Regional building group earthquake damage analysis model parameter updating method
Through the combination of Bayesian parameter learning algorithm and building geometric features and basic attribute parameters, an empirical calculation formula was established, which solved the problem of updating the parameter of earthquake damage analysis model in regional building complexes, and achieved efficient and accurate model parameters under a few monitoring data. Updated to improve the overall accuracy of earthquake damage analysis in building complexes.
Patent Information
- Application Number
- CN202410473582.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-19
- Publication Date
- 2025-01-14
AI Technical Summary
The existing technology is difficult to effectively update the earthquake damage analysis model parameters of regional building complexes, especially when the number of monitoring buildings is limited, resulting in limited improvement in the accuracy of building complex models.
Using the iterative update method based on Bayesian parameter learning algorithm, the geometric features and basic attribute parameters of the building are calculated through the actual measured data of some monitored buildings in the building complex, the empirical calculation formula is established, the model parameters are gradually updated, and the calibration and sum of the building complex model parameters are realized. renew.
Under the conditions of limited monitoring of building data, the model parameters of all buildings in the building complex can be efficiently updated. The calculation amount is small, the update result is high accuracy, and the model update is only needed for the actual measurement results of one building.
Smart Images
Figure CN119312010A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of building earthquake damage analysis, and in particular relates to a method for updating parameters of a regional building complex earthquake damage analysis model. Background Art
[0002] For building earthquake damage analysis, due to the necessary simplifications made when establishing earthquake damage analysis models and the randomness of the mechanical properties of engineering structural materials, it is usually necessary to calibrate or update the initial model of the building structure based on measured data to improve the accuracy of the earthquake response analysis of the building structure. At present, there are many model update methods and technologies for single buildings. These methods and technologies mainly use structural vibration monitoring data to calibrate or update the structural model parameters, such as the Kalman filter algorithm for updating the time-varying model parameters of the structural system, the deterministic direct update method for updating the time-invariant model parameters of the structural system, and the Bayesian method considering uncertainty.
[0003] Although there are many model parameter updating methods based on monitoring data for single building structures, due to the limited number of buildings with monitoring systems in the building complex, the existing single building structure model updating methods can only update the limited monitoring building structure models in the building complex, which has a very limited effect on improving the overall accuracy of regional building complex earthquake damage simulation. Therefore, there is an urgent need for a building complex model updating method based on the monitoring data of a few buildings to improve the accuracy of the building complex model. Summary of the invention
[0004] The technical problem to be solved by the present invention is to provide a method for updating parameters of a regional building group earthquake damage analysis model.
[0005] The objective of the present invention is achieved through the following technical solutions:
[0006] A method for updating parameters of a regional building complex earthquake damage analysis model includes an empirical calculation formula for key model parameters of the building complex, an updating method thereof, and an updating method for model parameters of the building complex. The empirical calculation formula for model parameters is gradually updated by using the measured model parameters of some monitored buildings in the building complex, thereby realizing calibration and updating of the model parameters of the building complex.
[0007] Furthermore, the empirical calculation formula for the model parameters is a calculation formula for calculating the building model parameters using the building geometric feature parameters and the building basic attribute parameters.
[0008] Furthermore, the building geometric characteristic parameters and building basic attribute parameters can be directly acquired through satellite remote sensing images or on-site investigations.
[0009] Furthermore, the model parameter empirical calculation formula updating method is an iterative updating method based on the Bayesian parameter learning algorithm.
[0010] Furthermore, each iterative update only requires the measured results of one building.
[0011] Furthermore, the direct object of the update is the empirical calculation formula of the model parameters.
[0012] Furthermore, the updated empirical calculation formula for model parameters is used to calculate the model parameters of the unmonitored buildings in the building complex.
[0013] Compared with the prior art, the present invention has the following beneficial effects:
[0014] Only the monitoring data of a limited number of monitoring buildings in the building complex are needed to update the parameters of the earthquake damage analysis model of all buildings in the building complex. Each time the building complex model is updated, only the parameter identification results of one building are required, and the calculation amount is small. There is an initial empirical calculation formula for the model parameters as a constraint. Even when the number of monitored buildings is small, the update results will not have large errors. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 It is a method flow chart of the parameter updating method of the regional building complex earthquake damage analysis model. DETAILED DESCRIPTION
[0016] The following will describe the implementation methods of the present application in detail with the help of accompanying drawings and examples, so that the implementation process of how the present application uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.
[0017] The parameter updating method of regional building group earthquake damage analysis model is as follows: Figure 1 As shown, the following steps include: establishing an empirical calculation formula for key model parameters of a building complex, iteratively updating the empirical calculation formula for model parameters, and updating the model parameters of the building complex. The following uses the basic period, a key parameter for earthquake damage simulation of building structures, as an example to illustrate the method for updating the parameters of the regional building complex earthquake damage analysis model.
[0018] (1) First, based on existing research literature, an empirical calculation formula for the basic period of a building is established. Referring to existing research, the empirical formula for the basic period of reinforced concrete structures is shown in formula (1).
[0019]
[0020] Where H, B are the height and width of the building structure respectively; and b are the empirical formula parameters, which are represented by the parameter vector θ(,b).
[0021] (2) Then, based on the Bayesian parameter learning algorithm, an update formula for the parameters of the basic period experience calculation formula is established. The steps are as follows:
[0022] 1) First, establish the posterior probability distribution of the building basic period and its empirical formula parameters. Specifically, by multiplying the prior probability distribution of the above parameters with the likelihood function, the posterior probability distribution after considering the basic period of monitoring and identification is obtained.
[0023] The prior probability distribution of the building basic period and its empirical formula parameters is shown in formula (2):
[0024] P(T 0 ,θ)=P(T 0 |θ)P(θ) (2)
[0026] Where is the prior probability distribution of the empirical formula parameters.
[0027] Assume that the average value is θ 0 , the covariance matrix is ∑ θ The Gaussian distribution of is, then the probability distribution of is as shown in formula (3):
[0028]
[0029] Where:
[0030]
[0031] Assuming that the distribution in equation (2) is also Gaussian, then:
[0032]
[0033] Where:
[0034]
[0035] Where, is the standard deviation of the error of the basic period empirical formula.
[0036] The likelihood function of the building basic period and its empirical formula parameters is shown in formula (7). Assume that it is a normal distribution with a mean of and a standard deviation of :
[0037]
[0038] Where, is the basic period of the building to be monitored and identified.
[0039]
[0040] The posterior probability distribution of the building basic period and its empirical formula parameters is shown in formula (9):
[0041]
[0042] 2) Then, the posterior expected value of the building basic period and its empirical formula parameters is calculated. Specifically, based on the above posterior probability distribution (Formula (9)), the formula parameter value when the posterior probability is the largest is calculated as the posterior expected value of the building basic period and its empirical formula parameters:
[0043] Assuming the posterior probability distribution of the building basic period and its empirical formula parameters to be normal distribution, equation (9) is equivalent to:
[0044]
[0045] Where:
[0046] To obtain the minimum value, calculate the first-order partial derivative of each variable and set the partial derivative to 0, and obtain the following three equations:
[0047]
[0048]
[0049]
[0050] By solving the three-variable linear equation composed of equations (12)-(14), we can obtain the updated building basic period and the posterior expectation of the empirical formula parameters.
[0051] 3) Finally, the posterior standard deviation of , , and b is calculated for the next parameter update calculation.
[0052] (3) Finally, the updated empirical formula parameter,b,is used to calculate the basic period of each building based on its geometric information.
[0053] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any simple modification, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention are within the scope of the technical solution of the present invention.
Claims
1. A method for updating parameters of a regional building complex earthquake damage analysis model, including an empirical calculation formula for key model parameters of a building complex, and an updating method thereof, as well as a method for updating model parameters of a building complex, characterized in that: The calibration and updating of the building complex model parameters are achieved by gradually updating the empirical calculation formula of the model parameters using the measured model parameters of some monitored buildings in the building complex.
2. The method for updating parameters of a regional building complex earthquake damage analysis model according to claim 1, characterized in that: The model parameter empirical calculation formula is a calculation formula for calculating the building model parameters using building geometric feature parameters and building basic attribute parameters.
3. The building geometric characteristic parameters and building basic attribute parameters according to claim 2, characterized in that: The parameters can be directly obtained through satellite remote sensing images or on-site surveys.
4. The method for updating parameters of a regional building complex earthquake damage analysis model according to claim 1, characterized in that: The model parameter empirical calculation formula updating method is an iterative updating method based on the Bayesian parameter learning algorithm.
5. The method for updating parameters of a regional building complex earthquake damage analysis model according to claim 1, characterized in that: Each iterative update only requires the measured results of one building.
6. The method for updating parameters of a regional building complex earthquake damage analysis model according to claim 1, characterized in that: The direct object of the update is the empirical calculation formula of the model parameters.
7. The method for updating parameters of a regional building complex earthquake damage analysis model according to claim 1, characterized in that: The updated empirical calculation formula for model parameters is used to calculate the model parameters of unmonitored buildings in the building complex.