A method for constructing and modifying a simplified model of a single building for calculating seismic response

Abstract

The present invention proposes a method for constructing and correcting a simplified model of a single building for calculating earthquake response, comprising: step 1: consulting architectural drawings or public data to obtain basic information of the building structure; step 2: constructing an initial simplified model of a single building based on the basic information of the building structure; step 3: obtaining monitoring information based on seismic sensors and preprocessing the monitoring information; step 4: correcting the initial simplified model of the single building in stages using a particle swarm optimization algorithm to obtain a simplified model of the single building. Step 5: inputting earthquake motion time history records into the simplified model of the single building to calculate the elastic and elastoplastic seismic responses of each layer of the structure, and evaluating the correction results. The present invention is based on the monitoring data correction model, which is more suitable for calculating the earthquake response of buildings with insufficient structural information. At the same time, the particle swarm algorithm in the model correction stage adopts parallel calculation to determine the structural parameters, which significantly shortens the calculation time.

Description

Technical Field

[0001] The invention relates to a method for constructing and correcting a simplified model of a single building for calculating earthquake response, and belongs to the technical field of civil engineering. Background Art

[0002] Earthquakes are one of the most common and harmful natural disasters in the world. Destructive earthquakes often bring serious economic losses and casualties to modern cities. Building structures are the basic facilities of cities with high population density. According to statistics, in the 130 major earthquake disasters in the world, 90%-95% of the casualties were caused by building collapse. Therefore, after an earthquake, how to timely and accurately assess the damage state of building structures is of great significance for personnel rescue and building maintenance.

[0003] At present, the calculation of the impact of earthquake disasters is mainly concentrated in the numerical simulation stage. By establishing a model equivalent to the structure and inputting the earthquake motion to calculate the response of the structure to predict the damage state of the structure. The simplified model is the most common single-building model. Due to its advantages such as simple model, few parameters and fast calculation speed, it has become the choice of most researchers to calculate the seismic response of the structure. However, since the model has been greatly simplified and the structural parameters in the model are difficult to accurately estimate, the calculation results of the model often have a large error from the actual situation. In addition, due to factors such as the aging of building materials, damage to the structure caused by earthquakes, and reinforcement and renovation of buildings, the mechanical properties of the building structure are in dynamic change.

[0004] With the development of new generation information technology, sensors and computers have provided new and powerful tools for quickly and accurately constructing urban earthquake damage scenarios. In recent years, monitoring data has been widely used in civil engineering research, such as dynamic response prediction, finite element model update, structural health monitoring and damage detection. Model update aims to "correct" or "update" the initial finite element model based on data processing. Existing research mainly focuses on bridges and single buildings. The main purpose of model update is to modify the iterative update parameters to produce a structural model that better reflects the monitoring data than the initial model. At present, important single structures, especially for super-high-rise and multi-story structures for public services, are equipped with a certain number of acceleration sensors on specific layers, which can provide real-time monitoring data of structural response after an earthquake. Therefore, how to use monitoring data to update the finite element model so that it can be closer to the actual building structure, thereby further improving the accuracy of earthquake damage simulation results, is particularly important for earthquake disaster simulation. Summary of the invention

[0005] The present invention aims to solve the technical problem of low accuracy of earthquake damage simulation results in the prior art, and further proposes a method for constructing and correcting a simplified model of a single building for calculating earthquake response.

[0006] The technical solution adopted by the present invention to solve the above problems is: the present invention proposes a method for constructing and correcting a simplified model of a single building for calculating seismic response, the steps comprising:

[0007] Step 1: Consult architectural drawings or public data to obtain basic information about the building structure;

[0008] Step 2: Construct an initial simplified model of a single building based on the basic information of the building structure;

[0009] Step 3: Acquire monitoring information based on seismic sensors and pre-process the monitoring information;

[0010] Step 4: Based on the pre-processed monitoring information, the natural vibration period of the building structure and the seismic motion time history records are obtained, and the initial single building simplified model is corrected in stages using the particle swarm optimization algorithm to obtain a single building simplified model;

[0011] Step 5: Input the seismic time history records into the simplified model of the single building to calculate the elastic and elastoplastic seismic responses of each layer of the structure, and evaluate the modified results.

[0012] Optionally, the basic information of the building structure in step 1 includes building height, number of building floors, basic building size, structure type, design intensity, site category and design earthquake grouping.

[0013] Optionally, the initial single building simplified model in step 2 includes an elastic stage and an elastoplastic stage, the elastic stage is used to determine the elastic initial bending stiffness and initial shear stiffness of the building under a small earthquake, and the elastoplastic stage is used to determine the elastic bending stiffness, shear stiffness and structural skeleton curve of the building under a large earthquake.

[0014] Optionally, in step 3, the seismic sensors are divided into top-layer sensors and ground-layer sensors;

[0015] The step of preprocessing the monitoring information includes: performing baseline correction and filtering on the acceleration data obtained by the seismic sensor, removing noise in the monitoring information, and obtaining the preprocessed monitoring information.

[0016] Optionally, the step of obtaining the natural vibration period of the building structure and the seismic motion time history record in step 4 includes:

[0017] Step 4.1: Perform modal analysis on the pre-processed acceleration data measured by the top-floor sensor to obtain the natural vibration periods of each order of the building structure;

[0018] Step 4.2: The preprocessed acceleration data measured by the ground sensor is used as the ground motion time history record.

[0019] Optionally, the steps of modifying the initial single building simplified model in stages in step 4 include:

[0020] Step 4.1: For high-rise bending-shear structures, the particle swarm objective function is the square sum of the errors of the first five natural vibration periods, and the optimization parameters are the bending stiffness and shear stiffness of each layer. The optimization parameter range and the number of particle swarms are determined and iterative updates are performed until the number of updates reaches the preset value, and the correction of the elastic stage is completed; for multi-layer shear structures, the particle swarm objective function is the first-order periodic error of the structure, and the optimization parameters are the bending stiffness and shear stiffness of each layer. The optimization parameter range and the number of particle swarms are determined and iterative updates are performed until the number of updates reaches the preset value, and the correction of the elastic stage is completed;

[0021] Step 4.2: Based on step 4.1, the optimal value of elastic stiffness of each layer of the building structure is obtained, and the elastic-plastic parameters are optimized while keeping the elastic stiffness unchanged. The objective function is to minimize the difference between the model calculation and the structure, that is, the sum of square errors of the time history and maximum values ​​of the acceleration and displacement of the top layer of the structure. The optimization parameters are yield strength ratio, peak strength ratio, post-yield stiffness reduction coefficient, ultimate strength reduction coefficient, ultimate displacement / rotation coefficient and hysteresis coefficient. The optimization parameter range and the number of particle swarms are determined and iterative updates are performed until the number of updates reaches the preset value to complete the correction of the elastic-plastic stage.

[0022] Optionally, the step of evaluating the correction result includes:

[0023] Step 5.1: Input the earthquake time history records into the simplified model of the single building before correction, in the corrected elastic stage and in the corrected elastoplastic stage to calculate the earthquake time history response and maximum value of each layer of the building structure respectively;

[0024] Step 5.2: Compare the time history responses of the top layer acceleration and displacement of the structure monitoring before and after correction, and evaluate the model correction results from the time history perspective. At the same time, compare the peak acceleration and peak displacement of the structure monitoring layer, and evaluate the model correction results from the peak perspective.

[0025] The beneficial effects of the present invention are:

[0026] 1. The present invention is based on easily determined structural information and parameters for modeling and correction, which fits the actual situation and is convenient for subsequent promotion and application.

[0027] 2. The present invention is based on the monitoring data correction model and is more suitable for earthquake response calculation of buildings with insufficient structural information. At the same time, the particle swarm algorithm in the model correction stage adopts parallel calculation to determine the structural parameters, which significantly shortens the calculation time.

[0028] 3. The present invention is verified on high-rise bending-shear structures and multi-layer shear structures. The accuracy of elastic seismic response and displacement response is good. Under a certain accuracy, it can be used to calculate elastic-plastic seismic response. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A flowchart of a method for constructing and modifying a simplified model of a single building for calculating seismic response provided by an embodiment of the present invention;

[0030] Figure 2 A comparison diagram of the acceleration time history response of the top floor of a high-rise bending-shear structure provided by an embodiment of the present invention;

[0031] Figure 3 A comparison diagram of the time-history response of the top floor displacement of a high-rise bending-shear structure provided by an embodiment of the present invention;

[0032] Figure 4 A comparison diagram of peak acceleration and peak displacement of each layer of a high-rise bending-shear structure provided by an embodiment of the present invention;

[0033] Figure 5 A comparison diagram of the acceleration time-history response of the top layer of the multi-layer shear structure provided by an embodiment of the present invention;

[0034] Figure 6 A comparison diagram of the acceleration time-history response of the top layer of the multi-layer shear structure provided by an embodiment of the present invention;

[0035] Figure 7 A comparison diagram of the peak acceleration and peak displacement of each layer of the multi-layer shear structure provided in an embodiment of the present invention. Example

[0036] Combination Figure 1-7 This embodiment is described as follows. Figure 1 As shown, the method for constructing and correcting a simplified model of a single building for calculating seismic response described in this embodiment includes:

[0037] S1: Establish an initial simplified model of a single building based on the basic information of the structure;

[0038] S101: Obtain basic information about the structure by consulting relevant information: building height, number of floors, basic building dimensions, structure type, fortification intensity, site category, and design earthquake grouping.

[0039] S102: The establishment of a simplified model can be divided into elastic and elastoplastic stages. For small earthquakes, the structure does not enter the plastic stage, and only elastic modeling is required to determine the elastic bending stiffness and shear stiffness. For large earthquakes, the structure enters the plastic deformation stage and elastoplastic modeling is required. In addition to determining the elastic stiffness, the structural skeleton curve still needs to be determined.

[0040] The specific process of elasticity modeling is as follows:

[0041] S10201: Estimate the fundamental period T1 of the structure according to an empirical formula, and determine the first-order circular frequency ω1 of the structure according to ω1=2π / T1;

[0042] S10202: Estimate the bending-shear stiffness ratio α0 according to the building structure type. The value range of the bending-shear stiffness ratio is shown in Table 1. According to formula (1), the eigenvalue γ1 corresponding to the first-order vibration mode is determined;

[0043] S10203: Determine the initial bending stiffness EI of the structure according to the first-order circular frequency ω1 of the structure and formula (2), and then determine the initial shear stiffness GA of the structure according to formula (3).

[0044]

[0045] Table 1

[0046]

[0047] For elastic-plastic modeling, it is necessary to determine the structural skeleton curve. Bilinear and trilinear curves are the most commonly used skeleton curves for simulating inter-floor hysteresis behavior. Compared with the bilinear skeleton curve, the trilinear skeleton curve contains more parameters that need to be calibrated and can better represent various nonlinear behaviors. Therefore, this embodiment uses the trilinear curve as the skeleton curve of the bending spring and the shear spring. The specific process is as follows:

[0048] S10204: Modeling of elastic phase and model modification;

[0049] S10205: Determine the design strength of each layer of the structure according to the mode decomposition method in China's "Code for Seismic Design of Buildings" (GB50011-2012);

[0050] S10206: Determine the structural yield point, peak point and limit point. The yield point is determined by the yield strength ratio and elastic stiffness, see equations (4) to (7); the peak point is determined by the peak strength ratio and the post-yield stiffness reduction factor, see equations (8) to (11); the limit point is determined by the ultimate strength reduction factor and the ultimate displacement coefficient / ultimate rotation coefficient, see equations (12) to (15);

[0051] V y =Ω y,V V d (4)

[0052]

[0053] M y =Ω y,M M d (6)

[0054]

[0055] In formulas (4)-(7): V d and M d are the design shear force and design bending moment respectively; Vy and M y are yield shear force and yield bending moment respectively; D y and θ y are the yield displacement and yield rotation respectively; Ω y,V and Ω y,M are the yield shear force ratio and yield bending moment ratio, GA and EI are the initial shear stiffness and bending stiffness, respectively; h is the interlayer height.

[0056] V p =Ω p,V V y (8)

[0057]

[0058] M p =Ω p,M M y (10)

[0059]

[0060] In formulas (8)-(11): V p and M p are the peak shear force and peak bending moment respectively; D p and θ p are the peak displacement and peak rotation angle respectively; Ω p,V and Ω p,M are the peak shear force ratio and peak bending moment ratio respectively; η is the stiffness reduction coefficient after yielding.

[0061] V u =γ V V P (12)

[0062] D u =μ V D P (13)

[0063] M u =γ M M P (14)

[0064] θ u =μ M θ P (15)

[0065] In formulas (12)-(15): V u and M u are the ultimate shear force and ultimate bending moment respectively; D u and θ u are the limit displacement and limit rotation angle respectively; γ V and γ Mare the ultimate shear force ratio and the ultimate bending moment ratio respectively; μ V and μ M are the limit displacement coefficient and the limit rotation coefficient respectively.

[0066] S10207: Since the determination of the yield point, peak point and limit point on the structural skeleton curve is often based on statistical data, the present invention does not determine the initial value. By giving the value range of each parameter and based on the monitoring data, an optimization method is used to determine the optimal value of each parameter.

[0067] The simplified model of the embodiment of the present invention is implemented using the opensees program, and each layer of the structure is simulated using a two-dimensional twoNodeLink unit. The mechanical behaviors in the three degrees of freedom directions are defined by materials (x, y, and rotation around z). Except for the shear stiffness of the shear springs (y direction) and the bending stiffness of the bending springs (rotation around z) of each layer, which are actual stiffnesses, the stiffness in other directions is set to a maximum value (1.0e20), and the damping is Rayleigh damping.

[0068] S2: Process the monitoring data and use the particle swarm optimization algorithm to correct and simplify the model in stages.

[0069] S201: Data processing. First, the acceleration data obtained by the seismic sensor is baseline corrected and filtered to remove noise. The top sensor acceleration data is modally analyzed to obtain the natural vibration periods of each order of the structure as the real period of the structure.

[0070] S202: Elastic stage model correction. The model correction parameters are the bending stiffness and shear stiffness of each layer of the structure. The particle swarm optimization algorithm is used, and the optimization parameters are the model correction parameters. The objective function is to minimize the difference between the model and the structure.

[0071] In this embodiment, for high-rise bending-shear structures, the particle swarm objective function is the sum of square errors of the first five cycles, the optimization parameters are the bending stiffness and shear stiffness of each layer, the optimization range is (0.01*initial stiffness, 10*initial stiffness), the number of particle swarms is 30, the number of iterations is 100, the value range of inertia weight is (0.9-0.4), the value range of learning factor is (2-0.5), and the value range of learning factor is (0.5-2). For multi-layer shear structures, the particle swarm objective function is the first-order periodic error of the structure, the optimization parameters are the bending stiffness and shear stiffness of each part, the optimization range is (0.01*initial stiffness, 10*initial stiffness), the number of particle swarms is 30, the number of iterations is 100, the value range of inertia weight is (0.9-0.4), the value range of learning factor is (2-1.5), and the value range of learning factor is (1.5-2).

[0072] S203: Model correction in the elastic-plastic stage. First, the optimal value of elastic stiffness of each layer of the structure is obtained according to the correction of the elastic stage model, and the structural skeleton curve is determined under the premise of keeping the elastic stiffness unchanged. The model correction parameters are the parameters for determining the structural skeleton curve: yield strength ratio, peak strength ratio, post-yield stiffness reduction coefficient, ultimate strength reduction coefficient, ultimate displacement / rotation coefficient and the coefficients for determining the structural hysteresis model: pinchx, pinchy (parameters representing hysteresis performance in opensees). The particle swarm optimization algorithm is used, and the optimization parameters are the model correction parameters. The objective function is to minimize the difference between the model calculation and the structure.

[0073] In the embodiment of the present invention, the particle swarm objective function is the time history of the top acceleration and displacement of the structure and the sum of square errors of the maximum values. The optimization parameters and optimization ranges are: yield strength ratio (1.1, 10), peak strength ratio (1.1, 10), post-yield stiffness reduction coefficient (0.1, 1), ultimate strength reduction coefficient (0.1, 1), ultimate displacement / rotation coefficient (1.1, 10) and hysteresis coefficient (0, 1). The other parameter settings of the particle swarm algorithm are the same as those in the elastic stage.

[0074] S3: Based on the seismic time history records of the building structure, the modified model is input to calculate the elastic and elastoplastic seismic responses of each layer of the structure, and the results are compared with those before and after modification and structural monitoring to evaluate the optimization results.

[0075] S301: Using the acceleration data of the ground sensor as the seismic time history record, inputting the pre-correction, corrected elastic stage and corrected elastoplastic stage models to respectively calculate the seismic time history response and maximum value of each layer of the structure.

[0076] S302: Compare the time history responses of the top acceleration and displacement of the structure monitoring layer before and after correction, and evaluate the model correction result from the time history perspective. At the same time, compare the peak acceleration and peak displacement of the structure monitoring layer, and evaluate the model correction result from the peak perspective.

[0077] Substitute the optimized bending stiffness and shear stiffness of each layer of the structure into the simplified model of the single building, use the acceleration data of the ground sensor as the seismic motion, input the models before and after correction to calculate the elastic time-history response and maximum value of each layer of the structure, and input the models after correcting only the elastic stage and correcting the elastic-plastic stage to calculate the elastic-plastic earthquake time-history response and maximum value of each layer of the structure. For the elastic stage model correction, a 54-story steel structure is used for verification. The calculation results of the top acceleration and displacement of the simplified model of the single building are compared with the monitoring data as shown below: Figure 2 He Ru Figure 3 As shown, Figure 4The comparison between the peak acceleration and peak displacement of the structure monitoring layer and the monitoring data is shown. Among them, the average error of the peak acceleration of each layer of the structure is 0.061, and the average relative error is 7.34%; the average error of the peak displacement is 0.018, and the average relative error is 5.30%. It can be seen that the optimization result of the displacement is significantly better than the acceleration. From the perspective of time history, this method can better predict the elastic seismic response of high-rise structures. For the model correction in the elastic-plastic stage, a 7-story reinforced concrete structure is used for verification. The calculation results of the acceleration and displacement of the top floor of the model are compared with the monitoring data as shown in the following figure. Figure 5 He Ru Figure 6 As shown, Figure 7 The comparison between the peak acceleration and peak displacement of the structure monitoring layer and the monitoring data is shown. Among them, the average error of the peak acceleration of each layer of the structure is 0.504, and the average relative error is 11.03%; the average error of the peak displacement is 0.011, and the average relative error is 12.13%. It can be seen that the optimization effect of the model in the elastic-plastic stage is slightly worse than that in the elastic stage. However, under certain accuracy requirements, this method can be used to predict the elastic-plastic seismic response of the structure.

[0078] The above is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this profession can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent replacement and improvement made to the above embodiments without departing from the content of the technical solution of the present invention, based on the technical essence of the present invention, within the spirit and principles of the present invention, still fall within the protection scope of the technical solution of the present invention.

Claims

1. A method for constructing and modifying a simplified model of a single building for calculating seismic response, characterized in that: The steps of constructing and correcting a simplified model of a single building for calculating seismic response include: Step 1: Consult architectural drawings and public data to obtain basic information about the building structure; Step 2: constructing an initial simplified model of a single building based on the basic information of the building structure; The initial single building simplified model includes an elastic stage and an elastic-plastic stage, wherein the elastic stage is used to determine the elastic initial bending stiffness and initial shear stiffness of the building under a small earthquake, and the elastic-plastic stage is used to determine the elastic bending stiffness, shear stiffness and structural skeleton curve of the building under a large earthquake; Step 3: Acquire monitoring information based on the seismic sensor and pre-process the monitoring information; Step 4: Based on the pre-processed monitoring information, the natural vibration period of the building structure and the earthquake time history record are obtained, and the initial single building simplified model is corrected in stages using the particle swarm optimization algorithm to obtain a single building simplified model; The steps to modify the initial single building simplified model in stages include: Step 4.1: For high-rise bending-shear structures, the particle swarm objective function is the square sum of the errors of the first five natural vibration periods, and the optimization parameters are the bending stiffness and shear stiffness of each layer. The optimization parameter range and the number of particle swarms are determined and iterative updates are performed until the number of updates reaches the preset value, and the correction of the elastic stage is completed; for multi-layer shear structures, the particle swarm objective function is the first-order periodic error of the structure, and the optimization parameters are the bending stiffness and shear stiffness of each layer. The optimization parameter range and the number of particle swarms are determined and iterative updates are performed until the number of updates reaches the preset value, and the correction of the elastic stage is completed; Step 4.2: Based on step 4.1, the optimal value of elastic stiffness of each layer of the building structure is obtained, and the elastic-plastic parameters are optimized under the premise of keeping the elastic stiffness unchanged. The objective function is to minimize the difference between the model calculation and the structure, that is, the sum of square errors of the time history and maximum values ​​of the acceleration and displacement of the top layer of the structure. The optimization parameters are yield strength ratio, peak strength ratio, post-yield stiffness reduction factor, ultimate strength reduction factor, ultimate displacement / rotation coefficient and hysteresis coefficient. The optimization parameter range and the number of particle swarms are determined and iterative updates are performed until the number of updates reaches the preset value to complete the correction of the elastic-plastic stage; Step 5: Input the earthquake motion time history record into the simplified model of the single building to calculate the elastic and elastoplastic seismic response of each layer of the structure, and evaluate the modified results.

2. The method for constructing and modifying a simplified model of a single building for calculating seismic response according to claim 1, characterized in that: The basic information of the building structure in step 1 includes building height, number of building floors, basic building size, structure type, design intensity, site category and design earthquake grouping.

3. The method for constructing and modifying a simplified model of a single building for calculating seismic response according to claim 1, characterized in that: In step 3, the seismic sensors are divided into top-level sensors and ground-level sensors; The step of preprocessing the monitoring information includes: performing baseline correction and filtering processing on the acceleration data obtained by the seismic sensor to remove noise in the monitoring information and obtain preprocessed monitoring information.

4. The method for constructing and modifying a simplified model of a single building for calculating seismic response according to claim 1, characterized in that: The steps of obtaining the natural vibration period and earthquake time history records of the building structure in step 4 include: Perform modal analysis on the pre-processed acceleration data measured by the top-floor sensor to obtain the natural vibration periods of each order of the building structure; The preprocessed acceleration data measured by the ground sensors are used as the seismic time history records.

5. The method for constructing and modifying a simplified model of a single building for calculating seismic response according to claim 1, characterized in that: The steps for evaluating the correction results in step 5 include: Step 5.1: Input the earthquake time history records into the simplified model of the single building before correction, in the corrected elastic stage and in the corrected elastoplastic stage to calculate the earthquake time history response and maximum value of each layer of the building structure respectively; Step 5.2: Compare the time history responses of the top layer acceleration and displacement of the structure monitoring before and after correction, and evaluate the model correction results from the time history perspective. At the same time, compare the peak acceleration and peak displacement of the structure monitoring layer, and evaluate the model correction results from the peak perspective.

Images