Corrosion steel surface appearance modeling method based on frequency spectrum generation

Abstract

The invention discloses a rusted steel surface topography modeling method based on frequency spectrum generation, which comprises the following steps of: measuring the thickness of rusted steel by a ruler to obtain a rusted steel point position thickness data set, calculating a representative value of the measured thickness of the rusted steel, and calculating the surface topography of the rusted steel. Further correcting the manual measurement thickness representative value of the rusted steel according to the determined manual measurement error correction proportionality coefficient to obtain the corrected thickness of the rusted steel plate, and approximately evaluating the uniform corrosion degree; the method comprises the following steps: acquiring a material category of rusted steel, searching a pre-constructed non-uniform corrosion evaluation data set of various types of steel, and calculating a simulated surface curve corresponding to non-uniform corrosion of the rusted steel; and generating an artificially synthesized surface by combining the corrected thickness of the rusted steel plate and the simulated surface curve of non-uniform corrosion, and constructing a model capable of being used for finite element analysis of the tensile property of the rusted steel. According to the method, only one ruler thickness parameter is needed, the three-dimensional surface matched with actual steel corrosion can be rapidly generated, and the problem that modeling of the rusted steel surface is complicated is solved.

Description

Technical Field

[0001] The invention relates to a corroded steel material analysis technology, and in particular to a corroded steel material surface morphology modeling method based on spectrum generation. Background Art

[0002] Steel structures or steel components are exposed to complex and changeable natural environments and are extremely susceptible to corrosion. The performance of steel materials corroded by the natural environment is seriously degraded, and in severe cases it may even lead to structural collapse, causing serious economic losses and casualties. In the current situation where the volume of newly built steel structures is rapidly expanding, in order to better expand the application of steel in building structures, it is necessary to conduct in-depth research on the mechanical properties of steel after corrosion. At this time, finite element modeling and analysis of corroded steel is particularly important.

[0003] At present, the modeling methods for corroded steel are mostly manual measurement and on-site 3D scanning. Steel corrosion has uniform corrosion and non-uniform corrosion components. For uniform corrosion, 3D scanning results are accurate, but the equipment requirements are high and the operation is difficult; while manual measurement is easy to operate, but its results usually overestimate its uniform corrosion components.

[0004] Although the above method can achieve modeling of corroded steel to a certain extent, the large error of manual measurement, the high cost and high operation difficulty of on-site 3D scanning are still not conducive to rapid and accurate modeling of corroded steel. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the above prior art and provide a method for modeling the surface morphology of corroded steel based on spectrum generation. This method for modeling the surface morphology of corroded steel based on spectrum generation can quickly perform finite element modeling analysis while ensuring accuracy.

[0006] The purpose of the present invention is achieved through the following technical solution: The surface morphology modeling method of corroded steel based on spectrum generation comprises the following steps:

[0007] S1. Measure the thickness of the corroded steel with a ruler to obtain a data set of the thickness of the corroded steel points, calculate the representative value of the thickness of the corroded steel measured with a ruler, further correct the representative value of the thickness of the corroded steel measured manually according to the determined manual measurement error correction proportional coefficient, obtain the corrected thickness of the corroded steel plate, and approximately evaluate the uniform corrosion degree;

[0008] S2, obtaining the material category of the corroded steel, searching for pre-built non-uniform corrosion assessment data sets of various types of steel, and calculating the simulated surface curve corresponding to the non-uniform corrosion of the corroded steel;

[0009] S3. An artificial synthetic surface is generated by combining the corrected thickness of the corroded steel plate and the simulated surface curve of uneven corrosion, and a model that can be used for finite element analysis of the tensile properties of corroded steel is constructed.

[0010] Preferably, before step S1, the following steps are performed:

[0011] The corrosion of steel is divided into uniform corrosion and uneven corrosion.

[0012] Preferably, step S1 includes the following specific steps:

[0013] S11, determining the manual measurement error correction proportional coefficient;

[0014] S12, using a vernier caliper to measure the thickness of several points of the corroded steel to be measured, to obtain a data set of thickness of the corroded steel points;

[0015] S13, taking an average value of the corroded steel point thickness data set to obtain a representative value of the measured thickness of the corroded steel to be measured;

[0016] S14, using the manual measurement error correction proportional coefficient, further correct the manual measurement thickness representative value of the corroded steel to be measured according to the measurement error, obtain the corrected thickness of the corroded steel plate, and approximately evaluate the uniform corrosion degree, wherein the corrected thickness of the corroded steel plate is:

[0017] h1=a1y0

[0018] Among them, h1 is the corrected thickness of the corroded steel plate, a1 is the correction coefficient of the manual measurement error, and y0 is the representative value of the manual measurement thickness.

[0019] Preferably, step S11 includes the following specific steps:

[0020] S111. Perform 3D scanning on a large number of different corroded steel plates to obtain the mean value of the point cloud thickness;

[0021] S112. Use a vernier caliper to measure the thickness of the steel plate at multiple locations;

[0022] S113, taking the average of the steel plate thickness values ​​measured at multiple locations to obtain a representative value of the thickness measured manually;

[0023] S114, the average value of the point cloud thickness of each corroded steel plate and the representative value of the thickness measured by manual ruler, the two measured values ​​are taken as a set of data, and a database is formed after obtaining several sets of data;

[0024] S115. Compare each group of measurement values ​​in the database, and average the ratio of each group to obtain a widely applicable artificial measurement error correction proportional coefficient.

[0025] Preferably, step S2 includes the following specific steps:

[0026] S21. Construct a dataset for evaluating non-uniform corrosion of various types of steel;

[0027] S22, obtaining the material category of the corroded steel to be tested;

[0028] S23, searching for pre-built non-uniform corrosion assessment data sets of various types of steels, and obtaining the frequency domain characteristics, each frequency and corresponding components of the non-uniform corrosion steel to be tested;

[0029] S24. Calculate the simulated surface curve of non-uniform corrosion by using the frequency domain characteristics of the non-uniformly corroded steel to be tested, each frequency and the corresponding components:

[0030]

[0031] Among them, A i is the frequency amplitude corresponding to the frequency; ω i is the frequency, ranging from 1 to 50; b1 is the amplitude-frequency correlation coefficient, which is related to the type of corroded steel material; b2 is the amplitude-frequency correlation index, which is related to the type of corroded steel material; y2 is the offset of the non-uniform corrosion coordinate; x is the length coordinate of the middle of the specimen; is a random number representing the phase.

[0032] Preferably, step S21 includes the following specific steps:

[0033] S211. Conduct sample testing and data collection on various types of corroded steel;

[0034] S212. Select 5 or more sample lines in the length direction on the upper and lower surfaces of each corroded steel sample;

[0035] S213, extracting the uneven corrosion surface concave-convex conditions on the sample line and recording them as a two-dimensional non-periodic signal;

[0036] S214. The non-periodic signal is decomposed into the sum of sine and cosine functions of different frequencies using Fourier transform.

[0037] Preferably, in step S314, the decomposition formula of the non-periodic signal is:

[0038]

[0039] Where ω is the frequency; F(ω) is the frequency domain function (phase spectrum, amplitude spectrum) with frequency as the independent variable; x represents the length coordinate of the middle part of the specimen; and f(x) is the function with x as the independent variable.

[0040] Preferably, step S3 includes the following specific steps:

[0041] S31, setting the corrected thickness of the corroded steel material as the reference thickness of the corroded steel material model;

[0042] S32, using the simulated surface curve of uneven corrosion as the surface morphology curve of the corroded steel model;

[0043] S33. The surface morphology curve of the corroded steel model fluctuates up and down with the base thickness of the corroded steel model as the base line to generate an artificial synthetic surface, thereby constructing a model that can be used for finite element analysis of the tensile properties of the corroded steel.

[0044] Preferably, the artificial synthetic surface is generated in step S33 using the following formula:

[0045] y=h1+y2

[0046] Among them, y is the total offset of each point on the artificial synthetic surface, h1 is the thickness of the corroded steel plate after correction, and y2 is the offset of the non-uniform corrosion coordinates.

[0047] The present invention has the following advantages over the prior art:

[0048] The method of the present invention determines the uniform corrosion thickness by measuring, and obtains the non-uniform corrosion thickness by means of Fourier spectrum inverse transformation. The uniform corrosion thickness and the non-uniform corrosion thickness are superimposed to generate an artificial synthetic corrosion surface to accurately characterize the corrosion morphology characteristics of the corroded steel, and to construct a finite element model that can be used to simulate the corroded steel and its structural analysis. The method of the present invention only requires one measuring thickness parameter to quickly generate a three-dimensional surface that matches the actual steel corrosion, solving the problem of complicated modeling of the corroded steel surface. Therefore, the present invention solves the problems of high equipment requirements, high operating difficulty, poor accuracy and high cost of traditional technologies. At the same time, the present application only requires one measuring thickness parameter to realize finite element model analysis, improving analysis efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic flow chart of a method for modeling the surface morphology of corroded steel based on spectrum generation provided in one embodiment of the present invention.

[0050] Figure 2 It is a regression curve of the frequency domain characteristics of Q275 low carbon steel obtained in one embodiment of the present invention.

[0051] Figure 3 This is a frequency domain characteristic regression curve of 600MPa grade high-strength steel obtained in one embodiment of the present invention.

[0052] Figure 4 Schematic diagram of five typical artificially generated surfaces based on a uniformly eroded surface in one embodiment of the present invention. DETAILED DESCRIPTION

[0053] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0054] This embodiment takes the establishment of a database, the determination of the measurement error correction proportional coefficient, and the construction and application of various types of steel non-uniform corrosion assessment data sets as examples. Figure 1 As shown, the surface morphology modeling method of corroded steel based on spectrum generation in this embodiment includes the following steps:

[0055] First, the corrosion of steel is divided into uniform corrosion and uneven corrosion. Then the following operations are performed:

[0056] S1, using a ruler to measure the thickness of the corroded steel, obtaining a data set of the thickness of the corroded steel points, calculating the representative value of the measured thickness of the corroded steel, further correcting the representative value of the manually measured thickness of the corroded steel according to the determined manual measurement error correction proportional coefficient, obtaining the corrected thickness of the corroded steel plate, and approximately evaluating the uniform corrosion degree; step S1 includes the following specific steps:

[0057] S11, determining the manual measurement error correction proportional coefficient; Step S11 includes the following specific steps:

[0058] S111. Perform 3D scanning on a large number of different corroded steel plates to obtain the mean value of the point cloud thickness;

[0059] S112. Use a vernier caliper to measure the thickness of the steel plate at multiple locations;

[0060] S113, taking the average of the steel plate thickness values ​​measured at multiple locations to obtain a representative value of the thickness measured manually;

[0061] S114, the average value of the point cloud thickness of each corroded steel plate and the representative value of the thickness measured by manual ruler, the two measured values ​​are taken as a set of data, and a database is formed after obtaining several sets of data;

[0062] S115. Compare each group of measurement values ​​in the database, and average the ratio of each group to obtain a widely applicable artificial measurement error correction proportional coefficient.

[0063] Specifically, the manual measurement error correction proportional coefficient determined in this embodiment, the database formed by measuring 18 corroded steel plate specimens and the manual measurement error correction proportional coefficient are shown in Table 1. The average thickness obtained by the point cloud is about 0.96 times the measured thickness, because the measurement ignores the existence of rust pits and overestimates the remaining thickness of the corroded steel plate.

[0064] Table 1 Mean value of point cloud thickness and measured thickness

[0065]

[0066] By correcting the thickness of the ruler, the deviation of the vernier caliper measurement result is corrected, and a more accurate uniform corrosion component is obtained, which is more accurate and convenient for evaluating the corrosion degree of steel structures in actual projects. When the database volume is large enough, the obtained manual ruler error correction proportional coefficient is more widely applicable, making the correction effect more accurate. As can be seen from Table 1 above, the manual ruler error correction proportional coefficient of this embodiment is 0.96.

[0067] S12, using a vernier caliper to measure the thickness of several points of the corroded steel to be measured, to obtain a data set of thickness of the corroded steel points;

[0068] S13, taking an average value of the corroded steel point thickness data set to obtain a representative value of the measured thickness of the corroded steel to be measured;

[0069] S14, using the manual measurement error correction proportional coefficient, further correct the manual measurement thickness representative value of the corroded steel to be measured according to the measurement error, obtain the corrected thickness of the corroded steel plate, and approximately evaluate the uniform corrosion degree, wherein the corrected thickness of the corroded steel plate is:

[0070] h1=a1y0

[0071] Among them, h1 is the corrected thickness of the corroded steel plate, a1 is the correction coefficient of the manual measurement error, and y0 is the representative value of the manual measurement thickness.

[0072] S2, obtaining the material category of the corroded steel, searching for pre-built non-uniform corrosion assessment data sets of various types of steel, and calculating the simulated surface curve corresponding to the non-uniform corrosion of the corroded steel; step S2 includes the following specific steps:

[0073] S21, constructing a dataset for evaluating non-uniform corrosion of various types of steels; Step S21 includes the following specific steps:

[0074] S211. Conduct sample testing and data collection on various types of corroded steel;

[0075] S212. Select 5 or more sample lines in the length direction on the upper and lower surfaces of each corroded steel sample;

[0076] S213, extracting the uneven corrosion surface concave-convex conditions on the sample line and recording them as a two-dimensional non-periodic signal;

[0077] S214, the non-periodic signal is decomposed into the sum of sine and cosine functions of different frequencies using Fourier transform. In step S214, the decomposition formula of the non-periodic signal is:

[0078]

[0079] Where ω is the frequency; F(ω) is the frequency domain function (phase spectrum, amplitude spectrum) with frequency as the independent variable; x represents the length coordinate of the middle part of the specimen; and f(x) is the function with x as the independent variable.

[0080] S22, obtaining the material category of the corroded steel to be tested;

[0081] S23, searching for pre-built non-uniform corrosion assessment data sets of various types of steels, and obtaining the frequency domain characteristics, each frequency and corresponding components of the non-uniform corrosion steel to be tested;

[0082] A dataset of non-uniform corrosion assessment of various types of steel was constructed. The frequency domain characteristics of 12 Q275 low-carbon steel corroded specimens and 6 600MPa high-strength steel corroded specimens were analyzed to obtain the amplitude spectrum law. After regression analysis, the frequency domain characteristics of the surface morphology of Q275 low-carbon steel and 600MPa high-strength steel after corrosion were obtained, such as Figure 2 and Figure 3 shown.

[0083] The relationship between each frequency and the corresponding component obtained by regression is:

[0084]

[0085] Among them, A i is the component corresponding to the frequency; ω i is the frequency, ranging from 1 to 50.

[0086] As the data set expands, more accurate and comprehensive frequency domain characteristics of non-uniformly corroded steel will be obtained after regression analysis, and the non-uniform corrosion coordinate offset will be more accurate, thereby generating more effective simulated surface curves of non-uniform corrosion. The frequency domain characteristics of non-uniformly corroded steel are related to the type and properties of steel, so the construction of the data set requires a large amount of data accumulation for each type of steel.

[0087] According to the type of the corroded steel to be tested, a pre-constructed dataset of various types of steel non-uniform corrosion assessment is searched to obtain the frequency domain characteristics of the surface morphology of the corroded steel to be tested after corrosion.

[0088] S24. Calculate the simulated surface curve of non-uniform corrosion by using the frequency domain characteristics of the non-uniformly corroded steel to be tested, each frequency and the corresponding components:

[0089]

[0090] Among them, A i is the frequency amplitude corresponding to the frequency; ω iis the frequency, ranging from 1 to 50; b1 is the amplitude-frequency correlation coefficient, which is related to the type of corroded steel material; b2 is the amplitude-frequency correlation index, which is related to the type of corroded steel material; y2 is the offset of the non-uniform corrosion coordinate; x is the length coordinate of the middle of the specimen; is a random number representing the phase.

[0091] S3, combining the corrected thickness of the corroded steel plate and the simulated surface curve of uneven corrosion to generate an artificial synthetic surface, and constructing a model that can be used for finite element analysis of tensile properties of corroded steel. Step S3 includes the following specific steps:

[0092] S31, setting the corrected thickness of the corroded steel material as the reference thickness of the corroded steel material model;

[0093] S32, using the simulated surface curve of uneven corrosion as the surface morphology curve of the corroded steel model;

[0094] S33. The surface morphology curve of the corroded steel model fluctuates up and down with the base thickness of the corroded steel model as the base line to generate an artificial synthetic surface, thereby constructing a model that can be used for finite element analysis of the tensile properties of the corroded steel.

[0095] The artificial synthetic surface is generated in step S33 using the following formula:

[0096] y=h1+y2

[0097] Among them, y is the total offset of each point on the artificial synthetic surface, h1 is the thickness of the corroded steel plate after correction, and y2 is the offset of the non-uniform corrosion coordinates.

[0098] Since the phase spectrum uses random numbers in the process of generating the surface topography, the model calculation results may fluctuate, so 30 different surface topography models were generated for the 18 specimens and settled, and the results showed that the average value of the peak force fluctuation of the specimens was about 4%, and the average value of the peak strain fluctuation was about 40%.

[0099] Exemplarily, the 18 corrosion specimens were subjected to tensile tests, and finite element results and experimental results were obtained and compared. Table 2 further compares the artificial surface modeling results and experimental results.

[0100] Table 2 Peak load and peak strain of steel plate

[0101]

[0102] The analysis results of artificially generated surface modeling, on the one hand, take into account the overestimation of uniform corrosion by measuring, and use the proportional coefficient corrected by artificial measuring error, so the peak force is closer to the test result; on the other hand, the non-uniform corrosion component is taken into account, and the peak strain is also closer to the test result.

[0103] The above specific implementation modes are preferred embodiments of the present invention and cannot be used to limit the present invention. Any other changes or other equivalent replacement methods that do not deviate from the technical solution of the present invention are included in the protection scope of the present invention.

Claims

1. A method for modeling the surface morphology of corroded steel based on spectrum generation, characterized in that: The following steps are involved: S1. Measure the thickness of the corroded steel with a ruler to obtain a data set of the thickness of the corroded steel points, calculate the representative value of the thickness of the corroded steel measured with a ruler, further correct the representative value of the thickness of the corroded steel measured manually according to the determined manual measurement error correction proportional coefficient, obtain the corrected thickness of the corroded steel plate, and approximately evaluate the uniform corrosion degree; S2, obtaining the material category of the corroded steel, searching for pre-built non-uniform corrosion assessment data sets of various types of steel, and calculating the simulated surface curve corresponding to the non-uniform corrosion of the corroded steel; S3. An artificial synthetic surface is generated by combining the corrected thickness of the corroded steel plate and the simulated surface curve of uneven corrosion, and a model that can be used for finite element analysis of the tensile properties of corroded steel is constructed.

2. The method for modeling the surface morphology of corroded steel based on spectrum generation according to claim 1, characterized in that: Before step S1, the following steps are performed: The corrosion of steel is divided into uniform corrosion and uneven corrosion.

3. The method for modeling the surface morphology of corroded steel based on spectrum generation according to claim 1, characterized in that: Step S1 includes the following specific steps: S11, determining the manual measurement error correction proportional coefficient; S12, using a vernier caliper to measure the thickness of several points of the corroded steel to be measured, to obtain a data set of thickness of the corroded steel points; S13, taking an average value of the corroded steel point thickness data set to obtain a representative value of the measured thickness of the corroded steel to be measured; S14, using the manual measurement error correction proportional coefficient, further correct the manual measurement thickness representative value of the corroded steel to be measured according to the measurement error, obtain the corrected thickness of the corroded steel plate, and approximately evaluate the uniform corrosion degree, wherein the corrected thickness of the corroded steel plate is: h1=a1y0 Among them, h1 is the corrected thickness of the corroded steel plate, a1 is the correction coefficient of the manual measurement error, and y0 is the representative value of the manual measurement thickness.

4. The method for modeling the surface morphology of corroded steel based on spectrum generation according to claim 3 is characterized in that: Step S11 includes the following specific steps: S111. Perform 3D scanning on a large number of different corroded steel plates to obtain the mean value of the point cloud thickness; S112. Use a vernier caliper to measure the thickness of the steel plate at multiple locations; S113, taking the average of the steel plate thickness values ​​measured at multiple locations to obtain a representative value of the thickness measured manually; S114, the average value of the point cloud thickness of each corroded steel plate and the representative value of the thickness measured by manual ruler, the two measured values ​​are taken as a set of data, and a database is formed after obtaining several sets of data; S115. Compare each group of measurement values ​​in the database, and average the ratio of each group to obtain a widely applicable artificial measurement error correction proportional coefficient.

5. The method for modeling the surface morphology of corroded steel based on spectrum generation according to claim 1, characterized in that: Step S2 includes the following specific steps: S21. Construct a dataset for evaluating non-uniform corrosion of various types of steel; S22, obtaining the material category of the corroded steel to be tested; S23, searching for pre-built non-uniform corrosion assessment data sets of various types of steels, and obtaining the frequency domain characteristics, each frequency and corresponding components of the non-uniform corrosion steel to be tested; S24. Calculate the simulated surface curve of non-uniform corrosion by using the frequency domain characteristics of the non-uniformly corroded steel to be tested, each frequency and the corresponding components: Among them, A i is the frequency amplitude corresponding to the frequency; ω i is the frequency, ranging from 1 to 50; b1 is the amplitude-frequency correlation coefficient, which is related to the type of corroded steel material; b2 is the amplitude-frequency correlation index, which is related to the type of corroded steel material; y2 is the offset of the non-uniform corrosion coordinate; x is the length coordinate of the middle of the specimen; is a random number representing the phase.

6. The method for modeling the surface morphology of corroded steel based on spectrum generation according to claim 5, characterized in that: Step S21 includes the following specific steps: S211. Conduct sample testing and data collection on various types of corroded steel; S212. Select 5 or more sample lines in the length direction on the upper and lower surfaces of each corroded steel sample; S213, extracting the uneven corrosion surface concave-convex conditions on the sample line and recording them as a two-dimensional non-periodic signal; S214. The non-periodic signal is decomposed into the sum of sine and cosine functions of different frequencies using Fourier transform.

7. The method for modeling the surface morphology of corroded steel based on spectrum generation according to claim 6, characterized in that: In step S314, the decomposition formula of the non-periodic signal is: Where ω is the frequency; F(ω) is the frequency domain function (phase spectrum, amplitude spectrum) with frequency as the independent variable; x represents the length coordinate of the middle part of the specimen; and f(x) is the function with x as the independent variable.

8. The method for modeling the surface morphology of corroded steel based on spectrum generation according to claim 1, characterized in that: Step S3 includes the following specific steps: S31, setting the corrected thickness of the corroded steel material as the reference thickness of the corroded steel material model; S32, using the simulated surface curve of uneven corrosion as the surface morphology curve of the corroded steel model; S33. The surface morphology curve of the corroded steel model fluctuates up and down with the base thickness of the corroded steel model as the base line to generate an artificial synthetic surface, thereby constructing a model that can be used for finite element analysis of the tensile properties of the corroded steel.

9. The method for modeling the surface morphology of corroded steel based on spectrum generation according to claim 7, characterized in that: The artificial synthetic surface is generated in step S33 using the following formula: y=h1+y2 Among them, y is the total offset of each point on the artificial synthetic surface, h1 is the thickness of the corroded steel plate after correction, and y2 is the offset of the non-uniform corrosion coordinates.

Images