Orthotropic steel bridge deck u rib fatigue performance evaluation method and system

Abstract

The application provides a kind of orthotropic steel bridge deck U rib fatigue performance evaluation method and system, related to U rib fatigue performance evaluation technical field, including: constructing U rib model, setting load and environmental parameter working condition combination, simulating residual stress evolution, screening out main control parameter;Residual stress variation coefficient theoretical equation is constructed, and the final formula is obtained after processing;Combination main control parameter, simulate the welding process of no cyclic load, extract the initial value of characteristic parameter;Find the most unfavorable combination, simulate the stress evolution under its action, extract the final value and compare with the initial value to get the change, generate fatigue comprehensive index, compared with the preset threshold to evaluate the fatigue performance of U rib.The application discloses the core mechanism that residual stress, load and environmental factors change the stress state of U rib together, provides a solid theoretical support for U rib fatigue performance evaluation, accurately quantifies the influence of main control parameter on U rib fatigue performance, significantly improves the evaluation result of fatigue performance.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of U rib fatigue performance evaluation, in particular to a fatigue performance evaluation method and system for orthotropic steel bridge deck U rib. BACKGROUND

[0002] Orthotropic steel decks (OSD) are widely used in steel bridges due to their light weight, high strength, ease of construction and manufacturing. Among various fatigue diseases of OSD, the fatigue cracks in the weld between the deck and the U rib occur most frequently. Considering that the residual stress of the OSD deck-U rib weld is close to or even exceeds the yield limit of the material, the stress generated by external load will superimpose on the residual stress inside the component in this case. This superposition effect will change the stress distribution state inside the component. During the service process of OSD, the wheel load borne by the OSD has strong randomness in terms of operating position and load amplitude, so the superposition effect of residual stress and external load is also very strong. It is necessary to reveal the interaction mechanism of residual stress and external load and accurately grasp the local stress state of the weld under the superposition of residual stress and external load.

[0003] In the prior art, a steel bridge deck U rib inner welding reinforcement fatigue performance evaluation method provided in CN118133606A includes the following steps: obtaining the real bridge fatigue load spectrum, and using finite element crack propagation simulation on the longitudinal rib and top plate structural details to obtain the stress history of each important fatigue failure mode; using rain flow counting method or water leakage method to process the stress history to obtain the fatigue damage accumulation of each important fatigue failure mode; determining the design parameters of the longitudinal rib and top plate structural details, designing and carrying out steel bridge deck fatigue test according to the stress characteristics of fatigue vulnerable details to obtain the actual fatigue resistance of the structure system under each important fatigue failure model; carrying out welding process test on the structural details after U rib inner welding reinforcement, using metallographic analysis to establish a defect deterioration effect theoretical analysis model, combining the fatigue damage accumulation of each important fatigue failure mode and the actual fatigue resistance of the structure system to quantify the fatigue resistance deterioration effect of micro-cracks on each important fatigue failure mode after U rib inner welding reinforcement; unifying the evaluation standards of finite element calculation, fatigue test and welding process test, taking the cumulative damage degree as an index to evaluate the fatigue resistance and residual life of the structure system after U rib inner welding reinforcement.

[0004] However, there are still the following shortcomings: on the one hand, the prior art only studies the influence of fatigue load or welding process on the fatigue of U ribs, without considering the coupling effect of residual stress, load amplitude and environmental temperature (such as the relaxation effect of cyclic load on residual stress under high temperature environment), which leads to the inability to quantify the stress redistribution mechanism under the synergistic action of the three, resulting in deviation of the evaluation model from the actual service state; on the other hand, the prior art relies on single finite element simulation or test data, lacks iterative optimization of the theoretical model based on data-driven, and is difficult to accurately quantify the nonlinear influence of load cycle number, temperature and other parameters on residual stress decay, resulting in inaccurate evaluation results.

[0005] The above information disclosed in the background section is only for the purpose of enhancing the understanding of the background of the present disclosure, and therefore it can include information that does not constitute the prior art known to those of ordinary skill in the art. SUMMARY

[0006] The purpose of the present application is to provide a fatigue performance evaluation method and system for orthotropic steel bridge deck U ribs to solve the problems raised in the background art.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0008] A fatigue performance evaluation method for orthotropic steel bridge deck U ribs, comprising the following specific steps:

[0009] S1. Construct a U rib model, set the load and environmental parameter range and gradient to form a working condition combination, simulate the residual stress evolution under each working condition, record the simulation data at each time for the weld area, extract the load, environmental parameter matrix and residual stress decay rate, and select the load cycle number, cyclic load amplitude and environmental temperature as the main control parameters through the Pearson coefficient;

[0010] S2. Fit the main control parameters to construct a residual stress change coefficient theoretical equation, divide the training set and the test set, use the training set to fit the equation parameters to obtain a preliminary empirical formula, and iterate and optimize the final formula through the test set;

[0011] S3. Combine the main control parameters to form a main control parameter combination, input the final formula to calculate the residual stress change coefficient, simulate the U rib welding process without cyclic load through a thermoelastic-plastic finite element model, and extract the initial value of the residual stress characteristic parameter of the weld;

[0012] S4. Traverse the parameter combination to find the most unfavorable combination corresponding to the maximum value of the residual stress change coefficient, input the thermoelastic-plastic finite element model to simulate the residual stress evolution of the weld under the action of cyclic load and environmental temperature, extract the final value of the characteristic parameter, compare it with the initial value to obtain the change amount, process the change amount and the residual stress change coefficient corresponding to the most unfavorable combination to generate a fatigue comprehensive index, and compare it with a preset threshold to evaluate the fatigue performance state of the U rib.

[0013] Further, the simulation data includes stress peak value, stress gradient, residual stress evolution curve, temperature field distribution data; the load parameter includes load cycle number, cycle load amplitude, stress ratio; the environmental parameter includes environmental temperature, environmental humidity, environmental pH value; the characteristic parameter includes the maximum residual stress value of the weld, the central point residual stress, the average residual stress.

[0014] Further, the specific steps of S2 are as follows:

[0015] It is determined that the load cycle number, the cycle load amplitude and the environmental temperature are input variables, and the residual stress variation coefficient is an output variable, to form one-to-one sample data, and the sample data is , wherein, The load cycle number, the cycle load amplitude, the environmental temperature and the residual stress variation coefficient of the i-th sample are respectively , wherein, is the index of the sample, is the number of samples;

[0016] The sample data set is divided into a training set and a test set according to a ratio of 7:3;

[0017] Define a basic function library: including addition, subtraction, multiplication, division, exponential, logarithm, power function, trigonometric function and variable;

[0018] A particle swarm optimization algorithm is adopted to randomly combine the basic functions and variables to generate a large number of candidate equations;

[0019] The fitness is used to measure the fitting degree of the candidate equation and the training set data, wherein the higher the fitness, the smaller the fitting error;

[0020] The candidate equations with high fitness are retained, and the low fitness equations are eliminated;

[0021] The retained candidate equations are subjected to function item crossing and random mutation to generate a new generation of candidate equations, and the iteration is stopped when a preset iteration number is reached;

[0022] The data in the test set is input into the candidate equation after selection, crossing and mutation to calculate the prediction error, and the mean square error is used as the loss function. For multiple candidate equations, the equation with the smallest test set error is selected as the final formula of the residual stress variation coefficient;

[0023] The final formula of the residual stress variation coefficient is:

[0024]

[0025] , wherein, is the residual stress variation coefficient;

[0026] Where, 、 、 are the number of load cycles, cyclic load amplitude, and ambient temperature, respectively. 、 、 are the weight coefficients of load cycle number, cyclic load amplitude and ambient temperature respectively. On this basis, .

[0027] Furthermore, the specific steps of S3 are as follows:

[0028] Combine the main control parameters to form a main control parameter combination, and build a main control parameter data set based on the main control parameter combination ,in, Respectively The number of load cycles, cyclic load amplitude, and ambient temperature of the main control parameter combination are is the index of the main control parameter combination, is the number of master control parameter combinations;

[0029] For each combination of main control parameters, the thermoelastic finite element model is used to simulate the U-rib welding process without cyclic load and calculate the The initial residual stress distribution of the weld corresponding to the main control parameter combination ,in Weld Axis coordinates, Axis coordinates, axis coordinates;

[0030] Extract the initial values ​​of the maximum residual stress value, center point residual stress, and average residual stress of the weld from the initial residual stress distribution;

[0031]

[0032] in, For the The initial value of the maximum residual stress of the weld corresponding to the main control parameter combination;

[0033]

[0034] in, For the The initial value of residual stress at the center of the weld corresponding to the main control parameter combination is: The center point of the weld Axis coordinates, Axis coordinates, axis coordinates;

[0035]

[0036] wherein, is the initial value of the average residual stress of the weld corresponding to the th master control parameter combination, is the total volume of the weld.

[0037] Further, the change amount of the characteristic parameter is calculated by comparing the final value with the initial value of the characteristic parameter, and the formula is as follows:

[0038]

[0039]

[0040]

[0041] wherein, , , are the change amounts of the maximum residual stress, the center point residual stress, and the average residual stress of the weld respectively, is the final value of the maximum residual stress of the weld corresponding to the most unfavorable combination, is the final value of the center point residual stress of the weld corresponding to the most unfavorable combination, is the final value of the average residual stress of the weld corresponding to the most unfavorable combination, is the initial value of the maximum residual stress of the weld corresponding to the most unfavorable combination, is the initial value of the center point residual stress of the weld corresponding to the most unfavorable combination, is the initial value of the average residual stress of the weld corresponding to the most unfavorable combination;

[0042] The change amount of the characteristic parameter and the residual stress change coefficient corresponding to the most unfavorable combination are data processed to generate a fatigue comprehensive index, and the formula is as follows:

[0043]

[0044]

[0045]

[0046]

[0047] wherein, , , are the first fatigue comprehensive index, the second fatigue comprehensive index, and the third fatigue comprehensive index respectively, is the residual stress change coefficient corresponding to the most unfavorable combination, that is, the maximum value of the residual stress change coefficient, a residual stress change coefficient corresponding to the most unfavorable combination.

[0048] Further, the fatigue comprehensive index is compared with the preset threshold value to evaluate the fatigue performance state of the U rib, and the specific steps are as follows:

[0049] When all the fatigue comprehensive indexes are less than or equal to the preset threshold value, that is, and and the fatigue performance of the U rib is excellent.

[0050] When at least one fatigue comprehensive index is greater than the preset threshold value, that is, or or the fatigue performance of the U rib is poor.

[0051] wherein, , , are the threshold values of the first fatigue comprehensive index, the second fatigue comprehensive index and the third fatigue comprehensive index respectively.

[0052] A fatigue performance evaluation system of a U rib of an orthotropic steel bridge deck panel, the system being used to perform the fatigue performance evaluation method of the orthotropic steel bridge deck panel U rib described in any of the above, comprising:

[0053] a simulation module, configured to construct a U rib model, set a load and an environmental parameter range and a gradient to form a working condition combination, simulate residual stress evolution under each working condition, record simulation data at each time of a weld joint, extract a load, an environmental parameter matrix and a residual stress decay rate, and screen out a load cycle number, a cyclic load amplitude and an environmental temperature as main control parameters through a Pearson coefficient;

[0054] a fitting module, configured to fit the main control parameters to construct a residual stress change coefficient theoretical equation, divide a training set and a test set, obtain a preliminary empirical formula through the training set fitting equation parameters, and obtain a final formula through test set iteration optimization;

[0055] a feature extraction module, configured to combine the main control parameters to form a main control parameter combination, input the final formula to calculate a residual stress change coefficient, simulate a U rib welding process without a cyclic load through a thermoelastic-plastic finite element model, and extract initial values of residual stress characteristic parameters of a weld joint;

[0056] a data evaluation module, configured to traverse the parameter combination to find a most unfavorable combination corresponding to a maximum residual stress change coefficient, input the thermoelastic-plastic finite element model to simulate weld joint residual stress evolution under the action of a cyclic load and an environmental temperature, extract final values of characteristic parameters, compare the final values with the initial values to obtain a change amount, process the change amount and a residual stress change coefficient corresponding to the most unfavorable combination to generate a fatigue comprehensive index, and compare the fatigue comprehensive index with a preset threshold value to evaluate the fatigue performance state of the U rib.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] The present invention constructs a U-rib model, simulates the residual stress evolution process with load and environmental parameters as variables, records the simulation data of the weld at each moment, extracts the parameter matrix and residual stress decay rate of the working condition combination, selects the main control parameters through the Pearson coefficient, fits the main control parameters, constructs a theoretical equation for the residual stress variation coefficient, and obtains its final formula after optimization. This reveals the core mechanism by which residual stress, load and environmental factors synergistically change the stress state of the U-rib, overcomes the defect of the existing technical evaluation lacking a theoretical basis, and provides a solid theoretical support for the fatigue performance evaluation of the U-rib.

[0059] By inputting the main control parameter combination into the final formula of the residual stress variation coefficient, the residual stress variation coefficient is calculated, the U-rib welding is simulated through the fatigue damage model, the initial value of the characteristic parameter of the weld residual stress is extracted, the main control parameter combination is traversed, and the main control parameter combination corresponding to the maximum value of the residual stress variation coefficient is found. The change of the characteristic parameter is obtained, and the change of the characteristic parameter and the residual stress variation coefficient corresponding to the most unfavorable combination are processed to generate a fatigue comprehensive index. The fatigue comprehensive index is compared with the preset threshold to evaluate the fatigue performance status of the U-rib. The deep integration of theoretical derivation and data mining is achieved, the influence of the main control parameters on the fatigue performance of the U-rib is comprehensively and accurately quantified, and the fatigue performance evaluation results are significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 Schematic diagram of the overall method flow of the present invention;

[0061] Figure 2 It is a block diagram of the module composition of the present invention;

[0062] Figure 3 Schematic diagram of the fitting of the load cycle number and the residual stress variation coefficient of the present invention;

[0063] Figure 4 Schematic diagram of the fitting of the cyclic load amplitude and the residual stress variation coefficient of the present invention;

[0064] Figure 5 Schematic diagram of the fitting of the ambient temperature and the residual stress variation coefficient of the present invention. DETAILED DESCRIPTION

[0065] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0066] It should be noted that, unless otherwise defined, technical terms or scientific terms used in the present application shall have the common meaning understood by one of ordinary skill in the art to which the present application pertains. The terms "first", "second", and similar terms used in the present application do not indicate any order, number, or importance, but are only used to distinguish different components. The terms "include", "contain", and similar terms mean that the elements or objects before the terms encompass the elements or objects listed after the terms and their equivalents, and do not exclude other elements or objects. The terms "connect" or "connected" and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. The terms "upper", "lower", "left", "right", and the like are only used to indicate relative positional relationships, which can change accordingly when the absolute positions of the described objects change.

[0067] Embodiment 1:

[0068] Referring to Figures 1 to 5 The present application provides a technical solution:

[0069] A fatigue performance evaluation method for orthotropic steel bridge deck U-rib, the specific steps comprising:

[0070] S1. Construct a U-rib model based on a cyclic plasticity constitutive relation, set load and environmental parameter ranges and gradients to form a working condition combination, simulate residual stress evolution under each working condition, record simulation data at each time for the weld area, extract load, environmental parameter matrix, and residual stress decay rate, and select load cycle number, cyclic load amplitude, and environmental temperature as the main control parameters through a Pearson coefficient;

[0071] On the basis of the above embodiment, the simulation data includes stress peak value, stress gradient, residual stress evolution curve, and temperature field distribution data; the load parameters include load cycle number, cyclic load amplitude, and stress ratio; and the environmental parameters include environmental temperature, environmental humidity, and environmental pH value.

[0072] On this basis, it should be noted that:

[0073] The simulation data, load parameters, and environmental parameters are all normalized, and subsequent data processing and analysis are all based on normalized data.

[0074] On the basis of the above embodiment, the U-rib model is constructed based on a cyclic plasticity constitutive relation, and the specific steps are as follows:

[0075] According to the actual engineering size of the orthotropic steel bridge deck U-rib, the cross-sectional parameters of the U-rib are determined, including height, thickness, web slope, cover plate thickness, and weld geometry characteristics, and the weld geometry characteristics include weld leg size and bevel angle.

[0076] The cyclic plastic constitutive theory model suitable for fatigue analysis of steel structure is selected, which needs to contain the parameters of dynamic hardening and isotropic hardening to reflect the plastic accumulation, Bauschinger effect and stress amplitude attenuation characteristics of the material under cyclic load;

[0077] According to the steel grade of U-rib and deck plate, the basic mechanical parameters such as elastic modulus and Poisson's ratio are input; combined with the material cyclic test data, the key parameters in the cyclic plastic constitutive model are determined, including yield strength, cyclic hardening coefficient and saturated plastic strain amplitude;

[0078] The geometric model is meshed, and the mesh accuracy of the weld and stress concentration area is highlighted; according to the actual stress state of the U-rib in the steel deck plate, the constraint boundary conditions are set, including fixing the edge displacement of the deck plate and limiting the rotation of the U-rib end to simulate the connection relationship with the overall structure;

[0079] By applying monotonic tensile load or simple cyclic load, the stress-strain curve of the model is calculated, which is compared with the material test data, and the constitutive parameters are adjusted until the deviation between the simulation results and the test results is within the allowable range.

[0080] On the basis of the above embodiment, the load and environmental parameter range and gradient are set to form a combination of working conditions, the residual stress evolution under each working condition is simulated, and the simulation data at each time of the weld part are recorded. The specific steps are as follows:

[0081] The load type focuses on cyclic load, which is used to simulate vehicle load;

[0082] Load cycle number: according to the bridge design life and daily traffic volume, take times, divided into 5-8 levels according to logarithmic gradient;

[0083] Cyclic load amplitude: reference actual vehicle load standard value (such as 100-500kN), divided into 4-6 levels according to arithmetic gradient (such as 100kN, 200kN…);

[0084] Stress ratio: according to the load distribution characteristics of the bridge deck, take 0.1-0.5, divided into gradient according to 0.1 interval;

[0085] According to the bridge service environment, the key environmental parameters are selected:

[0086] Environmental temperature: covering the extreme temperature difference in the engineering area (such as-30℃-60℃), divided into 8-10 levels according to 10℃ interval;

[0087] Environmental humidity: considering the atmospheric humidity fluctuation (such as 30%-90%), divided into 4 levels according to 20% interval;

[0088] Environmental pH value: simulate the influence of industrial environment or acid rain, take 4.0-8.0, divided into 5 levels according to 1.0 interval;

[0089] Orthogonal test method is used to reduce the test amount, and representative combinations are selected from the gradient of load and environmental parameters to ensure that the water level of each parameter is evenly covered;

[0090] Each group of working conditions contains one group of load parameters (load cycle number + cycle load amplitude + stress ratio) and one group of environmental parameters (environmental temperature + environmental humidity + environmental pH), forming a complete working condition matrix;

[0091] The working condition combination parameters are input into the U rib model, and numerical simulation is carried out for each working condition combination. During the simulation process, the evolution process of the residual stress at the U rib weld position under the combined action of cyclic load and environmental factors is calculated, and the simulation data at each time is recorded in real time:

[0092] Stress peak value and position;

[0093] Stress gradient, i.e. stress change rate along the length direction of the weld;

[0094] Residual stress evolution curve with time;

[0095] Temperature field distribution data, i.e. instantaneous temperature value of the weld and surrounding area.

[0096] On the basis of the above embodiment, the parameter matrix and the residual stress decay rate are extracted from the simulation data at each time of the weld position, and the load cycle number, cycle load amplitude and environmental temperature are selected as the main control parameters through the Pearson coefficient. The specific steps are as follows:

[0097] For each working condition combination, the specific values of load and environmental parameters are sorted from the recorded simulation data to form a structured parameter matrix. Each row of the matrix corresponds to a group of working conditions, and the columns successively contain the load cycle number (such as ), cycle load amplitude (such as 100 kN), stress ratio (such as 0.1), environmental temperature (such as -30℃), environmental humidity (such as 30%), and environmental pH (such as 4.0), to ensure that the parameter values correspond to the working condition combination one by one, and completely reflect the input conditions of the test;

[0098] From the residual stress evolution curve of the weld position, the residual stress value at the initial time (before the application of cyclic load) and the residual stress value at each time are extracted. The decay rate of each working condition is calculated according to the formula "residual stress decay rate = (initial residual stress value - residual stress value at each time) / initial residual stress value x 100%". If there are multiple monitoring points, take the average of the decay rates of each point as the representative decay rate of the working condition, and quantify the decay degree of residual stress under the action of cyclic load and environment;

[0099] The six parameters in the parameter matrix (number of load cycles, cyclic load amplitude, stress ratio, ambient temperature, ambient humidity, and ambient pH) are used as independent variables, and the corresponding residual stress decay rate is used as the dependent variable. The Pearson correlation coefficient r between each independent variable and the dependent variable is calculated. The closer the absolute value of the coefficient r is to 1, the stronger the correlation between the parameter and the residual stress decay; the closer it is to 0, the weaker the correlation.

[0100] Compare the absolute values ​​of the Pearson coefficients of the six parameters, sort them from large to small, and select the top three with absolute values ​​of coefficients ≥ The parameters, namely the number of load cycles, cyclic load amplitude and ambient temperature, are determined as the main control parameters, among which, is the preset Pearson coefficient threshold.

[0101] S2. Fit the main control parameters to construct a theoretical equation for the residual stress variation coefficient, divide the set into training and test sets, use the training set to fit the equation parameters to obtain a preliminary empirical formula, and then iteratively optimize the test set to obtain the final formula;

[0102] Based on the above embodiment, the specific steps of S2 are as follows:

[0103] Clearly take the number of load cycles, cyclic load amplitude, and ambient temperature as input variables, and the residual stress variation coefficient as output variable to form a one-to-one corresponding sample data. The sample data is ,in, Respectively The number of load cycles, cyclic load amplitude, ambient temperature, and residual stress variation coefficient of each sample are is the index of the sample, is the number of samples;

[0104] The sample data set is divided into training set and test set in a ratio of 7:3;

[0105] Define basic function library: including addition, subtraction, multiplication, division, exponent, logarithm, power function, trigonometric function and variables;

[0106] Using the particle swarm optimization algorithm, the basic functions and variables are randomly combined to generate a large number of candidate equations;

[0107] The fitness is used to measure the degree of fit between the candidate equation and the training set data, where a higher fitness means a smaller fitting error;

[0108] Retain candidate equations with high fitness and eliminate equations with low fitness;

[0109] Perform function term crossover and random mutation on the retained candidate equations to generate a new generation of candidate equations, and stop the iteration when the preset number of iterations is reached;

[0110] The data in the test set is input into the selected, crossed and mutated candidate equation to calculate the prediction error, and the mean square error is used as the loss function. For multiple groups of candidate equations, the equation with the minimum test set error is selected as the final formula of the residual stress change coefficient;

[0111] Table 1. Change of residual stress coefficient with load cycle number, cyclic load amplitude and environmental temperature

[0112]

[0113] According to Table 1, the residual stress change coefficient is positively correlated with the load cycle number, the cyclic load amplitude and the environmental temperature, which is specifically shown as follows:

[0114] With the increase of the load cycle number from 200 to 10000, the cyclic load amplitude from 10 to 50 and the environmental temperature from 0℃ to 40℃, the residual stress change coefficient gradually increases from 6 to 56.5, showing a continuous upward trend.

[0115] The synergistic effect of the three parameters makes the residual stress change coefficient significantly increase with the increase of each parameter, which reflects that the more the load cycle number, the larger the load amplitude and the higher the environmental temperature, the more obvious the residual stress attenuation of the U-rib weld, the more serious the material fatigue performance degradation, and the higher the risk of fatigue failure.

[0116] According to Figures 3-5 It can be seen that the black block represents the "residual stress change coefficient" data point, and the black straight line is the fitting line of the corresponding parameters and the residual stress change coefficient, which shows that the three parameters are positively correlated with the residual stress change coefficient. The higher the load cycle number, the cyclic load amplitude and the environmental temperature, the larger the residual stress change coefficient, the more obvious the residual stress attenuation of the U-rib weld, the more serious the material fatigue performance degradation and the higher the risk of fatigue failure.

[0117] On the basis of the above embodiment, the final formula of the residual stress change coefficient is:

[0118]

[0119] wherein, is the residual stress change coefficient, which is used to evaluate the residual stress degradation of the U-rib weld under the action of cyclic load and environment in combination with the load cycle number, the cyclic load amplitude and the environmental temperature. The larger the residual stress change coefficient, the more obvious the residual stress attenuation of the U-rib weld, the more serious the material fatigue performance degradation and the higher the risk of fatigue failure under the corresponding load and environmental conditions;

[0120] wherein, , , respectively are the number of load cycles, the cyclic load amplitude, and the ambient temperature;

[0121] On this basis, it should be noted that:

[0122] The number of load cycles refers to the total number of cyclic loads that the U-rib welds bear during service, and its essence is to accumulate material fatigue damage through repeated stress action. From the physical mechanism, each load cycle will cause a small plastic deformation inside the weld, and with the increase of the number of cycles, this plastic deformation is continuously superimposed, leading to an increase in the dislocation density inside the material, and then triggering the cyclic softening effect, i.e., the material's ability to resist deformation decreases with the increase of the number of cycles, making it difficult to maintain a stable state of the residual stress, showing a continuous release trend. The degree of this decay intensifies with the accumulation of the number of cycles, directly leading to an increase in the residual stress variation coefficient;

[0123] The cyclic load amplitude refers to the maximum change amplitude of the load in each cycle, and its core is to determine the damage depth of a single cycle through the stress amplitude. When the amplitude increases, the alternating stress amplitude borne by the weld site also increases, and when it exceeds the fatigue limit of the material, it will accelerate the initiation and propagation of internal micro-defects. These defects, as stress concentration points, will break the balance of residual stress and promote the redistribution of stress to the defect area, accelerating the relaxation process of residual stress. Therefore, the larger the cyclic load amplitude, the faster the decay rate of residual stress, and the corresponding increase in the residual stress variation coefficient.

[0124] The ambient temperature refers to the temperature value of the U-rib service environment, which affects the stability of residual stress through the coupling of thermal stress and mechanical stress. From the thermal elastic-plastic principle, an increase in temperature will intensify the thermal motion of steel atoms, reduce the yield strength of the material, and weaken its ability to maintain residual stress. At the same time, the thermal stress generated by thermal expansion and contraction and the load stress superimposed will increase the total stress level at the weld site, further promoting dislocation movement and micro-defect propagation, and accelerating the release of residual stress. Therefore, when the ambient temperature rises, the residual stress decay process is intensified, leading to an increase in the residual stress variation coefficient.

[0125] Therefore, the residual stress variation coefficient and the number of load cycles, the cyclic load amplitude, and the ambient temperature are positively correlated.

[0126] From the physical mechanism, the number of load cycles, the cyclic load amplitude, and the ambient temperature have an impact on the residual stress, although there is coupling. The core is to act on the residual stress decay process through their own independent paths: the number of cycles dominates the "cumulative damage", the cyclic load amplitude dominates the "single damage intensity", and the temperature dominates the "material performance weakening degree". The linear function is used to express the relationship between the residual stress variation coefficient and the number of load cycles, the cyclic load amplitude, and the ambient temperature through the weight coefficient 、 、 Quantifying the contribution of each of the three, not only reflects the independent influence of each parameter, but also reflects the total effect of the superposition of multiple factors through the additive form, which is consistent with the logic of the fatigue theory that "damage accumulation has additivity";

[0127] The foregoing selects three linear parameters (load cycle number, cyclic load amplitude, and ambient temperature) that are strongly related to residual stress decay through Pearson correlation, indicating that the basic relationship between the three and the coefficient of change in residual stress is closer to linear correlation.

[0128] Therefore, the functional relationship between the coefficient of change in residual stress and the load cycle number, cyclic load amplitude, and ambient temperature is expressed in the above functional form.

[0129] In the formula, , , are the weight coefficients of the load cycle number, cyclic load amplitude, and ambient temperature, respectively.

[0130] The cyclic load amplitude directly determines the alternating stress amplitude that the weld joint is subjected to, and is the core factor that triggers the initiation and propagation of material micro-defects. When the amplitude exceeds the fatigue limit of the material, a single cycle can cause significant residual stress relaxation, and its influence on residual stress decay has "strength dominance" - even with fewer cycles, high amplitude loads can still quickly exacerbate residual stress deterioration. Therefore, As a coefficient quantifying this "high strength damage", it usually takes the maximum value.

[0131] The load cycle number affects the residual stress decay through cumulative effect, and its effect depends on "number superposition": when the amplitude does not exceed the fatigue limit, a sufficient number of cycles is needed to cause significant residual stress decay; when the amplitude is high, the cumulative effect of the number of cycles is amplified. However, compared with the "single strong impact" of the amplitude, the influence of the number of cycles is more inclined to "gradual accumulation", so its weight coefficient is less than but greater than the influence of ambient temperature.

[0132] Ambient temperature indirectly affects residual stress mainly by weakening material performance, and its effect can only be manifested when coupled with load action - pure temperature change (without load) has weak influence on residual stress. Even under load, the influence of temperature is more of "auxiliary strengthening" rather than "dominant triggering", so its independent contribution to residual stress decay is the lowest, and its weight coefficient is the smallest.

[0133] Therefore, based on , let .

[0134] As an embodiment, The value range is 0.3-0.4, The value range is 0.4-0.6, The value range is 0.1-0.2. The specific value is set by technical personnel according to actual conditions and is not limited here.

[0135] S3. Combine the master control parameters to form a master control parameter combination, input the final formula to calculate the residual stress variation coefficient, simulate the U-rib welding process without cyclic loading using a thermo-elastic-plastic finite element model, and extract the initial value of the weld residual stress characteristic parameter;

[0136] Based on the above embodiment, the specific steps of S3 are as follows:

[0137] Combine the main control parameters to form a main control parameter combination, and construct a main control parameter data set based on the main control parameter combination ,in, Respectively The number of load cycles, cyclic load amplitude, and ambient temperature of the main control parameter combination are is the index of the main control parameter combination, is the number of master control parameter combinations;

[0138] For each combination of main control parameters, the thermoelastic finite element model is used to simulate the U-rib welding process without cyclic loading. The purpose is to eliminate the subsequent load and environmental interference and obtain the initial residual stress distribution of the weld caused by the welding process itself. The welding process will produce inherent residual stress due to uneven cooling from high temperature. This is the "original state" for evaluating fatigue performance. The subsequent cyclic load and ambient temperature changes need to be calculated based on this initial state. For the The initial residual stress distribution of the weld corresponding to the main control parameter combination is: Weld Axis coordinates, Axis coordinates, axis coordinates;

[0139] The initial values ​​of the maximum residual stress, center point residual stress, and average residual stress of the weld are extracted from the initial residual stress distribution. The maximum residual stress reflects the initial stress level at the stress concentration point of the weld, which is the key location where fatigue cracks are prone to initiation. The geometric center of the weld is selected as a representative monitoring point to obtain the center point residual stress. The center point residual stress reflects the average initial state of the core area of ​​the weld. The overall stress mean is calculated by volume integration as the average residual stress. The average residual stress reflects the macroscopic distribution characteristics of the residual stress in the weld.

[0140]

[0141] wherein, is the initial value of the maximum residual stress of the weld corresponding to the nth main control parameter combination;

[0142]

[0143] wherein, is the initial value of the central point residual stress of the weld corresponding to the nth main control parameter combination, are the X-axis coordinate, Y-axis coordinate, Z-axis coordinate of the central point of the weld, respectively;

[0144]

[0145] wherein, is the initial value of the average residual stress of the weld corresponding to the nth main control parameter combination, is the total volume of the weld;

[0146] wherein, the formula is obtained by triple integration , the stress values of each point in the three-dimensional space are accumulated, and the essence is to calculate the "total effect" of the residual stress in the weld volume, and then divide the total volume of the total volume of the weld to obtain the average stress value per unit volume, i.e. the initial value of the average residual stress on the macro level.

[0147] S4. Find the most unfavorable combination corresponding to the maximum residual stress change coefficient by traversing the parameter combination, input the thermal elastoplastic finite element model to simulate the residual stress evolution of the weld under the action of cyclic load and environmental temperature, extract the final value of the characteristic parameter, compare the change with the initial value to obtain the change, process the change with the residual stress change coefficient corresponding to the most unfavorable combination to generate the fatigue comprehensive index, and compare it with the preset threshold to evaluate the fatigue performance state of the U rib.

[0148] On the basis of the above embodiment, the most unfavorable combination corresponding to the maximum residual stress change coefficient is found by traversing the parameter combination, the thermal elastoplastic finite element model is input to simulate the residual stress evolution of the weld under the action of cyclic load and environmental temperature, and the final value of the characteristic parameter is extracted. The specific steps are as follows:

[0149] traverse the main control parameter data set , substitute each combination into the final formula of the residual stress change coefficient, calculate the corresponding , and select the combination with the largest residual stress change coefficient , which is defined as the most unfavorable combination - the combination represents the most significant influence of the load and environmental conditions on the residual stress deterioration;

[0150] ​​​​The most unfavorable combination parameters are input into the thermoelastic-plastic finite element model, and the boundary conditions and loading system are set: the cycle load number, cycle load amplitude, and environmental temperature are , respectively. The dynamic evolution process of the weld residual stress under the coupling action of the two is simulated, and the residual stress distribution at the end of the evolution is recorded , at the end of the cycle;

[0151] From the final residual stress distribution , the maximum residual stress value , the center point residual stress , and the final value of the average residual stress are extracted.

[0152] On the basis of the above embodiment, the final value of the characteristic parameter is compared with the initial value to calculate the variation of the characteristic parameter, and the formula is as follows:

[0153]

[0154]

[0155]

[0156] Among them, , , are the variation of the maximum residual stress, the center point residual stress, and the average residual stress of the weld, respectively, is the final value of the maximum residual stress of the weld corresponding to the most unfavorable combination, is the final value of the center point residual stress of the weld corresponding to the most unfavorable combination, is the final value of the average residual stress of the weld corresponding to the most unfavorable combination, is the initial value of the maximum residual stress of the weld corresponding to the most unfavorable combination, is the initial value of the center point residual stress of the weld corresponding to the most unfavorable combination, is the initial value of the average residual stress of the weld corresponding to the most unfavorable combination;

[0157] The variation of the characteristic parameter and the residual stress variation coefficient corresponding to the most unfavorable combination are data processed to generate a fatigue comprehensive index, and the formula is as follows:

[0158]

[0159]

[0160]

[0161]

[0162] wherein, , , are the first fatigue comprehensive index, the second fatigue comprehensive index, the third fatigue comprehensive index respectively, is the maximum value of the residual stress variation coefficient corresponding to the most unfavorable combination, that is, the maximum value of the residual stress variation coefficient, is the residual stress variation coefficient corresponding to the most unfavorable combination;

[0163] The first fatigue comprehensive index is used to combine the maximum residual stress variation of the weld and the maximum value of the residual stress variation coefficient as two index parameters to evaluate the fatigue damage degree of the stress concentration part of the weld, and the greater the first fatigue comprehensive index, the more significant the residual stress variation at the stress concentration part of the weld under the most unfavorable combination, the more serious the fatigue damage of the part, and the higher the risk of fatigue failure.

[0164] The second fatigue comprehensive index is used to combine the variation of the central point residual stress of the weld and the maximum value of the residual stress variation coefficient as two index parameters to evaluate the fatigue damage degree of the core region of the weld, and the greater the second fatigue comprehensive index, the more obvious the residual stress change in the core region of the weld under the influence of the most unfavorable combination, the more prominent the fatigue performance degradation of the region, and the greater the possibility of fatigue failure.

[0165] The third fatigue comprehensive index is used to combine the variation of the average residual stress of the weld and the maximum value of the residual stress variation coefficient as two index parameters to evaluate the fatigue damage degree of the whole weld, and the greater the third fatigue comprehensive index, the more significant the change in the residual stress distribution of the whole weld under the most unfavorable combination, the more serious the fatigue performance degradation of the whole structure, and the higher the risk of fatigue problem of the whole.

[0166] On this basis, it should be noted that:

[0167] The maximum value of the residual stress variation coefficient corresponding to the most unfavorable combination is taken as the "quantitative benchmark of the most unfavorable condition", and the variation of the characteristic parameter is compared with the benchmark to obtain the fatigue comprehensive index, which is essentially a "relative quantitative value of the actual stress change under the most unfavorable condition";

[0168] If the variation of the characteristic parameter is smaller, it means that the residual stress state of the U-rib weld under extreme conditions is more stable, and the ability to resist external adverse factors is stronger, that is, the fatigue performance of the U-rib is better; otherwise, the fatigue performance of the U-rib is worse;

[0169] Therefore, in the formula of the first fatigue comprehensive index, the maximum residual stress variation of the weld is taken as the numerator, and the maximum value of the residual stress variation coefficient is taken as the denominator.

[0170] In the formula of the second fatigue comprehensive index, the change amount of the center point residual stress of the weld is taken as the numerator, and the maximum value of the residual stress change coefficient is taken as the denominator.

[0171] In the formula of the third fatigue comprehensive index, the change amount of the average residual stress of the weld is taken as the numerator, and the maximum value of the residual stress change coefficient is taken as the denominator.

[0172] On the basis of the above embodiment, the fatigue comprehensive index is compared with the preset threshold value, and the fatigue performance state of the U rib is evaluated, and the specific steps are as follows:

[0173] When all the fatigue comprehensive indexes are less than or equal to the preset threshold value, i.e. and and , the fatigue performance of the U rib is excellent;

[0174] When at least one fatigue comprehensive index is greater than the preset threshold value, i.e. or or , the fatigue performance of the U rib is poor;

[0175] wherein, , , are the threshold values of the first fatigue comprehensive index, the second fatigue comprehensive index, and the third fatigue comprehensive index, respectively;

[0176] For the threshold value of the first fatigue comprehensive index:

[0177] The U rib weld fatigue test is carried out, the maximum residual stress change amount and the corresponding maximum residual stress change coefficient under different cycle numbers are recorded, and the fatigue comprehensive index under each state is calculated; taking “first macroscopic crack” as the failure criterion, the index distribution range at failure is counted, and the lower limit value of the distribution is taken as the threshold value, so as to ensure that the threshold value has a failure warning accuracy rate of more than 95%.

[0178] For the threshold value of the second fatigue comprehensive index:

[0179] When carrying out the U rib weld fatigue test, the residual stress change of the core area (center point) of the weld is focused on - the change amount of the center point residual stress under different cycle numbers is recorded, the second fatigue comprehensive index is calculated in combination with the corresponding maximum residual stress change coefficient, and the distribution characteristics of the second fatigue comprehensive index in all failure cases are counted taking “first macroscopic crack” as the failure criterion. For example, if 85% of the failure cases in the test occur when the second fatigue comprehensive index is greater than 0.65, it indicates that the failure risk significantly increases when the index exceeds 0.65, and therefore the second threshold value can be initially set to 0.65.

[0180] Threshold value of the third fatigue comprehensive index:

[0181] The average residual stress change of the whole focused weld in the test is calculated, the third fatigue comprehensive index in different cycle stages is calculated, and the "first macroscopic crack" is used as the failure criterion. If the statistics find that the third fatigue comprehensive index is greater than 0.75 in 90% of the failure cases, it indicates that the structure has entered a high-risk state when the third fatigue comprehensive index is greater than 0.75, and therefore the third threshold value can be preliminarily set to 0.75.

[0182] Referring to Figure 2 The application also provides a technical solution:

[0183] A fatigue performance evaluation system of the orthotropic steel bridge deck U rib, the system is used to execute the fatigue performance evaluation method of the orthotropic steel bridge deck U rib described above, comprising:

[0184] A simulation module is configured to construct a U rib model, set a load and environmental parameter range and gradient to form a working condition combination, simulate residual stress evolution under each working condition, record simulation data at each time of the weld part, extract a load, environmental parameter matrix and residual stress decay rate, and screen out the load cycle number, cyclic load amplitude and environmental temperature as the main control parameters through the Pearson coefficient;

[0185] A fitting module is configured to fit the main control parameters to construct a residual stress change coefficient theoretical equation, divide a training set and a test set, use the training set to fit the equation parameters to obtain a preliminary empirical formula, and obtain a final formula through iteration optimization of the test set;

[0186] A feature extraction module is configured to combine the main control parameters to form a main control parameter combination, input the final formula to calculate the residual stress change coefficient, simulate the U rib welding process without cyclic load through a thermoelastic-plastic finite element model, and extract initial values of residual stress characteristic parameters of the weld;

[0187] A data evaluation module is configured to traverse the parameter combination to find a most unfavorable combination corresponding to a maximum value of the residual stress change coefficient, input the thermoelastic-plastic finite element model to simulate the residual stress evolution of the weld under the action of cyclic load and environmental temperature, extract final values of the characteristic parameters, compare the final values with the initial values to obtain a change amount, process the change amount and the residual stress change coefficient corresponding to the most unfavorable combination to generate a fatigue comprehensive index, and compare the fatigue comprehensive index with a preset threshold value to evaluate the fatigue performance state of the U rib.

[0188] The above formulas are dimensionless values, the formulas are obtained by collecting a large amount of data to simulate the most real situation, and the preset parameters in the formulas are set by a person skilled in the art according to the actual situation.

[0189] The above-described embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the above-described embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of the examples described in connection with the embodiments disclosed herein can be realized by electronic hardware, or a combination of computer software and electronic hardware. Whether the functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solutions.

[0190] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, and can be located in one place or distributed on multiple network units. Part or all of the units can be selected to achieve the purpose of the embodiments according to actual needs.

[0191] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application.

Claims

1. A method of fatigue performance evaluation of orthotropic steel bridge deck U-rib, characterized in that, The specific steps include: S1. Constructing a U-rib model, setting load and environmental parameter range and gradient forming working condition combination, simulating residual stress evolution under each working condition, recording simulation data at each time of the weld site, extracting load, environmental parameter matrix and residual stress decay rate, selecting load cycle number, cyclic load amplitude and environmental temperature as the main control parameters through the Pearson coefficient; S2. Fitting the main control parameters to construct the residual stress change coefficient theoretical equation, dividing the training set and the test set, using the training set to fit the equation parameters to obtain the preliminary empirical formula, and iterating and optimizing the test set to obtain the final formula; S3. Combining the main control parameters to form the main control parameter combination, inputting the final formula to calculate the residual stress change coefficient, simulating the U-rib welding process without cyclic load through the thermal elastic-plastic finite element model, and extracting the initial value of the weld residual stress characteristic parameter; S4. Traverse the parameter combination to find the most unfavorable combination corresponding to the maximum value of the residual stress change coefficient, input the thermal elastic-plastic finite element model to simulate the weld residual stress evolution under the action of cyclic load and environmental temperature, extract the final value of the characteristic parameter, compare with the initial value to obtain the change, process the change and the residual stress change coefficient corresponding to the most unfavorable combination to generate the fatigue comprehensive index, and compare with the preset threshold to evaluate the U-rib fatigue performance state.

2. The method of fatigue performance evaluation of orthotropic steel bridge deck U- ribs according to claim 1, characterized in that, The simulation data includes stress peak value, stress gradient, residual stress evolution curve, and temperature field distribution data; the load parameters include load cycle number, cyclic load amplitude, and stress ratio; the environmental parameters include environmental temperature, environmental humidity, and environmental pH; the characteristic parameters include the maximum residual stress value of the weld, the central point residual stress, and the average residual stress.

3. The method of fatigue performance evaluation of orthotropic steel bridge deck U- ribs according to claim 2, characterized in that, The specific steps of S2 are as follows: With the load cycle number, the cyclic load amplitude, and the ambient temperature as input variables, and the residual stress variation coefficient as the output variable, a one-to-one corresponding sample data is formed, and the sample data is wherein, the load cycle number, the cyclic load amplitude, the ambient temperature, and the residual stress variation coefficient of the i th sample, respectively, is the index of the sample, is the number of samples. Divide the sample data set into training set and test set in the ratio of 7:3; Define the basic function library: including addition, subtraction, multiplication, division, exponential, logarithm, power function, trigonometric function and variable; Use particle swarm optimization algorithm to randomly combine basic functions and variables to generate candidate equations; Measure the fitting degree of candidate equations and training set data through fitness, where the higher the fitness, the smaller the fitting error; Keep the candidate equations with high fitness and eliminate the low fitness equations; Cross and randomly mutate the retained candidate equations to generate a new generation of candidate equations, and stop iteration when the preset iteration number is reached; Input the data in the test set into the selected, crossed and mutated candidate equations to calculate the prediction error, use mean square error as the loss function, select the equation with the smallest test set error from multiple candidate equations as the final formula of the residual stress change coefficient; The final formula of the residual stress change coefficient is: wherein is the residual stress change coefficient; In the formula, , , are the number of load cycles, the cycle load amplitude, and the ambient temperature, respectively, , , are the weight coefficients of the number of load cycles, the cycle load amplitude, and the ambient temperature, respectively, and , on the basis of which .

4. The method of fatigue performance evaluation of orthotropic steel bridge deck U- ribs according to claim 3, characterized in that, The specific steps of S3 are as follows: The combination master parameters form a master parameter combination, and a master parameter data set is constructed according to the master parameter combination wherein, are the number of load cycles, the cyclic load amplitude and the ambient temperature of the i th master parameter combination, respectively, is the index of the master parameter combination, is the number of master parameter combinations; For each master parameter combination, the U-rib welding process under non-cyclic load is simulated using a thermo-elastoplastic finite element model, and the initial residual stress distribution of the weld corresponding to the first master parameter combination is calculated , wherein x, y, z are the x, y, z axial coordinates of the weld, respectively. Extract the initial values of the maximum residual stress value, the central point residual stress and the average residual stress of the weld from the initial residual stress distribution; wherein, is the initial value of the maximum residual stress of the weld corresponding to the i-th master parameter combination; and is the initial value of the maximum residual stress of the weld corresponding to the i-th master parameter combination; and wherein, is the initial value of the residual stress at the center point of the weld corresponding to the jth main control parameter combination, is the initial value of the residual stress at the center point of the weld corresponding to the jth main control parameter combination, is the initial value of the residual stress at the center point of the weld corresponding to the jth main control parameter combination, is the initial value of the residual stress at the center point of the weld corresponding to the jth main control parameter combination, is the initial value of the residual stress at the center point of the weld corresponding to the jth main control parameter combination, is the initial value of the residual stress at the center point wherein, is the initial value of the average residual stress of the weld corresponding to the i-th master parameter combination, is the initial value of the average residual stress of the weld corresponding to the i-th master parameter combination, is the total volume of the weld.

5. The method of fatigue performance evaluation of orthotropic steel bridge deck U- ribs according to claim 4, characterized in that, Compare the final value and the initial value of the characteristic parameter to calculate the change of the characteristic parameter, and the formula is as follows: wherein, , , are the change amount of the maximum residual stress, the center point residual stress, the average residual stress of the weld respectively, is the final value of the maximum residual stress of the weld corresponding to the most unfavorable combination, is the final value of the center point residual stress of the weld corresponding to the most unfavorable combination, is the final value of the average residual stress of the weld corresponding to the most unfavorable combination, is the initial value of the maximum residual stress of the weld corresponding to the most unfavorable combination, is the initial value of the center point residual stress of the weld corresponding to the most unfavorable combination, is the initial value of the average residual stress of the weld corresponding to the most unfavorable combination; Data process the change of the characteristic parameter and the residual stress change coefficient corresponding to the most unfavorable combination to generate the fatigue comprehensive index, and the formula is as follows: wherein, , , are a first fatigue comprehensive index, a second fatigue comprehensive index, a third fatigue comprehensive index, respectively, is a residual stress variation coefficient corresponding to the most unfavorable combination, i.e. the maximum value of the residual stress variation coefficient, is a residual stress variation coefficient corresponding to the most unfavorable combination.

6. The method of fatigue performance evaluation of orthotropic steel bridge deck U- ribs according to claim 5, characterized in that, Compare the fatigue comprehensive index with the preset threshold to evaluate the fatigue performance state of the U-rib, and the specific steps are as follows: When all the fatigue comprehensive indexes are less than or equal to a preset threshold, i.e. and and the fatigue performance of the U-rib is excellent. When at least one fatigue composite index is greater than a preset threshold, i.e. or or the fatigue performance of the U-rib is poor. wherein, , , are threshold values for the first, second, and third fatigue composite indices, respectively.

7. A system for evaluating fatigue performance of a U-rib of an orthotropic steel bridge deck, the system being used to perform the method for evaluating fatigue performance of a U-rib of an orthotropic steel bridge deck according to any one of claims 1 to 6, characterized in that, ​ The simulation module is configured to build a U-rib model, set a load and an environmental parameter range and gradient to form a working condition combination, simulate residual stress evolution under each working condition, record simulation data of each time at a weld position, extract a load, an environmental parameter matrix and a residual stress decay rate, and screen a load cycle number, a cyclic load amplitude and an environmental temperature as main control parameters through a Pearson coefficient. The fitting module is configured to fit the main control parameters to build a residual stress change coefficient theoretical equation, divide a training set and a test set, obtain a preliminary empirical formula through a training set fitting equation parameter, and obtain a final formula through test set iteration optimization. The feature extraction module is configured to combine the main control parameters to form a main control parameter combination, input the final formula to calculate a residual stress change coefficient, simulate a U-rib welding process without a cyclic load through a thermoelastic-plastic finite element model, and extract an initial value of a weld residual stress characteristic parameter. The data evaluation module is configured to traverse the parameter combination to find a most unfavorable combination corresponding to a maximum residual stress change coefficient, input a thermoelastic-plastic finite element model to simulate weld residual stress evolution under the action of a cyclic load and an environmental temperature, extract a final value of a characteristic parameter, compare the final value with the initial value to obtain a change amount, process the change amount and a residual stress change coefficient corresponding to the most unfavorable combination to generate a fatigue comprehensive index, and compare the fatigue comprehensive index with a preset threshold to evaluate a U-rib fatigue performance state.

Images