Steel structure wind-induced fatigue evaluation method based on meteorological data and wind tunnel data

Abstract

The invention discloses a method for evaluating wind-induced fatigue of a steel structure based on meteorological data and wind tunnel data. According to the method, an extreme value distribution model and a mixed Von-mises distribution theory are utilized to respectively fit local wind speed and wind direction probability density functions, and then a wind speed and wind direction joint probability density function is obtained; combining wind tunnel test data, carrying out structural dynamic time history analysis through a scale ratio theory and a mechanical equivalence principle, screening key components, and extracting a stress time history; and finally, calculating the fatigue life by using a rain flow counting method and an SN curve. According to the method, multi-source data are fused, the unsteady wind effect is considered, compared with a traditional method, the accuracy and rationality of wind-induced fatigue life prediction are improved, a reliable basis is provided for wind resistance design of an engineering structure, and the method has high engineering application value.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of wind disaster analysis of engineering structures, in particular to a steel structure wind-induced fatigue evaluation method based on meteorological data and wind tunnel data. BACKGROUND

[0002] Large-span and cantilever steel structures are usually wind-sensitive structures, and the design wind speed in coastal areas of China is usually large. At the same time, if the building shape is complex, under the action of unsteady wind, complex flow phenomena will be further generated, and the structure will bear difficult-to-predict fluctuating wind load, and the large strain low-cycle fatigue effect generated thereby will seriously affect the normal service of the structure.

[0003] The purpose of the present application is to overcome the shortcomings of the prior art and provide a steel structure wind-induced fatigue evaluation method based on meteorological data and wind tunnel data. SUMMARY

[0004] The purpose of the present application is to provide a steel structure wind-induced fatigue evaluation method based on meteorological data and wind tunnel data to solve the problems raised in the background.

[0005] To achieve the above purpose, the present application provides the following technical scheme: a steel structure wind-induced fatigue evaluation method based on meteorological data and wind tunnel data, comprising the following steps: Step 1: According to the data of the local measured daily maximum wind speed, use the extreme value distribution model to fit the local wind speed probability density function; Step 2: According to the wind direction data corresponding to the daily maximum wind speed measured in step 1, use the mixed Von-mises distribution theory to fit the local wind direction probability density function; Step 3: Combine the wind speed probability density function and the wind direction probability density function, based on the conditional probability theory and the polynomial fitting method, obtain the local wind speed and wind direction joint probability density function: multiply the wind speed probability density function and the wind direction probability density function directly to obtain the wind speed and wind direction joint probability density function , wherein the interval of the wind speed is , and the interval of the wind direction is ; wherein the wind direction represents north wind, and the wind direction represents east wind; Step 4: According to the basic wind speed converted from the local basic wind pressure and the wind direction definition method in the wind tunnel test, integrate the wind speed and wind direction joint probability density function in the corresponding wind speed and wind direction interval to obtain the discrete wind speed and wind direction joint probability distribution function, and determine the working condition of the structure dynamic time history analysis therefrom: assuming that the wind tunnel test is divided into groups of wind directions for testing, and the wind direction in the wind tunnel test is The angle between the wind direction and the northerly wind direction at the project site is The wind direction range is also divided into... The interval for each wind direction is [number]. ,in It is a non-negative integer; the basic wind speed is calculated based on the local basic wind pressure, and the wind speed range is... Evenly divided into The wind speed at the midpoint of each wind direction interval is [value missing]. ,in The value is a non-negative integer, and the length of each wind direction interval is not less than 4 m / s; Integrating the wind speed and direction probability density function over the corresponding wind speed and direction intervals yields the discrete joint probability distribution function of wind speed and direction. , These are discrete points, and there are a total of Let there be _ function values, denoted as _ ... , among which express One wind direction, express wind speed, that is The working conditions for structural dynamic time history analysis; Step 5: Using wind pressure time history data measured in wind tunnel tests, and applying the scaling theory and mechanical equivalence principle, apply dynamic wind loads to the finite element model of the actual engineering structure, and perform structural dynamic time history analysis for multiple wind directions and speeds. Based on the stress time history results, select several locations with maximum structural stress, consider them as key components for fatigue analysis, and extract their stress time history calculation results under all calculation conditions. Use wind pressure time history data measured in wind tunnel tests for loading, assuming the time scaling ratio of the wind tunnel test is... The sampling frequency of the measured data is The time step for dynamic time history analysis is Total computation time The total duration is determined based on the length of the wind pressure time history data measured in the wind tunnel test in step 4. Greater than 60 seconds; Dynamic wind loads were applied to the finite element model of the actual engineering structure, and structural dynamic time history analysis was performed under multiple wind directions and speeds. A total of [number missing] analyses were conducted. Calculations for each working condition; for each working condition in the dynamic time history analysis, extract the component with the highest stress, and record the component number and location of the highest stress; a total of records are made. One data point; Again Frequency statistics were performed on the data, and the five components with the highest stress levels that appeared most frequently were considered critical components. The stress time history data of these locations were recorded, and their values ​​were extracted from all calculated operating conditions. One stress time history; Step 6, using the rainflow counting method to process the stress time history data, the stress amplitude and cycle number of each wind speed and wind direction condition are obtained; Step 7, according to the SN curve and using the Goodman correction, the fatigue life of each wind speed and wind direction condition is calculated; then combined with the discrete wind speed and wind direction joint probability distribution function, the wind-induced fatigue life of the engineering structure is calculated.

[0006] Preferably, based on the extreme value distribution theory, the specific implementation steps of fitting the local wind speed probability density function are as follows: Through meteorological data, the ten-year daily average maximum wind speed of the project location is extracted, wherein the meteorological data includes the maximum wind speed value, the wind direction corresponding to the maximum wind speed, and the occurrence date; and the wind direction recorded in the meteorological data should be 16; Sort the maximum wind speed, remove about 10 wind speed data from the maximum value; then record the maximum value of the remaining data, and take the integer of the maximum wind speed fitting value from the back; the minimum value should be taken from the front according to the minimum value of the sorted data, and recorded as ; wherein The minimum value of Apply the extreme value distribution theory to data fitting of all wind speed data, then use the correlation coefficient to judge the fitting degree of different extreme value distribution models, and select the maximum extreme value distribution model as the local wind speed probability density function, wherein is the measured wind speed value, is the wind speed fitting value, is the average value of the measured wind speed value, represents the th data, and the selectable extreme value distribution model includes: Generalized extreme value distribution: , the parameters to be solved are ; Gumbel distribution: , the parameters to be solved are ; Lognormal distribution: , the parameters to be solved are ; Among them, in the generalized extreme value distribution and Gumbel distribution, is the shape parameter, is the scale parameter, is the location parameter; in the lognormal distribution, is the log mean, is the log standard deviation; is the wind speed independent variable, and The parameter values ​​of the three extreme value distribution models are obtained by using arbitrary parameter estimation methods.

[0007] Preferably, the specific implementation steps for fitting the local wind direction probability density function based on the mixed Von-Mises distribution theory are as follows: First, obtain the wind direction data corresponding to the measured daily maximum wind speed in step 1. Then, fit the data using a mixed Von-Mises distribution model, where the mixed Von-Mises distribution model takes the following form: , It is the wind direction angle, which is the independent variable. It is the first Weight coefficients of a unimodal Von-Mises distribution, express The mixed von Mises distribution, where It is the th in the mixed Von-Mises distribution A single-peaked Von-Mises distribution that simultaneously satisfies , ,and It is a 0th-order Bessel function, specifically... ; The average angle weighted by wind direction probability. The scaling function is obtained from the average vector length of the samples, and the estimation formula is as follows: , The calculation method is as follows ; Subsequently, the wind speed data was divided into 16 groups according to wind direction, and the parameter value sequence for each wind direction was obtained according to the extreme value distribution model determined in step 1. Converting wind direction to radians, for each parameter in the extreme value distribution model... This is equivalent to obtaining 16 discrete values ​​with wind direction as the independent variable, where the parameter In the representative extreme value distribution model Parameters; each parameter is calculated using a polynomial fitting method. exist Functional relationship between up and wind direction Then the parameters Substituting into the extreme value distribution model of wind speed, we obtain... That is, the wind speed probability density function.

[0008] Preferably, step 7 is implemented in the following ways: Based on the Goodman formula, correct the stress amplitude and corresponding number of cycles in step 6; the calculation method is as follows: In the formula, is the corrected stress amplitude, is the uncorrected stress amplitude, is the average stress, is the tensile strength of the material; Then, combined with the SN curve of the material itself, the fatigue damage of a single component under a single working condition is calculated using the Mises cumulative damage theory; that is, the loading times of the component when it is damaged under a certain constant amplitude stress level is , the fatigue damage of the component when it is subjected to cycles is: ， The damages caused by various levels of stress are linearly added up to the total damage of the component, and the total damage should be calculated according to the following formula: The fatigue damage value of a single component in Ts under the working condition of wind direction and wind speed is obtained ; ; For a single component, the total damage of the single component in Ts considering the distribution of wind speed and wind direction is obtained , and the fatigue life (years) of the single component is: ; the fatigue life of all five key components is calculated, and the minimum value is the fatigue life of the entire engineering structure.

[0009] Compared with the prior art, the beneficial effects of the present application are: Fusion of multi-source data improves evaluation accuracy and solves the limitations of traditional methods: the present application constructs a complete wind-induced fatigue evaluation system for steel structures by fusing the daily maximum wind speed and wind direction data measured by local meteorological stations and the wind pressure time history data obtained by wind tunnel tests. Compared with the traditional evaluation method which depends on wind speed data or simplifies the wind load model, the present application fully considers the probability distribution characteristics of wind speed, the directional influence of wind direction and the non-stationary effect of wind load. For example, in an embodiment of the present application, the local wind speed data are fitted by generalized extreme value distribution (GEV) (fitting goodness R²=0.9696), and a 4-peak mixed Von-Mises distribution is used to describe the wind direction probability density, so that the joint probability density function of wind speed and wind direction can accurately reflect the local wind environment characteristics, and the wind pressure time history data measured by wind tunnel tests are used for finite element time history analysis. This multi-source data fusion method avoids the evaluation deviation caused by ignoring the relationship between wind speed and wind direction and the spatial variation of fluctuating wind load in traditional methods, and provides a more reliable basis for wind-resistant design of complex steel structures.

[0010] ​Multi-dimensional feature modeling realizes accurate fatigue analysis and improves engineering safety: The application constructs a full-process analysis framework from wind environment data processing to fatigue life calculation through seven core steps. In the process of constructing the joint probability density function of wind speed and wind direction, the polynomial fitting method is used to establish the extreme value distribution parameter function with wind direction as the independent variable, so that the wind speed probability density function can dynamically change with the wind direction, thereby accurately depicting the wind speed distribution characteristics under different wind directions. In dynamic time history analysis, wind tunnel test data is mapped to the actual structure through scale ratio theory and mechanical equivalence principle, combined with rain flow counting method and Goodman correction, to realize the calculation of stress amplitude and cycle number of key components. Taking an embodiment of the application as an example, the method successfully selects the top five key components with the largest stress, and calculates the total damage considering the wind speed and wind direction distribution through the Mises cumulative damage theory, and finally determines the minimum value of the engineering fatigue life, which provides a clear time node for structure maintenance and safe operation.

[0011] Systematic process design improves engineering applicability and reduces evaluation cost: The application provides a standardized analysis process and calculation model, including wind speed data preprocessing (removing large maximum values), extreme value distribution model selection, wind direction mixed Von-Mises fitting, joint probability density function construction, dynamic time history analysis condition determination, key component screening and fatigue life calculation, etc. The systematic design makes the evaluation process repeatable and operable, without relying on complex field tests or customized models. In an embodiment of the application, by dividing the wind speed interval [0, 24 m / s] into 12 parts and the wind direction into 36 groups, only 12x36 calculation conditions are needed to cover the full wind environment scenario. At the same time, the method clearly defines the parameter calculation method and model selection standard of each step, reducing the dependence on the experience of professional engineers.

[0012] Considering the coupling of non-stationary wind effects and multiple working conditions, adapting to the evaluation needs of complex structures: The application fully considers the complex non-stationary effects of wind passing through buildings through wind tunnel test data loading and dynamic time history analysis of multiple wind directions and multiple wind speeds, solving the problem that existing methods cannot accurately evaluate the spatial distribution of fluctuating wind loads. In the treatment of wind-sensitive structures such as large-span and cantilever steel structures, the method can capture the dynamic changes of wind pressure on the structure surface under different wind directions, and through stress time history analysis, it can screen out the real dangerous components. For example, in an embodiment of the application, by frequency statistics of the stress maximum components of 12x36 working conditions, the top five key components with the most occurrences are successfully located, avoiding the omission of dangerous components caused by the simplification of wind load in traditional methods. In addition, the application realizes the cumulative evaluation of fatigue damage of the structure under natural wind environment through the coupling of wind speed and wind direction joint probability density function and fatigue life calculation, making the evaluation results more consistent with the actual service conditions, and providing key technical support for wind-resistant design of complex steel structures in coastal areas. BRIEF DESCRIPTION OF DRAWINGS

[0013] Figure 1 Flow chart of the method of the present application; Figure 2 Fitting of wind speed and direction of the present application; Figure 3 Joint probability density function of wind speed and direction of the present application; Figure 4 Stress time history calculation results of unfavorable position of the present application; Figure 5 Rainflow counting results of the present application; Figure 6 One of the embodiments of the present application; Figure 7 Parameter value table of embodiment 1; Figure 8 Parameters required for polynomial fitting; Figure 9 Stress time history results table under all calculation conditions; Figure 10 Wind-induced fatigue assessment results table of steel structure. DETAILED DESCRIPTION

[0014] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of the present application.

[0015] Please refer to Figures 1-5 The present application provides a technical solution: a steel structure wind-induced fatigue assessment method based on meteorological data and wind tunnel data, comprising the following steps: Step 1: According to the data of local measured daily maximum wind speed, the extreme value distribution model is used to fit the local wind speed probability density function; wherein the specific implementation steps are as follows: Through meteorological data, the ten-year average daily maximum wind speed of the project site is extracted, wherein the meteorological data includes the maximum wind speed value, the wind direction corresponding to the maximum wind speed, and the occurrence date; and the wind direction recorded in the meteorological data should be 16; Sort the maximum wind speed, remove about 10 wind speed data from the maximum value, then record the maximum value of the remaining data and take the integer towards the back to get the maximum value of the maximum wind speed fitting The minimum value should be obtained by taking the integer towards the front according to the minimum value of the sorted data, denoted as ​The minimum value should be the floor value of the basic wind speed calculated from the local basic wind pressure. The extreme value distribution theory was applied to fit the data of all wind speeds, and then the correlation coefficient was used. To determine the goodness of fit of different extreme value distribution models, select The maximum extreme value distribution model is used as the local wind speed probability density function, where This is the measured wind speed value. The fitted value is the wind speed. This represents the average of the measured wind speed values. Indicates the first For the given data, the selected extreme value distribution models include: Generalized extreme value distribution: The parameters to be determined are ; Gumbel distribution: The parameters to be determined are ; Log-normal distribution: The parameters to be determined are ; Among them, in the generalized extreme value distribution and the Gumbel distribution, For shape parameters, For scale parameters, For location parameters; in a log-normal distribution, The logarithmic mean is... The standard deviation is the logarithm. Let wind speed be the independent variable, and The parameter values ​​for the three extreme value distribution models are obtained by using arbitrary parameter estimation methods. The wind speed data processed by the extreme value distribution model is calculated using the arbitrary parameter estimation method to obtain the parameter values ​​of the three extreme value distribution models. The specific implementation of the arbitrary parameter estimation method can be selected from the gevfit() function in MATLAB, depending on the actual needs.

[0016] Step 2: Based on the wind direction data corresponding to the measured daily maximum wind speed in Step 1, the local wind direction probability density function is obtained by fitting using the mixed Von-Mises distribution theory; the specific implementation steps are as follows: First, obtain the wind direction data corresponding to the measured daily maximum wind speed in step 1. Then, fit the data using a mixed Von-Mises distribution model, where the mixed Von-Mises distribution model takes the following form: , It is the wind direction angle, which is the independent variable. It is the first Weight coefficients of a unimodal Von-Mises distribution, denotes a mixture of von-mises distributions, where is the i-th unimodal von-mises distribution in the mixture, while satisfying , , and is the 0-th order Bessel function, specifically ; is the average angle of the wind direction weighted by the probability of the wind direction, is the scale function, which is obtained by the average vector length of the samples, and the estimation formula is , the calculation method of ; Subsequently, the wind speed data is divided into 16 groups according to the wind direction, and the parameter value sequence of each wind direction is obtained according to the extreme value distribution model determined in step 1. Convert the wind direction to radian, and for each parameter in the extreme value distribution model, 16 discrete values with wind direction as the independent variable are obtained, where the parameter represents the parameter in the extreme value distribution model; using the polynomial fitting method, the functional relationship of each parameter with the wind direction is calculated , where ; then the parameter is substituted into the extreme value distribution model of the wind speed, and the , i.e. the wind speed probability density function, is obtained.

[0017] Step 3, combine the wind speed probability density function and the wind direction probability density function, and based on the conditional probability theory and the polynomial fitting method, obtain the local wind speed and wind direction joint probability density function; the local wind speed and wind direction joint probability density function is specifically: multiply the wind speed probability density function and the wind direction probability density function directly to obtain the wind speed and wind direction joint probability density function , where the interval of the wind speed is , and the interval of the wind direction is ; where wind direction represents north wind, represents east wind.

[0018] Step 4, according to the basic wind speed converted from the local basic wind pressure and the wind direction definition method in the wind tunnel test, integrate the wind speed and wind direction joint probability density function in the corresponding wind speed and wind direction interval to obtain the discrete wind speed and wind direction joint probability distribution function, and determine the working condition of the structure dynamic time history analysis according to the discrete wind speed and wind direction joint probability distribution function; the specific implementation steps are as follows:​ Suppose the wind tunnel test is divided into groups of wind direction to test, the wind direction in the wind tunnel test is , the angle between the wind direction and the north wind direction of the project location is ; the wind direction interval is also divided into , the interval of each wind direction is , where is a non-negative integer; according to the basic wind speed converted from the local basic wind pressure, the wind speed interval is evenly divided into , the midpoint wind speed of each wind direction interval is , where is a non-negative integer, and the length of each wind direction interval is not less than 4m / s; Integrate the wind speed and wind direction probability density function over the corresponding wind speed and wind direction interval, that is, integrate to get the discrete wind speed and wind direction joint probability distribution function , where the integral domain is the wind speed interval and the complete wind direction interval determined earlier , is a discrete point, and there are function values, denoted as , where represents wind directions, represents wind speeds, that is is the working condition for structural dynamic time history analysis.

[0019] Step 5, using the wind pressure time history data measured by wind tunnel test, through the scale ratio theory and mechanical equivalent principle, the dynamic wind load is applied to the finite element model of the actual engineering structure, and the multi-direction and multi-speed structure dynamic time history analysis is carried out accordingly; according to the stress time history results, a number of structural stress maximum positions are selected as the key components for fatigue analysis, and the stress time history calculation results under all calculation conditions are extracted; the specific implementation is as follows: Load using the wind pressure time history data measured by wind tunnel test, assuming that the time scale ratio of wind tunnel test is , the sampling frequency of measured data is , and the time step of dynamic time history analysis is ; the total calculation time According to the length of the wind pressure time history data measured by wind tunnel test in step 4, the total time is greater than 60s; Scale ratio theory: In wind tunnel test, the time scale ratio of wind tunnel test is generally given. Or calculate the time scale ratio through the geometric scale ratio and the wind speed scale ratio The sampling frequency of the data collector is also given in the wind tunnel test. For this scheme, the data needed is the time history data of the wind pressure measuring points, and the sampling frequency of the time history data is . Then the time step of the finite element time history analysis should be . The mechanical equivalent principle shows that the measured values of each measuring point in the wind tunnel test are often processed into wind pressure coefficients time history. This time history needs to be multiplied by the actual wind pressure to obtain the actual wind load borne by the structure.

[0020] The dynamic wind load is applied to the finite element model of the actual engineering structure, and the dynamic time history analysis of the structure under multiple wind directions and multiple wind speeds is carried out, a total of calculations are carried out; for each dynamic time history analysis, the component with the maximum stress is extracted, and the component number and the position of the maximum stress are recorded; a total of data are recorded; The frequency statistics are carried out on the data, and the top five components with the maximum stress that appear most frequently are regarded as the key components; the stress time history data of the above positions are extracted from all the calculation conditions, and a total of stress time histories are obtained.

[0021] The finite element model of the actual engineering structure is a finite element model established according to the actual engineering structure, and is related to the specific engineering.

[0022] Step 6, the rainflow counting method is used to process the stress time history data, and the stress amplitude and the cycle number under each wind speed and wind direction condition are obtained; Step 7, according to the SN curve and using the Goodman correction, the fatigue life under each wind speed and wind direction condition is calculated; then combined with the discrete wind speed and wind direction joint probability distribution function, the wind-induced fatigue life of the engineering structure is calculated; the specific implementation steps are as follows: According to the Goodman formula, the stress amplitude and the corresponding cycle number in step 6 are corrected; the calculation method is: , wherein is the corrected stress amplitude, is the uncorrected stress amplitude, is the average stress, is the tensile strength of the material; wherein is the stress amplitude obtained by the rainflow counting method, is the average stress obtained by the rainflow counting method.

[0023] Then combined with the SN curve of the material itself, the Mises cumulative damage theory is used to calculate the fatigue damage of a single component under a single condition; that is, the loading number when the component is destroyed under a certain constant amplitude stress level is which is subjected to fatigue damage in the secondary cycle is: ， The damage caused by various levels of stress is linearly accumulated to form the total damage of the component, and the total damage should be calculated according to the following formula: The fatigue damage value of the component in Ts is obtained The wind direction under the working condition of wind speed ; The total damage of the single component in Ts considering the distribution of wind speed and wind direction is obtained by calculating , and the fatigue life (years) of the single component is: ; the fatigue life of all five key components is calculated, and the minimum value is the fatigue life of the entire engineering structure.

[0024] Example 1 The following is an example of a practical engineering project, which introduces the specific implementation process: Step 1: The project site is Sanya City, Hainan Province, and the maximum wind speed recorded in Sanya City from 2008 to 2021 is selected as the original data. After excluding 4 wind speed data with large deviation (maximum 39 m / s, minimum 30 m / s), the analysis wind speed interval is determined as 0 m / s-24 m / s.

[0025] For wind speed data, the generalized extreme value distribution (GEV distribution), lognormal distribution (LN distribution) and Gumbel distribution are used for fitting. The goodness of fit is judged by the correlation coefficient . Finally, the GEV distribution is determined as the fitting function of the wind speed data in Sanya City. (The LN distribution corresponds to 1, and the Gumble distribution corresponds to 620)

[0026] Step 2: For wind direction data, a hybrid Von-mises distribution model is used for fitting. After changing the fitting parameters and trying several times, it is found that the 4-peak hybrid Von-mises distribution is in good agreement with the actual data. Finally, the 4-peak hybrid Von-mises distribution is determined as the fitting function of the wind direction data in Sanya City. The specific parameter values are shown in the parameter value table of Figure 7 .

[0027] Step 3: The processed wind speed data is divided into 16 groups according to the wind direction, and the parameter value sequence of the GEV distribution under each wind direction is obtained, i.e. 16 sequences of three parameters.

[0028] ​​After several attempts, a fifth-order polynomial was used to fit the three parameters. Figure 8 The parameters required for polynomial fitting were obtained, and the fitting yielded... In wind direction Functional relationship on.

[0029] parameters Substituting into the GEV distribution, we obtain the wind speed probability density function. .

[0030] Wind speed probability density function and wind direction probability density function Direct multiplication yields the joint probability density function of wind speed and wind direction. .

[0031] Step 4: Wind tunnel test uniform setting Group wind direction conditions. Wind tunnel test. Wind direction and meteorological data The wind angle is approximately This determined the wind direction range as follows. .in It is a non-negative integer. Based on computing resources, the wind speed range... Divide the wind into 12 equal parts, with the wind speed at the midpoint of each wind direction interval being [missing value]. ,in It is a non-negative integer.

[0032] Integrating the wind speed and direction probability density function over the corresponding wind speed and direction intervals yields the discrete joint probability distribution function of wind speed and direction. There are a total of Let there be _ function values, denoted as _ ... .

[0033] The time scaling ratio of wind tunnel testing is If the sampling frequency of the measured data is 300Hz, then the time step for the dynamic time history analysis should be . Total computation time .

[0034] Dynamic wind loads were applied to the finite element model of the actual engineering structure, and structural dynamic time history analysis was performed under multiple wind directions and speeds. A total of [number missing] analyses were conducted. Calculation of each working condition.

[0035] For each dynamic time-history analysis case, extract the component with the highest stress and record its number and location. A total of [number] records are generated. Data points.

[0036] right The first five component stress maximum positions with the most occurrences are regarded as key components, and the positions are recorded, and stress time history data thereof is extracted from all calculation conditions, and a total of stress time histories.

[0037] Figure 9 The table is the calculation result of 12x36 conditions, and the first five stress maximum rods with the most occurrences are regarded as key components, i.e., rods 72, 91, 103, 111 and 126.

[0038] Step 6: The rainflow counting method is used to process each stress time history data, and the stress amplitude and corresponding cycle number under each calculation condition are obtained.

[0039] Step 7: The stress amplitude and corresponding cycle number are corrected according to the Goodman formula. Combined with the SN curve of the material itself, the Mises cumulative damage theory is used to calculate the fatigue damage of a single component under a single condition. When a component is subjected to a constant amplitude stress level , the loading number when the component is damaged is , the fatigue damage of the component when subjected to cycles is ， The damage caused by various levels of stress is linearly accumulated to obtain the total damage of the component, and the total damage is . The fatigue damage value of a single component in wind direction wind speed condition is obtained. For a single component, the total damage of the component in is calculated to obtain the total damage of the single component in

[0040] . The fatigue life (years) of a single component is .

[0041] The fatigue life of all five key components is calculated, and the minimum value is the fatigue life of the entire engineering structure.

[0042] The table is the calculation result of each rod, and the wind-induced fatigue life of the structure in this case is 380 years. Figure 10

[0043] The present application considers the influence of local wind speed and wind direction distribution on the fatigue life of the structure; combined with wind tunnel test data, the complex unsteady effect of wind passing through the building is considered, and the problem that the precise evaluation of the spatial distribution of fluctuating wind load cannot be considered in the existing structure fatigue evaluation method is solved; the present application clearly defines the analysis and calculation process, provides a reliable practical application case, and can be applied to steel structure wind-induced fatigue evaluation.

[0044] ​​While embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely exemplary of the principles and application of the present application. Numerous modifications and adaptions can be effected without departing from the spirit and scope of the present application, which is not limited to the exact construction and arrangement described. It is intended, therefore, to cover all modifications and adaptions that fall within the scope of the claims and their equivalents.

Claims

1. A method for wind-induced fatigue assessment of steel structures based on meteorological data and wind tunnel data, characterized by, The method comprises the following steps: Step 1, according to the data of local measured daily maximum wind speed, using the extreme value distribution model, fitting to obtain the local wind speed probability density function; Step 2, according to the wind direction data corresponding to the measured daily maximum wind speed in step 1, using the mixed Von-mises distribution theory, fitting to obtain the local wind direction probability density function; Step 3, combine the wind speed probability density function and the wind direction probability density function, based on conditional probability theory and polynomial fitting method, to obtain the local wind speed and wind direction joint probability density function: multiply the wind speed probability density function and the wind direction probability density function directly to obtain the wind speed and wind direction joint probability density function , where the interval of wind speed is , and the interval of wind direction is ; where wind direction represents north wind, represents east wind; Step 4: Based on the basic wind speed converted from the local basic wind pressure and the wind direction definition method in the wind tunnel test, integrate the joint probability density function of wind speed and wind direction over the corresponding wind speed and wind direction intervals to obtain the discrete joint probability distribution function of wind speed and wind direction, and thereby determine the working conditions for structural dynamic time history analysis: assuming the wind tunnel test is divided into a total of... Group wind direction tests were conducted in the wind tunnel. The angle between the wind direction and the northerly wind direction at the project site is The wind direction range is also divided into... The interval for each wind direction is [number]. ,in It is a non-negative integer; the basic wind speed is calculated based on the local basic wind pressure, and the wind speed range is... Evenly divided into The wind speed at the midpoint of each wind direction interval is [value missing]. ,in The value is a non-negative integer, and the length of each wind direction interval is not less than 4 m / s; The wind speed and wind direction probability density function is integrated over the corresponding wind speed and wind direction intervals to obtain the discrete wind speed and wind direction joint probability distribution function , are discrete points, and there are function values, denoted as , where indicates wind directions, indicates wind speeds, i.e. is the working condition of the structural dynamic time history analysis; Step 5, using the wind pressure time history data measured by the wind tunnel test, the dynamic wind load is applied on the finite element model of the actual engineering structure through the scale ratio theory and the mechanical equivalent principle, and the structure dynamic time history analysis of multiple wind directions and multiple wind speeds is carried out accordingly; according to the stress time history results, a number of stress maximum positions of the structure are selected as the key components for fatigue analysis, and the stress time history calculation results of the key components under all calculation conditions are extracted: the wind pressure time history data measured by the wind tunnel test is used for loading, the time scale ratio of the wind tunnel test is assumed to be , the sampling frequency of the measured data is , and the time step of the dynamic time history analysis is ; total computation time total time length greater than 60 s; The wind load is applied to the finite element model of the actual engineering structure, the dynamic time-history analysis of the structure is carried out under multiple wind directions and multiple wind speeds, and a total of calculations are carried out; for each dynamic time-history analysis condition, the component with the maximum stress is extracted, and the component number and the maximum stress position are recorded; a total of data are recorded; Again to The first five stress maximum components with the most occurrences are regarded as the key components; the positions are recorded, and the stress time history data of all calculation conditions are extracted, a total of stress time history Step 6, using rain flow counting method to process stress time history data, obtaining stress amplitude and cycle number under each wind speed and wind direction condition; Step 7, according to the SN curve, and using Goodman correction, calculating the fatigue life under each wind speed and wind direction condition; Combined with the discrete wind speed and wind direction joint probability distribution function, the wind-induced fatigue life of the engineering structure is calculated. 2.The method for wind-induced fatigue assessment of steel structure based on meteorological data and wind tunnel data according to claim 1, characterized in that: Based on the extreme value distribution theory, the specific implementation steps of fitting the local wind speed probability density function are as follows: Through meteorological data, the ten-year daily average maximum wind speed of the project location is extracted, wherein the meteorological data includes maximum wind speed value, wind direction corresponding to maximum wind speed, occurrence date; and the wind direction recorded in the meteorological data should be 16; Sort the maximum wind speed, remove about 10 wind speed data from the maximum value, then record the maximum value of the remaining data and round it backward to get the maximum value of the maximum wind speed fitting ; the minimum value should be rounded forward according to the minimum value of the sorted data, recorded as ; wherein The minimum value of the minimum value should be the basic wind speed converted from the local basic wind pressure and rounded backward. The extreme value distribution theory was applied to fit the data of all wind speeds, and then the correlation coefficient was used. To determine the goodness of fit of different extreme value distribution models, select The maximum extreme value distribution model is used as the local wind speed probability density function, where This is the measured wind speed value. This is the fitted value for wind speed. This represents the average of the measured wind speed values. Indicates the first For the given data, the selected extreme value distribution models include: Generalized extreme value distribution: , the parameter to be found is ; Gumbel distribution: , the parameter to be solved is ; Lognormal distribution: , the parameter to be solved is ; wherein, in the generalized extreme value distribution and the Gumbel distribution, is the shape parameter, is the scale parameter, is the location parameter; in the lognormal distribution, is the log mean, is the log standard deviation; is the wind speed argument, and ; using any parameter estimation method, the final parameter values of the three extreme value distribution models are obtained. 3.The method for wind-induced fatigue assessment of steel structure based on meteorological data and wind tunnel data according to claim 1, characterized in that: Based on the mixed Von-mises distribution theory, the specific implementation steps of fitting the local wind direction probability density function are as follows: First, the wind direction data corresponding to the measured daily maximum wind speed in step 1 is obtained, and then a mixed Von-mises distribution model is used for fitting, wherein the form of the mixed Von-mises distribution model is as follows: , is a wind direction angle independent variable, is a weight coefficient of the th unimodal Von-Mises distribution, represents order mixed von-mises distribution, wherein is the th unimodal Von-mises distribution in the mixed Von-Mises distribution, and satisfies , , and is a 0-order Bessel function, specifically ; is a wind direction probability weighted average angle, is a scale function, which is obtained by the average vector length of the sample, and the estimation formula is , the calculation method of ; Then, the wind speed data is divided into 16 groups according to the wind direction, and the parameter value sequence under each wind direction is obtained according to the extreme value distribution model determined in step 1; Converting wind direction to radians, for each parameter in the extreme value distribution model... This is equivalent to obtaining 16 discrete values ​​with wind direction as the independent variable, where the parameter In the representative extreme value distribution model Parameters; each parameter is calculated using a polynomial fitting method. exist Functional relationship between up and wind direction Then the parameters Substituting into the extreme value distribution model of wind speed, we obtain... That is, the wind speed probability density function. 4.The method for wind-induced fatigue assessment of steel structure based on meteorological data and wind tunnel data according to claim 1, characterized in that: The specific implementation steps of step 7 are as follows: According to Goodman formula, the stress amplitude in step 6 and the corresponding cycle number are corrected; the calculation method is as follows: , wherein, is the corrected stress amplitude, is the uncorrected stress amplitude, is the average stress, is the tensile strength of the material; Then combined with the SN curve of the material itself, the fatigue damage of single component under single working condition is calculated using Mises cumulative damage theory; that is, the loading times of the component under a certain constant amplitude stress level are , the fatigue damage of the component under times of cyclic loading is: ， The total damage of the component is linearly accumulated by the damage caused by various levels of stress, and the total damage should be calculated according to the following formula: . The fatigue damage value of the single component under wind direction wind speed working condition, is obtained ; For individual member calculation The total damage of individual member within Ts considering the distribution of wind speed and wind direction is obtained, and the fatigue life (year) of individual member is: The fatigue life of the whole engineering structure is the minimum value of the fatigue life of all five key members.

Images