Method and system for optimizing overall performance of fabricated concrete frame structure

Abstract

The invention relates to a method and system for optimizing the overall performance of a fabricated concrete frame structure. The method for optimizing the overall performance of the fabricated concrete frame structure comprises the steps that a multi-scale digital twinborn model is constructed; obtaining design parameters; performing sensitivity analysis on the design parameters to screen key parameters; performing multi-objective optimization on the key parameters to obtain an optimal solution design scheme combination; performing construction feasibility verification on the optimal solution design scheme combination to obtain an optimal design scheme; construction is conducted based on the optimal design scheme, real-time monitoring analysis is conducted in the construction process, and construction is optimized and adjusted based on the real-time monitoring analysis result. The problem of overall performance coordination of the fabricated concrete in the design and construction stages can be solved, and therefore the overall performance and reliability of the fabricated concrete frame structure are remarkably improved.

Description

Technical Field

[0001] This application relates to the field of building performance optimization technology, and in particular to a method and system for optimizing the overall performance of prefabricated concrete frame structures. Background Technology

[0002] Prefabricated concrete frame structures are widely used in the construction industry, and their high degree of industrialization and short construction cycle significantly improve construction efficiency. However, these structures face a key challenge in actual engineering: due to insufficient coordination between the stiffness of the joints and the overall structure after component assembly, stress concentration is prone to occur, leading to a decline in the overall structural performance. This problem stems from the fragmentation of data across the design, construction, and operation stages, and the lack of a unified digital platform to span the entire lifecycle. Current technology cannot accurately predict the impact of construction errors on overall performance during the design phase, nor can it effectively assess the structural integrity and perform dynamic optimization in real time during construction. This limitation prevents the full potential of prefabricated concrete frame structures from being realized, necessitating a solution that can integrate data from the entire process and achieve collaborative optimization. Summary of the Invention

[0003] Therefore, it is necessary to provide a method and system for optimizing the overall performance of prefabricated concrete frame structures to address the problems in related technologies.

[0004] To achieve the above objectives, in a first aspect, this application provides a method for optimizing the overall performance of a prefabricated concrete frame structure, the method comprising: Constructing a multi-scale digital twin model includes: integrating multi-source data including parametric data structures, finite element analysis results, manufacturing process data, and 4D construction progress data, wherein the parametric data structures are obtained by converting geometric information from the BIM model; and constructing a multi-scale digital twin model based on the multi-source data, wherein the multi-scale digital twin model includes a macroscopic framework model, a mesoscopic node model, and a microscopic material model. Obtain design parameters; perform sensitivity analysis on the design parameters to screen key parameters; perform multi-objective optimization on the key parameters to obtain the optimal solution design scheme combination; verify the construction feasibility of the optimal solution design scheme combination to obtain the optimal design scheme. Construction is carried out based on the optimal design scheme, and real-time monitoring and analysis are conducted during the construction process. The construction is then optimized and adjusted based on the results of the real-time monitoring and analysis.

[0005] The aforementioned method for optimizing the overall performance of prefabricated concrete frame structures involves constructing a multi-scale digital twin model, performing sensitivity analysis on design parameters to screen key parameters, conducting multi-objective optimization of these key parameters to obtain the optimal solution design scheme combination, and verifying the construction feasibility of the optimal solution design scheme combination to obtain the optimal design scheme. This method enables performance prediction and optimization during the design phase. Furthermore, by conducting real-time monitoring and analysis during construction, and optimizing and adjusting the construction based on the results, the overall performance of the prefabricated concrete frame structure can be evaluated in real-time and dynamically optimized during construction. This method can solve the overall performance coordination problem of prefabricated concrete in the design and construction phases, thereby significantly improving the overall performance and reliability of prefabricated concrete frame structures.

[0006] In some embodiments, the integration includes multi-source data such as parametric data structures, finite element analysis results, manufacturing process data, and 4D construction progress data, including: An intermediate data format converter based on the IFC standard is used to convert the geometric information in the BIM model into the parametric data structure. The finite element analysis results are stored in HDF5 format, preserving field variable data including stress, strain, and displacement. Manufacturing process data, including the processing precision of components and the specifications of connectors in prefabricated concrete frame structures, are recorded using XML format. By associating timestamps with spatial coordinates, 4D construction progress data including a spatiotemporal matrix is ​​formed, where each row of the spatiotemporal matrix records the three-dimensional coordinates of the component at time t.

[0007] In some embodiments, constructing a multi-scale digital twin model based on the multi-source data includes: A beam-shell hybrid element is used to establish the overall frame to obtain the macroscopic frame model; The mesoscopic node model is obtained by using entity-contact hybrid modeling for the node region; The strength dispersion of concrete material is described using the Weibull distribution to obtain the micromaterial model.

[0008] In some embodiments, after constructing a multi-scale digital twin model based on the multi-source data, the method further includes: Establish a parameter mapping table, which includes a one-to-one correspondence between design variables and mechanical parameters in the multi-scale digital twin model; When the design variables change, the multi-scale digital twin model is updated using the region decomposition method based on the changes in the design variables.

[0009] In some embodiments, the sensitivity analysis of the design parameters to screen key parameters includes: Establish the parameter influence matrix; The design parameters are classified based on the parameter influence matrix to filter the key parameters.

[0010] In some embodiments, the multi-objective optimization of the key parameters to obtain the optimal solution design scheme combination includes: Construct the objective function vector of the key parameters; The improved MOEA / D-DE algorithm is used to optimize the objective function vector to obtain the optimal solution design scheme combination.

[0011] In some embodiments, the step of verifying the construction feasibility of the optimal solution design scheme combination to obtain the optimal design scheme includes: Establish construction constraints; Monte Carlo sampling is used to sample the optimal solution design scheme combination to generate candidate design schemes; The feasibility of each candidate design scheme is verified through 4D construction simulation in order to obtain a feasible candidate design scheme. Calculate the feasibility index of each feasible candidate design scheme, and obtain the optimal design scheme based on the feasibility index.

[0012] In some embodiments, the real-time monitoring and analysis during construction includes: Data is collected from the components assembled during construction using multi-source sensors to obtain measured point clouds; A non-rigid registration relationship is established between the measured point cloud and the multi-scale digital twin model to determine whether there are assembly deviations in the component; When assembly deviations exist in the components, an error propagation model is constructed based on a graph neural network to predict the chain effect of the component deviations on the prefabricated concrete frame structure.

[0013] In some embodiments, the step of acquiring data from components assembled during construction using multi-source sensors to obtain measured point clouds includes: Construct a spatial positioning network; Obtain the static and dynamic attitude data of the component; The measured point cloud is obtained based on the spatial positioning network, the static data, and the dynamic attitude data.

[0014] Secondly, this application also provides a system for optimizing the overall performance of prefabricated concrete frame structures, the system comprising: A multi-scale digital twin model construction module is used to construct a multi-scale digital twin model, including: integrating multi-source data including parametric data structures, finite element analysis results, manufacturing process data, and 4D construction progress data, wherein the parametric data structures are obtained by converting geometric information in the BIM model; and constructing a multi-scale digital twin model based on the multi-source data, wherein the multi-scale digital twin model includes a macroscopic framework model, a mesoscopic node model, and a microscopic material model. The optimal design scheme acquisition module is used to acquire design parameters; perform sensitivity analysis on the design parameters to screen key parameters; perform multi-objective optimization on the key parameters to obtain the optimal solution design scheme combination; and perform construction feasibility verification on the optimal solution design scheme combination to obtain the optimal design scheme. The construction monitoring, analysis and optimization module is used to carry out construction based on the optimal design scheme, perform real-time monitoring and analysis during the construction process, and optimize and adjust the construction based on the real-time monitoring and analysis results.

[0015] The aforementioned optimization system for the overall performance of prefabricated concrete frame structures can achieve performance prediction and optimization during the design phase, and can conduct real-time evaluation and dynamic optimization of the overall performance of prefabricated concrete frame structures during construction. It can solve the coordination problem of the overall performance system in the design and construction phases of prefabricated concrete, thereby significantly improving the overall performance and reliability of prefabricated concrete frame structures. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the conventional technology, the drawings used in the description of the embodiments or the conventional technology will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of a method for optimizing the overall performance of a prefabricated concrete frame structure provided in one embodiment of this application; Figure 2 This is a structural block diagram of a system for optimizing the overall performance of a prefabricated concrete frame structure, provided in another embodiment of this application.

[0018] Figure labeling: 10, Multi-scale digital twin model construction module; 20, Optimal design scheme acquisition module; 30, Construction monitoring, analysis and optimization module. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0020] In one embodiment, see Figure 1 This application also provides a method for optimizing the overall performance of a prefabricated concrete frame structure, the method comprising the following steps: S10~S30.

[0021] S10: Construct a multi-scale digital twin model, including: integrating multi-source data including parametric data structure, finite element analysis results, manufacturing process data and 4D construction progress data, wherein the parametric data structure is obtained by converting geometric information in the BIM model; constructing a multi-scale digital twin model based on the multi-source data, wherein the multi-scale digital twin model includes a macroscopic framework model, a mesoscopic node model and a microscopic material model.

[0022] S20: Obtain design parameters based on the multi-scale digital twin model; perform sensitivity analysis on the design parameters to screen key parameters; perform multi-objective optimization on the key parameters to obtain the optimal solution design scheme combination; verify the construction feasibility of the optimal solution design scheme combination to obtain the optimal design scheme.

[0023] S30: Carry out construction based on the optimal design scheme, conduct real-time monitoring and analysis during the construction process, and optimize and adjust the construction based on the results of the real-time monitoring and analysis.

[0024] The method for optimizing the overall performance of prefabricated concrete frame structures in this embodiment involves constructing a multi-scale digital twin model, performing sensitivity analysis on design parameters to screen key parameters, performing multi-objective optimization on these key parameters to obtain the optimal solution design scheme combination, and verifying the construction feasibility of the optimal solution design scheme combination to obtain the optimal design scheme. This method enables performance prediction and optimization during the design phase. Furthermore, by conducting real-time monitoring and analysis during construction, and optimizing and adjusting the construction based on the results, the overall performance of the prefabricated concrete frame structure can be evaluated in real-time and dynamically optimized during construction. This method can solve the overall performance coordination problem of prefabricated concrete in the design and construction phases, thereby significantly improving the overall performance and reliability of prefabricated concrete frame structures.

[0025] In step S10, please refer to Figure 1Step S10 of the process involves constructing a multi-scale digital twin model, which includes: integrating multi-source data including parametric data structures, finite element analysis results, manufacturing process data, and 4D construction progress data, wherein the parametric data structures are obtained by converting geometric information from the BIM model; and constructing a multi-scale digital twin model based on the multi-source data, wherein the multi-scale digital twin model includes a macroscopic framework model, a mesoscopic node model, and a microscopic material model.

[0026] As an example, in step S10, the integration of multi-source data including parametric data structures, finite element analysis results, manufacturing process data, and 4D construction progress data may include the following: using an intermediate data format converter based on the IFC (Industry Foundation Classes) standard to convert the geometric information in the BIM (Building Information Modeling) model into the parametric data structure; storing the finite element analysis results in HDF5 (Hierarchical Data Format Version 5) format, retaining field variable data including stress, strain, and displacement; recording manufacturing process data including component processing accuracy and connector specifications in the prefabricated concrete frame structure in XML (Extensible Markup Language) format; and forming a spatiotemporal matrix by associating timestamps with spatial coordinates. The 4D construction progress data, wherein each row of the spatiotemporal matrix records the three-dimensional coordinates of the component at time t.

[0027] As an example, the geometric information in the BIM model can be target geometric information that matches the prefabricated concrete frame structure, or it can be pre-collected and set geometric information, which can include, but is not limited to, the three-dimensional shape of beams, the cross-sectional profile of columns, etc. The corresponding parametric data structure can include, but is not limited to, length, cross-sectional dimensions, node position coordinates, etc.

[0028] As an example, parameterized data structures can be stored in key-value format.

[0029] As an example, the finite element analysis results can be the analysis results of finite element software (such as ABAQUS or ANSYS), focusing on retaining field variable data directly related to the structural performance of the prefabricated concrete frame structure. Specifically, these can include stress (e.g., Von Mises stress, principal stress, and shear stress), strain (e.g., normal strain, shear strain, and plastic strain), and displacement field (linear displacement, rotational displacement, etc. in the x / y / z directions).

[0030] As an example, manufacturing process data can include data on the machining accuracy of prefabricated components during the factory production stage (e.g., length tolerance, cross-sectional dimension tolerance, surface flatness, etc.) and the specifications of connectors (e.g., bolt type / registration / preload, gusset plate thickness / material / tolerance, etc.); it can also include production information (e.g., processing equipment, production batch, and quality inspection results, etc.).

[0031] As an example, the spacetime matrix In the table, n represents the number of time nodes or components (which can be selected according to the scenario, such as one time node per hour or one record per component). The four columns correspond to the time t, x-axis coordinate, y-axis coordinate, and z-axis coordinate, respectively. That is, each row of records is (t, x, y, z). The time t can be in days, and the coordinates (x, y, z) can be in millimeters.

[0032] As an example, the multi-source data can be used as input data for a multi-scale digital twin model.

[0033] As an example, in step S10, constructing a multi-scale digital twin model based on multi-source data may include the following: establishing an overall framework using beam-shell hybrid elements to obtain the macroscopic framework model; using solid-contact hybrid modeling for the node regions to obtain the mesoscopic node model; and using Weibull distribution to describe the strength dispersion of concrete materials to obtain the microscopic material model.

[0034] As an example, beam elements in the beam-shell hybrid element are used to simulate frame beams and frame columns; the beam element only needs to input the cross-sectional dimensions (such as column 500×500mm) and material properties (such as C50 concrete elastic modulus 34.5GPa) to accurately calculate the bending, axial and shear deformation of the structure, which is highly efficient.

[0035] As an example, shell elements in a beam-shell hybrid element are used to simulate floor slabs and shear walls (if the precast concrete frame structure includes shear walls); shell elements can accommodate both in-plane (axial, shear) and out-of-plane (complete) deformations, making them more accurate than beam elements and more efficient than solid elements.

[0036] As an example, the stiffness matrix in the macroscopic framework model can be obtained by the following formula: in, Here is the stiffness matrix, and n is the number of components. k i Let be the element stiffness matrix of the i-th component. T i Let be the coordinate transformation matrix of the i-th component. Coordinate transformation matrix Ti The transpose of .

[0037] As an example, a solid-contact hybrid modeling approach is used for the node region to obtain the meso-level node model. Specifically, the solid elements in the solid-contact hybrid element model can be used to simulate the concrete in the core area of ​​the node (e.g., the core area of ​​a beam-column node), the node plate (e.g., a steel plate), and bolts, etc. Solid elements can accurately simulate the stress distribution of complex geometries (e.g., stress concentration around bolt holes) and are the core of meso-level node modeling. The contact elements in the solid-contact hybrid element model are used to simulate the contact surface between the bolt and the node plate and the contact surface between the node plate and the concrete. The contact elements can define the friction coefficient (e.g., the friction coefficient of steel-to-steel contact is 0.3) and contact stiffness of the contact surface to avoid stress calculation deviations caused by assuming no slippage at the contact surface.

[0038] As an example, the contact surface stiffness is a key parameter describing the ability of a contact surface to resist relative deformation. The formula for the contact surface stiffness in a mesoscopic nodal model can be as follows: in, k c This refers to the contact surface stiffness (the unit can be N / mm, which is the force required to produce a 1mm relative deformation). E c The elastic modulus of the contact material (e.g., grout). E c =25GPa, steel gasket E c =206GPa). A c This refers to the contact area (e.g., the contact area between the bolt and the gusset plate). t c The thickness of the contact layer (for example, the grout layer is 20mm thick and the steel gasket is 3mm thick).

[0039] As an example, micromaterial models can be used to describe the strength dispersion of concrete materials. The uniaxial compressive strength of concrete of the same batch and strength grade is not a fixed value (e.g., the strength of C30 concrete may fluctuate between 28-35 MPa). This dispersion will lead to differences in the bearing capacity of meso-level nodes, which need to be quantified by micromaterial models.

[0040] As an example, concrete is a multiphase composite material, and its strength follows a Weibull distribution, which better reflects the actual situation that low-strength samples account for a small proportion, high-strength samples account for a small proportion, and medium-strength samples account for a large proportion compared to a normal distribution.

[0041] As an example, in a microscopic material model, the probability density function of the Weibull distribution can be: in, f(s) For intensity s The probability density at time (the unit can be 1 / MPa). s This refers to the uniaxial compressive strength of concrete (the unit can be MPa). m This is a shape parameter (unitless, typically ranging from 3 to 10). s 0 This is the scale parameter (also known as feature intensity, the unit can be MPa). e It is a natural constant.

[0042] As an example, after constructing a multi-scale digital twin model based on the multi-source data in step S10, the following may also be included: establishing a parameter mapping table, which includes a one-to-one correspondence between design variables and mechanical parameters in the multi-scale digital twin model; when the design variables change, updating the multi-scale digital twin model using the region decomposition method based on the changes in the design variables.

[0043] As an example, the parameter mapping table can map design variables to mechanical parameters in a multi-scale digital twin model. For instance, the node plate thickness in the design variable corresponds to the contact surface stiffness in the meso-scale node model, the component length in the design variable corresponds to the element stiffness matrix in the macro-scale frame model, the concrete strength grade in the design variable corresponds to the Weibull distribution in the micro-scale material model, and the bolt preload in the design variable corresponds to the contact properties in the meso-scale node model, and so on.

[0044] As an example, when the design variables change, during the process of updating the multi-scale digital twin model using the domain decomposition method based on the changes in the design variables, taking the stiffness matrix as an example, the stiffness matrix can be updated based on the following formula: in, K new The updated stiffness matrix, K old The stiffness matrix before the update. For the increment of the stiffness matrix, , S This represents the set of affected elements (e.g., elements whose material properties have changed, elements whose stress state has changed, etc.). T j For the first j The left transformation matrix of each unit, This is the transpose of the coordinate transformation matrix. For the first j The increment of the local stiffness matrix of each element.

[0045] In the above example, by establishing a parameter mapping table, when the design variables change, the multi-scale digital twin model is updated based on the changes in the design variables using the region decomposition method. This allows calculations to be performed only on the affected areas of the multi-scale digital twin model, avoiding a full recalculation of the entire multi-scale digital twin model, significantly shortening the optimization cycle and reducing the cost of using the multi-scale digital twin model.

[0046] In step S20, please refer to Figure 1 In step S20, design parameters are obtained based on the multi-scale digital twin model; sensitivity analysis is performed on the design parameters to screen key parameters; multi-objective optimization is performed on the key parameters to obtain the optimal solution design scheme combination; and construction feasibility is verified on the optimal solution design scheme combination to obtain the optimal design scheme.

[0047] As an example, in step S20, obtaining design parameters and performing sensitivity analysis on the design parameters to screen key parameters may include the following: establishing a parameter influence matrix; classifying the design parameters based on the parameter influence matrix to screen the key parameters.

[0048] As an example, the obtained design parameters can be the design parameters to be optimized.

[0049] As an example, construct the parameter influence matrix. , where is the number of design parameters and is the number of performance indicators.

[0050] As an example, the formula for how parameters affect the elements of a matrix can be: in, S ij The parameters affect the elements of the matrix. x i For the i-th design parameter (e.g., the thickness of the gusset plate or the bolt preload, etc.). P j For the j-th performance index (e.g., nodal stiffness or stress concentration factor), the partial derivative is... The absolute rate of change of response, i.e., the change in the j-th performance index when the i-th design parameter changes. P j The absolute range of change.

[0051] As an example, the design parameters are graded based on the parameter influence matrix to filter the key parameters. Specifically, this can be done by: if the absolute values ​​of the elements of the parameter influence matrix are... > First preset value (the first preset value can be, but is not limited to, 0.3), then the i-th parameter at this time This is a key parameter; if the second preset value (which can be, but is not limited to, 0.1) is less than the absolute value of the elements of the parameter-affected matrix... If the first preset value is less than or equal to the i-th parameter, then the i-th parameter at this time... x i These are minor parameters; if the parameter affects the absolute value of the matrix elements... If the value is less than or equal to the second preset value, then the i-th parameter at this time... x i These values ​​can be fixed to empirical values. Key parameters require focused optimization, while minor parameters can be adjusted appropriately.

[0052] In the example above, by performing sensitivity analysis on the design parameters to identify key parameters, the number of subsequent optimization parameters can be greatly reduced, the amount of computation can be significantly reduced, and the optimization efficiency can be significantly improved.

[0053] As an example, in step S20, multi-objective optimization of the key parameters to obtain the optimal solution design scheme combination may include the following: constructing the objective function vector of the key parameters; and using the improved MOEA / D-DE algorithm to optimize the objective function vector to obtain the optimal solution design scheme combination.

[0054] As an example, the constructed objective function vector can be: in, Key parameters x The stiffness matrix (i.e., the overall stiffness, can be obtained from a multi-scale digital twin model) The maximum displacement of all nodes. C total (x) The total project cost.

[0055] As an example, the improved MOEA / D-DE algorithm is used to optimize the objective function vector to obtain the optimal solution design scheme combination, which may include the following: using the Chebyshev decomposition method to transform the multi-objective problem into multiple single-objective sub-problems, the corresponding formulas are as follows: in, The comprehensive deviation objective function for a single subproblem is used to measure the deviation between the current design scheme and the ideal scheme. The smaller the value, the better the comprehensive performance of the scheme. x For example, a key parameter vector. x =[gusset plate thickness d, beam section height h, bolt spacing s], vector of each key parameter xA design scheme is possible; l For the weight vector, λ= [ l 1 , l 2 , l 3 ], l 1 + λ 2 +λ 3 =1, which reflects the degree of importance attached to the three objectives of stiffness, displacement and cost. For example, if the prefabricated concrete frame structure is located in an earthquake zone, then stiffness is the priority; if the prefabricated concrete frame structure is a regular factory building, then cost is the priority. l i Let i be the weight vector; For the ideal target vector, , This can yield the best results for each objective selected from a large number of initial design options. It can be the smallest among all design options. It can be the minimum maximum displacement of the node among all design schemes. It can be the lowest total cost among all design options; It can be the i-th ideal target vector; f i (x) This represents the current value of the i-th objective function; Let be the absolute deviation of the i-th target; The weighted bias for the i-th objective; To take the maximum value among the three weighted deviations.

[0056] As an example, for the key parameters identified through sensitivity analysis, multiple sets of different weight vectors from the above formula can be generated to cover different trade-off strategies; by minimizing each set of weight vectors... This allows us to obtain multiple optimal solution design schemes (each set of weight vectors yields one optimal solution design scheme), thus resulting in a combination of optimal solution design schemes.

[0057] As an example, step S20, which verifies the construction feasibility of the optimal solution design scheme combination to obtain the optimal design scheme, may include the following: establishing construction constraints; using Monte Carlo sampling on the optimal solution design scheme combination to generate candidate design schemes; verifying the feasibility of each candidate design scheme through 4D construction simulation to obtain feasible candidate design schemes; calculating the feasibility index of each feasible candidate design scheme, and obtaining the optimal design scheme based on the feasibility index.

[0058] As an example, the construction constraints can be established as follows: in, Due to time constraints, t assembly (x) To determine the assembly time, the hoisting sequence can be simulated using a spatiotemporal matrix of 4D construction progress data in a multi-scale digital twin model, and the time consumed in each step can be calculated to obtain the total assembly time. T max The longest planned time for the project; For accuracy constraints, The cumulative tolerance can be obtained by predicting the cumulative effect of installation deviation through error propagation analysis of the meso-level node model in the multi-scale digital twin model. To allow the maximum deviation; Due to cost constraints, For special process costs, the costs of special processes such as customized node plates and high-precision bolts can be extracted from the manufacturing process data in the multi-scale digital twin model to obtain the special process costs. To reserve budget for special processes; the project's longest planned time and maximum allowable deviation. and reserve budget for special processes This can be pre-set based on experience.

[0059] As an example, in step S20, Monte Carlo sampling is used on the combination of optimal solution design schemes to generate candidate design schemes. This can evenly cover design schemes with different trade-off strategies, ensuring that no potential feasible schemes are missed, and avoiding the possibility of omissions due to only verifying a few schemes.

[0060] As an example, in step S20, the specific method for verifying the feasibility of each candidate design scheme through 4D construction simulation to obtain a feasible candidate design scheme can be as follows: combining the 3D spatial model and the time schedule, the construction process of the candidate design scheme is pre-drilled in a multi-scale digital twin model to verify whether the above three constraints are met. If so, the corresponding candidate design scheme is feasible.

[0061] As an example, in step S20, the formula for calculating the feasibility index of each feasible candidate design scheme can be as follows: in, FI Feasibility index; If a candidate design violates the constraints, the excess portion is taken; if the constraints are satisfied, 0 is taken. This is the maximum permissible value for the constraint; The degree of satisfaction of a single constraint; The product of the satisfaction of the three constraints.

[0062] Specifically, the optimal design scheme can be selected based on the feasibility index, meaning the candidate design scheme with the highest feasibility index can be chosen as the optimal design scheme. Of course, in other examples, the post-meeting selected scheme with an index greater than a preset value (which can be, but is not limited to, 0.8) can also be recommended as the optimal design scheme.

[0063] In step S30, please refer to Figure 1 In step S30, construction is carried out based on the optimal design scheme, and real-time monitoring and analysis are conducted during the construction process. Based on the results of the real-time monitoring and analysis, the construction is optimized and adjusted.

[0064] As an example, in step S30, the real-time monitoring and analysis during construction may include the following: collecting data from the assembled components during construction using multi-source sensors to obtain a measured point cloud; establishing a non-rigid registration relationship between the measured point cloud and the multi-scale digital twin model to determine whether the components have assembly deviations; and when the components have assembly deviations, constructing an error propagation model based on a graph neural network to predict the chain effect of the component deviations on the prefabricated concrete frame structure.

[0065] As an example, in step S30, the step of collecting data on the components assembled during construction through multi-source sensors to obtain a measured point cloud may include the following: constructing a spatial positioning network; acquiring static data and dynamic attitude data of the components; and obtaining the measured point cloud based on the spatial positioning network, the static data, and the dynamic attitude data.

[0066] As an example, a global coordinate system can be established using a distributed laser scanner array, with each distributed laser scanner covering a spherical region with a radius of 30m. Point cloud registration can then be achieved using the ICP (Iterative Closest Point) algorithm. Where E(R,t) is the total error of point cloud alignment, R is the rotation matrix, and t is the translation vector. p i Let i be the i-th measurement point (which can be in three dimensions) in the distributed laser scanner A (the point cloud to be aligned). q i For the i-th measurement point in the distributed laser scanner B (reference scanner) (and) p i The corresponding matching point is the measurement point of distributed laser scanner A. p iCoordinates after rotation and translation w i is the weighting coefficient, with a value ranging from 0 to 1, and n is the total number of measurement points.

[0067] Specifically, the essence of the ICP algorithm is to find the optimal rotation matrix and translation vector that aligns the point clouds of two distributed laser scanners in the global coordinate system, i.e., minimizing the difference between the two distributed laser scanners after rotation and translation. Specifically, the Levenberg-Marquardt optimization algorithm can be used to solve for the optimal rotation matrix and translation vector.

[0068] As an example, static data such as component ID, production batch, and material parameters can be recorded using RFID tags.

[0069] As an example, acceleration and angular velocity can be measured in real time using an IMU sensor (sampling frequency of 100Hz), and attitude can be calculated using quaternion integration: in, q These are the four attitude elements, used to describe the current spatial attitude of a component. ,satisfy ; The quaternion is the rate of change of time, which describes the instantaneous rate of change of attitude; It is the quaternion tensor product; oh This is the angular velocity vector measured by the IMU, which is a 3×1 vector, and the unit can be rad / s; This is the scaling factor; q k+1 Let be the attitude quaternion at time k+1; and let be the attitude quaternion at time k. The time step is expressed in seconds (s). Let be the rate of change of the four elements at time k.

[0070] As an example, in step S30, establishing a non-rigid registration relationship between the measured point cloud and the multi-scale digital twin model to determine whether the component has assembly deviations may include: establishing a non-rigid registration relationship between the measured point cloud and the multi-scale digital twin model (specifically, BIM model-related parameter data in the multi-scale digital twin model), and using an improved CPD (Coherent Point Drift) algorithm to achieve optimal registration of the virtual and real point clouds by minimizing the following error function: in, This represents the total registration error; i Optimize parameters for the core; p nFor the nth measured point (the point cloud on the surface of the component collected by the distributed laser scanner, such as the corner points of the column or the center of the bolt hole); q m For the m-th model point (the surface point cloud of the multi-scale digital twin model, and...) p n The theoretical point at the corresponding location is also the m-th input point, which is the input to the multi-scale digital twin model. Conditional probability, i.e., the actual measured point p n With model points q m The probability of a match; T( q m , i ) is a thin plate spline transformation (transforming model points) q m Through parameters i (Mapped to predicted points that fit the measured points) The square of the positional deviation between the measured point and the predicted point; This is a regularization term to prevent excessive transformation; l The regularization coefficient is used. G This is the regularized weight matrix; denoted as the trace of the matrix; N represents the number of measured points; and M represents the number of model points.

[0071] As an example, in step S30, when the component has assembly deviations, an error propagation model is constructed based on a graph neural network to predict the chain effect of the component deviations on the prefabricated concrete frame structure. This may include: when the component has assembly deviations, an error propagation model is constructed based on a graph neural network, where nodes represent components and edges represent mechanical connection relationships, and the deviation characteristics of each component are updated using the following formula: in, For nodes v In the l+1 Layer deviation feature vector; It is a non-linear activation function; For nodes v The neighbor feature aggregation operation represents the aggregation of nodes. v All neighboring nodes u The features are weighted and summed to integrate neighbor information into the node. v ; For nodes u In the l Layer deviation feature vector; For the first l Learnable bias vectors of a layer.

[0072] Specifically, the output of each network layer includes key indicators such as displacement deviation and stress redistribution.

[0073] Specifically, when a component has an assembly deviation, an error propagation model based on a graph neural network can be constructed to identify the adjacent components affected by the deviation. During the optimization and adjustment of the construction, in addition to optimizing and adjusting the current component with the assembly deviation, the adjacent components affected by it are also optimized and adjusted accordingly.

[0074] In another embodiment, please refer to Figure 1 See Figure 2 This application also provides an optimization system for the overall performance of prefabricated concrete frame structures. The optimization system includes: a multi-scale digital twin model construction module 10, used to construct a multi-scale digital twin model, comprising: integrating multi-source data including parametric data structures, finite element analysis results, manufacturing process data, and 4D construction progress data, wherein the parametric data structures are obtained by converting geometric information from a BIM model; constructing a multi-scale digital twin model based on the multi-source data, wherein the multi-scale digital twin model includes a macroscopic frame model, a mesoscopic node model, and a microscopic material model; an optimal design scheme acquisition module 20, used to obtain design parameters based on the multi-scale digital twin model; performing sensitivity analysis on the design parameters to screen key parameters; performing multi-objective optimization on the key parameters to obtain an optimal solution design scheme combination; verifying the construction feasibility of the optimal solution design scheme combination to obtain an optimal design scheme; and a construction monitoring, analysis, and optimization module 30, used to carry out construction based on the optimal design scheme, perform real-time monitoring and analysis during construction, and optimize and adjust the construction based on the real-time monitoring and analysis results.

[0075] The aforementioned optimization system for the overall performance of prefabricated concrete frame structures can achieve performance prediction and optimization during the design phase, and can conduct real-time evaluation and dynamic optimization of the overall performance of prefabricated concrete frame structures during construction. It can solve the coordination problem of the overall performance system in the design and construction phases of prefabricated concrete, thereby significantly improving the overall performance and reliability of prefabricated concrete frame structures.

[0076] As an example, the optimization system for the overall performance of the prefabricated concrete frame structure in this embodiment can be used to perform, for example... Figure 1 The method for optimizing the overall performance of the prefabricated concrete frame structure in its corresponding embodiments.

[0077] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features of the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0078] The above embodiments merely illustrate several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for optimizing the overall performance of a fabricated concrete frame structure, characterized by, The method comprises the following steps: Constructing a multi-scale digital twin model, comprising: integrating multi-source data including parameterized data structure, finite element analysis results, manufacturing process data and 4D construction progress data, wherein the parameterized data structure is converted from geometric information in a BIM model; constructing a multi-scale digital twin model based on the multi-source data, wherein the multi-scale digital twin model comprises a macroscopic framework model, a mesoscopic node model and a microscopic material model; Obtaining design parameters; performing sensitivity analysis on the design parameters to screen key parameters; performing multi-objective optimization on the key parameters to obtain an optimal solution design scheme combination; and verifying construction feasibility of the optimal solution design scheme combination to obtain an optimal design scheme; Based on the optimal design scheme, construction is carried out, real-time monitoring and analysis are carried out during construction, and the construction is optimized and adjusted based on the real-time monitoring and analysis results.

2. The method for optimizing the overall performance of a fabricated concrete frame structure according to claim 1, wherein, The multi-source data including parameterized data structure, finite element analysis results, manufacturing process data and 4D construction progress data comprises: An intermediate data format converter based on IFC standard is used to convert the geometric information in the BIM model into the parameterized data structure; The finite element analysis results are stored in HDF5 format, which retains field variable data including stress, strain and displacement; The manufacturing process data including the processing precision of components and the specifications of connecting pieces in the fabricated concrete frame structure are recorded in XML format; The 4D construction progress data is formed by associating time stamps with spatial coordinates to form a space-time matrix, and each row of the space-time matrix records the three-dimensional coordinates of the component at time t.

3. The method for optimizing the overall performance of a fabricated concrete frame structure according to claim 1, wherein The multi-scale digital twin model is constructed based on the multi-source data, comprising: A global framework is established by using beam-shell hybrid elements to obtain the macroscopic framework model; A solid-contact hybrid modeling is used for the node area to obtain the mesoscopic node model; The strength dispersion of the concrete material is described by using Weibull distribution to obtain the microscopic material model.

4. The method for optimizing the overall performance of a fabricated concrete frame structure according to claim 1, wherein After constructing the multi-scale digital twin model based on the multi-source data, the following steps are further included: A parameter mapping table is established, which includes a one-to-one correspondence between design variables and mechanical parameters in the multi-scale digital twin model; When the design variables change, the multi-scale digital twin model is updated based on the changes in the design variables using a domain decomposition method.

5. The method for optimizing the overall performance of a fabricated concrete frame structure according to claim 1, wherein The sensitivity analysis on the design parameters to screen key parameters comprises: A parameter influence matrix is established; The design parameters are classified based on the parameter influence matrix to screen the key parameters.

6. The method for optimizing the overall performance of a fabricated concrete frame structure according to claim 1, wherein The multi-objective optimization on the key parameters to obtain an optimal solution design scheme combination comprises: A target function vector of the key parameters is constructed; An improved MOEA / D-DE algorithm is used to optimize the target function vector to obtain an optimal solution design scheme combination.

7. The method for optimizing the overall performance of a fabricated concrete frame structure according to claim 1, wherein The construction feasibility verification of the optimal solution design scheme combination to obtain an optimal design scheme comprises: Construction constraints are established; Monte Carlo sampling is used on the optimal solution design scheme combination to generate candidate design schemes; The feasibility of each candidate design scheme is verified through a 4D construction simulation to obtain a candidate design scheme with feasibility; A feasibility index of each candidate design scheme with feasibility is calculated, and an optimal design scheme is obtained based on the feasibility index.

8. The method for optimizing the overall performance of a fabricated concrete frame structure according to claim 1, wherein, The real-time monitoring and analysis during the construction process includes: Data of a component assembled during the construction process is collected by a multi-source sensor to obtain a measured point cloud; A non-rigid registration relationship between the measured point cloud and the multi-scale digital twin model is established to determine whether the component has an assembly deviation; When the component has an assembly deviation, an error propagation model is constructed based on a graph neural network to predict the cascading effect of the deviation of the component on the assembly type concrete frame structure.

9. The method for optimizing the overall performance of a fabricated concrete frame structure according to claim 8, wherein, The data of the component assembled during the construction process is collected by the multi-source sensor to obtain the measured point cloud, including: A spatial positioning network is constructed; Static data and dynamic attitude data of the component are obtained; The measured point cloud is obtained based on the spatial positioning network, the static data and the dynamic attitude data.

10. A system for optimizing the overall performance of a fabricated concrete frame structure, characterized by, It includes: A multi-scale digital twin model construction module is configured to construct a multi-scale digital twin model, including: integrating multi-source data including parameterized data structure, finite element analysis result, manufacturing process data and 4D construction progress data, the parameterized data structure being converted from geometric information in a BIM model; constructing a multi-scale digital twin model based on the multi-source data, the multi-scale digital twin model including a macro framework model, a meso node model and a micro material model; An optimal design scheme acquisition module is configured to obtain design parameters; perform sensitivity analysis on the design parameters to screen key parameters; perform multi-objective optimization on the key parameters to obtain an optimal solution design scheme combination; and perform construction feasibility verification on the optimal solution design scheme combination to obtain an optimal design scheme; A construction monitoring and analysis optimization module is configured to perform construction based on the optimal design scheme, perform real-time monitoring and analysis during the construction process, and perform optimization and adjustment of the construction based on the real-time monitoring and analysis results.

Images