculated. s Thus, different corrosion rates η were obtained. s R under different rebar diameters D and different cut lengths L λ data;

[0046] 4)R λ Establishment of probability models

[0047] Under the same conditions, R λ The data is treated as a sample population. Hypothesis testing is used to determine that its distribution is a Type I extreme value distribution. Maximum likelihood estimation is used to determine the distribution parameters of different sample populations.

[0048] 5) Establish R λ Functional relationship of probability distribution parameters affected by different factors

[0049] With average corrosion rate η s The diameter D and cut length L of the reinforcing bar are the influencing factors, and R is established. λ Mathematical model of probability distribution parameters: μ Rλ =f(η s , D, L), σ Rλ =f(η s (D, L);

[0050] Through derivation, μ can be obtained. Rλ The expression:

[0051]

[0052] In the formula, c5 is a composite parameter, and its expression is as follows:

[0053] c5 = 0.32ln(η) s +2.41

[0054] σ Rλ It did not show any change with η s The obvious patterns of change in D and L show an overall random distribution, with a mean of 0.038 and a standard deviation of 0.016. In practical applications, the quantile value of 0.064, which has a 95% guarantee rate, is taken.

[0055] 6) Mechanical properties of corroded steel bars

[0056] Tensile tests were performed on the corroded steel reinforcement specimens, as shown in the attached document. Figure 5 Obtain the stress-strain curve of the test section, and conduct numerical tensile tests on the computer-generated solid model of the same test section. Refer to the attached document. Figure 6 Comparative studies revealed that the actual mechanical properties of the material did not degrade; the degradation of the steel reinforcement's mechanical properties was actually a nominal degradation caused by uneven corrosion, i.e., the uneven surface of the corroded steel reinforcement (see attached document). Figure 1 ) Machining into attachments Figure 2 The standard specimens shown were subjected to tensile tests, and the measured steel properties were consistent with those before corrosion.

[0057] 7) Nominal Material Property Analysis Model

[0058] Based on the above principles and methods, an analytical approach was used to establish R. λ Nominal yield strength f′ of corroded steel bars y Nominal ultimate strength f′ u Nominal elastic modulus E′ s and nominal limiting strain ε′ u Mathematical model between:

[0059]

[0060]

[0061]

[0062]

[0063]

[0064] 8) Using reliability analysis methods, the probability distribution parameter models of each degradation coefficient or degradation parameter are derived.

[0065] The specific method is: [Regarding R...] λ The function is expanded using Taylor series, omitting higher-order minor terms and retaining only the first two terms, transforming it into a linear functional relationship, thereby establishing R.λ The mapping relationship between the probability model and the normal distribution model of each degradation parameter is determined, and the first moment (mean), second moment (variance) and standard deviation of each parameter are finally solved.

[0066] Nominal yield strength f′ y mean μ FY and standard deviation σ FY The calculation formula (3) is as follows:

[0067]

[0068] Nominal ultimate strength f′ u The mean σ FU and standard deviation σ FU The calculation formula (4) is as follows:

[0069]

[0070] Nominal elastic modulus E s The mean μ of ′ ES and standard deviation σ ES The calculation formula (5) is as follows:

[0071]

[0072] Nominal limit strain ε′ u mean μ EU and standard deviation σ EU The calculation formula (6) is as follows:

[0073]

[0074] Example 1

[0075] This example uses a naturally corroded steel bar, 100mm in length. After acid pickling and rust removal, the residual mass m1 = 190.8g was obtained. The original diameter of the steel bar was 18mm. It was an HRB400 ribbed steel bar with a yield strength of 400MPa and an elastic modulus of 2.1×10⁻⁶ before corrosion. 5 MPa, ultimate strength 570MPa, hardening modulus 1572MPa, ultimate strain 0.11.

[0076] 1) Calculate the corrosion rate

[0077] The steel bar has a diameter D = 18 mm, a standard linear density of 1.998 kg / m before corrosion, and a length L = 100 mm. Therefore, its mass before corrosion is m0 = 199.8 g, and the corrosion rate is η. s for:

[0078]

[0079] 2) Calculate the corrosion non-uniformity coefficient R λ

[0080] coefficient R λ It conforms to a type I extreme value distribution, based on the average corrosion rate η s The diameter D and length L of the reinforcing bar can determine its positional parameter μ. Rλ and scale parameter σ Rλ :

[0081] c5 = 0.32ln(η) S ) + 2.41 = 0.32 × ln(0.045) + 2.41 = 1.4177

[0082]

[0083] σ Rλ =0.064 is taken as the quantile value for the 95% guarantee rate;

[0084] 3) Calculate the nominal parameters after degradation

[0085] Nominal yield strength f′ y mean μ FY and standard deviation σ FY :

[0086]

[0087] Nominal ultimate strength f u The mean μ of ′ FU and standard deviation σ FU :

[0088]

[0089] Nominal elastic modulus E′ s mean μ ES and standard deviation σ ES :

[0090]

[0091]

[0092] Nominal limit strain ε′ u mean μ EU and standard deviation σ EU :

[0093]

[0094]

[0095] The calculation results of various nominal material properties after degradation are summarized in the table below:

[0096]

[0097] Example 2

[0098] A corroded steel rebar specimen was obtained through accelerated corrosion in the laboratory. The measured corrosion rate (weight loss rate) was 11.0%. The rebar had a diameter of 20 mm and a length of 150 mm. Based on historical data, the corrosion non-uniformity of the rebar obtained through accelerated corrosion in the laboratory was relatively small. Compared with natural corrosion, the corrosion non-uniformity coefficient R under the same conditions was significantly lower. λ The value is about 2.8% lower than expected.

[0099] Before corrosion, the steel bar's properties were: HRB400 grade, yield strength 400 MPa, elastic modulus 2.1 × 10⁻⁶. 5 MPa, ultimate strength 570MPa, hardening modulus 1572MPa, ultimate strain 0.11.

[0100] 1) Calculate the corrosion non-uniformity coefficient R λ

[0101] The corrosion rate is known, so the location parameter μ can be calculated directly. Rλ and scale parameter σ Rλ :

[0102] c5 = 0.32ln(η) s ) + 2.41 = 0.32 × ln(0.11) + 2.41 = 1.7037

[0103]

[0104] σ Rλ =0.064 is taken as the quantile value for the 95% guarantee rate;

[0105] 2)R λ Adjustment

[0106] Based on the known conditions, the corrosion unevenness coefficient of this corroded steel bar is 2.8% lower than that of natural corrosion.

[0107] μ Rλ = (1 - 2.8%) × 1.124 = 1.093

[0108] σ Rλ =0.064 remains unchanged.

[0109] 3) Calculate the nominal parameters after degradation

[0110] Nominal yield strength f′ y mean μ FY and standard deviation σ FY :

[0111]

[0112] Nominal ultimate strength f u The mean μ of ′ FU and standard deviation σ FU :

[0113]

[0114] Nominal elastic modulus E′ s mean μ ES and standard deviation σ ES :

[0115]

[0116]

[0117] Nominal limit strain ε′ u mean μ EU and standard deviation σ EU :

[0118]

[0119]

[0120] The calculation results of various nominal material properties after degradation are summarized in the table below:

[0121]

[0122] Example 3

[0123] A batch of 100 corroded steel rebar specimens were obtained under artificial climatic conditions of alternating wet and dry conditions in the laboratory. The specimens showed severe corrosion, with measured corrosion rates (weight loss rates) all around 21%. The initial diameter of the rebar was 14 mm, and its length was 120 mm. The material properties of the rebar before corrosion were: HRB400 grade steel, yield strength 400 MPa, elastic modulus 2.1 × 10⁻⁶. 5 MPa, ultimate strength 570MPa, hardening modulus 1572MPa, ultimate strain 0.11.

[0124] For all 100 specimens, the ratio of the maximum residual cross-sectional area to the minimum residual cross-sectional area of ​​each section was measured to obtain the measured R for all specimens. λ The location parameter μ of the population is obtained by using the maximum likelihood estimation method. Rλ =1.24, scale parameter σ Rλ =0.09.

[0125] 1) Determine the corrosion non-uniformity coefficient R λ

[0126] Based on the empirical formula proposed in this patent, with D=14, L=120, η s =21% Calculated to obtain μ Rλ =1.319, σ Rλ =0.064. In contrast, μ is obtained directly from sample data through parameter estimation. Rλ =1.24, σ Rλ =0.09. The comparison shows that the corrosion unevenness of the actual rusted specimens is lower than that of natural rust. Due to the large sample size, the estimated result of the actual parameters, i.e., μ, is used here. Rλ =1.24, σ Rλ =0.09.

[0127] 2) Calculate the nominal parameters after degradation

[0128] Nominal yield strength f′ y mean μ FY and standard deviation σ FY :

[0129]

[0130] Nominal ultimate strength f′ u mean μ FU and standard deviation σ FU :

[0131]

[0132] Nominal elastic modulus E′ s mean μ ES and standard deviation σ ES :

[0133]

[0134]

[0135] Nominal limit strain ε′ u mean μ EU and standard deviation σ EU :

[0136]

[0137]

[0138] The calculation results of various nominal material properties after degradation are summarized in the table below:

[0139]

[0140] Through the above Examples 1-3, it is demonstrated that under three representative conditions, based on the average corrosion rate (weight loss rate) and initial material property data of the corroded steel bars, the mean and standard deviation of the nominal yield strength, ultimate strength, elastic modulus, and ultimate strain of the corroded steel bars after degradation can be conveniently calculated using the method of this application. The results can provide crucial basic data for calculating the residual bearing capacity of corroded reinforced concrete members, predicting their remaining service life, and designing for member repair and reinforcement. The method of this application can also provide a reference for scientific research related to corroded steel bars.

[0141] This invention is based on the following scientific research findings and mechanisms:

[0142] 1) After the steel bars corrode, the steel does not undergo fundamental degradation of its mechanical properties at the microscopic material level; that is, unevenly corroded steel bars (see attached image) Figure 1 Machining (with attachments) Figure 2 After obtaining the standard mechanical property test specimens shown in the figure, a standard tensile test was performed, and the obtained mechanical properties were consistent with those of the steel bars before corrosion.

[0143] 2) The main reason for the degradation of the macroscopic mechanical properties of corroded steel bars is the uneven corrosion: Traditional experimental methods measure the deformation and applied tensile force of a section, calculate the stress through the average cross-sectional area, and calculate the strain through the total deformation of the section. The obtained stress and strain are nominal stress and nominal strain, rather than material property indicators in the strict sense.

[0144] 3) The effects of uneven corrosion on the mechanical properties of steel bars include: uneven corrosion causes the residual cross-sectional area of ​​the steel bars to change continuously along the length direction, and the stress level at each point is different when under stress. The stress level is the highest at the most severely corroded point, which is the first to yield and eventually breaks at this point; circumferential corrosion unevenness causes eccentric stress, and stress concentration will occur in rust pits.

[0145] 4) In scientific research and practical engineering applications, due to the great difficulty in measuring and quantifying the non-uniformity of corrosion, the average corrosion rate (weight loss rate) is still the most important indicator for evaluating the degree of corrosion. The assessment and quantification of the degradation of the macroscopic mechanical properties of corroded steel bars based on nominal material properties still have irreplaceable practical significance and are a basic variable for performance evaluation, life prediction and repair and reinforcement design of corroded concrete structural components.

[0146] 5) Based on physical and mechanical principles and correlation analysis, the optimal parameter for quantifying corrosion non-uniformity is proposed as the ratio of the maximum residual cross-sectional area to the minimum residual cross-sectional area of ​​the sample, defined as the non-uniform corrosion coefficient R. λ ; Establish different corrosion rates η through big data statistical analysis s R below λ Probability distribution model;

[0147] 6) According to Rλ The physical meaning of this can be quantified analytically by considering the influence of the diameter and cut length of the corroded steel bar sample, and corrected using experimental data. This allows for the determination of R. λ The probability distribution parameters are corrected, which greatly improves the accuracy and adaptability of the model;

[0148] 7) Define the degradation coefficient of the mechanical properties of corroded steel bars as the ratio of each nominal material property after degradation to the original material property. Through analytical analysis, derive the relationship between each degradation coefficient and R. λ The mathematical model is obtained; using the reliability analysis method, the probability distribution model of the degradation coefficient of each mechanical property is finally obtained through the analytical expression of the degradation model;

[0149] 8) There is no unified standard for commonly used laboratory accelerated corrosion techniques. Various methods exist, such as full immersion, partial immersion, and alternating wet and dry methods, with varying applied current densities. These differences result in significant variations in the uniformity of steel reinforcement corrosion even at the same degree of corrosion, leading to excessive dispersion in the degradation of mechanical properties of corroded steel reinforcement across different research results. The key intermediate variable R introduced in this invention… λ This can effectively solve this defect.

Claims

1. A reliability assessment method for the degradation of mechanical properties of corroded steel bars based on average corrosion rate, characterized in that... Includes the following steps: 1) The diameter of the uncorroded steel bar sample is D, the length is L, and the mass is m0. After the rust is removed, the steel bar sample is weighed to obtain its residual mass m1. The average corrosion rate η is then calculated. s The calculation formula (1) is as follows: 2) Corrosion unevenness coefficient R λ It conforms to a type I extreme value distribution, based on the average corrosion rate η s Given the diameter D and length L of the reinforcing bar, determine its positional parameter μ. Rλ and scale parameter σ Rλ The calculation formula (2) is as follows: In the formula, c5 is a comprehensive parameter, c5 = 0.32ln(η) s +2.41; Based on statistical analysis, σ Rλ The overall distribution is normal, with a mean of 0.038 and a standard deviation of 0.

016. In practical applications, the quantile value of 0.064, which has a 95% guarantee rate, is taken. 3) Based on the obtained position parameter μ Rλ and scale parameter σ Rλ Based on the yield strength f of the steel bar before corrosion y Ultimate strength f u Elastic modulus E s Strengthening modulus k and ultimate strain ε u The mean and standard deviation of the normal distribution of various mechanical property parameters of the corroded steel bars after degradation were calculated. Nominal yield strength f′ y mean μ FY and standard deviation σ FY The calculation formula (3) is as follows: Nominal ultimate strength f′ u mean μ FU and standard deviation σ FU The calculation formula (4) is as follows: Nominal elastic modulus E′ s mean μ ES and standard deviation σ ES The calculation formula (5) is as follows: Nominal limit strain ε′ u mean μ EU and standard deviation σ EU The calculation formula (6) is as follows: 。 2. The reliability assessment method for the degradation of mechanical properties of corroded steel bars based on average corrosion rate according to claim 1, characterized in that... The following steps are also included between step 2) and step 3): If the corroded steel bars are obtained through laboratory accelerated corrosion methods, the resulting location parameter μ Rλ and scale parameter σ Rλ The following adjustments will be made: When the number of specimens n < 30, the position parameter μ Rλ Corrected by multiplying by a coefficient γ, the scaling parameter σ Rλ Keeping it constant, the coefficient γ is determined as follows: The R of each specimen is measured. λ Let η be the measured corrosion rate of the i-th specimen. s,i , through η s,i The theoretical value μ was calculated. Rλ,i In fact, the coefficient of corrosion non-uniformity is R. λ,i The correction coefficient γ for the i-th specimen is obtained. i =R λ，i / μ Rλ，i For all γ i The final correction coefficient is obtained by taking the average value. The number of specimens n≥30, and the measured R of each specimen λ The position parameter μ is obtained by directly estimating the value using the maximum likelihood estimation method. Rλ and scale parameter σ Rλ .

Images