Environmental adaptability growth model construction method of industrial product in natural environment

Abstract

The invention discloses a method for constructing an environmental adaptability growth model of an industrial product in a natural environment. The method comprises the following steps: extracting apparent damage characteristics and mechanical property characteristics of the industrial product; establishing a first relation model of the apparent damage characteristics and the environmental stress; establishing a second relation model of the mechanical property characteristics and the environmental stress; establishing a coupling relationship between the apparent damage characteristics and the mechanical property characteristics, wherein the coupling relationship represents interaction between the apparent damage characteristics and the mechanical property characteristics; and through the first relation model, the second relation model and the coupling relation, determining an environmental adaptability growth model of the industrial product under the natural environment condition. The method provides a scientific basis for product design, maintenance and life prediction.

Description

Technical Field

[0001] The present invention belongs to the technical field of environmental adaptability of industrial products, and particularly relates to a method for constructing an environmental adaptability growth model of industrial products in the natural environment. Background Art

[0002] When industrial products are exposed in the natural environment for a long time, they will be affected by the combined action of various environmental stresses, such as temperature change, humidity, salt spray, ultraviolet radiation, acid rain, sand and dust, etc. These environmental stresses will cause irreversible damage to the surface and internal structure of the products, thereby affecting their mechanical properties and service life. For example, metal materials may corrode, composite materials may delaminate or age, and coating materials may peel off or fail, etc. These damages not only reduce the performance of the products, but may also cause safety hazards and increase the maintenance and replacement costs.

[0003] At present, for the environmental adaptability assessment of industrial products in the natural environment, there have been some related researches and technologies, but there are still some problems and deficiencies. For example, the limitations of the apparent damage assessment are mainly manifested in the traditional method relying on manual observation, lack of systematicness, insufficient dynamic monitoring, etc. The deficiencies of the mechanical property assessment are mainly manifested in limited test samples, single performance index and lack of environmental stress correlation; the deficiencies of the environmental adaptability model are mainly manifested in model simplification, insufficient data driving and unsatisfactory prediction accuracy, etc. Moreover, the existing technologies usually separate the apparent damage assessment from the mechanical property test, do not establish the correlation between the two, and lack intelligent methods. The existing technologies mostly rely on traditional methods and fail to make full use of intelligent technologies such as image processing and machine learning for data analysis and model construction. Summary of the Invention

[0004] In view of this, the object of the present invention is to provide a method for constructing an environmental adaptability growth model, which includes:

[0005] Step 1: Extract the apparent damage characteristics and mechanical property characteristics of industrial products;

[0006] Step 2: Establish a first relationship model between the apparent damage characteristics and environmental stresses;

[0007] Step 3: Establish a second relationship model between the mechanical property characteristics and environmental stresses;

[0008] Step 4: Establish a coupling relationship between the apparent damage characteristics and the mechanical property characteristics, and the coupling relationship characterizes the interaction existing between the apparent damage characteristics and the mechanical property characteristics;

[0009] Step 5: Determine the environmental adaptability growth model of industrial products under natural environmental conditions through the first relationship model, the second relationship model and the coupling relationship.

[0010] Furthermore, the extraction of apparent damage features includes:

[0011] Regularly take multi-angle photos of industrial products under different lighting conditions to obtain images of industrial products;

[0012] Preprocess the images;

[0013] Identify damage on the preprocessed images. The damage categories include one or a combination of more than one of corrosion, cracks, coating peeling, etc.;

[0014] For corrosion damage, the apparent damage features include corrosion area, corrosion depth, corrosion type, and corrosion degree;

[0015] For crack damage, the apparent damage features include crack length, crack width, and crack distribution density;

[0016] For coating peeling damage, the apparent damage features include the peeling area and peeling ratio of the coating peeling area.

[0017] Furthermore, the extraction of mechanical property features includes:

[0018] Extract strength features, toughness features, hardness features, and fatigue features. Among them, the strength features include extraction through tensile strength tests, extraction through compressive strength tests, and extraction through flexural strength tests; the toughness features include extraction through impact toughness tests and extraction through fracture toughness tests; the hardness features are extracted through hardness tests; the fatigue features include extraction through fatigue life tests and extraction through fatigue crack growth tests.

[0019] Furthermore, the mechanical property features also include parameters such as creep limit and creep rate extracted through creep property tests, as well as elastic modulus, etc. extracted through elastic modulus tests.

[0020] Furthermore, the first relationship model includes linear models, exponential models, power-law models, logarithmic models, piecewise models, models based on physical mechanisms, etc.

[0021] Furthermore, the second relationship model includes linear degradation models, exponential degradation models, power-law degradation models, logarithmic degradation models, piecewise degradation models, models based on physical mechanisms, etc.

[0022] Furthermore, the environmental adaptability growth model includes linear growth models, exponential growth models, power-law growth models, logarithmic coupling models, etc.

[0023] The beneficial effects of the present invention are:

[0024] The present invention systematically extracts core features, namely, through image processing and mechanical testing technologies, it systematically extracts the surface damage features (such as corrosion area, corrosion degree) and mechanical property features (such as strength, toughness) of industrial products; and based on long-term environmental exposure data, constructs a mathematical model describing the variation of apparent damage features and mechanical property features with environmental stress; moreover, explores the coupling relationship between apparent damage features and mechanical property features, and based on this, establishes an environmental adaptability growth model; at the same time, through dynamic monitoring and model prediction, evaluates the environmental adaptability performance of products under different environmental conditions, providing a scientific basis for product design, maintenance, and life prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings, where:

[0026] Figure 1 is a flowchart of the method for constructing an environmental adaptability growth model shown in the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] The following will refer to the accompanying drawings to describe the preferred embodiments of the present invention in detail. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the protection scope of the present invention.

[0028] Figure 1 is a flowchart of the method for constructing an environmental adaptability growth model. As shown, the method may include the following steps:

[0029] Step 1: Extract the apparent damage features and mechanical property features of industrial products;

[0030] Step 2: Establish a first relationship model between the apparent damage features and environmental stress;

[0031] Step 3: Establish a second relationship model between the mechanical property features and environmental stress;

[0032] Step 4: Establish a coupling relationship between the apparent damage features and the mechanical property features, and the coupling relationship characterizes the interaction existing between the apparent damage features and the mechanical property features;

[0033] Step 5: Determine the environmental adaptability growth model of industrial products under natural environmental conditions through the first relationship model, the second relationship model, and the coupling relationship.

[0034] It should be noted that Figure 1 the execution order of Steps 2, 3, and Step 4 shown in Figure 1The steps may be executed in the shown order, or in other suitable orders. For example, step 4 may be executed first, and then steps 2 and 3 may be executed simultaneously or in sequence; or steps 2 and 3 may be executed simultaneously first, and then step 4 may be executed.

[0035] In some embodiments, the apparent damage feature extraction includes:

[0036] Regularly taking multi-angle pictures of the industrial product under different lighting conditions to obtain images of the industrial product;

[0037] Preprocessing the images;

[0038] Performing damage identification on the preprocessed images, and the damage categories include one or a combination of more of the following: corrosion, crack, material delamination, coating peeling, etc.

[0039] For example, a high-resolution digital camera or industrial-grade optical equipment can be used to regularly take pictures of the surface of the industrial product to ensure that the images are clear and distortion-free; and, images are collected under different lighting conditions (such as natural light, polarized light) to enhance the visibility of damage features; moreover, for the same area, pictures can be taken from multiple angles to ensure comprehensive capture of damage features.

[0040] The preprocessing of the images may include denoising, image enhancement, image segmentation, etc. For example, filtering algorithms (such as Gaussian filtering, median filtering) can be used to remove noise in the images to facilitate subsequent image processing; through image enhancement methods such as contrast adjustment and histogram equalization, the contrast between the damaged area and the background can be enhanced to facilitate the identification of the damaged area; in addition, through image segmentation, the damaged area in the image can be separated from the undamaged area to facilitate subsequent feature extraction.

[0041] For corrosion damage, the apparent damage features may include corrosion area, corrosion depth, corrosion type, corrosion degree, etc.

[0042] For corrosion damage, edge detection algorithms (such as Canny edge detection) or threshold segmentation techniques can be used to identify the boundary of the corrosion area; then, morphological operations (such as erosion) are used to optimize the corrosion area to ensure that the boundary is continuous and accurate; further, the pixel area of the corrosion area is calculated by pixel statistics method. Furthermore, the pixel area can be converted into the actual physical area of the corrosion area and calibrated in combination with the image resolution.

[0043] For corrosion damage, the depth information of the corrosion area can also be obtained by a three-dimensional surface profiler or a laser scanner. Through the depth distribution map, the severity of corrosion can be quantified.

[0044] For corrosion damage, it is also possible to classify the corrosion types (such as pitting corrosion, uniform corrosion, etc.) based on image texture features (such as gray-level co-occurrence matrix, local binary pattern). For the classification of corrosion types, machine learning algorithms (such as support vector machines, convolutional neural networks) can be used to automatically identify the corrosion types.

[0045] According to the corrosion area, corrosion depth, and corrosion type, through the corrosion degree rating standard, it is also possible to quantitatively rate the corrosion degree.

[0046] For crack damage, the apparent damage features can include crack length, crack width, and crack distribution density. For example, crack detection algorithms can be used to identify cracks on the surface of industrial products, thereby calculating crack length, crack width, and crack distribution density, etc.

[0047] For coating peeling damage, the apparent damage features can include the peeling area and peeling ratio of the coating peeling area. For example, through color differences or texture changes, the coating peeling area can be identified, and then the area of the peeling area and the ratio of the coating peeling can be calculated.

[0048] In some embodiments, a surface roughness meter or image texture analysis technology can also be used for quantitative analysis of the change in surface roughness of industrial products. The change in surface roughness can also be used as an apparent damage feature.

[0049] In some embodiments, the extraction of mechanical property features can include: extracting strength features, toughness features, hardness features, and fatigue features.

[0050] Strength features include extraction through tensile strength testing, extraction through compressive strength testing, and extraction through flexural strength testing.

[0051] For tensile strength testing, a universal material testing machine can be used to perform tensile testing on the sample, record the stress-strain curve of the sample during the tensile process, and thereby extract key parameters such as the maximum tensile strength and yield strength.

[0052] For compressive strength testing, the sample can be subjected to compressive testing, record the stress-strain curve during compression, and extract parameters such as the maximum compressive strength and compression modulus.

[0053] For flexural strength testing, the sample can be subjected to three-point bending or four-point bending testing, record the load-displacement curve during bending, and thereby extract parameters such as flexural strength and flexural modulus.

[0054] Toughness features include extraction through impact toughness testing and extraction through fracture toughness testing.

[0055] For the impact toughness test, an impact testing machine with a pendulum can be used to conduct the impact test, and the energy absorption value of the sample during the impact process is recorded, so as to extract parameters such as impact toughness (such as impact power, impact strength).

[0056] For the fracture toughness test, a fracture mechanics testing machine can be used to conduct the fracture toughness test, and the load-displacement curve during the crack propagation process is recorded, so as to extract parameters such as fracture toughness (such as stress intensity factor).

[0057] The hardness characteristics are extracted through hardness tests. For example, the surface of the sample can be tested for hardness using a hardness tester (such as Brinell hardness tester, Rockwell hardness tester, Vickers hardness tester), and the hardness value is recorded, so as to evaluate the change in the surface hardness of the material.

[0058] The fatigue characteristics include extraction through fatigue life tests and extraction through fatigue crack propagation tests.

[0059] For the fatigue life test, a fatigue testing machine can be used to conduct cyclic loading tests on the sample, and the fatigue life of the sample under cyclic loading (i.e., the number of stress cycles experienced by the material before fatigue failure) is recorded, so as to extract parameters such as fatigue limit and fatigue strength.

[0060] For the fatigue crack propagation test, the sample can be tested for fatigue crack propagation, and the relationship between the crack propagation rate and the stress intensity factor is recorded, so as to extract parameters such as the fatigue crack propagation threshold value.

[0061] In some embodiments, the mechanical property characteristics may further include parameters such as creep limit and creep rate extracted through creep property tests and elastic modulus extracted through elastic modulus tests.

[0062] For example, a high-temperature creep test is conducted on the sample, and the change of creep strain with time is recorded, so as to extract parameters such as creep limit and creep rate; for the elastic modulus test, through tensile or compression tests, the elastic modulus of the sample is calculated, so as to evaluate the stiffness of the material in the elastic deformation stage.

[0063] In step 2, the established first relationship model can be used to reflect the change of apparent damage characteristics with environmental stress. The first relationship model can be expressed as: A(t) = f(E, t), where A represents the apparent damage characteristics, E represents the environmental stress, t represents time, and f represents the relationship between the apparent damage characteristics and the environmental stress.

[0064] The first relationship model may include a linear model, an exponential model, a power-law model, a logarithmic model, a piecewise model, a model based on physical mechanisms, a machine learning model, etc. The specific model selection should be based on the damage development law, the characteristics of environmental stress, and the characteristics of experimental data.

[0065] The linear model is applicable to the situation where the range of environmental stress variation is small and the damage development is relatively uniform. The linear model can be expressed as:

[0066] A(t) = A0 + k E ·E·t

[0067] In the above formula, A0 represents the initial damage value; k E represents the damage growth coefficient, that is, the damage increment caused by unit environmental stress.

[0068] The exponential model is applicable to the situation where environmental stress has an accelerating effect on damage development. The exponential model can be expressed as:

[0069]

[0070] In the above formula, A0 represents the initial damage value, and k2 represents the damage growth rate coefficient.

[0071] The power-law model is applicable to the situation where there is a non-linear relationship between damage development and environmental stress and the damage characteristics are in a power function relationship with environmental stress. The power-law model can be expressed as:

[0072] A(t) = k3·E n ·t

[0073] In the above formula, k3 and n represent the model parameters, and n represents the degree of non-linearity of damage development.

[0074] The logarithmic model is applicable to the situation where the damage development is fast in the initial stage and tends to be gentle in the later stage (that is, the damage development is sensitive to environmental stress in the initial stage and approaches saturation in the later stage), and the apparent damage characteristics increase logarithmically with environmental stress. The logarithmic model can be expressed as:

[0075] A = a·ln(E·t) + b

[0076] In the above formula, a and b represent the model parameters.

[0077] The piecewise model divides the damage development process into multiple stages, and each stage is described by a different mathematical model. It is applicable to the situation where the damage development process is complex. The piecewise model can be expressed as:

[0078]

[0079] In the above formula, a1, a2, a3 and b1, b2, b3 represent the parameters of each stage. It should be noted that the number of stages described in the above formula and the form of the model in each stage are only illustrative and are not used to limit the number of stages of the piecewise model and the form of the model in each stage.

[0080] The physics-based model starts from the physical process of damage development and combines the action mechanism of environmental stress to construct a more accurate model. The physics-based model can be expressed as:

[0081] A = f1(E, t, C),

[0082] In the above formula, C represents material parameters, including chemical composition, microstructure, etc. of the material, and f1 represents the influence of environmental stress E, time t, and material parameter C on the apparent damage characteristic A.

[0083] For example, assuming that the corrosion area increases exponentially with environmental stress (such as temperature, salt spray concentration, etc.), the first relationship model is:

[0084]

[0085] where, A(t) is the corrosion area, A0 represents the initial corrosion area, k E represents the corrosion growth coefficient, E is the environmental stress, and t is the time.

[0086] In step 3, the established second relationship model is a mathematical model that characterizes the change law of mechanical property characteristics (such as strength, hardness, toughness, etc.) of materials or products under the action of environmental stress (temperature, humidity, pollutant concentration, mechanical load, etc.).

[0087] The second relationship model can include a linear degradation model, an exponential degradation model, a power-law degradation model, a logarithmic degradation model, a piecewise degradation model, a physics-based model, a machine learning model, etc.

[0088] The linear degradation model assumes that the mechanical property characteristics change linearly with environmental stress, and is applicable to the situation where the environmental stress is small and the performance degradation is slow and uniform. The linear degradation model can be expressed as:

[0089] M = M0 - k1'·E·t,

[0090] In the above formula, M represents the mechanical property characteristics (such as strength, toughness, etc.), M0 represents the initial value of the mechanical property characteristics, and k1' represents the degradation rate of the performance per unit environmental stress and unit time.

[0091] The exponential degradation model assumes that the mechanical property characteristics decay exponentially with environmental stress, and is applicable to the situation where the environmental stress has an accelerating effect on the performance degradation. The exponential degradation model can be expressed as:

[0092]

[0093] In the above formula, M0 represents the initial value of the mechanical property characteristics, and k2' represents the performance degradation rate coefficient.

[0094] The applicable scenario of the exponential degradation model is that environmental stress has an obvious accelerating effect on the degradation of mechanical properties, such as the strength degradation of materials in high-temperature environments.

[0095] The power-law degradation model assumes that the mechanical property characteristics change as a power function of the environmental stress and is applicable to the case where there is a non-linear relationship between property degradation and environmental stress. The power-law degradation model can be expressed as:

[0096]

[0097] In the above formula, M0 represents the initial mechanical property characteristic value, and k3′ represents the degree of non-linearity of property degradation.

[0098] The applicable scenario of the power-law degradation model is that the influence of environmental stress on the degradation of mechanical properties has obvious non-linear characteristics, such as fatigue crack propagation.

[0099] The logarithmic degradation model assumes that the mechanical property characteristics change logarithmically with the environmental stress and is applicable to the case where property degradation is relatively fast in the initial stage and tends to level off in the later stage. The logarithmic degradation model can be expressed as:

[0100] M = M0 - k4′·ln(E·t)

[0101] In the above formula, M0 represents the initial mechanical property characteristic value, and k4′ represents the model parameter.

[0102] The applicable scenario of the logarithmic degradation model is that the mechanical properties are sensitive to environmental stress in the initial stage of degradation and tend to saturate in the later stage.

[0103] The piecewise degradation model divides the mechanical property degradation process into multiple stages, and each stage is described by a different mathematical model, which is applicable to the case where the property degradation process is complex. The piecewise degradation model can be expressed as:

[0104]

[0105] In the above formula, D1, D2, and D3 represent the initial mechanical property characteristic values of each stage, and q1, q2, and q3 represent the property degradation coefficients of each stage.

[0106] The applicable scenario of the piecewise degradation model is that the mechanical property degradation process has obvious stage characteristics, such as the aging process of materials.

[0107] The model based on physical mechanism starts from the physical process of mechanical property degradation and combines the action mechanism of environmental stress to construct a more accurate model. The model based on physical mechanism can be expressed as:

[0108] M = f2(E, t, C),

[0109] In the above formula, f2 represents the influence of environmental stress E, time t, and material parameter C (such as chemical composition and microstructure of the material) on the mechanical property characteristic M.

[0110] For example, the fatigue life model (based on the S-N curve): N = C′·(Δσ) -m , where N represents the fatigue life, Δσ represents the stress amplitude, and C′ and m represent material constants.

[0111] For another example, the creep damage model: where ε represents the creep strain, σ represents the stress, p, n1, and m1 represent material parameters, Q represents the activation energy, R represents the gas constant, and T represents the temperature.

[0112] The applicable scenarios of the model based on physical mechanisms are those that require an accurate description of the mechanical property degradation mechanism, such as high-temperature creep, fatigue fracture, etc.

[0113] For example, the strength of a metal material degrades linearly with environmental stress, and the second relationship model is M(t) = M0 - k M ·E·t, where M(t) is the strength, M0 represents the initial strength, and k M is the strength degradation coefficient.

[0114] In step 4, a coupling relationship between the apparent damage characteristic and the mechanical property characteristic is established. This is because there may be an interaction between the apparent damage characteristic and the mechanical property characteristic. For example: surface damage caused by corrosion will reduce the strength of the material; crack propagation will accelerate the fatigue failure of the material.

[0115] The coupling relationship can be expressed as:

[0116] M(t) = M0′ - αA(t)

[0117] In the above formula, M0′ represents the value of the mechanical property characteristic at the initial moment; α represents the coupling coefficient, indicating the degree of influence of the apparent damage on the mechanical property; A(t) represents the value of the apparent damage characteristic at time t.

[0118] Suppose the initial strength of a certain metal material is M0 = 500 MPa, the corrosion area increases with time, and the influence coefficient of corrosion on the strength is α = 0.1. When the corrosion area A(t) = 100 mm 2 , the strength M(t) of the material can be calculated through the coupling relationship:

[0119] M(t) = 500 - 0.1×100 = 490 MPa

[0120] This shows that due to the increase in the corrosion area by 100 mm 2 , the strength of the material decreases from the initial 500 MPa to 490 MPa.

[0121] Next, in step 5, an environmental adaptability growth model is constructed through the first relationship model, the second relationship model, and the coupling relationship. The environmental adaptability growth model can be expressed as: S = h(A, M, t), where S represents the environmental adaptability index, and h represents the mapping relationship between the apparent damage feature A, the mechanical property feature M, and the environmental adaptability index S.

[0122] The environmental adaptability growth model aims to comprehensively describe the adaptability change law of industrial products in the atmospheric natural environment. This model not only considers the development of apparent damage (such as corrosion, cracks, etc.), but also considers the degradation of mechanical properties (such as strength, toughness, etc.), as well as the interaction between the two.

[0123] According to the forms of the first relationship model, the second relationship model, and the coupling relationship, the constructed environmental adaptability growth model can have multiple expression forms, specifically depending on the assumptions and application scenarios of the model. The following are several common model expressions.

[0124] Linear growth model: S = a1A + a2M + a3t + a4

[0125] Exponential growth model:

[0126] Power-law growth model:

[0127] Logarithmic growth model: S = d1 ln(A) + d2ln(M) + d3t + d4

[0128] It should be noted that for some parameters in the environmental adaptability growth model (such as the coefficients or exponents in the front), the same symbols are used as those in the expressions of the first relationship model, the second relationship model, and / or the coupling relationship mentioned above, but their specific values are related to the actual situation. For example, the specific value of a1 in the first relationship model may be different from the value of a1 in the linear growth model here, and the same represented characters can also take different values, which will not be elaborated here.

[0129] After constructing the environmental adaptability growth model, it can be verified through experimental data. The test environment can represent the typical characteristics of a certain type of environment, the test conditions are natural environmental conditions, and the environmental factor values can be monitored in real time. Then, through on-site measurement experiments, data on the apparent damage characteristics and mechanical property characteristics under the actual environment are obtained, and model prediction is carried out based on the obtained data. Finally, by comparing the model prediction results with the test data, the accuracy and applicability of the model are verified.

[0130] The following is illustrated through specific embodiments.

[0131] Example: Construction of the environmental adaptability growth model of metal materials under corrosive environments

[0132] 1. Actual data assumptions

[0133] Suppose there is a set of experimental data of metal materials under corrosive environments, including:

[0134] Environmental stress E: Salt spray concentration (unit: g / m 3 )

[0135] Time t: Exposure time (unit: days)

[0136] Apparent damage characteristic A: Corrosion area (unit: mm 2 )

[0137] Mechanical property characteristic M: Tensile strength (unit: MPa)

[0138] The experimental data are as follows:

[0139] Time (days) <![CDATA[Salt spray concentration (g / m 3 )]]> <![CDATA[Corrosion area (mm 2 )]]> Tensile strength (MPa) 10 5 2.5 450 20 5 5.0 430 30 5 7.5 410 40 5 10 390

[0140] 2. First relationship model

[0141] Suppose the apparent damage characteristic A grows linearly with the environmental stress E and time t, then the first relationship model can be expressed as: A(t) = A0 + k1·E·t.

[0142] According to the experimental data, by fitting, we get A(t) = 0.05·E·t,

[0143] where:

[0144] k1 = 0.05 (unit: mm 2 / (g / m 3 ·day),

[0145] A0 = 0 (the initial corrosion area is 0).

[0146] 3. Second relationship model

[0147] Suppose the mechanical property characteristic M degrades linearly with the environmental stress E and time t, then the second relationship model can be expressed as: M(t) = M0 - k2·E·t.

[0148] According to the experimental data, by fitting, we get M(t) = 470 - 0.4·E·t,

[0149] where: M0 = 470, k2 = 0.4 (MPa / (g / m 3 ·day).

[0150] 4. Coupling relationship

[0151] Assume that there is a linear coupling relationship between the apparent damage feature A and the mechanical property feature M:

[0152] M(t) = M0′ - αA(t),

[0153] According to the experimental data, the fitting result is M(t) = 470 - 8A(t),

[0154] where:

[0155] α = 8 MPa / mm 2 .

[0156] 5. Environmental adaptability growth model

[0157] Combining the first relationship model, the second relationship model and the form of the coupling relationship, an environmental adaptability growth model is constructed. Assume that the environmental adaptability index S(t) is defined as the degradation degree of the tensile strength M(t), taking into account the coupling effect of the apparent damage and the mechanical properties.

[0158] Substitute the first relationship model into the coupling relationship:

[0159] S(t) = M(t) = 470 - 8A(t) = 470 - 8(0.05·E·t) = 470 - 0.4·E·t,

[0160] General expression (including environmental stress E):

[0161] S(t) = 470 - 0.4·E·t,

[0162] Simplified expression (when E = 5 g / m3):

[0163] S(t) = 470 - 2·t.

[0164] 6. Model verification

[0165] Use the experimental data to verify the accuracy of the model. For example, when t = 20 days and E = 5 g / m 3 .

[0166] 1. Corrosion area: A = 0.05·5·20 = 5.0 mm 2 (Consistent with the experimental data)

[0167] 2. Tensile strength: M = 470 - 0.4·5·20 = 430 MPa (Consistent with the experimental data)

[0168] 3. Environmental adaptability index: S = 430 MPa (Directly reflecting the strength degradation)

[0169] Through the method proposed by the present invention, core features can be systematically extracted. For example, through image processing and mechanical testing techniques, surface damage features (corrosion area, corrosion degree) and mechanical property features (such as strength, toughness) can be systematically extracted; an environmental adaptability growth model is established. For example, based on long-term environmental exposure data, a mathematical model describing the variation of apparent damage features and mechanical property features with environmental stress is constructed; moreover, the method of the present invention can perform real-time dynamic monitoring and prediction. Through dynamic monitoring and model prediction, the environmental adaptability performance of the product under different environmental conditions can be evaluated, providing a scientific basis for product design, maintenance, and life prediction.

[0170] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the purpose and scope of the present technical solution, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A method for constructing an environmental adaptability growth model, characterized in that: include: Step 1: Extract the apparent damage characteristics and mechanical properties characteristics of industrial products; Step 2: Establish the first relationship model between apparent damage characteristics and environmental stress; Step 3: Establish a second relationship model between mechanical performance characteristics and environmental stress; Step 4: Establish the coupling relationship between the apparent damage characteristics and the mechanical properties characteristics. The coupling relationship characterizes the interaction between the apparent damage characteristics and the mechanical properties characteristics. Step 5: Determine the environmental adaptability growth model of industrial products under natural environmental conditions through the first relationship model, the second relationship model and the coupling relationship.

2. The method for constructing an environmental adaptability growth model according to claim 1, characterized in that: Apparent damage feature extraction includes: Take regular photos of industrial products at multiple angles under different lighting conditions to obtain images of industrial products; Preprocess the image; Perform damage identification on the preprocessed image, where the damage categories include: one or more combinations of corrosion, cracks, coating peeling, etc.; For corrosion damage, the apparent damage characteristics include corrosion area, corrosion depth, corrosion type and corrosion degree; For crack damage, the apparent damage characteristics include crack length, crack width, and crack distribution density; For coating spalling damage, the apparent damage characteristics include the spalling area and spalling ratio of the coating spalling area.

3. The method for constructing an environmental adaptability growth model according to claim 1, characterized in that: The extraction of mechanical properties characteristics includes: Strength features, toughness features, hardness features and fatigue features are extracted, wherein the strength features include extraction through tensile strength test, extraction through compression strength test and extraction through bending strength test; the toughness features include extraction through impact toughness test and extraction through fracture toughness test; the hardness features are extracted through hardness test; the fatigue features include extraction through fatigue life test and extraction through fatigue crack extension test.

4. The method for constructing an environmental adaptability growth model according to claim 3, characterized in that: Mechanical performance characteristics also include parameters such as creep limit and creep rate extracted through creep performance testing, and elastic modulus extracted through elastic modulus testing.

5. The method for constructing an environmental adaptability growth model according to claim 1, characterized in that: The first relationship model includes a linear model, an exponential model, a power-law model, a logarithmic model, a piecewise model, a model based on a physical mechanism, and the like.

6. The method for constructing an environmental adaptability growth model according to claim 1, characterized in that: The second relationship model includes a linear degradation model, an exponential degradation model, a power-law degradation model, a logarithmic degradation model, a piecewise degradation model, a model based on a physical mechanism, and the like.

7. The method for constructing an environmental adaptability growth model according to claim 1, characterized in that: Environmental adaptability growth models include linear growth models, exponential growth models, power-law growth models, logarithmic growth models, etc.

Images