Method and system for predicting and optimizing time-varying strength degradation of steel structure

Abstract

The invention provides a steel structure time-varying strength degradation prediction and optimization method and system, and belongs to the technical field of structure health monitoring and service life evaluation. The prediction method comprises the following steps: establishing a steel time-varying strength degradation model according to corrosion and fatigue factors influencing steel strength degradation; the strength degradation process is simulated through Monte Carlo simulation, the steel strength degradation value of each year is calculated according to a steel time-varying strength degradation model, and result statistics is conducted; and performing strength degradation law analysis according to a statistical result of Monte Carlo simulation, and performing steel structure time-varying strength degradation prediction according to the strength degradation law. According to the method, the prediction precision is improved, the actual effect is evaluated through quantitative verification of optimization measures, and a scientific basis is provided for engineering practice.

Description

Technical Field

[0001] The present invention relates to the technical field of structural health monitoring and life assessment, and in particular to a method for predicting time-varying strength degradation of steel structures, a safety optimization method and a system. Background Art

[0002] During long-term use, the strength of steel structures will gradually degrade due to factors such as environmental corrosion and fatigue damage, affecting the safety and service life of the structure. Traditional strength degradation analysis methods have the following problems in steel structure strength degradation analysis:

[0003] 1. Analyzing the strength degradation caused by a single factor such as corrosion or fatigue alone cannot accurately reflect the coupled effect of corrosion and fatigue.

[0004] 2. The calculation is based on a deterministic model formula, without considering the uncertainty of the actual degradation process.

[0005] 3. Without quantitative analysis of the importance of influencing factors, it is difficult to guide the formulation of optimization measures.

[0006] 4. There is a lack of quantitative verification of optimization measures, making it impossible to evaluate their actual effects. Summary of the Invention

[0007] To solve the technical problems existing in the prior art, the present invention provides a method and system for predicting and optimizing the time-varying strength degradation of steel structures. A comprehensive model considering the coupled effects of corrosion and fatigue is established, and the uncertainty of steel strength degradation is quantified through Monte Carlo simulation to improve prediction accuracy. The contribution of initial strength, corrosion amount, and fatigue damage to strength degradation is quantified through importance analysis of influencing factors. The actual effect of optimization measures is evaluated through quantitative verification, providing a scientific basis for engineering practice.

[0008] The present invention provides a method for predicting time-varying strength degradation of steel structures, comprising the following steps:

[0009] Based on the corrosion and fatigue factors that affect the strength degradation of steel, a time-varying strength degradation model of steel is established;

[0010] Monte Carlo simulation is used to simulate the strength degradation process. The annual steel strength degradation value is calculated based on the time-varying strength degradation model of steel, and the results are statistically analyzed.

[0011] The strength degradation law is analyzed based on the statistical results of Monte Carlo simulation, and the time-varying strength degradation of steel structures is predicted based on the strength degradation law.

[0012] Preferably, the time-varying strength degradation model of steel is as follows:

[0013] S(t)=S0·f c(t)·f f (t)

[0014] in,

[0015] S(t): steel strength at time t;

[0016] S0: initial strength of steel;

[0017] f c (t): Corrosion factor affecting strength;

[0018] f f (t): Fatigue influence factor on strength.

[0019] Preferably, the initial strength S0 is defined as the initial strength of the steel before use, which obeys the following normal distribution:

[0020]

[0021] in,

[0022] mean;

[0023] Standard deviation.

[0024] Preferably, the corrosion impact factor f c (t) is defined as the loss of steel cross-section due to corrosion:

[0025]

[0026] h0: initial thickness of steel;

[0027] Δd(t): cumulative corrosion amount at time t;

[0028] μ d : Annual average corrosion amount;

[0029] σ d : Discrete degree of corrosion amount;

[0030] δ d (i): The corrosion amount in the i-th year. The corrosion amount in each year is independent and obeys the normal distribution. The corrosion rate is constant without considering environmental changes.

[0031] Preferably, the fatigue influence factor f f (t) is defined as fatigue-induced internal damage to steel:

[0032] f f (t) = exp(-k·D(t))

[0033]

[0034] in,

[0035] k: fatigue damage coefficient;

[0036] D(t): cumulative fatigue damage at time t;

[0037] a is the year in which the steel begins to be used;

[0038] D(i): fatigue damage in year i, fatigue damage is independent every year and obeys log-normal distribution;

[0039] μ D : Annual average fatigue damage value;

[0040] σ D : Discrete degree of fatigue damage.

[0041] Preferably, the time-varying strength degradation model of steel is specifically: Preferably, the Monte Carlo simulation of the strength degradation process specifically includes the following steps:

[0042] Define the parameters of the steel time-varying strength degradation model;

[0043] Generate initial intensity samples;

[0044] Generate corrosion volume samples;

[0045] Generate fatigue damage samples;

[0046] Calculate the cumulative corrosion amount;

[0047] Calculate the accumulated fatigue loss;

[0048] Calculate the annual steel strength degradation value based on the steel time-varying strength degradation model;

[0049] Calculate the mean and quantile, and draw the strength degradation curve.

[0050] The present invention provides a steel structure optimization method, which uses the above-mentioned steel structure time-varying strength degradation prediction method to predict the strength degradation of the steel structure and optimizes the steel structure based on the prediction results, specifically comprising:

[0051] Based on the Monte Carlo simulation results, the standardized regression coefficient method is used to quantify the contribution of each influencing factor to strength degradation;

[0052] According to the contribution of each influencing factor to strength degradation, corresponding measures are taken to optimize the steel structure.

[0053] Preferably, quantifying the contribution of each influencing factor to strength degradation by using the standardized regression coefficient method based on the Monte Carlo simulation results specifically includes the following steps:

[0054] Extract Monte Carlo simulation results;

[0055] Perform preset standardization on simulation results;

[0056] Construct a multiple linear regression model;

[0057] Calculate standardized regression coefficients;

[0058] Sort the influencing factors according to their regression coefficients;

[0059] Draw a comparison chart of the importance of each influencing factor.

[0060] Preferably, the initial strength, corrosion amount, fatigue damage and ultimate strength are normalized as follows:

[0061]

[0062] X: original data;

[0063] μ x : mean;

[0064] σ x :variance;

[0065] X std : Data after standardization.

[0066] Preferably, after standardization, a multiple linear regression model is established as follows:

[0067] S std (T) = β0 + β1S 0,std +β2Δd std (T)+β3D std (T)+ε

[0068] S std (T): final intensity after normalization;

[0069] S 0,std : Normalized initial strength;

[0070] Δd std (T): normalized cumulative corrosion amount;

[0071] D std (T): normalized cumulative fatigue damage;

[0072] β0: intercept term;

[0073] β1,β2,β3: standardized regression coefficients;

[0074] ε: error term.

[0075] Preferably, the least squares method is used to fit the regression model to obtain standardized regression coefficients β1, β2, and β3. The factors are ranked according to the absolute values of the standardized regression coefficients. The larger the absolute value of the regression coefficient, the greater the impact of the corresponding influencing factor on strength degradation.

[0076] Preferably, Monte Carlo simulation is used to quantitatively analyze the effects of optimization measures and provide customized optimization solutions for different engineering scenarios.

[0077] Preferably, the optimization measures include: anti-corrosion coating, structural reinforcement and load testing.

[0078] Preferably, the effects of the optimization measures are quantitatively analyzed through Monte Carlo simulation, specifically including:

[0079] Define optimization measures in Monte Carlo simulations;

[0080] Adjust the parameters of the steel time-varying strength degradation model based on the optimization measures;

[0081] Run Monte Carlo simulations to calculate annual steel strength degradation values based on the steel time-varying strength degradation model;

[0082] Statistical means and quantiles;

[0083] Compare the original results with the results after optimization measures are taken, quantitatively analyze the optimization effect, and draw a comparison chart of the optimization effect.

[0084] The present invention also provides a steel structure optimization system, comprising a processor, wherein the processor is capable of executing the steps of the above-mentioned method for predicting time-varying strength degradation of steel structures.

[0085] Compared with the prior art, the present invention has the following beneficial effects:

[0086] This invention innovatively establishes a comprehensive model that takes into account the coupled effects of corrosion and fatigue. It quantifies the uncertainty of steel strength degradation through Monte Carlo simulation to improve prediction accuracy. It quantifies the contribution of initial strength, corrosion amount and fatigue damage to strength degradation through importance analysis of influencing factors. It also quantitatively verifies optimization measures to evaluate their actual effects, providing a scientific basis for engineering practice. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying any creative work.

[0088] Figure 1This is a flow chart of a method for predicting time-varying strength degradation of steel structures according to an embodiment of the present invention;

[0089] Figure 2 A flow chart of a Monte Carlo simulation of strength degradation process according to an embodiment of the present invention;

[0090] Figure 3 This is a steel strength degradation curve diagram according to an embodiment of the present invention;

[0091] Figure 4 A flow chart of Monte Carlo influence factor importance analysis according to an embodiment of the present invention;

[0092] Figure 5 A histogram of the importance of influencing factors according to an embodiment of the present invention;

[0093] Figure 6 A flowchart of Monte Carlo optimization verification according to an embodiment of the present invention;

[0094] Figure 7 This is a diagram showing the effect of optimization measures of an embodiment of the present invention. DETAILED DESCRIPTION

[0095] The specific embodiments of the present invention are described in detail below.

[0096] like Figure 1 As shown, the present invention provides a method for predicting time-varying strength degradation of steel structures, comprising the following steps:

[0097] Based on the corrosion and fatigue factors that affect the strength degradation of steel, a time-varying strength degradation model of steel is established;

[0098] Monte Carlo simulation is used to simulate the strength degradation process. The annual steel strength degradation value is calculated based on the time-varying strength degradation model of steel, and the results are statistically analyzed.

[0099] The strength degradation law is analyzed based on the statistical results of Monte Carlo simulation, and the time-varying strength degradation of steel structures is predicted based on the strength degradation law.

[0100] According to a specific embodiment of the present invention, the time-varying strength degradation model of steel is as follows:

[0101] S(t)=S0·f c (t)·f f (t)

[0102] in,

[0103] S(t): steel strength at time t;

[0104] S0: initial strength of steel;

[0105] f c (t): Corrosion factor affecting strength;

[0106] f f (t): Fatigue influence factor on strength.

[0107] According to a specific embodiment of the present invention, the initial strength S0 is defined as the initial strength of the steel before use, which obeys the following normal distribution:

[0108]

[0109] in,

[0110] mean;

[0111] Standard deviation.

[0112] According to a specific embodiment of the present invention, the corrosion impact factor f c (t) is defined as the loss of steel cross-section due to corrosion:

[0113]

[0114] h0: initial thickness of steel;

[0115] Δd(t): cumulative corrosion amount at time t;

[0116] μ d : Annual average corrosion amount;

[0117] σ d : Discrete degree of corrosion amount;

[0118] δ d (i): The corrosion amount in the i-th year. The corrosion amount in each year is independent and obeys the normal distribution. The corrosion rate is constant without considering environmental changes.

[0119] According to a specific embodiment of the present invention, the fatigue impact factor f f (t) is defined as fatigue-induced internal damage to steel:

[0120] f f (t) = exp(-k·D(t))

[0121]

[0122] in,

[0123] k: fatigue damage coefficient;

[0124] D(t): cumulative fatigue damage at time t;

[0125] a is the year in which the steel begins to be used;

[0126] D(i): fatigue damage in year i, fatigue damage is independent every year and obeys log-normal distribution;

[0127] μ D : Annual average fatigue damage value;

[0128] σ D : Discrete degree of fatigue damage.

[0129] According to a specific embodiment of the present invention, the time-varying strength degradation model of steel is specifically:

[0130] According to a specific embodiment of the present invention, the Monte Carlo simulation of the strength degradation process specifically includes the following steps:

[0131] Define the parameters of the steel time-varying strength degradation model;

[0132] Generate initial intensity samples;

[0133] Generate corrosion volume samples;

[0134] Generate fatigue damage samples;

[0135] Calculate the cumulative corrosion amount;

[0136] Calculate the accumulated fatigue loss;

[0137] Calculate the annual steel strength degradation value based on the steel time-varying strength degradation model;

[0138] Calculate the mean and quantile, and draw the strength degradation curve.

[0139] The present invention provides a steel structure optimization method, which uses the above-mentioned steel structure time-varying strength degradation prediction method to predict the strength degradation of the steel structure and optimizes the steel structure based on the prediction results, specifically comprising:

[0140] Based on the Monte Carlo simulation results, the standardized regression coefficient method is used to quantify the contribution of each influencing factor to strength degradation;

[0141] According to the contribution of each influencing factor to strength degradation, corresponding measures are taken to optimize the steel structure.

[0142] According to a specific embodiment of the present invention, quantifying the contribution of each influencing factor to strength degradation by using the standardized regression coefficient method based on the Monte Carlo simulation results specifically includes the following steps:

[0143] Extract Monte Carlo simulation results;

[0144] Perform preset standardization on simulation results;

[0145] Construct a multiple linear regression model;

[0146] Calculate standardized regression coefficients;

[0147] Sort the influencing factors according to their regression coefficients;

[0148] Draw a comparison chart of the importance of each influencing factor.

[0149] According to a specific embodiment of the present invention, the initial strength, corrosion amount, fatigue damage and ultimate strength are normalized as follows:

[0150]

[0151] X: original data;

[0152] μ x : mean;

[0153] σ x :variance;

[0154] X std : Data after standardization.

[0155] According to a specific embodiment of the present invention, after standardization, a multiple linear regression model is established as follows:

[0156] S std (T) = β0 + β1S 0,std +β2Δd std (T)+β3D std (T)+ε

[0157] S std (T): final intensity after normalization;

[0158] S 0,std : Normalized initial strength;

[0159] Δd std (T): normalized cumulative corrosion amount;

[0160] D std (T): normalized cumulative fatigue damage;

[0161] β0: intercept term;

[0162] β1,β2,β3: standardized regression coefficients;

[0163] ε: error term.

[0164] According to a specific embodiment of the present invention, the least squares method is used to fit the regression model to obtain standardized regression coefficients β1, β2, and β3. The factors are ranked according to the absolute values of the standardized regression coefficients. The larger the absolute value of the regression coefficient, the greater the influence of the corresponding influencing factor on strength degradation.

[0165] According to a specific embodiment of the present invention, Monte Carlo simulation is used to quantitatively analyze the effects of optimization measures and provide customized optimization solutions for different engineering scenarios.

[0166] According to a specific embodiment of the present invention, the optimization measures include: anti-corrosion coating, structural reinforcement and load detection.

[0167] According to a specific embodiment of the present invention, quantitative analysis of the effects of optimization measures through Monte Carlo simulation specifically includes:

[0168] Define optimization measures in Monte Carlo simulations;

[0169] Adjust the parameters of the steel time-varying strength degradation model based on the optimization measures;

[0170] Run Monte Carlo simulations to calculate annual steel strength degradation values based on the steel time-varying strength degradation model;

[0171] Statistical means and quantiles;

[0172] Compare the original results with the results after optimization measures are taken, quantitatively analyze the optimization effect, and draw a comparison chart of the optimization effect.

[0173] The present invention also provides a steel structure optimization system, comprising a processor, wherein the processor is capable of executing the steps of the above-mentioned method for predicting time-varying strength degradation of steel structures.

[0174] Example 1

[0175] The present invention provides a method for predicting time-varying strength degradation of steel structures, comprising the following steps:

[0176] Based on the corrosion and fatigue factors that affect the strength degradation of steel, a time-varying strength degradation model of steel is established;

[0177] Monte Carlo simulation is used to simulate the strength degradation process. The annual steel strength degradation value is calculated based on the time-varying strength degradation model of steel, and the results are statistically analyzed.

[0178] The strength degradation law is analyzed based on the statistical results of Monte Carlo simulation, and the time-varying strength degradation of steel structures is predicted based on the strength degradation law.

[0179] Example 2

[0180] The present invention provides a method for predicting time-varying strength degradation of steel structures, comprising the following steps:

[0181] Based on the corrosion and fatigue factors that affect the strength degradation of steel, a time-varying strength degradation model of steel is established;

[0182] Monte Carlo simulation is used to simulate the strength degradation process. The annual steel strength degradation value is calculated based on the time-varying strength degradation model of steel, and the results are statistically analyzed.

[0183] The strength degradation law is analyzed based on the statistical results of Monte Carlo simulation, and the time-varying strength degradation of steel structures is predicted based on the strength degradation law.

[0184] In this embodiment, the time-varying strength degradation model of steel is as follows:

[0185] S(t)=S0·f c (t)·f f (t)

[0186] in,

[0187] S(t): steel strength at time t;

[0188] S0: initial strength of steel;

[0189] f c (t): Corrosion factor affecting strength;

[0190] f f (t): Fatigue influence factor on strength.

[0191] In this embodiment, the initial strength S0 is defined as the initial strength of the steel before use, which obeys the following normal distribution:

[0192]

[0193] in,

[0194] mean;

[0195] Standard deviation.

[0196] In other embodiments, the initial strength of the steel may be preset according to actual production conditions, or may be set according to expert experience.

[0197] In this embodiment, the corrosion impact factor f c (t) is defined as the loss of steel cross-section due to corrosion:

[0198]

[0199] h0: initial thickness of steel;

[0200] Δd(t): cumulative corrosion amount at time t;

[0201] μ d : Annual average corrosion amount;

[0202] σ d : Discrete degree of corrosion amount;

[0203] δ d (i): The corrosion amount in year i. The corrosion amount in each year is independent and obeys the normal distribution. The corrosion rate is constant without considering environmental changes.

[0204] In other embodiments, the volume loss or mass loss of steel caused by corrosion may also be defined as the corrosion influencing factor.

[0205] In this embodiment, the fatigue influence factor f f (t) is defined as fatigue-induced internal damage to steel:

[0206] f f (t) = exp(-k·D(t))

[0207]

[0208] in,

[0209] k: fatigue damage coefficient;

[0210] D(t): cumulative fatigue damage at time t;

[0211] a is the year in which the steel begins to be used;

[0212] D(i): fatigue damage in year i, fatigue damage is independent every year and obeys log-normal distribution;

[0213] μ D : Annual average fatigue damage value;

[0214] σ D : Discrete degree of fatigue damage.

[0215] In this embodiment, the time-varying strength degradation model of steel is specifically: In this embodiment, the Monte Carlo simulation of the strength degradation process specifically includes the following steps:

[0216] Define the parameters of the steel time-varying strength degradation model;

[0217] Generate initial intensity samples;

[0218] Generate corrosion volume samples;

[0219] Generate fatigue damage samples;

[0220] Calculate the cumulative corrosion amount;

[0221] Calculate the accumulated fatigue loss;

[0222] Calculate the annual steel strength degradation value based on the steel time-varying strength degradation model;

[0223] Calculate the mean and quantile, and draw the strength degradation curve.

[0224] Example 3

[0225] The present invention provides a method for predicting time-varying strength degradation of steel structures, comprising the following steps:

[0226] 1. Establishment of time-varying strength degradation model for steel structures

[0227] Steel strength degradation may involve multiple random variables, such as time t, annual corrosion depth, fatigue damage caused by the number of load cycles, and variation in the material's initial strength. Traditional analysis methods typically analyze the impact of individual factors on strength degradation under specific environments or operating conditions. This invention establishes a mathematical model that considers the coupled effects of corrosion and fatigue, the primary factors in strength degradation, and incorporates factors such as mean and dispersion into the formula to account for their randomness. This model is unrestricted by environmental or operating conditions, significantly improving the accuracy and applicability of predictions.

[0228] In this embodiment, the process of establishing the time-varying strength degradation model of steel structures is as follows:

[0229] (1) Model Overview

[0230] During the service life of steel structures, corrosion and fatigue are the main factors causing strength degradation. In order to quantify this process, the coupling effect of the two is considered and the following time-varying strength model is established:

[0231] S(t)=S0·f c (t)·f f (t)

[0232] S(t): Steel strength at time t.

[0233] S0: Initial strength.

[0234] f c (t): Corrosion factor affecting strength.

[0235] f f (t): Fatigue influence factor on strength.

[0236] (2) Initial strength S0.

[0237] Definition: The initial strength of unused steel follows a normal distribution, and the initial strengths of different samples are independent of each other.

[0238]

[0239] Mean.

[0240] Standard deviation.

[0241] (3) Corrosion influence factor f c (t).

[0242] Definition: Corrosion causes loss of cross-sectional area in steel, thereby reducing strength.

[0243]

[0244] h0: initial thickness of steel.

[0245] Δd(t): Cumulative corrosion amount at time t.

[0246] μ d : Average annual corrosion amount.

[0247] σ d : Discrete degree of corrosion amount.

[0248] δ d (i): The corrosion amount in year i. The corrosion amount follows a normal distribution independently each year, and the corrosion rate is constant without considering environmental changes.

[0249] (4) Fatigue influence factor f f (t).

[0250] Definition: Fatigue causes internal damage to steel, which reduces its strength.

[0251] f f (t) = exp(-k·D(t))

[0252]

[0253]

[0254] k: fatigue damage coefficient.

[0255] D(t): Cumulative fatigue damage at time t.

[0256] D(i) is the fatigue damage in year i. The fatigue damage is independent for each year and follows a log-normal distribution. The fatigue damage is related to the load history but simplified to a constant rate.

[0257] μ D: Annual average fatigue damage value.

[0258] σ D : Discrete degree of fatigue damage.

[0259] Assumption: obey lognormal distribution.

[0260] (5) Finally, the time-varying strength degradation model of steel structure is obtained.

[0261] By combining the effects of corrosion and fatigue, a time-varying strength degradation model of steel structures is obtained. The meanings and values of the parameters are shown in Table 1.

[0262]

[0263] Table 1. Example of parameter representation for time-varying strength degradation model of steel structure

[0264]

[0265]

[0266] 2. Use Monte Carlo simulation to simulate the strength degradation process and perform statistical analysis of the results;

[0267] Monte Carlo simulation is a numerical method based on random sampling. It simulates the behavior of a system by generating a large number of random samples, thereby evaluating the statistical characteristics of the system. Traditional methods mostly use deterministic models, ignoring the randomness of degradation factors such as corrosion and fatigue. The present invention introduces Monte Carlo simulation. This embodiment uses MATLAB programming to quantify the uncertainty of strength degradation by generating a large number of random samples. Figure 2 , the specific simulation analysis process is as follows:

[0268] (1) Define model parameters;

[0269] The parameter settings are shown in Table 1 Model Parameter Table, where the time span is set to T = 50 years and the number of simulations is set to N = 10,000 times.

[0270] (2) Generate random samples;

[0271] Initial intensity S0: Generate N initial intensity samples.

[0272] Corrosion amount δ d (i): Generate an N×T erosion matrix.

[0273] Fatigue damage D(i): Generates an N×T fatigue damage matrix.

[0274] (3) Calculate the cumulative degradation amount;

[0275] Cumulative corrosion amount:

[0276] Cumulative fatigue damage:

[0277] The annual strength degradation value is calculated based on the total strength model.

[0278] (4) Statistical characteristics analysis;

[0279] Statistical mean:

[0280] Statistical quantile: 5% quantile S 0.05 (t): lower limit of intensity; 95% quantile: S 0.95 (t): Upper limit of intensity.

[0281] 3. Analysis of strength degradation law;

[0282] Monte Carlo simulation results such as steel strength degradation curves are professionally analyzed.

[0283] (1) Overall degradation trend:

[0284] Reference Figure 3 The mean curve (solid line) shows that the strength of steel continues to decline from the initial 400MPa to about 270MPa after 50 years, a decrease of 32.5%. The two dotted lines are the upper and lower limits of strength, respectively.

[0285] (2) Staged degradation characteristics:

[0286] Initial stage (0-10 years): The strength decreases fastest, with an average annual decrease of 2.75 MPa, corresponding to the cumulative damage stage of environmental erosion.

[0287] Medium-term stage (10-30 years): The rate of decline slows down to 2.625MPa / year and enters a stable degradation period.

[0288] Late stage (30-50 years): The average annual decrease is only 2.435MPa, which may reach the critical stable state of material damage.

[0289] Based on the analysis results, steel structure degradation prediction can be performed.

[0290] Example 4

[0291] The present invention provides a steel structure optimization method, which uses any of the above-mentioned steel structure time-varying strength degradation prediction methods to predict the strength degradation of the steel structure and optimizes the steel structure based on the prediction results, specifically comprising:

[0292] Based on the Monte Carlo simulation results, the standardized regression coefficient method is used to quantify the contribution of each influencing factor to strength degradation;

[0293] According to the contribution of each influencing factor to strength degradation, corresponding measures are taken to optimize the steel structure.

[0294] Importance analysis quantifies the contribution of each influencing factor to strength degradation, determining which factors play a dominant role. Traditional methods lack quantitative analysis of the importance of influencing factors. This method uses a standardized regression coefficient method to quantify the contribution of initial strength, corrosion, and fatigue damage to strength degradation. This method standardizes the influencing factors and strength degradation values, calculates regression coefficients, and thus quantifies the impact of each factor.

[0295] In this embodiment, referring to Figure 4 ,According to the Monte Carlo simulation results, the standardized regression coefficient method is used to quantify the contribution of each influencing factor to strength degradation. The specific steps include the following:

[0296] (1) Data extraction. The following data were extracted from the Monte Carlo simulation: initial strength S0, cumulative corrosion amount Δd(T), cumulative fatigue damage D(T), and final strength S(T).

[0297] (2) Standardization. Standardize the initial strength, corrosion, fatigue damage and final strength:

[0298]

[0299] in,

[0300] X: original data.

[0301] μ x : mean.

[0302] σ x :variance.

[0303] X std : Normalized data.

[0304] (3) Multiple linear regression. Establish a standardized multiple linear regression model:

[0305] S std (T) = β0 + β1S 0,std +β2Δd std (T)+β3D std (T)+ε

[0306] S std (T): Final intensity after normalization.

[0307] S 0,std : Normalized initial intensity.

[0308] Δd std (T): Normalized cumulative corrosion amount.

[0309] Dstd (T): Normalized cumulative fatigue damage.

[0310] β0: intercept term.

[0311] β1,β2,β3: standardized regression coefficients.

[0312] ε: error term.

[0313] (4) Calculate the regression coefficient: Use the least squares method to fit the regression model and obtain the standardized regression coefficients β1, β2, and β3. Sort the factors according to the absolute value of the standardized regression coefficient. The larger the absolute value, the greater the influence of the factor on strength degradation. Figure 5 , draw a bar chart of the importance analysis of each factor based on the regression coefficient corresponding to each influencing factor.

[0314] Traditional methods usually propose optimization measures but lack quantitative verification. This invention uses Monte Carlo simulation to quantitatively analyze the effects of three optimization measures and provide customized optimization solutions for different engineering scenarios.

[0315] In this embodiment, the specific steps are as follows:

[0316] (1) Optimization measures and parameter settings.

[0317] Table 2. Optimization measures and parameter setting examples

[0318]

[0319]

[0320] (2) Verification of optimization measures.

[0321] For each optimization measure, rerun the Monte Carlo simulation process to calculate the optimized strength degradation value. Compare the optimized strength degradation curve with the original curve to evaluate the optimization effect. The optimization effect is shown in Figure 7 The analysis of the three optimization measures is shown in Table 3. Figure 6 The specific verification process is as follows:

[0322] Through Monte Carlo simulation, the effects of optimization measures are quantitatively analyzed, including:

[0323] Define optimization measures in Monte Carlo simulations;

[0324] Adjust the parameters of the steel time-varying strength degradation model according to the optimization measures;

[0325] Run Monte Carlo simulations to calculate annual steel strength degradation values based on the steel time-varying strength degradation model;

[0326] Statistical means and quantiles;

[0327] Compare the original results with the results after optimization measures are taken, quantitatively analyze the optimization effect, and draw a comparison chart of the optimization effect.

[0328] Table 3. Optimization measures effect analysis table

[0329]

[0330] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A method for predicting time-varying strength degradation of steel structures, characterized in that: The steps include: Based on the corrosion and fatigue factors that affect the strength degradation of steel, a time-varying strength degradation model of steel is established; Monte Carlo simulation is used to simulate the strength degradation process. The annual steel strength degradation value is calculated based on the time-varying strength degradation model of steel, and the results are statistically analyzed. The strength degradation law is analyzed based on the statistical results of Monte Carlo simulation, and the time-varying strength degradation of steel structures is predicted based on the strength degradation law.

2. The method for predicting time-varying strength degradation of steel structures according to claim 1, characterized in that: The time-varying strength degradation model of steel is as follows: S(t)=S0·f c (t)·f f (t) in, S(t): steel strength at time t; S0: initial strength of steel; f c (t): Corrosion factor affecting strength; f f (t): Fatigue influence factor on strength.

3. The method for predicting time-varying strength degradation of steel structures according to claim 2, characterized in that: The initial strength S0 is defined as the initial strength of the steel before use and follows the following normal distribution: in, mean; Standard deviation.

4. The method for predicting time-varying strength degradation of steel structures according to claim 3, characterized in that: Corrosion impact factor f c (t) is defined as the loss of steel cross-section due to corrosion: h0: initial thickness of steel; Δd(t): cumulative corrosion amount at time t; μ d : Annual average corrosion amount; σ d : Discrete degree of corrosion amount; δ d (i): The corrosion amount in the i-th year. The corrosion amount in each year is independent and obeys the normal distribution. The corrosion rate is constant without considering environmental changes.

5. The method for predicting time-varying strength degradation of steel structures according to claim 4, characterized in that: Fatigue influence factor f f (t) is defined as fatigue-induced internal damage to steel: f f (t)=exp(-k·D(t)) in, k: fatigue damage coefficient; D(t): cumulative fatigue damage at time t; a is the year in which the steel begins to be used; D(i): fatigue damage in year i, fatigue damage is independent every year and obeys log-normal distribution; μ D : Annual average fatigue damage value; σ D : Discrete degree of fatigue damage.

6. The method for predicting time-varying strength degradation of steel structures according to claim 5, characterized in that: The time-varying strength degradation model of steel is as follows:

7. The method for predicting time-varying strength degradation of steel structures according to any one of claims 1 to 6, characterized in that: The Monte Carlo simulation of the strength degradation process specifically includes the following steps: Define the parameters of the steel time-varying strength degradation model; Generate initial intensity samples; Generate corrosion volume samples; Generate fatigue damage samples; Calculate the cumulative corrosion amount; Calculate the accumulated fatigue loss; Calculate the annual steel strength degradation value based on the steel time-varying strength degradation model; Calculate the mean and quantile, and draw the strength degradation curve.

8. A steel structure optimization method, characterized in that: The method for predicting time-varying strength degradation of steel structures according to any one of claims 1 to 7 is used to predict strength degradation of steel structures, and the steel structure is optimized according to the prediction results, specifically comprising: Based on the Monte Carlo simulation results, the standardized regression coefficient method is used to quantify the contribution of each influencing factor to strength degradation; According to the contribution of each influencing factor to strength degradation, corresponding measures are taken to optimize the steel structure.

9. The steel structure optimization method according to claim 8, characterized in that: Based on the Monte Carlo simulation results, the standardized regression coefficient method is used to quantify the contribution of each influencing factor to strength degradation. The specific steps include the following: Extract Monte Carlo simulation results; Perform preset standardization on simulation results; Construct a multiple linear regression model; Calculate standardized regression coefficients; Sort the influencing factors according to their regression coefficients; Draw a comparison chart of the importance of each influencing factor.

10. The steel structure optimization method according to claim 8, characterized in that: The initial strength, corrosion amount, fatigue damage and ultimate strength are normalized as follows: X: original data; μ x : mean; σ x :variance; X std : Data after standardization.

11. The steel structure optimization method according to claim 8, characterized in that: After standardization, the multiple linear regression model is established as follows: S std (T)=β0+β1S 0,std +β2Δd std (T)+β3D std (T)+e S std (T): final intensity after normalization; S 0,std : Normalized initial strength; Δd std (T): normalized cumulative corrosion amount; D std (T): normalized cumulative fatigue damage; β0: intercept term; β1,β2,β3: standardized regression coefficients; ε: error term.

12. The steel structure optimization method according to claim 11, characterized in that: The least squares method is used to fit the regression model to obtain the standardized regression coefficients β1, β2, and β3. The factors are ranked according to the absolute values of the standardized regression coefficients. The larger the absolute value of the regression coefficient, the greater the influence of the corresponding factor on strength degradation.

13. The steel structure optimization method according to claim 8, characterized in that: Through Monte Carlo simulation, we can quantitatively analyze the effects of optimization measures and provide customized optimization solutions for different engineering scenarios.

14. The steel structure optimization method according to claim 13, characterized in that: Optimization measures include: anti-corrosion coating, structural reinforcement and load testing.

15. The steel structure optimization method according to any one of claims 13-14, characterized in that: Through Monte Carlo simulation, the effects of optimization measures are quantitatively analyzed, including: Define optimization measures in Monte Carlo simulations; Adjust the parameters of the steel time-varying strength degradation model according to the optimization measures; Run Monte Carlo simulations to calculate annual steel strength degradation values based on the steel time-varying strength degradation model; Statistical means and quantiles; Compare the original results with the results after optimization measures are taken, quantitatively analyze the optimization effect, and draw a comparison chart of the optimization effect.

16. A steel structure optimization system, characterized in that: The method comprises a processor capable of executing the steps of the method for predicting time-varying strength degradation of steel structures according to any one of claims 1 to 7.

Images