CN115270238B - Dynamic load-based bridge static behavior prediction method - Google Patents

Abstract

The invention discloses a dynamic load-based bridge static behavior prediction method which comprises the following steps of establishing a full-bridge structure analysis model according to design data of an existing bridge, carrying out dynamic load test on the existing bridge to obtain a bridge dynamic response value, carrying out sensitivity analysis on each design parameter in an initial structure analysis model to obtain and determine key design parameters to be corrected, which influence a bridge structure, constructing training samples of the key design parameters to be corrected by utilizing a machine learning intelligent algorithm based on a uniform design sampling method, establishing a proxy prediction model, correcting the initial structure analysis model by utilizing the prediction results of the machine learning and the intelligent algorithm, carrying out bridge static behavior prediction based on the corrected structure analysis model, and evaluating the prediction results of bridge static force by introducing an error analysis method. The method has the advantages of reducing the cost of the bridge static load test, reducing the damage to the structure of the bridge and having high accuracy of the prediction result.

Description

Dynamic load-based bridge static behavior prediction method

Technical Field

The invention relates to the technical fields of civil engineering, structural engineering, bridge detection and health monitoring, in particular to a dynamic load-based bridge static behavior prediction method.

Background

With the rapid development of global traffic construction, the total number of bridges has increased rapidly in the world since this century, and only the highway bridge of China has 87.83 ten thousand seats according to statistics in 2019. The method is particularly important to the rapid performance evaluation, damage identification, intelligent analysis and control of safety performance of the existing bridge structure in the face of huge bridge base numbers.

The existing effective method for evaluating the technical state of the bridge structure is to pass a load test, but the bridge load test has the defects of higher test cost, long test process, large workload, certain damage to the bridge structure and need to seal roads, and seriously affects normal transportation and travel. In the face of huge quantity of bridges which are required to be rapidly evaluated in technical conditions of bridges, how to improve the bridge detection efficiency, reduce the bridge detection cost, avoid the influence of traffic travel caused by road closure, achieve the accuracy and reliability of detection results, realize the intelligent recognition rapid detection and intelligent analysis evaluation decision of bridge structures, and become the problem to be solved in the current research. The dynamic load test of the bridge can only identify the dynamic performance of the whole bridge structure, but compared with the static load test, the dynamic load test of the bridge has the advantages of relatively lower cost, relatively shorter test time, simpler test process and less influence on traffic.

Therefore, researchers in the field aim to provide a dynamic load-based bridge static behavior prediction method, which accurately predicts the static load result of a bridge by using dynamic load test results with lower cost, higher efficiency and smaller traffic influence.

Disclosure of Invention

In view of the defects and shortcomings of the existing bridge in the aspects of health monitoring, damage identification, intelligent analysis and performance evaluation, the invention aims to solve the problems of achieving the aim of rapidly predicting the static behavior of the bridge based on a relatively simple, low-cost, convenient and efficient dynamic load test result combined with a machine learning intelligent algorithm and a model correction technology.

In order to achieve the above purpose, the invention provides a dynamic load-based bridge static behavior prediction method, which is characterized by comprising the following steps:

Step 1, building a structural analysis model of a full bridge according to design data of an existing bridge, and taking the structural analysis model as an initial structural analysis model for subsequent model correction;

step 2, carrying out dynamic load test on the existing bridge to obtain a bridge dynamic response value;

step 3, performing sensitivity analysis on each design parameter in the initial structure analysis model to obtain and determine key design parameters to be corrected, which affect the bridge structure;

Step 4, based on a uniform design sampling method, constructing training samples of key design parameters to be corrected by utilizing a machine learning intelligent algorithm and constructing a proxy prediction model;

step 5, correcting the initial structural analysis model by using the prediction results of the machine learning and the intelligent algorithm to obtain a corrected structural analysis model;

Step 6, predicting bridge static behaviors based on the corrected structural analysis model;

and 7, introducing an error analysis method to evaluate the bridge static prediction result, wherein the error analysis adopts a Root Mean Square Error (RMSE) analysis method, and the root mean square error has the following calculation formula:

Wherein,The test response value and the intelligent prediction model prediction response value for the ith group of samples respectively,Is the average value of the experimental response values.

Further, the structural analysis model of step 1 is typically modeled using numerical methods, which are modeling methods of finite elements, boundary elements, discrete elements, and/or infinite elements.

Further, in the step 2, a measuring method (such as a machine vision measuring method) of measuring the bridge and the like by adopting a contact or non-contact method, a direct method or an indirect method is adopted in the dynamic load test of the bridge, so as to obtain the dynamic characteristic parameters of the bridge, wherein the dynamic characteristic parameters comprise the frequency, the vibration mode, the damping, the impact coefficient, the dynamic deflection, the dynamic strain and the like of the bridge.

Further, in the step 3, sensitivity analysis is carried out on each design parameter of the initial structural analysis model of the bridge in the step 1 by adopting a sensitivity analysis method, sensitivity weight indexes of different design parameters are analyzed, and key design parameters to be corrected of the bridge are determined. The sensitivity analysis method can also adopt a gray correlation degree method to realize weight analysis of the key parameters to be corrected.

Further, a uniform design sampling method is adopted in step 4, training samples of key design parameters to be corrected, which are uniformly distributed in space, are constructed, and a prediction model is built by combining an intelligent algorithm, wherein the intelligent algorithm comprises a Bayesian theory, a Gaussian process method, a Kriging model, various agent models and other prediction methods.

And further, step 5, based on the prediction model constructed in step 4, calling the bridge dynamic load test result obtained in step 2 to predict each key design parameter to be corrected, and substituting the prediction result of each key design parameter to be corrected into the initial structure analysis model of the bridge constructed in step 1 to realize correction of the initial structure analysis model.

Further, step 6 is based on the corrected structural analysis model, and load working conditions are applied to the corrected structural analysis model according to the static load test scheme of the bridge, so that prediction of the static force result of the bridge is achieved. The static load prediction result can be used for respectively carrying out deformation, internal force, stress and the like on the whole bridge or local bridge.

And 7, introducing an error analysis method, and evaluating the predicted result of the bridge static load. The error analysis generally adopts a root mean square error RMSE (Root Mean Squared Error) analysis method.

Based on the dynamic load test result of the existing bridge, the invention combines the structural analysis model correction method and the intelligent algorithm technology, realizes the accurate prediction of the static behavior of the existing bridge, and achieves the following technical effects:

(1) The static behavior of the bridge is predicted based on the dynamic load test result of the existing bridge, so that the cost of the static test of the bridge is reduced to a great extent, the detection efficiency is improved, the closed traffic problem caused by the static load test is avoided, and the damage to the structure itself in the loading process of the bridge is reduced;

(2) And the intelligent algorithm is adopted, and the model correction method is combined to predict the static behavior of the existing bridge, so that the accuracy of the obtained prediction result is higher, and the actual bridge health condition is more fitted.

(3) The method can realize rapid static behavior analysis on a large number of bridges, can carry out omnibearing safety evaluation on the whole structure of the bridge, and provides a new method for health monitoring, operation and maintenance of the bridge.

Drawings

Fig. 1a is a flow chart of the present invention.

FIG. 1b is a flow chart of a dynamic load based bridge static behavior prediction method of the invention;

FIG. 2 is a graph of a dynamic load test result (sampling time on the abscissa and amplitude on the ordinate) for a bridge according to an embodiment of the present invention;

FIG. 3 is a graph of the results of a parameter sensitivity analysis in accordance with one embodiment of the present invention;

FIG. 4 is a flow chart of the combination of analytical model modification and intelligent algorithm of the present invention;

FIG. 5 is a graph comparing predicted static load results of a bridge with actual static load test results in situ in one embodiment of the invention.

Detailed Description

The following describes the detailed implementation of the embodiments of the present invention with reference to the drawings. It should be understood that the detailed description and specific examples, while indicating and illustrating the invention, are not intended to limit the invention.

It should be noted that, without conflict, the embodiments of the present invention and features of the embodiments may be combined with each other.

The invention will be described in detail below with reference to the drawings in connection with exemplary embodiments.

Step 1, establishing an initial structure analysis model of the bridge according to existing bridge design data;

According to the design data of the existing bridge, the values of the geometric shapes, specific sizes and material properties of all parts of the bridge and the form of boundary conditions are defined, and the structural analysis software is utilized to build an initial analysis model of the bridge structure, wherein the analysis model is used as a reference model for the correction of a follow-up model. The analysis model is usually a numerical model, such as finite element model, boundary element model, discrete element model, infinite element model, etc., and common analysis software is ANSYS, ABAQUS, midas.

Step 2, carrying out dynamic load test on the existing bridge to obtain a bridge dynamic response value;

In order to obtain the dynamic response value of the bridge, in a specific embodiment, an on-site dynamic load test is performed on an existing bridge by adopting an environmental excitation method in a direct measurement method. During testing, vibration response of the bridge under the ground pulse is obtained through the L/8, L/4 and L/2 typical sections of the bridge, which are arranged on the vibration pickup. FIG. 2 is a graph showing the results of dynamic load testing on a bridge, wherein the abscissa represents the test time and the ordinate represents the amplitude. And carrying out Fourier transform on the time domain result to obtain a frequency domain result reflecting the bridge frequency characteristics.

The fourier transform formula is:

Wherein j is a virtual unit, j 2 = -1 is no unit, T is a period, unit is second, X is an original function of X, T is time, unit is second, ω is frequency, and X (T) is a continuous time signal.

Step 3, performing sensitivity analysis on each design parameter in the initial structure analysis model, and obtaining and determining key design parameters to be corrected, which affect the bridge structure:

And (3) selecting design parameters to be corrected on the basis of the bridge initial structure model established in the step (1). And carrying out sensitivity analysis on different design parameters by using a global sensitivity analysis method, and determining sensitivity weight indexes of different design parameters by changing a change threshold of one design parameter to 10% each time through a variable control method, keeping other design parameters unchanged, and calculating influence degrees of corresponding first two-order frequency response results f1 and f2 of the bridge structure when different design parameters change.

The key parameters comprise design parameters such as structural geometric dimensions, elastic modulus of concrete materials, concrete volume weight and the like. And respectively changing structural design parameters one by using a sensitivity analysis method, calculating corresponding structural response values, constructing a key parameter sensitivity analysis model, quantifying the specific influence effect of different parameters on the bridge structure, and determining the key design parameters to be corrected. Fig. 3 shows the result of parameter sensitivity analysis in one embodiment, where the parameters in this embodiment are K1, K2, K3, r1, r2, and r3, respectively, and the corresponding structure response results are the first 2 nd order frequencies f1 and f2.

Step 4, based on a uniform design sampling method, constructing training samples of key design parameters to be corrected by utilizing a machine learning intelligent algorithm, and constructing a proxy prediction model:

And (3) respectively setting variable thresholds for the parameters according to the key design parameters to be corrected determined in the step (3), writing the variable thresholds into macro files, reading the macro files of the parameter thresholds by utilizing structural analysis software, and sequentially calculating structural response values corresponding to each group of parameters in the thresholds, wherein the response values can be the frequencies of each step of the bridge or the deflection value of a certain measuring point of the bridge, training samples are generated through limited times of calculation, and a prediction model is established by utilizing the training samples. FIG. 4 is a flow chart of the combination of the structural analysis model modification and the intelligent algorithm of the present invention.

Specifically, step4 is based on a uniform design sampling method, and an intelligent algorithm is utilized to establish training samples of design parameters and response results, and to establish a Kriging prediction model:

according to the key design parameters to be corrected determined by the sensitivity analysis in the step 3, K1, K2, K3, r1, r2 and r3 are taken as parameters to be corrected in the embodiment. By a uniform design sampling method, through a uniform design table U_n(m^r), where U represents the uniform design table, n represents the number of required uniform tests, m represents the number of acceptable factor levels, and r represents the number of factors that can be arranged at most.

This example creates a uniform design table for U₃₀(30⁶), i.e., 30 trials, 30 levels, 6 parameters. Training samples of the intelligent algorithm are formed by the corresponding structure response frequency results f1 and f2, and specific training samples are shown in table 1.

Substituting the training sample into a Kriging theoretical model:

y(x)=f(x)^Tβ+Z(x) (3);

Wherein y (x) is a Kriging model function, T is the transposed meaning, f (x) is a polynomial model, and beta is a regression coefficient. Z (x) is a stochastic process called a variational function or correlation model, which in turn builds a Kriging proxy predictive model.

Step 5, correcting the initial structural analysis model by using the prediction results of the machine learning and the intelligent algorithm to obtain a corrected structural analysis model;

And (3) inputting the result data obtained by the dynamic load test in the step (2) according to the training sample established in the step (4), predicting the optimal value of a group of parameters to be corrected by a machine learning intelligent algorithm, and substituting the predicted value into the initial structural analysis model to realize the correction of the structural analysis model.

The machine learning intelligent algorithm comprises a Kriging model algorithm, a Gaussian process algorithm, a Bayesian algorithm, a random forest algorithm, a cloud theory algorithm, various agent models and the like.

This embodiment employs a Kriging model algorithm, where first the Kriging model comprises two parts, a polynomial and a random distribution, i.e., y (x) =f (x)^T β+z (x), where:

f(x)^Tβ＝[f₁(x),f₂(x),...,f_p(x)]β＝f₁(x)β₁+f₂(x)β₂+...+f_p(x)β_p (4);

f (x) is a polynomial model, p is the number of polynomials, and β is the regression coefficient.

Z (x) is a stochastic process called a variational function or correlation model, and the covariance matrix of Z (x) is:

wherein Cov () is covariance, sigma is standard deviation, and theta is hyper-parameter. x_i and x_j are sample points; A spatial correlation function for any two sample points x_i and x_j of the sample points, the function form of which is:

In the Kriging regression function model prediction process, the problem is converted into the minimum optimized problem, namelyAnd (3) obtaining a parameter theta by solving the minimum optimization problem of the formula, and constructing an optimal Kriging prediction model, wherein theta is a super parameter, m is a natural number, m=1, 2,3, and m and sigma are standard deviations.

Step 6, predicting bridge static behaviors based on the corrected structural analysis model;

Based on the structural analysis model corrected in the step 5, the numerical model is matched with the performance of the actual bridge, the loading is simulated in the corrected numerical model according to the loading position of the static load test of the actual bridge, and the results of the bridge such as displacement, stress and the like of typical sections such as bridge span, pivot and the like are calculated through structural analysis software. In the embodiment, under the concentrated load effect in the midspan of a 3-span continuous beam bridge, the deflection W1-W18 of each longitudinal measuring point of the bridge is predicted and compared with the measured result to verify.

Step 7, an error analysis method is introduced to evaluate the bridge static prediction result, wherein the error analysis adopts a Root Mean Square Error (RMSE) analysis method;

the invention provides a root mean square error RMSE (Root Mean Squared Error) as an evaluation index of a prediction result.

The root mean square error calculation formula is as follows:

Wherein,The test response value and the intelligent prediction model prediction response value for the ith group of samples respectively,Is the average value of the experimental response values. RMSE is used to evaluate the accuracy of the smart prediction model, the closer this value is to 0, the smaller the error between the test response value and the smart prediction model predicted value.

Table 1U₃₀(30⁶) design matrix

TABLE 2 initial value, measured value, predicted value (mm) for each deflection measurement point

Table 3 deflection error analysis

In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings are merely for convenience in describing the present invention and simplifying the description, and do not indicate or imply that the device or element being referred to must have a specific orientation, be configured and operated in a specific orientation, and therefore should not be construed as limiting the present invention.

Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present invention, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise.

In the present invention, unless explicitly specified and limited otherwise, the terms "mounted," "connected," "secured," and the like are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally formed, mechanically connected, electrically connected, or communicable with each other, directly connected, indirectly connected through an intervening medium, or in communication between two elements or in an interactive relationship between two elements, unless otherwise explicitly specified. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.

In the present invention, unless expressly stated or limited otherwise, a first feature "up" or "down" a second feature may be the first and second features in direct contact, or the first and second features in indirect contact via an intervening medium. Moreover, a first feature being "above," "over" and "on" a second feature may be a first feature being directly above or obliquely above the second feature, or simply indicating that the first feature is level higher than the second feature. The first feature being "under", "below" and "beneath" the second feature may be the first feature being directly under or obliquely below the second feature, or simply indicating that the first feature is less level than the second feature.

In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.

While embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the invention, and that variations, modifications, alternatives and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the invention.

Claims (9)

Translated fromChinese

1.基于动载的桥梁静力行为预测方法，其特征在于，包括以下步骤：1. A method for predicting static behavior of a bridge based on dynamic loads, characterized in that it comprises the following steps:步骤1、根据既有桥梁的设计资料，建立全桥的结构分析模型；并将所述结构分析模型作为后续模型修正的初始结构分析模型；Step 1: Establish a structural analysis model of the entire bridge based on the design data of the existing bridge; and use the structural analysis model as the initial structural analysis model for subsequent model correction;步骤2、对既有桥梁进行动载测试，获取桥梁动力响应值；Step 2: Perform a dynamic load test on the existing bridge to obtain the dynamic response value of the bridge;步骤3、对初始结构分析模型中的各设计参数进行敏感性分析，获取并确定影响桥梁结构的待修正关键设计参数；Step 3: Perform sensitivity analysis on each design parameter in the initial structural analysis model to obtain and determine the key design parameters to be corrected that affect the bridge structure;步骤4、基于均匀设计抽样法，利用机器学习智能算法，构建待修正关键设计参数的训练样本并建立代理预测模型；Step 4: Based on the uniform design sampling method, using machine learning intelligent algorithms, construct training samples of key design parameters to be corrected and establish a proxy prediction model;步骤5、利用机器学习和智能算法的预测结果对初始结构分析模型进行修正，获取修正后的结构分析模型；Step 5: Use the prediction results of machine learning and intelligent algorithms to correct the initial structural analysis model to obtain a corrected structural analysis model;步骤6、基于修正后的结构分析模型进行桥梁静力行为预测；Step 6: Predict the static behavior of the bridge based on the modified structural analysis model;步骤7、引入误差分析方法，对桥梁静力的预测结果进行评价；其中，误差分析采用均方根误差RMSE分析方法；均方根误差计算公式如下：Step 7: Introduce the error analysis method to evaluate the prediction results of bridge statics; the error analysis adopts the root mean square error RMSE analysis method; the root mean square error calculation formula is as follows:其中，分别为第i组样本所对用的试验响应值和智能预测模型预测响应值，为实验响应值的平均值。in, are the experimental response value and the predicted response value of the intelligent prediction model for the i-th group of samples, is the average of the experimental responses.2.如权利要求1所述的基于动载的桥梁静力行为预测方法，其特征在于：步骤1的结构分析模型采用数值方法建模，所述数值方法为有限元、边界元、离散元和/或无限元建模方法。2. The method for predicting static behavior of a bridge based on dynamic loads as described in claim 1 is characterized in that the structural analysis model in step 1 is modeled using a numerical method, and the numerical method is a finite element, boundary element, discrete element and/or infinite element modeling method.3.如权利要求2所述的基于动载的桥梁静力行为预测方法，其特征在于：步骤2中对桥梁动载测试时采用接触式或非接触式、直接法或间接法测量桥梁的动力性能；其中，所述动力性能的参数包括桥梁的频率、振型、阻尼、冲击系数、动挠度及动应变。3. The method for predicting static behavior of a bridge based on dynamic loads as described in claim 2 is characterized in that: in step 2, the dynamic load test of the bridge adopts contact or non-contact, direct or indirect method to measure the dynamic performance of the bridge; wherein the parameters of the dynamic performance include the frequency, vibration mode, damping, impact coefficient, dynamic deflection and dynamic strain of the bridge.4.如权利要求3所述的基于动载的桥梁静力行为预测方法，其特征在于：步骤2中通过采用直接测量法中的环境激励法对一座既有桥梁进行现场动载测试；测试时通过拾振器布置在桥梁的L/8、L/4、L/2典型截面获取桥梁在地脉动下的振动响应，并对获取的时域结果进行傅里叶变换，既可得到反映桥梁频率特征的频域结果；4. The method for predicting static behavior of a bridge based on dynamic loads as claimed in claim 3 is characterized in that: in step 2, an on-site dynamic load test is performed on an existing bridge by using the environmental excitation method in the direct measurement method; during the test, vibration pickups are arranged at typical sections of the bridge at L/8, L/4, and L/2 to obtain the vibration response of the bridge under ground pulsation, and the obtained time domain results are Fourier transformed to obtain frequency domain results reflecting the frequency characteristics of the bridge;傅里叶变换公式为：The Fourier transform formula is:式中：j为虚单位，j^2＝-1，无单位；T为周期，单位为秒；X为x的原函数；t为时间，单位为秒；ω为频率，x(t)为连续时间信号。Where: j is an imaginary unit, j^2=-1, no unit; T is the period, in seconds; X is the original function of x; t is the time, in seconds; ω is the frequency, and x(t) is a continuous time signal.5.如权利要求3所述的基于动载的桥梁静力行为预测方法，其特征在于：步骤3中采用灵敏度分析方法对步骤1桥梁的初始结构分析模型的各设计参数进行敏感性分析，分析不同设计参数的灵敏度权重指标，确定桥梁的待修正关键设计参数。5. The method for predicting static behavior of a bridge based on dynamic loads as described in claim 3 is characterized in that: in step 3, a sensitivity analysis method is used to perform sensitivity analysis on each design parameter of the initial structural analysis model of the bridge in step 1, and the sensitivity weight indexes of different design parameters are analyzed to determine the key design parameters of the bridge to be corrected.6.如权利要求5所述的基于动载的桥梁静力行为预测方法，其特征在于：步骤4采用均匀设计抽样方法，构建空间均匀满布的待修正关键设计参数的训练样本，并结合智能算法建立预测模型；其中，所述智能算法包括贝叶斯理论、高斯过程方法和/或Kriging模型。6. The method for predicting static behavior of a bridge based on dynamic loads as described in claim 5 is characterized in that: step 4 adopts a uniform design sampling method to construct a training sample of key design parameters to be corrected that is uniformly distributed in space, and establishes a prediction model in combination with an intelligent algorithm; wherein the intelligent algorithm includes Bayesian theory, Gaussian process method and/or Kriging model.7.如权利要求6所述的基于动载的桥梁静力行为预测方法，其特征在于：步骤4采用的智能算法为Kriging模型算法；7. The method for predicting static behavior of a bridge based on dynamic loads as claimed in claim 6, characterized in that: the intelligent algorithm used in step 4 is a Kriging model algorithm;其中，Kriging模型包含多项式和随机分布两部分，即y(x)＝f(x)^Tβ+Z(x)，其中：The Kriging model consists of two parts: polynomial and random distribution, i.e., y(x) = f(x)^T β + Z(x), where:f(x)^Tβ＝[f₁(x),f₂(x),...,f_p(x)]β＝f₁(x)β₁+f₂(x)β₂+...+f_p(x)β_p (4)；f(x)^T β＝[f₁ (x), f₂ (x),..., f_p (x)]β＝f₁ (x)β₁ +f₂ (x)β₂ +...+f_p (x)β_p (4);f(x)为多项式向量，p为多项式数目，β为回归系数；f(x) is the polynomial vector, p is the number of polynomials, and β is the regression coefficient;Z(x)是一个随机过程，称之为变异函数或相关模型，Z(x)的协方差矩阵为：Z(x) is a random process, called a variogram or correlation model, and the covariance matrix of Z(x) is:式中：Cov()为协方差；σ为标准差；θ为超参数；x_i和x_j为样本点；为样本点中任何两个样本点x_i和x_j的空间相关函数，其函数形式为：Where: Cov() is the covariance; σ is the standard deviation; θ is the hyperparameter;_xi and_xj are sample points; is the spatial correlation function of any two sample points_xi and_xj in the sample points, and its function form is:在Kriging回归函数模型预测过程中，将问题转化问最小优化问题，即通过求解公式的最小优化问题，得到参数θ，便可构建最优的Kriging预测模型，其中θ为超参数，m为自然数，m＝1,2,3,…m，σ为标准差。In the prediction process of the Kriging regression function model, the problem is transformed into a minimum optimization problem, that is, By solving the minimum optimization problem of the formula and obtaining the parameter θ, the optimal Kriging prediction model can be constructed, where θ is a hyperparameter, m is a natural number, m = 1, 2, 3, ... m, and σ is the standard deviation.8.如权利要求7所述的基于动载的桥梁静力行为预测方法，其特征在于：步骤5基于步骤4构建的预测模型，调用步骤2所获取的桥梁动载测试结果，对各待修正关键设计参数进行预测，并将各待修正关键设计参数的预测结果代入步骤1所构建的桥梁的初始结构分析模型，实现对初始结构分析模型的修正。8. The method for predicting static behavior of a bridge based on dynamic loads as described in claim 7 is characterized in that: step 5 is based on the prediction model constructed in step 4, calls the bridge dynamic load test results obtained in step 2, predicts each key design parameter to be corrected, and substitutes the prediction results of each key design parameter to be corrected into the initial structural analysis model of the bridge constructed in step 1 to realize the correction of the initial structural analysis model.9.如权利要求6所述的基于动载的桥梁静力行为预测方法，其特征在于：步骤6基于修正后的结构分析模型，根据桥梁的静载试验方案，在修正后的结构分析模型中进行荷载工况的施加，实现对桥梁静力结果的预测。9. The method for predicting the static behavior of a bridge based on dynamic loads as described in claim 6 is characterized in that: step 6 is based on the modified structural analysis model, and according to the static load test plan of the bridge, the load condition is applied in the modified structural analysis model to achieve the prediction of the static results of the bridge.