Low-temperature fatigue crack propagation life prediction method and system based on steel bridge deck

Abstract

The invention belongs to the technical field of industrial detection, and provides a low-temperature fatigue crack propagation life prediction method and system based on a steel bridge deck, and the method comprises the steps: judging whether the crack direction deflection risk assessment needs to be started or not through the sharp reduction of steel plasticity and the non-uniform analysis of local stress gradient; if yes, the crack direction deflection risk of the steel bridge deck is evaluated by analyzing the grain orientation difference and the strengthening phase of the welding seam and the HAZ area; if the risk is high, a stress gradient gentle area is determined through a weld toe stress gradient cloud picture, a resistance-gradient coupling equation is established according to the minimum resistance principle and in combination with the stress gradient, and a crack deflection path is determined; dividing the crack crossing area according to the crack deflection path, calculating the actual propagation rate of the crack by combining the path proportion and stress gradient distribution of each stress gradient area after division, and correcting the life prediction output result, thereby improving the crack propagation path prediction precision and the fatigue life prediction evaluation precision.

Description

Technical Field

[0001] This invention belongs to the field of industrial testing technology, specifically a method and system for predicting the low-temperature fatigue crack propagation life of steel bridge decks. Background Technology

[0002] In laboratory fatigue crack propagation tests, the stress field of standard specimens (such as CT specimens) is a preset "uniform gradient distribution", and the crack propagation direction is fixed by the specimen geometry and load direction. However, when steel bridge decks are actually in service, the low temperature environment will exacerbate the "non-uniformity of stress gradient in local structural details", causing the crack propagation direction to deflect as the temperature decreases. Existing life prediction models all assume that cracks propagate along a preset direction (such as perpendicular to the load direction), which ultimately causes a huge deviation between the predicted life and the actual life. For example, after the longitudinal ribs of the steel bridge deck are welded to the top plate, there are micro-defects such as "weld toe undercut" and "lack of fusion" at the weld root. In addition, due to the superposition of welding residual stress and bridge deck bending stress, a "local high stress gradient zone" is formed (the stress drops rapidly from 300MPa at the weld toe to 100MPa in the base material area, with a gradient difference of 200MPa / mm). At room temperature: Steel exhibits good plasticity. After cracks initiate from weld toe defects, they propagate stably along the direction perpendicular to the principal stress (i.e., the bridge deck thickness direction), consistent with the propagation direction preset by the laboratory model. However, at low temperatures (e.g., -60℃): Steel plasticity decreases sharply, dislocation movement is hindered, and crack tips struggle to propagate along the principal stress direction through "plastic passivation." At this point, the non-uniformity of the local stress gradient is amplified—if there are minute "grain orientation differences" or "strengthening phase aggregation regions" near the weld toe, the crack will deviate from the principal stress direction, turning towards "subgrain boundaries or weakened regions with gentler stress gradients" (e.g., propagating along the longitudinal rib length direction). This deflection causes the crack to continuously traverse "regions with different stress gradients" during propagation, and its propagation rate (da / dN) fluctuates with the gradient. However, the propagation rate da / dN calculated by the model based on a "uniform stress gradient" is constant, ultimately causing a deviation between the predicted and actual lifespan.

[0003] Therefore, the present invention provides a method and system for predicting the low-temperature fatigue crack propagation life based on steel bridge deck. Summary of the Invention

[0004] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0005] The technical solution adopted by this invention to solve its technical problem is: a method for predicting the low-temperature fatigue crack propagation life based on steel bridge decks, comprising the following steps:

[0006] Step S10: By performing correlation analysis of the sharp decrease in steel plasticity and the non-uniformity of local stress gradient, determine whether it is necessary to initiate a crack direction deflection risk assessment.

[0007] Step S20: If necessary, assess the risk of crack direction deflection in the steel bridge deck by analyzing the grain orientation differences and strengthening phases in the weld and HAZ region.

[0008] Step S30: If the risk of crack deflection in the steel bridge deck is high, the stress gradient gradient cloud map obtained by finite element simulation is used to determine the stress gradient flat zone. Based on the principle of minimum resistance and combined with the stress gradient, a resistance-gradient coupling equation is established, and the crack deflection angle is calculated. Based on the crack deflection angle, the crack path propagation length is analyzed and predicted, and the crack deflection path is determined.

[0009] Step S40: Combining the finite element stress gradient, the crack crossing region determined by the crack deflection path is divided into stress gradient zones. The actual crack propagation rate is calculated based on the path proportion and stress gradient distribution of each divided stress gradient zone. The life prediction output of the original fatigue life prediction model is corrected based on the actual crack propagation rate.

[0010] A further technical solution of the present invention is as follows: the analysis process for the sharp decrease in the plasticity of the steel is as follows:

[0011] Tensile specimens were prepared from the base material and butt weld, and after being kept at each temperature node, they were loaded until fracture. The elongation after fracture δ at each temperature was calculated, and the δ-temperature curve was obtained by fitting. The critical temperature at which the plasticity drops sharply was determined from the temperature curve based on the preset elongation after fracture.

[0012] The process of analyzing the non-uniform local stress gradient is as follows:

[0013] Residual stress tests were performed on the weld specimens to obtain longitudinal and transverse residual stresses. Strain gauges were attached to the weld joint between the longitudinal rib and the top plate of the actual steel bridge deck to collect traffic loads and obtain bending stress through strain-stress conversion.

[0014] A weld model is established, residual stress and bending stress are applied, the stress distribution at the weld toe at different temperatures is extracted, and the stress gradient is calculated.

[0015] As a further technical solution of the present invention, the process of determining whether to initiate a crack direction deflection risk assessment is as follows:

[0016] At the critical temperature where plasticity decreases sharply, the decrease in δ and the increase in stress gradient are obtained;

[0017] If both the decrease in δ and the increase in stress gradient are greater than or equal to the preset threshold, it indicates that the non-uniformity of stress gradient is significant, and a crack direction deflection risk assessment needs to be initiated.

[0018] As a further technical solution of the present invention, the process of analyzing the grain orientation differences and strengthening phases in the weld and HAZ region is as follows:

[0019] Weld and HAZ samples were taken and scanned after pretreatment to obtain grain orientation maps. The proportion of high-angle grain boundaries in the grain orientation maps was statistically analyzed.

[0020] Thin slices were taken from around the weld toe and thinned by double-spray electrolysis to prepare TEM samples. The samples were observed using a transmission electron microscope, and the density of the reinforcing phase particles was statistically analyzed. The reinforcing phase aggregation area was determined based on the particle density, and the distance from the weld toe was measured.

[0021] If the proportion of high-angle grain boundaries and the distance from the strengthening phase aggregation zone to the weld toe are both greater than or equal to the preset threshold, it indicates a high risk of crack direction deflection in the steel bridge deck.

[0022] As a further technical solution of the present invention, the process of determining the stress gradient flattening region by obtaining the weld toe stress gradient cloud map through finite element simulation is as follows:

[0023] Based on the weld model, the mesh around the weld toe is refined. After applying residual stress and bending stress at the critical temperature of sharp plasticity decrease, the stress gradient cloud map of the weld toe is extracted. The stress region where the stress gradient meets the preset requirements is determined according to the stress gradient cloud map of the weld toe, which is the stress gradient flat region.

[0024] A further technical solution of the present invention is as follows: the process of establishing a resistance-gradient coupling equation based on the principle of minimum resistance and combined with finite element stress gradient data, and calculating and predicting the crack deflection angle is as follows:

[0025] Obtain the crack propagation resistance for different crystal orientations in the grain orientation diagram, and construct the resistance-gradient coupling equation by combining the finite element stress gradient:

[0026] Crack deflection angle θ = arctan(gradient of the stress gradient flat zone / gradient of the principal stress direction) × (resistance of the principal stress direction / resistance of the crystal orientation of the stress gradient flat zone), where the principal stress direction is the initial crack propagation direction.

[0027] A further technical solution of the present invention is as follows: the process of determining the crack deflection path by obtaining the crack path propagation length based on crack deflection angle analysis is as follows:

[0028] Using an in-situ fatigue stage of a microscope, cyclic loading was applied to the weld specimen, and the entire process of crack initiation, deflection, propagation to the stress gradient flattening zone was recorded in real time. Multiple crack morphologies were also captured.

[0029] The crack path of each shot is fitted, and the crack deflection angle and crack path propagation length at each stage are calculated to obtain the dynamic curve of deflection angle-propagation length.

[0030] Traverse all data points on the curve, extract the crack propagation length corresponding to the predicted crack deflection angle, and integrate them into a crack propagation sequence.

[0031] Calculate the coefficient of variation of the crack propagation sequence, then average the crack propagation lengths contained in the crack propagation sequence to obtain the predicted crack propagation length under the crack deflection angle. Otherwise, perform crack propagation sequence clustering through the K-means algorithm, and select the average crack propagation length corresponding to the cluster with the highest proportion of data points in the cluster as the predicted crack propagation length.

[0032] The crack deflection path is determined based on the crack deflection direction and the predicted crack propagation length.

[0033] As a further technical solution of the present invention, the result of dividing the stress gradient region of the crack crossing region determined according to the crack deflection path is a high gradient region, a medium gradient region and a low gradient region.

[0034] Fatigue crack propagation tests were conducted on the stress gradient region at the critical temperature of rapid decrease in plasticity to obtain the da / dN-ΔK curve, where da / dN is the crack propagation rate and ΔK is the stress intensity factor amplitude. The Paris formula was fitted to calculate the crack propagation rate in different stress gradient regions.

[0035] Calculate the path percentage of the stress gradient region based on the path in the stress gradient region;

[0036] The actual crack propagation rate is calculated by weighting the crack propagation rate and path proportion in different stress gradient zones.

[0037] A further technical solution of the present invention is as follows: the process of correcting the life prediction output of the original fatigue life prediction model based on the actual crack propagation rate is as follows:

[0038] Substitute the actual extension rate into the original fatigue life prediction model to obtain the corrected life prediction output. Then, calculate the ratio of the corrected life prediction output to the life prediction output of the original fatigue life prediction model to obtain the correction coefficient.

[0039] A low-temperature fatigue crack propagation life prediction system based on steel bridge decks includes the following modules:

[0040] Crack deflection analysis module: By performing correlation analysis of sharp decrease in steel plasticity and non-uniform local stress gradient, it determines whether a crack direction deflection risk assessment needs to be initiated.

[0041] Crack deflection assessment module: If necessary, the risk of crack deflection in the steel bridge deck can be assessed by analyzing the grain orientation differences and strengthening phases in the weld and HAZ region.

[0042] Crack deflection path determination module: If the risk of crack deflection in the steel bridge deck is high, the stress gradient gradient cloud map obtained by finite element simulation is used to determine the stress gradient flat zone. Based on the principle of minimum resistance and combined with the stress gradient, a resistance-gradient coupling equation is established, and the crack deflection angle is calculated. Based on the crack deflection angle, the crack path propagation length is analyzed and predicted to determine the crack deflection path.

[0043] Fatigue life prediction correction module: Combining finite element stress gradient, the stress gradient region is divided into stress gradient zones based on the crack deflection path. The actual crack propagation rate is calculated based on the path proportion and stress gradient distribution of each stress gradient zone. The life prediction output of the original fatigue life prediction model is corrected based on the actual crack propagation rate.

[0044] The beneficial effects of this invention are as follows: This invention mainly focuses on the prediction of low-temperature fatigue crack life of steel bridge deck welds, constructing a closed-loop process of risk assessment, evaluation, path prediction, rate calculation, and life correction: First, the characteristics of the sharp decrease in the plasticity of steel are quantified through low-temperature tensile tests. Combined with X-ray diffraction measured residual stress and finite element simulation, the plasticity and local stress gradient of steel are analyzed. When the plasticity index decreases and the stress gradient increase exceeds the standard, crack deflection risk assessment is initiated. By characterizing the grain orientation and strengthening phase of the weld and HAZ region, the crack deflection risk is determined. When the risk is high, the stress gradient flat area is located based on the finite element weld toe stress gradient cloud map. The deflection angle is calculated according to the principle of minimum resistance and the resistance-gradient coupling equation is established to determine the crack deflection path. High, medium, and low stress gradient areas are divided. Through low-temperature fatigue tests, the Paris formula is fitted and the actual propagation rate is calculated by weighting. Finally, the prediction results of the original fatigue life model are corrected by the actual propagation rate and the correction coefficient is output, which improves the accuracy of crack propagation path prediction and fatigue life prediction and evaluation. Attached Figure Description

[0045] The invention will now be further described with reference to the accompanying drawings.

[0046] Figure 1 This is a flowchart illustrating the steps of a method for predicting the low-temperature fatigue crack propagation life based on a steel bridge deck, as described in an embodiment of the present invention.

[0047] Figure 2 This is a schematic diagram of the logic judgment of a method for predicting the low-temperature fatigue crack propagation life based on a steel bridge deck, as described in an embodiment of the present invention.

[0048] Figure 3This is a flowchart of a low-temperature fatigue crack propagation life prediction system based on a steel bridge deck, as described in an embodiment of the present invention. Detailed Implementation

[0049] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0050] Example 1: Please refer to Figures 1-2 As shown in the embodiment of the present invention, a method for predicting the low-temperature fatigue crack propagation life based on a steel bridge deck includes the following steps:

[0051] Step S10: By performing correlation analysis of the sharp decrease in steel plasticity and the non-uniformity of local stress gradient, determine whether it is necessary to initiate a crack direction deflection risk assessment.

[0052] In step S10, the analysis process for the sharp decrease in the plasticity of steel is as follows:

[0053] Standard tensile specimens were prepared from the base material and butt weld according to GB / T228.1-2021 "Metallic materials - Tensile testing - Part 1: Test at room temperature". After being kept at each temperature node, they were loaded until fracture and the elongation after fracture δ at each temperature was calculated.

[0054] The critical temperature at which plasticity decreases sharply was determined by fitting the δ-temperature curve using Origin software and based on the preset elongation after fracture (50%).

[0055] For example: From Q345qD / Q370qE base material and butt weld (welding process: submerged arc welding, welding wire H08MnA, flux HJ431), standard tensile specimens (diameter 10mm, gauge length 50mm) are processed according to GB / T228.1-2021 "Metallic materials, tensile testing - Part 1: Test at room temperature"; 15 specimens are prepared for each material (5 temperature nodes, 3 parallel specimens for each node);

[0056] An MTSC64.104 electro-hydraulic servo tensile testing machine was used, equipped with a TH-800 low-temperature environmental chamber (temperature control accuracy ±1℃). The data acquisition system had a sampling frequency of 100Hz. The specimens were placed in the environmental chamber and stabilized at -60℃, -40℃, -20℃, 0℃, and room temperature (23±2℃) for 30 min. Tensile loading was applied at a rate of 5 mm / min until fracture. The elongation after fracture δ at each temperature was calculated (e.g., δ=22% for Q345qD weld at room temperature, δ=9% for -60℃). The “δ-temperature” curve was fitted using Origin software. When the δ decrease exceeded 50%, the corresponding temperature was the “critical temperature for a sharp decrease in plasticity” (Q345qD weld -45℃, Q370qE base material -50℃).

[0057] Understandably, the core function of the critical temperature for a sharp decrease in plasticity is to pinpoint the key temperature node for the plasticity deterioration of a material—at low temperatures, materials will transform from tough to brittle, and a "sharp decrease in plasticity" is the core indicator of this transformation (e.g., the plasticity index of Q345qD steel drops sharply at around -60℃). Determining this critical temperature provides the environmental premise for subsequent judgment of the non-uniformity of stress gradient.

[0058] In step S10, the process of analyzing the non-uniformity of local stress gradients is as follows:

[0059] The weld specimens were subjected to residual stress testing in accordance with GB / T 7704-2017 "Non-destructive testing - X-ray stress determination method" to obtain longitudinal and transverse residual stress.

[0060] HBM1-LY11-3 / 350 strain gauges were attached to the weld joints of the longitudinal ribs and top plate of the actual steel bridge deck. Traffic load data were collected, and bending stress was obtained through strain-stress conversion.

[0061] A 1:1 weld model was established, residual stress and bending stress were applied, the stress distribution at the weld toe at different temperatures was extracted, and the stress gradient was calculated.

[0062] For example, an X-ray diffractometer (Brook D8 Discover) was used to test the residual stress of the weld specimen according to GB / T 7704-2017 "Nondestructive Testing - X-ray Stress Measurement Method". The test points were distributed 5 mm on both sides of the weld toe centerline, with one point taken every 0.5 mm. The longitudinal residual stress (210 MPa at the weld toe and 80 MPa in the base metal area) and the transverse residual stress (140 MPa at the weld toe and 60 MPa in the base metal area) were obtained.

[0063] HBM1-LY11-3 / 350 strain gauges were attached to the weld between the longitudinal ribs and the top plate of the actual steel bridge deck. Traffic load data for 3 days (morning peak 7:00-9:00, evening peak 17:00-19:00, and off-peak hours) were collected. The bending stress (static 90MPa, dynamic peak 170MPa) was obtained by strain-stress conversion (elastic modulus of 206GPa at room temperature; elastic modulus of 215GPa at -60℃).

[0064] A 1:1 weld model was created using ABAQUS (element type C3D8R, mesh size: 0.1mm for weld toe and 1mm for base material).

[0065] Apply residual stress (as initial stress) and bending stress (as external load).

[0066] Extract the stress distribution at the weld toe at different temperatures and calculate the stress gradient. For example, the stress at the weld toe (x=0) is 300 MPa, and the stress at x=1 mm is 180 MPa. The stress gradient is (300-180) / 1 mm = 120 MPa / mm. At -60℃, the stress at x=0 is 330 MPa, and the stress at x=1 mm is 110 MPa. The stress gradient is 220 MPa / mm.

[0067] In step S10, the process of determining whether to initiate a crack direction deflection risk assessment is as follows:

[0068] At the critical temperature where plasticity decreases sharply, the decrease in δ and the increase in stress gradient are obtained;

[0069] If both the decrease in δ and the increase in stress gradient are greater than or equal to the preset threshold, it indicates that the non-uniformity of stress gradient is significant, and a crack direction deflection risk assessment needs to be initiated.

[0070] If both the decrease in δ and the increase in stress gradient are less than the preset threshold or the non-uniformity is greater than or equal to the preset threshold, it indicates that the non-uniformity of stress gradient is not significant, and therefore it is not necessary to initiate a crack direction deflection risk assessment.

[0071] Among them, the preset thresholds corresponding to the δ reduction and stress gradient increase are both obtained with reference to the low-temperature fatigue design code for steel structures.

[0072] Step S20: If necessary, assess the risk of crack direction deflection in the steel bridge deck by analyzing the grain orientation differences and strengthening phases in the weld and HAZ region.

[0073] In step S20, the process of analyzing the grain orientation differences and strengthening phases in the weld and HAZ region is as follows:

[0074] Weld and HAZ samples were taken and scanned after pretreatment to obtain grain orientation maps. The proportion of high-angle grain boundaries in the grain orientation maps was statistically analyzed.

[0075] Thin sections were taken from around the weld toe and thinned by double-spray electrolysis to prepare TEM samples. The samples were observed using a Tecnai G2F20TEM (transmission electron microscope). The field of view was randomly selected, the density of the reinforcing phase particles was statistically analyzed, the reinforcing phase aggregation area was determined based on the particle density, and the distance from the weld toe was measured.

[0076] If the proportion of high-angle grain boundaries and the distance from the strengthening phase aggregation region to the weld toe are both greater than or equal to the preset threshold, it indicates a high risk of crack direction deflection in the steel bridge deck.

[0077] If the proportion of high-angle grain boundaries and the distance from the strengthening phase aggregation zone to the weld toe are both less than the preset threshold or are not greater than or equal to the preset threshold, it indicates that the risk of crack direction deflection in the steel bridge deck is low.

[0078] For example, take weld seam and (heat-affected zone) HAZ samples (size 10mm×10mm×5mm), and grind them with sandpaper (400#-2000#), polish them (diamond polishing paste 1μm), and electropolish them (electrolyte: perchloric acid + ethanol = 1:9, voltage 20V).

[0079] Using a ZEISS Sigma 500 SEM (with EBSD accessory), with an accelerating voltage of 20 kV, a step size of 0.1 μm, and a scanning area of ​​5 mm × 3 mm at the weld toe, a grain orientation map was obtained.

[0080] Channel5 software was used to analyze and statistically determine the proportion of high-angle grain boundaries (orientation difference > 15°) (32% in the weld toe HAZ region and 12% in the base material region).

[0081] A 3mm×3mm×0.5mm thin sheet was taken from around the weld toe and thinned by double-spray electrolysis (electrolyte same as EBSD, temperature -20℃) to prepare a TEM sample;

[0082] Using a Tecnai G2F20TEM (accelerating voltage 200kV), five fields of view (each with an area of ​​1μm²) were randomly selected. The particle density of the strengthening phase (NbC, TiN) was statistically analyzed (7×10³ particles / μm² in the weld toe HA region and 2.5×10³ particles / μm² in the base metal region) and their distance from the weld toe (45μm from the weld toe in the HAZ region and 120μm from the weld toe in the base metal region).

[0083] The aggregation zone is determined by statistically analyzing the average density ρ_avg and standard deviation σ of the reinforcing phase particles across all fields of view.

[0084] If the density of the reinforcing phase particles in the field of view is greater than ρ_avg+2σ, it is considered an aggregation region.

[0085] Step S30: If the risk of crack deflection in the steel bridge deck is high, the stress gradient gradient cloud map obtained by finite element simulation is used to determine the stress gradient flat zone. Based on the principle of minimum resistance and combined with the stress gradient, a resistance-gradient coupling equation is established, and the crack deflection angle is calculated. Based on the crack deflection angle, the crack path propagation length is analyzed and predicted, and the crack deflection path is determined.

[0086] In step S30, the process of determining the stress gradient flattening region using the weld toe stress gradient cloud map obtained from finite element simulation is as follows:

[0087] Based on the weld model, the mesh around the weld toe is refined, and after applying residual stress and bending stress at the critical temperature of sharp decrease in plasticity, the stress gradient cloud map of the weld toe is extracted. Based on the stress gradient cloud map of the weld toe, the stress region where the stress gradient meets the preset requirements is determined, which is the stress gradient flat region.

[0088] For example, in the weld model, the mesh of the 5mm×5mm area around the weld toe is refined to 0.05mm, and a stress load (residual stress + bending stress) at -60℃ is applied. The stress gradient cloud map is extracted, and the area with stress gradient <100MPa / mm is found - located in the transition zone between the weld HAZ and the base material (along the longitudinal rib length direction, 200-300μm from the weld toe, 1-2mm in the thickness direction). This area is marked as the stress gradient flat area, and the coordinates of the stress gradient flat area are (x=250μm, y=1.5mm, z=0-5mm).

[0089] In step S30, based on the principle of minimum resistance and combined with finite element stress gradient data, the process of establishing the resistance-gradient coupling equation and calculating the predicted crack deflection angle is as follows:

[0090] Obtain the crack propagation resistance for different crystal orientations in the grain orientation diagram, and construct the resistance-gradient coupling equation by combining the finite element stress gradient:

[0091] Crack deflection angle θ = arctan(gradient of stress gradient in gentle region / gradient of principal stress direction) × (resistance of principal stress direction / resistance of crystal orientation in gentle stress gradient region), where the principal stress direction is the initial crack propagation direction;

[0092] For example, the crack propagation resistances in different crystal orientations are obtained (85 MPa·m^(1 / 2) along the

[110] crystal orientation and 130 MPa·m^(1 / 2) along the

[100] crystal orientation), and combined with finite element stress gradient data (80 MPa / mm in the gentle region and 220 MPa / mm in the principal stress direction), a resistance-gradient coupling equation is established:

[0093] θ = arctan(80 / 220) × (130 / 85) ≈ 15° × 1.53 ≈ 23°, that is, the crack deflects 23° from the direction of the principal stress and propagates along the length of the longitudinal rib;

[0094] In step S30, the crack path propagation length is obtained based on the crack deflection angle analysis, and the process of determining the crack deflection path is as follows:

[0095] Using an in-situ fatigue stage of a scanning electron microscope (SEM) or an optical microscope (OM), cyclic loads (residual stress + bending stress) were applied to the weld specimens. The entire process of crack initiation, deflection, and propagation from the weld toe to the stress gradient flattening zone was recorded in real time, and multiple crack morphologies were captured (one image was taken at 100 load cycles).

[0096] Using Image-Pro image analysis software, the crack path of each shot is fitted, and the crack deflection angle (relative to the principal stress direction) and crack path propagation length at each stage are calculated to obtain a dynamic curve of deflection angle-propagation length.

[0097] For example, the way to fit the crack path for each shot can be: when fitting the crack path with Image-Pro, select piecewise curve fitting (instead of a broken line) - the crack path in the high gradient area (such as 0~4.5mm) is tortuous, so use a 3rd order polynomial fitting; the path in the medium and low gradient area (such as 4.5~15mm) is flat, so use a 2nd order polynomial fitting.

[0098] Traverse all data points on the curve (e.g., 100 points in total, corresponding to crack propagation lengths of 0~15mm), extract the crack propagation lengths corresponding to the predicted crack deflection angles, and integrate them into a crack propagation sequence.

[0099] Calculate the coefficient of variation of the crack propagation sequence. If the coefficient of variation is less than the coefficient of variation threshold, it means that the crack propagation length is stable under the predicted crack deflection angle. Then, the crack propagation length contained in the crack propagation sequence is averaged to obtain the predicted crack propagation length under the crack deflection angle.

[0100] If the coefficient of variation is greater than or equal to the coefficient of variation threshold, it indicates that the crack propagation length is unstable under the predicted crack deflection angle, and cluster analysis is performed.

[0101] The specific process of cluster analysis includes:

[0102] A1. Determine the number of clusters K, which is set according to the physical laws of crack propagation and the predicted crack deflection angle. For example, when the crack deflection angle is 23°, it is usually concentrated in a single interval transitioning from the high stress gradient region to the medium stress gradient region, rather than multiple intervals. Set the number of clusters K=2 (to avoid over-splitting).

[0103] A2 uses the K-means algorithm to perform crack propagation sequence clustering, dividing the data into two classes and calculating the average length within each class and the percentage of data points within each class. For example, Class 1: 3.1, 3.2 mm (40% of the total, average length 3.15 mm); Class 2: 3.5, 3.8, 3.6 mm (60% of the total, average length 3.63 mm).

[0104] A3, select the cluster with the highest percentage of data points within the cluster (Cluster 2, 60%), and the average crack propagation length within the cluster (3.63 mm) is the predicted crack propagation length;

[0105] The crack deflection path is determined based on the crack deflection direction and the predicted crack propagation length.

[0106] Step S40: Combining the finite element stress gradient, the crack crossing region determined by the crack deflection path is divided into stress gradient zones. The actual crack propagation rate is calculated based on the path proportion and stress gradient distribution of each divided stress gradient zone. The life prediction output of the original fatigue life prediction model is corrected based on the actual crack propagation rate.

[0107] In step S40, the stress gradient region is divided into a high gradient region, a medium gradient region, and a low gradient region based on the crack deflection path.

[0108] For example, the high gradient region is 0-4.5 mm with a gradient of 200-220 MPa / mm.

[0109] Medium gradient region: 4.5-10.5mm, gradient 100-200MPa / mm;

[0110] Low gradient region: 10.5-15mm, gradient <100MPa / mm;

[0111] In step S40, the process of calculating the actual crack propagation rate based on the path proportion and stress gradient distribution of each divided stress gradient region is as follows:

[0112] Fatigue crack propagation tests were conducted on the stress gradient region at the critical temperature of rapid decrease in plasticity according to GB / T3075-2008 "Method for controlling axial force in fatigue testing of metallic materials". The da / dN-ΔK (crack propagation rate - stress intensity factor amplitude) curve was obtained, and the Paris formula (da / dN=C(ΔK)^m) was fitted, where C is a constant and m is the propagation exponent. The crack propagation rate in different stress gradient regions was calculated.

[0113] The path percentage of the stress gradient region is calculated based on the path of the stress gradient region. The path percentage of the stress gradient region is calculated as: path length of the stress gradient region / predicted crack propagation length. For example, the path percentages for high gradient region are 30%, 40%, and 30%, respectively, for high gradient region: 0-4.5 mm, medium gradient region: 4.5-10.5 mm, and low gradient region: 10.5-15 mm.

[0114] The actual crack propagation rate is calculated by weighting the crack propagation rate and path proportion in different stress gradient zones, as follows:

[0115] ;

[0116] For example:

[0117] In the high gradient region: C1 = 1.2 × 10⁻¹¹, m1 = 3.2, when ΔK = 30 MPa・m^(1 / 2), da / dN1 = 1.2 × 10⁻¹¹ × 30³.² = 1.8 × 10⁻ 6 mm / cycle;

[0118] In the medium gradient region: when C2 = 0.8 × 10⁻¹¹, m2 = 2.9, and ΔK = 25 MPa・m^(1 / 2), da / dN2 = 0.8 × 10⁻¹¹ × 25². 9 =1.1×10⁻ 6 mm / cycle;

[0119] In the low gradient region: when C3 = 0.5 × 10⁻¹¹, m3 = 2.6, and ΔK = 20 MPa・m^(1 / 2), da / dN3 = 0.5 × 10⁻¹¹ × 20². 6 =0.6×10⁻ 6 mm / cycle;

[0120] ④ Weighted average actual expansion rate da / dN = 0.3 × 1.8 × 10⁻ 6 + 0.4×1.1×10⁻ 6 + 0.3×0.6×10⁻ 6 =1.1×10⁻ 6 mm / cycle;

[0121] In step S40, the process of correcting the life prediction output of the original fatigue life prediction model based on the actual crack propagation rate is as follows:

[0122] Substitute the actual expansion rate into the original fatigue life prediction model to obtain the corrected life prediction output. Calculate the ratio of the corrected life prediction output to the original fatigue life prediction model's life prediction output to obtain the correction coefficient (corrected life / original model life). This coefficient is used to quickly correct the original fatigue life prediction model's life prediction output in subsequent engineering applications.

[0123] For example, the original fatigue life prediction model uses da / dN1=1.8×10⁻ in the high gradient region. 6 Based on the mm / cycle calculation, as the crack propagates from 0.5 mm to 20 mm, the life prediction output N1 of the original fatigue life prediction model is 1.1 × 10⁻¹⁰. 6 In the next cycle, based on the actual expansion rate of 1.1 × 10⁻ 6 The corrected lifetime prediction output N² is 0.7 × 10⁻⁶ based on the mm / cycle calculation. 6 If the cycle continues, the correction factor = N2 / N1 = 0.64;

[0124] The technical solution of this invention is as follows: This invention mainly focuses on the prediction of low-temperature fatigue crack life of steel bridge deck welds, constructing a closed-loop process of risk assessment, evaluation, path prediction, rate calculation, and life correction: First, the characteristics of the sharp decline in steel plasticity are quantified through low-temperature tensile tests. Combined with X-ray diffraction-measured residual stress and finite element simulation, the plasticity and local stress gradient of the steel are analyzed. When the plasticity index decreases and the stress gradient increase exceeds the standard, the crack deflection risk assessment is initiated. By characterizing the grain orientation and strengthening phase of the weld and HAZ region, the crack deflection risk is determined. When the risk is high, the crack deflection risk is determined based on the finite element weld toe stress. Gradient cloud maps are used to locate areas with gentle stress gradients. A resistance-gradient coupling equation is established based on the principle of minimum resistance to calculate the deflection angle and determine the crack deflection path. High (0-4.5mm, 30%), medium (4.5-10.5mm, 40%), and low (10.5-15mm, 30%) gradient zones are defined. Low-temperature fatigue tests are conducted, the Paris formula is fitted, and the actual propagation rate is calculated using weighted averages. Finally, the prediction results of the original fatigue life model are corrected based on the actual propagation rate, and correction coefficients are output. This improves the accuracy of crack propagation path prediction and enhances the precision of fatigue life prediction and assessment.

[0125] Example 2, please refer to Figure 3 As shown in the embodiment of the present invention, a low-temperature fatigue crack propagation life prediction system based on steel bridge deck includes the following modules:

[0126] Crack deflection analysis module: By performing correlation analysis of sharp decrease in steel plasticity and non-uniform local stress gradient, it determines whether a crack direction deflection risk assessment needs to be initiated.

[0127] Crack deflection assessment module: If necessary, the risk of crack deflection in the steel bridge deck can be assessed by analyzing the grain orientation differences and strengthening phases in the weld and HAZ region.

[0128] Crack deflection path determination module: If the risk of crack deflection in the steel bridge deck is high, the stress gradient gradient cloud map obtained by finite element simulation is used to determine the stress gradient flat zone. Based on the principle of minimum resistance and combined with the stress gradient, a resistance-gradient coupling equation is established, and the crack deflection angle is calculated. Based on the crack deflection angle, the crack path propagation length is analyzed and predicted to determine the crack deflection path.

[0129] Fatigue life prediction correction module: Combining finite element stress gradient, the stress gradient region is divided into stress gradient zones based on the crack deflection path. The actual crack propagation rate is calculated based on the path proportion and stress gradient distribution of each stress gradient zone. The life prediction output of the original fatigue life prediction model is corrected based on the actual crack propagation rate.

[0130] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for predicting the low-temperature fatigue crack propagation life of steel bridge decks, characterized in that: Includes the following steps: Step S10: By performing correlation analysis of the sharp decrease in steel plasticity and the non-uniformity of local stress gradient, determine whether it is necessary to initiate a crack direction deflection risk assessment. Step S20: If necessary, assess the risk of crack direction deflection in the steel bridge deck by analyzing the differences in grain orientation and strengthening phases in the weld and HAZ region. Step S30: If the risk of crack deflection in the steel bridge deck is high, the stress gradient gradient cloud map obtained by finite element simulation is used to determine the stress gradient flat zone. Based on the principle of minimum resistance and combined with the stress gradient, a resistance-gradient coupling equation is established, and the crack deflection angle is calculated. Based on the crack deflection angle, the crack path propagation length is analyzed and predicted, and the crack deflection path is determined. Step S40: Combining the finite element stress gradient, the crack crossing region determined by the crack deflection path is divided into stress gradient zones. The actual crack propagation rate is calculated based on the path proportion and stress gradient distribution of each divided stress gradient zone. The life prediction output of the original fatigue life prediction model is corrected based on the actual crack propagation rate.

2. The method for predicting the low-temperature fatigue crack propagation life based on steel bridge decks according to claim 1, characterized in that: The analytical process for the sharp decrease in the plasticity of the steel is as follows: Tensile specimens were prepared from the base material and butt weld, and after being kept at each temperature node, they were loaded until fracture. The elongation after fracture δ at each temperature was calculated, and the δ-temperature curve was obtained by fitting. The critical temperature at which the plasticity drops sharply was determined from the temperature curve based on the preset elongation after fracture. The process of analyzing the non-uniform local stress gradient is as follows: Residual stress tests were performed on the weld specimens to obtain longitudinal and transverse residual stresses. Strain gauges were attached to the weld joint between the longitudinal rib and the top plate of the actual steel bridge deck to collect traffic loads and obtain bending stress through strain-stress conversion. A weld model is established, residual stress and bending stress are applied, the stress distribution at the weld toe at different temperatures is extracted, and the stress gradient is calculated.

3. The method for predicting the low-temperature fatigue crack propagation life based on steel bridge decks according to claim 2, characterized in that: The process for determining whether a crack direction deflection risk assessment needs to be initiated is as follows: At the critical temperature where plasticity decreases sharply, the decrease in δ and the increase in stress gradient are obtained; If both the decrease in δ and the increase in stress gradient are greater than or equal to the preset threshold, it indicates that the non-uniformity of stress gradient is significant, and a crack direction deflection risk assessment needs to be initiated.

4. The method for predicting the low-temperature fatigue crack propagation life based on steel bridge decks according to claim 3, characterized in that: The process of analyzing the grain orientation differences and strengthening phases in the weld and HAZ region is as follows: Weld and HAZ samples were taken and scanned after pretreatment to obtain grain orientation maps. The proportion of high-angle grain boundaries in the grain orientation maps was statistically analyzed. Thin slices were taken from around the weld toe and thinned by double-spray electrolysis to prepare TEM samples. The samples were observed using a transmission electron microscope, and the density of the reinforcing phase particles was statistically analyzed. The reinforcing phase aggregation area was determined based on the particle density, and the distance from the weld toe was measured. If the proportion of high-angle grain boundaries and the distance from the strengthening phase aggregation zone to the weld toe are both greater than or equal to the preset threshold, it indicates a high risk of crack direction deflection in the steel bridge deck.

5. The method for predicting the low-temperature fatigue crack propagation life based on steel bridge decks according to claim 4, characterized in that: The process of determining the stress gradient flattening region using the weld toe stress gradient cloud map obtained through finite element simulation is as follows: Based on the weld model, the mesh around the weld toe is refined. After applying residual stress and bending stress at the critical temperature of sharp plasticity decrease, the stress gradient cloud map of the weld toe is extracted. The stress region where the stress gradient meets the preset requirements is determined according to the stress gradient cloud map of the weld toe, which is the stress gradient flat region.

6. The method for predicting the low-temperature fatigue crack propagation life based on steel bridge decks according to claim 5, characterized in that: The process of establishing the resistance-gradient coupling equation based on the principle of minimum resistance and combined with finite element stress gradient data, and calculating and predicting the crack deflection angle, is as follows: Obtain the crack propagation resistance for different crystal orientations in the grain orientation diagram, and construct the resistance-gradient coupling equation by combining the finite element stress gradient: Crack deflection angle θ = arctan(gradient of the stress gradient flat zone / gradient of the principal stress direction) × (resistance of the principal stress direction / resistance of the crystal orientation of the stress gradient flat zone), where the principal stress direction is the initial crack propagation direction.

7. The method for predicting the low-temperature fatigue crack propagation life based on steel bridge decks according to claim 6, characterized in that: The process of determining the crack deflection path by obtaining the crack path propagation length based on crack deflection angle analysis is as follows: Using an in-situ fatigue stage of a microscope, cyclic loading was applied to the weld specimen, and the entire process of crack initiation, deflection, propagation to the stress gradient flattening zone was recorded in real time. Multiple crack morphologies were also captured. The crack path of each shot is fitted, and the crack deflection angle and crack path propagation length at each stage are calculated to obtain the dynamic curve of deflection angle-propagation length. Traverse all data points on the curve, extract the crack propagation length corresponding to the predicted crack deflection angle, and integrate them into a crack propagation sequence. Calculate the coefficient of variation of the crack propagation sequence, then average the crack propagation lengths contained in the crack propagation sequence to obtain the predicted crack propagation length under the crack deflection angle. Otherwise, perform crack propagation sequence clustering through the K-means algorithm, and select the average crack propagation length corresponding to the cluster with the highest proportion of data points in the cluster as the predicted crack propagation length. The crack deflection path is determined based on the crack deflection direction and the predicted crack propagation length.

8. The method for predicting the low-temperature fatigue crack propagation life based on steel bridge decks according to claim 7, characterized in that: The stress gradient region division of the crack crossing region determined according to the crack deflection path is divided into a high gradient region, a medium gradient region, and a low gradient region. Fatigue crack propagation tests were conducted on the stress gradient region at the critical temperature of rapid decrease in plasticity to obtain the da / dN-ΔK curve, where da / dN is the crack propagation rate and ΔK is the stress intensity factor amplitude. The Paris formula was fitted to calculate the crack propagation rate in different stress gradient regions. Calculate the path percentage of the stress gradient region based on the path within the stress gradient region; The actual crack propagation rate is calculated by weighting the crack propagation rate and path proportion in different stress gradient zones.

9. The method for predicting the low-temperature fatigue crack propagation life based on steel bridge decks according to claim 8, characterized in that: The process of correcting the fatigue life prediction output of the original fatigue life prediction model based on the actual crack propagation rate is as follows: Substitute the actual extension rate into the original fatigue life prediction model to obtain the corrected life prediction output. Then, calculate the ratio of the corrected life prediction output to the life prediction output of the original fatigue life prediction model to obtain the correction coefficient.

10. A low-temperature fatigue crack propagation life prediction system based on steel bridge deck, characterized in that: Includes the following modules: Crack deflection analysis module: By performing correlation analysis of sharp decrease in steel plasticity and non-uniform local stress gradient, it determines whether a crack direction deflection risk assessment needs to be initiated. Crack deflection assessment module: If necessary, the risk of crack deflection in the steel bridge deck can be assessed by analyzing the differences in grain orientation and strengthening phases in the weld and HAZ region. Crack deflection path determination module: If the risk of crack deflection in the steel bridge deck is high, the stress gradient gradient cloud map obtained by finite element simulation is used to determine the stress gradient flat zone. Based on the principle of minimum resistance and combined with the stress gradient, a resistance-gradient coupling equation is established, and the crack deflection angle is calculated. Based on the crack deflection angle, the crack path propagation length is analyzed and predicted to determine the crack deflection path. Fatigue life prediction correction module: Combining finite element stress gradient, the stress gradient region is divided into stress gradient zones based on the crack deflection path. The actual crack propagation rate is calculated based on the path proportion and stress gradient distribution of each stress gradient zone. The life prediction output of the original fatigue life prediction model is corrected based on the actual crack propagation rate.

Images