A method and system for converting bridge finite element models based on multi-algorithm matching

Abstract

This disclosure provides a method and system for converting bridge finite element models based on multi-algorithm matching. The method includes: S1, extracting component numbers, three-dimensional geometric information, and mechanically related attachment attributes from the bridge BIM model, and calculating cross-sectional mechanical properties; using the component number as the primary key, associating the three-dimensional geometric information, mechanically related attachment attributes, and cross-sectional mechanical properties to obtain structured data; S2, converting the structured data into finite element components using a component conversion algorithm; S3, connecting the finite element components to generate a complete bridge finite element model and adding load information. This method enables efficient and stable conversion between BIM models and finite element skeleton models, while improving the reuse efficiency of BIM model data.

Description

Technical Field

[0001] This disclosure relates to the field of engineering digital technology, and in particular to a method and system for converting bridge finite element models based on multi-algorithm matching. Background Technology

[0002] Currently, Building Information Modeling (BIM) software has significant advantages in ease of use, modeling efficiency, and model data management. Meanwhile, Finite Element Analysis (FEA) software is widely used in structural stress analysis in civil engineering. In engineering practice, BIM modeling software focuses on expressing the model's geometric structure and engineering attributes, while finite element models concentrate on structural mechanics analysis and calculation; their focuses differ.

[0003] However, most BIM modeling software currently lacks the ability to directly perform finite element analysis and does not provide an effective interface with mainstream finite element software (such as Midas Civil). Finite element software cannot directly use the native data of BIM models. In other words, when using finite element software to perform structural mechanics analysis calculations, users need to manually recreate the model's geometry and mechanics-related data, which limits the reuse efficiency of BIM model data.

[0004] In engineering modeling applications, especially in scenarios such as bridge engineering, designers urgently need BIM models to be integrated throughout the entire process of structural modeling, stress analysis, and structural optimization. Therefore, how to achieve efficient and stable conversion between BIM models and finite element skeleton models, and expand the functionality of BIM modeling software in the integrated process, has become a key research direction in the current field of BIM modeling software integration and development. Summary of the Invention

[0005] This disclosure provides a method and system for converting bridge finite element models based on multi-algorithm matching, which can achieve efficient and stable conversion between BIM models and finite element skeleton models, and improve the reuse efficiency of BIM model data.

[0006] Firstly, this disclosure provides a method for transforming a bridge finite element model based on multi-algorithm matching, including:

[0007] S1. Based on the bridge BIM model, extract the component number, the three-dimensional geometric information of the component, and the mechanical connection attributes, and calculate the mechanical properties of the section. Using the component number as the primary key, associate the three-dimensional geometric information, the mechanical connection attributes, and the mechanical properties of the section to obtain structured data.

[0008] S2, through a component transformation algorithm, finite element components are obtained based on structured data transformation;

[0009] S3 connects the finite element components, establishes the anchorage connection between the foundation structure and the ground, generates a complete bridge finite element model, and adds load information.

[0010] In some embodiments, step S2 includes:

[0011] S21. For components in the superstructure, substructure, and foundation, identify the current component type and geometric characteristics, and select candidate component conversion algorithms.

[0012] S22. Execute the conversion algorithms for each candidate component and monitor the execution efficiency, and remove the candidate component conversion algorithms that have timed out.

[0013] S23. After step S22, if there is at least one candidate component conversion algorithm that has not been removed, then perform a preliminary convergence check on the candidate component conversion algorithm that has completed the conversion, and remove the candidate component conversion algorithm that has not passed the preliminary convergence check.

[0014] S24. After step S23, if there are at least two candidate component conversion algorithms that have not been removed, then determine the suitability of the candidate component conversion algorithms; and calculate the score for the candidate component conversion algorithms that have passed the preliminary convergence check according to the following formula:

[0015]

[0016] Where α is the final score; t is the calculation time of the candidate component conversion algorithm in seconds; δ is the ranking of the degree of fit between the cross-section processing method and the corresponding algorithm, with the complete variable cross-section modeling processing method ranking as 1, the middle complex cross-section replaced by simple cross-section modeling processing method ranking as 2, and the equal cross-section equivalent replacement modeling processing method ranking as 3. The minimum spacing between finite element nodes in the transformation algorithm for this component is given in meters.

[0017] S25. The algorithm with the highest score is selected as the optimal component conversion algorithm. The processing result of the optimal component conversion algorithm is selected as the current finite element component conversion result. The optimal component conversion algorithm is also used as the optimal component conversion algorithm for subsequent components with the same component type and component geometric feature type as the current component.

[0018] S26. For all components in the support structure, a unique preset component conversion algorithm is used to convert the structured data into finite element components.

[0019] In some embodiments, the algorithm fit includes the degree of fit of the cross-section processing method and the minimum spacing of finite element nodes.

[0020] In some embodiments, the degree of fit of the cross-section processing methods, from high to low, is as follows: complete variable cross-section modeling, modeling of complex cross-sections in the middle by replacing them with simple cross-sections, and modeling of equivalent replacement of cross-sections.

[0021] In some embodiments, step S23, the preliminary convergence check of the conversion algorithm for the candidate components after the conversion is completed, includes:

[0022] Preliminary convergence verification is performed based on the constraints of the spacing between adjacent finite element nodes and the cross-sectional consistency constraint of the finite element component. If all constraints are met, the verification is passed; if any constraint is not met, the verification is failed.

[0023] In some embodiments, step S2 further includes:

[0024] If all alternative component conversion algorithms are removed after step S22 or step S23, then the preset backup component conversion algorithm in this component type is adopted as the optimal component conversion algorithm for all components with the same component type and component geometric feature type as the current component.

[0025] In some embodiments, step S2 further includes:

[0026] If, after step S23, only one candidate component transformation algorithm passes the preliminary convergence check, then that candidate component transformation algorithm is taken as the optimal component transformation algorithm, and the optimal component transformation algorithm is taken as the optimal component transformation algorithm for subsequent components with the same component type and component geometric feature type as that component.

[0027] In some embodiments, the alternative component conversion algorithm includes a component conversion algorithm for the complete variable cross-section modeling processing method corresponding to the component geometric feature type under the current component type, and also includes algorithm variants of the component conversion algorithm for the complete variable cross-section modeling processing method in terms of cross-section processing method and / or finite element node minimum spacing refinement.

[0028] In some embodiments, the finite element model is further formatted to generate a data format text for a specific finite element software, including an MCT file for Midas software, a TCL file for OpenSEES, and a CDB file for ANSYS.

[0029] Secondly, this disclosure provides a bridge finite element model conversion system based on multi-algorithm matching, used to execute the bridge finite element model conversion method based on multi-algorithm matching, the system comprising:

[0030] The structured data generation module is used to extract component numbers, three-dimensional geometric information of components, and mechanically related attachment attributes based on the bridge BIM model, and to calculate the mechanical properties of the cross sections; using the component number as the primary key, it associates the three-dimensional geometric information, mechanically related attachment attributes, and cross-sectional mechanical properties to obtain structured data;

[0031] The finite element component conversion module is used to convert structured data into finite element components through a component conversion algorithm.

[0032] The finite element model conversion module is used to connect finite element components, establish anchorage between the foundation structure and the ground, generate a complete bridge finite element model, and add load information.

[0033] The beneficial effects of this disclosure are that, compared with the prior art, this disclosure has the following advantages:

[0034] 1. This disclosure provides a general framework for the conversion of bridge finite element models based on multi-algorithm matching, which realizes efficient and stable conversion between BIM models and finite element skeleton models, improves the reuse efficiency of BIM model data, and expands the application of BIM modeling software in finite element stress analysis.

[0035] 2. This disclosed solution starts from BIM model data, can automatically extract key data, and reprocess some data to generate structured data that can be directly called as standardized input data when converting finite element components, thereby improving the reuse efficiency of BIM model data.

[0036] 3. This scheme pre-sets multiple component conversion algorithm templates for different types of components in bridge structures. Taking into account modeling accuracy, algorithm efficiency, and computational convergence, a dynamic comparison mechanism is used to select the optimal component conversion algorithm. Thus, while ensuring model accuracy and stability, it achieves synergistic optimization of modeling efficiency and computational performance, obtaining the optimal finite element modeling scheme that simultaneously meets accuracy requirements, good convergence, and high conversion efficiency.

[0037] 4. In this solution, the system maintains the coordinate information of the first and last nodes of the component and the sequential index table of the finite element elements in real time, so as to realize the rapid positioning of the component connection position and the insertion position of the new element, avoid repeated traversal in the modeling process, improve the conversion efficiency of finite element components, and thus improve the running efficiency of finite element conversion under high concurrency conditions. Attached Figure Description

[0038] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0039] Figure 1 A flowchart illustrating a bridge finite element model conversion method based on multi-algorithm matching is provided for embodiments of this disclosure;

[0040] Figure 2 This is a schematic diagram of the structural types of various parts of the bridge in the BIM model provided in this embodiment of the disclosure;

[0041] Figure 3 A schematic diagram of the segmentation method for extracting key points and cross-sectional information from a BIM model according to an embodiment of this disclosure;

[0042] Figure 4 This disclosure provides a schematic diagram of obtaining the cross-section of a bridge main beam using a segmented partitioning method and converting it into a finite element component.

[0043] Figure 5 This is a schematic diagram illustrating a structured JSON data storage method provided in an embodiment of this disclosure;

[0044] Figure 6 A flowchart illustrating an optional implementation of step S2 provided in this disclosure;

[0045] Figure 7 This is a schematic diagram of a bridge finite element skeleton model provided in an embodiment of the present disclosure;

[0046] Figure 8 A schematic diagram of a bridge finite element model conversion system based on multi-algorithm matching is provided for embodiments of this disclosure.

[0047] The accompanying drawings have illustrated specific embodiments of this disclosure, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this disclosure to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0048] The present disclosure will be further described below with reference to the accompanying drawings. The following embodiments are only used to illustrate the technical solutions of the present disclosure more clearly, and should not be used to limit the scope of protection of the present disclosure.

[0049] Figure 1 This is a flowchart illustrating a bridge finite element model conversion method based on multi-algorithm matching, according to an embodiment of this disclosure. This method can be executed by a bridge finite element model conversion system based on multi-algorithm matching. This system can be implemented in software and / or hardware, and can be integrated as a functional module into existing BIM modeling applications, or deployed and run independently.

[0050] The bridge finite element model conversion method based on multi-algorithm matching proposed in this disclosure mainly includes two stages of data processing: First, key data is extracted from the BIM model data and further processed to obtain structured data, which can be used as standardized input data for subsequent finite element component conversion algorithms. Second, the structured data is input into the finite element component conversion module, processed by the algorithm to obtain finite element components. These components are then connected through the finite element model construction module to generate a complete bridge finite element model. This establishes a general framework for bridge finite element model conversion based on multi-algorithm matching, enabling efficient and stable conversion between BIM models and finite element skeleton models.

[0051] like Figure 1 As shown in this embodiment, the bridge finite element model transformation method based on multi-algorithm matching can include:

[0052] Step S1: Based on the bridge BIM model, extract component numbers, 3D geometric information of the components, and mechanically related attachment attributes, and calculate cross-sectional mechanical properties. Using the component number as the primary key, associate the 3D geometric information, mechanically related attachment attributes, and cross-sectional mechanical properties to obtain structured data. Specifically, this includes the following steps:

[0053] Step S11: Based on the bridge BIM model, extract the component number and the three-dimensional geometric information of the component;

[0054] Specifically, each component in the BIM model has a unique number (i.e., component ID). A data table is established using this number as the primary key field. When associating the component with its three-dimensional geometric information, mechanically related attachment attributes, cross-sectional mechanical properties, and other data in the future, the association is achieved through this ID number.

[0055] like Figure 2 As shown, the bridge BIM model construction type includes five parts: superstructure, supports, substructure, foundation structure, and auxiliary structures.

[0056] The superstructure includes, for example, the following component types: main beams, end beams, middle beams, and cast-in-place layers. The main beams are usually precast beams or cast-in-place box beams.

[0057] A bearing is a connecting support structure between the upper and lower structures. It is usually a plate rubber bearing or a pot rubber bearing and includes the following component types: wedge block, brace and pad. The bearing can be omitted when the upper and lower structures are fixed together.

[0058] The substructure includes two types of structures: piers and abutments. Pier types include column piers, wall piers, and vase-shaped piers, while abutment types include seat plate abutments and U-shaped abutments. Piers include components such as abutment caps and approach slabs, while abutments include components such as pier columns, cap beams, and tie beams.

[0059] The foundation is the part buried deep underground and connected to the strata, including components such as pile caps and pile foundations;

[0060] Ancillary structures refer to structures that do not participate in the structural stress but only provide loads, such as bridge decks, guardrails, and decorative structures on the bridge surface such as streetlights.

[0061] In one embodiment, extracting the three-dimensional geometric information of the component specifically includes:

[0062] 1) For supports, the three-dimensional geometric information of the components is obtained by direct extraction method;

[0063] The direct extraction method uses the secondary development interface provided by BIM software to directly extract the three-dimensional geometric information of the component, including the coverage area of ​​the component (the smallest cuboid that can enclose it), the component's length, width, surface area, volume, and centroid coordinates, which are then stored in association with the component's ID number. This method yields relatively basic three-dimensional geometric information, and in this embodiment, it is only used for support structures.

[0064] Understandably, since the components obtained by the direct extraction method do not need to focus on specific geometric structures, the centroid of the component obtained by the direct extraction method is the key point of the component.

[0065] 2) For the superstructure, substructure, and foundation, the three-dimensional geometric information of the components is obtained by segmentation method;

[0066] like Figure 3 , Figure 4 As shown, the segmentation method includes: along the length extension direction of the three-dimensional geometric information of the component, cutting at intervals along the direction perpendicular to the extension direction to obtain multiple cross sections, extracting the cross section contour shape information, calculating the centroid of the cross section based on the contour line, and using it as the key point corresponding to the cross section, recording the key point information and the corresponding cross section contour shape information.

[0067] Specifically, after obtaining multiple cross-sections through sectioning, the contour lines of the cross-sections are extracted using the secondary development interface provided by the BIM software. The centroid of the cross-section is then calculated based on the contour lines and used as the key point corresponding to that cross-section. The key point information is combined with the corresponding section contours into data groups, which are then numbered sequentially according to the order of extension direction of the sectioning position. All data groups are written into a data table and associated with the corresponding component IDs.

[0068] Understandably, the 3D geometric information extracted by the segmentation method includes at least the key point information of the component and the cross-sectional contour shape information corresponding to each key point. The key point information includes the key point number and coordinate data.

[0069] Step S12: Extract the mechanically related connection attributes of each component;

[0070] Understandably, to construct a finite element model, it's often necessary to extract some geometry-independent but mechanically relevant connection attributes. Mechanically relevant connection attributes include material properties and connection parameters, such as bearing capacity, concrete elastic modulus, and stiffness of the pile foundation-soil connection. This data can be directly extracted from the connection attributes of BIM model components; the types of connection attributes that need to be recorded are usually predefined in a data table.

[0071] Step S13: Calculate the cross-sectional mechanical properties of the component based on its three-dimensional geometric information and mechanically related connection properties.

[0072] In this embodiment, the mechanical properties of the cross section are the cross section mechanical property data required for subsequent stress analysis based on the finite element model (e.g., when calculating the structural stiffness matrix), including parameters such as cross section area, cross section profile shape, cross section moment of inertia, cross section rotational inertia, and cross section shear modulus in the XY direction. Specifically, the cross section profile shape can be obtained from the three-dimensional geometric information of the component extracted in step S11, while other mechanical properties can be calculated based on the cross section profile shape information and material properties.

[0073] In practical implementation, the 3D model manager first filters components containing cross-sections and calculates the mechanical properties of all cross-sections. For simple cross-sections with regular geometry (such as square or circular cross-sections), the mechanical properties are solved using standard calculation formulas defined in industry-standard specifications. For cross-sections with complex geometry (such as octagonal or square cross-sections with holes), conventional methods commonly used in this field, such as the equivalent piecewise method (i.e., calculating mechanical properties by dividing the cross-section into simple cross-sections and then superimposing them) and the numerical integration method (i.e., discretizing the complex cross-section into a set of points and using numerical approximation algorithms such as Monte Carlo integration), can be used for calculation. The above calculation methods are all existing technologies in this field and will not be elaborated further here.

[0074] Finally, the obtained cross-sectional mechanical properties are associated with the key points determined in step S11 and stored in the data table.

[0075] Understandably, the mechanical property data of the cross-sections extracted using the segmentation method cannot be directly parsed from the original BIM model because they are not linked to it. This step effectively completes the cross-sectional mechanical property information of the component on the segment, providing the necessary mechanical property data for subsequent finite element modeling and structural mechanics analysis, thus ensuring the accuracy and reliability of the finite element model.

[0076] Step S14: Using the component number as the primary key, associate the three-dimensional geometric information, mechanically related attachment attributes, and cross-sectional mechanical attributes to obtain structured data.

[0077] Specifically, structured data includes the three-dimensional geometric information, cross-sectional mechanical properties, and mechanically related attachment attributes obtained in the above steps, which are stored in different files in the form of JSON data. For example... Figure 5 As shown, the structured data file includes a 3D geometric information JSON file, a cross-sectional mechanical property information JSON file, and a mechanically related attachment property information JSON file.

[0078] It's worth noting that in existing finite element method (FEM) software, to convert data into a finite element model, users need to manually input the 3D geometric information, cross-sectional mechanical properties, and mechanically related connection attributes of each component. This means re-inputting the geometric and mechanical data of the original model into the FEM software, which then converts it into a finite element model based on the user's input and performs further structural stress analysis. As you can imagine, this method is complex and cannot effectively reuse BIM model data.

[0079] This embodiment starts from BIM model data, can automatically extract key data, and reprocess some data to generate structured data that can be directly called as standardized input data when converting finite element components. This is one of the innovations of this disclosure.

[0080] Step S2: Obtain finite element components based on structured data through a component transformation algorithm.

[0081] The specific implementation method of the component-level component transformation algorithm is as follows: Based on structured data containing cross-sectional mechanical properties, the key point information in the three-dimensional geometric information of the component is extracted, and finite element elements and finite element nodes are generated according to the key point information and the finite element node spacing set by the component transformation algorithm. Corresponding finite element nodes, mechanically related attachment attributes and cross-sectional mechanical properties are added to the finite element elements. The link relationship between finite element elements is established according to the shared nodes, thereby generating finite element components.

[0082] It should be noted that there is a general method for the specific implementation of the component conversion algorithm in structural stress analysis engineering applications, which is an existing technology in this field, and this disclosure does not make any specific limitations on it.

[0083] The component transformation algorithm processes components, that is, it transforms finite element components based on structured data, taking the component level as the unit, during the transformation process.

[0084] In this embodiment, for the support structure, since its structure is relatively simple, it is usually modeled using only a single elastic connection method. That is to say, only one component conversion algorithm is provided for the support structure, which can be directly used for the finite element component conversion of the support structure.

[0085] For components in the superstructure, substructure, and foundation, multiple optional conversion algorithms are typically preset. The algorithm manager, for each component type and each type of component geometric feature, presets a component conversion algorithm with a complete variable cross-section modeling processing method in its corresponding component conversion algorithm sub-library. These geometric features include, but are not limited to, contour line form and cross-sectional shape distribution features.

[0086] For example, based on the differences in the geometric structural characteristics of the main beam components, they are classified into simply supported precast beams, continuous precast beams, and cast-in-place box girders. Specifically, precast beams are usually composed of multiple parallel main beams. In simply supported precast beams, the main beam segments in each span are not connected, while in continuous precast beams, the main beam segments in each span are connected by crossbeams. Cast-in-place box girders are integral continuous cast-in-place structures with closed or partially closed box-shaped cross sections. As another example, for pier components, based on the differences in the geometric structural characteristics of the pier components, they are classified into bifurcated single-column vase piers and multi-column piers with tie beams. Specifically, bifurcated single-column vase piers consist of a single upright pier column with a streamlined vase-shaped profile that narrows at the waist. The pier top or middle section usually bifurcates into 2-4 legs. Multi-column piers with tie beams are usually composed of 2-4 linearly arranged columns with horizontal tie beams between them. It is evident that even among components of the same type (such as main beams or piers), differences in their geometric structural characteristics can lead to variations in the finite element unit organization framework and node connection methods. Therefore, the system needs to match different component transformation algorithms for them.

[0087] Furthermore, in the component conversion algorithm sub-library corresponding to each component geometric feature type, in addition to the preset component conversion algorithm for the complete variable cross-section modeling processing method, it further includes algorithm variants of the component conversion algorithm for the complete variable cross-section modeling processing method in terms of cross-section processing method and / or finite element node minimum spacing refinement. In this embodiment, the preset component conversion algorithm for the complete variable cross-section modeling processing method refers to the algorithm that adopts the complete variable cross-section modeling processing method during the modeling process, and the finite element node spacing is a default preset value (for example, for pier components, the default preset value of its finite element node spacing can be set to 1m).

[0088] It should be noted that in specific engineering implementations, when transforming the finite element node spacing of the algorithm, the node spacing is usually further refined only for complex cross-sectional segments with significant stress changes in the component. Different degrees of refinement result in different algorithm variants. For other segments with relatively uniform stress distribution (such as the main body of the component), further refinement of the node spacing is usually unnecessary (i.e., other segments still use the default preset values). For example, when refining the minimum finite element node spacing of the algorithm for a bifurcated single-column vase-shaped pier, only the bifurcated segment is modeled with a node spacing of 0.5m; other segments still use 1m as the node spacing. Therefore, the minimum finite element node spacing, i.e., the finite element node spacing corresponding to the complex cross-sectional segment with significant stress changes in the component, is used here as a classification index for the algorithm.

[0089] Specifically, for components in the superstructure, substructure, and foundation, the algorithm manager pre-sets component conversion algorithm libraries corresponding to different component types. Each component conversion algorithm library pre-sets multiple component conversion algorithm sub-libraries corresponding to different component geometric structural feature types. For example, the main beam component algorithm library includes a simply supported precast beam algorithm sub-library, a continuous precast beam algorithm sub-library, and a cast-in-place box girder algorithm sub-library, etc. The pier column component algorithm library includes a single-column vase pier component conversion algorithm sub-library with bifurcations and a multi-column pier component conversion algorithm sub-library with tie beams, etc.

[0090] Meanwhile, to ensure comprehensive modeling capabilities, for components in the superstructure, substructure, and foundation, the system pre-defines at least one basic component conversion algorithm in the component conversion algorithm library for each component type. This algorithm is simple in structure, has strong convergence, but low computational accuracy. This ensures that at least one algorithm can pass the initial convergence check, guaranteeing that all components can successfully complete the finite element conversion. This basic component conversion algorithm does not require fitting degree, execution efficiency calculations, or initial convergence checks in step S2. For example, for pier components, the component conversion algorithm library pre-defines a single-point hard-connection component conversion algorithm as a basic conversion algorithm. Specifically, the single-point hard-connection component conversion algorithm directly converts the pier into an elastic structure connected to the ground.

[0091] In practical modeling, various component conversion algorithms offer different levels of model fit and computational efficiency. Generally, higher model fit results in a more detailed finite element model, but also increases modeling and subsequent solution computation time, and makes model convergence more difficult. For example, when performing full-condition modeling and simulation analysis on some complex bridge models, a single complete calculation cycle may exceed 5 hours, but the improvement in computational accuracy may not be significant. Therefore, designers typically need to balance model fit and computational efficiency. For instance, while ensuring that the model fit meets engineering requirements, a modeling method with a higher degree of structural simplification can be selected to shorten the overall modeling and simulation time.

[0092] Therefore, in this embodiment, when selecting component conversion algorithms for components in the superstructure, substructure, and foundation, both modeling fit and computational efficiency are comprehensively considered. The optimal component conversion algorithm is determined through dynamic comparison and analysis, thereby ensuring the reliability of modeling while achieving high efficiency in the finite element modeling process.

[0093] In one optional implementation, step S2 is carried out as follows:

[0094] 1) In the superstructure, substructure and foundation, for a certain component geometric feature type under a component type, select the optimal component conversion algorithm, use the optimal component conversion algorithm, and convert the structured data to obtain the finite element component; 2) For all components in the support structure, use a unique preset component conversion algorithm to convert the structured data to obtain the finite element component.

[0095] like Figure 6 As shown, the specific implementation steps of step S2 are as follows:

[0096] Step S21: For components in the superstructure, substructure, and foundation, identify the current component type and component geometric features, and select candidate component conversion algorithms;

[0097] The alternative component conversion algorithm includes a component conversion algorithm for the complete variable cross-section modeling processing method corresponding to the component geometric feature type under the current component type, and also includes an algorithm variant of the component conversion algorithm for the complete variable cross-section modeling processing method in terms of cross-section processing method and / or finite element node minimum spacing refinement.

[0098] In the specific implementation, for the component to be converted, the algorithm manager first identifies the component type and selects the component conversion algorithm library for that component type from the algorithm library. Then, it analyzes the geometric structural features of the current component, including the outline form, cross-sectional shape distribution features, etc., and selects the component conversion algorithm sub-library that matches the geometric structural features of the current component from the component conversion algorithm library and adds it to the candidate component conversion algorithm library.

[0099] Step S22: Execute the conversion algorithms for each candidate component and monitor the execution efficiency, and remove the candidate component conversion algorithms that have timed out. The specific steps are as follows:

[0100] The computation manager executes multiple alternative component conversion algorithms, monitors and records the computation time of each alternative algorithm;

[0101] If the calculation time of a candidate component conversion algorithm exceeds the preset time limit, the calculation task of the candidate component conversion algorithm will be forcibly stopped, and the candidate component conversion algorithm will be removed from the candidate component conversion algorithm library.

[0102] Preferably, for candidate component conversion algorithms whose execution time does not exceed the preset time limit, the algorithm is executed twice on the same component, and the average calculation time is recorded as an evaluation index for subsequent candidate component conversion algorithms.

[0103] In practice, the computation manager starts a corresponding number of threads based on the number of algorithms in the candidate component conversion algorithm library. For performance reasons, the number of threads is generally limited to no more than 5. If the number of algorithms exceeds this value, calculations are performed in batches, ensuring that each batch has the same number of calculations. For example, if the candidate component conversion algorithm library contains 12 algorithms, they are divided into 3 groups of 4 algorithms each. The computation manager can start multiple threads to execute multiple component conversion algorithms simultaneously. Once the computation task of one group is completed, the computation task of the next group automatically starts. The computation manager monitors each computation task in real time. If a computation task takes more than 60 seconds, it forcibly stops the task and removes the corresponding component conversion algorithm from the candidate component conversion algorithm library.

[0104] Step S23: After step S22, if there is at least one candidate component conversion algorithm that has not been removed, then perform a preliminary convergence check on the candidate component conversion algorithm that has completed the conversion, and remove the candidate component conversion algorithm that has not passed the preliminary convergence check.

[0105] In practice, preliminary convergence verification is performed based on the constraints on the spacing between adjacent finite element nodes and the cross-sectional consistency of the finite element component. If all constraints are met, the verification is passed; if any constraint is not met, the verification is failed.

[0106] In this embodiment, the spacing constraint between adjacent finite element nodes means that the spacing between adjacent finite element nodes does not exceed a first preset range. The cross-sectional consistency constraint means that when adjacent nodes have different cross-sectional types, the difference in their cross-sectional profile area does not exceed a second preset range.

[0107] For example, if the distance between adjacent finite element nodes is greater than 3m, the verification is considered to have failed. If the cross-section types corresponding to adjacent nodes are different, it is necessary to check whether the difference in the outer contour area of ​​the cross-section is less than 50%. If it does not meet this requirement, the verification is considered to have failed.

[0108] This step performs a preliminary convergence check on the finite element components after each algorithm transformation. If any algorithm fails the convergence check, it is directly removed from the candidate algorithm library.

[0109] Step S24: After step S23, if there are at least two candidate component conversion algorithms that have not been removed, then determine the suitability of the candidate component conversion algorithms; and calculate the score for the candidate component conversion algorithms that have passed the preliminary convergence check according to the following formula:

[0110]

[0111] Where α is the final score; t is the calculation time of the candidate component conversion algorithm in seconds; δ is the ranking of the degree of fit between the cross-section processing method and the corresponding algorithm, with the complete variable cross-section modeling processing method ranking as 1, the middle complex cross-section replaced by simple cross-section modeling processing method ranking as 2, and the equal cross-section equivalent replacement modeling processing method ranking as 3. The minimum spacing between finite element nodes in the transformation algorithm for this component is given in meters.

[0112] Understandably, the scoring method follows the principle of "fit first, computational efficiency second".

[0113] In one embodiment, the algorithm fit includes two parameters: the degree of fit of the cross-section processing method and the minimum spacing between finite element nodes.

[0114] The degree of fit of the cross-section processing methods, from high to low, is as follows: complete variable cross-section modeling, modeling with a simple cross-section replacing a complex cross-section in the middle, and modeling with an equivalent replacement of a uniform cross-section. Optionally, the algorithm ranking corresponding to the complete variable cross-section modeling processing method is 1, the algorithm ranking corresponding to the modeling with a simple cross-section replacing a complex cross-section in the middle is 2, and the algorithm ranking corresponding to the modeling with an equivalent replacement of a uniform cross-section is 3.

[0115] Among them, complete variable cross section modeling refers to directly based on the original structural form, retaining the characteristics of its cross section changing along the length of the component, and establishing a variable cross section model that reflects the original structural line shape;

[0116] Modeling a complex section in the middle by replacing it with a simple section means simplifying the complex section in the middle part while preserving the overall structural characteristics of the component, that is, replacing it with a simple section.

[0117] In bridge component modeling, since the complex cross-section in the middle is usually located in a region with relatively uniform stress distribution, its details have little impact on the overall stress response. Therefore, a simple cross-section can be reasonably used to replace it locally with an equivalent one, thereby improving the efficiency of modeling while ensuring the accuracy of calculation.

[0118] Equivalent replacement modeling refers to converting a variable cross-section member into a uniform cross-section member model with equivalent mechanical properties (such as stiffness and mass distribution) to simplify the calculation process.

[0119] It should be noted that not all candidate algorithms for the components to be converted include the three cross-section processing methods mentioned above. The above ranking is only used to clarify the order of the compatibility of the three cross-section processing methods and as a reference for judging the degree of algorithm compatibility.

[0120] The minimum spacing between finite element nodes refers to the minimum spacing between finite element nodes, that is, the spacing between finite element nodes corresponding to complex cross-sections with significant stress changes in a component. The smaller the minimum spacing between finite element nodes, the higher the modeling accuracy.

[0121] In this embodiment, the fit of the component transformation algorithm is jointly determined by the fit of the algorithm's cross-section processing method and the minimum spacing index of finite element nodes. It is understandable that a higher fit of the algorithm's cross-section processing method and a smaller minimum node spacing indicate a higher algorithm fit.

[0122] Step S25: The algorithm with the highest score is selected as the optimal component conversion algorithm. The processing result of the optimal component conversion algorithm is selected as the current finite element component conversion result. The optimal component conversion algorithm is also used as the optimal component conversion algorithm for subsequent components with the same component type and geometric feature type as the current component.

[0123] Specifically, in the subsequent component transformation, if the geometric structure features of a certain component are the same as the component type and geometric structure feature type currently being transformed in this step, then the optimal component transformation algorithm can be directly selected as the component transformation algorithm for the subsequent component.

[0124] Step S26: For all components in the support structure, a unique preset component conversion algorithm is used to convert the structured data into finite element components.

[0125] For the support structure, due to its relatively simple structure, it is usually modeled using only a single elastic connection method. That is to say, for all components in the support structure, only one general component conversion algorithm is used. Therefore, for the support structure, there is no need to perform algorithm comparison and convergence verification in step S2. The preset, unique component conversion algorithm applicable to all support components can be directly called to complete the modeling calculation.

[0126] It should be noted that the auxiliary structures themselves hardly participate in mechanical calculations and are usually only provided as gravity loads. Therefore, the finite element modeling conversion process in this step only applies to the four parts: the superstructure, the substructure, the foundation structure, and the supports. The auxiliary structures are only added to the finite element model as load information.

[0127] Furthermore, in real-world engineering environments, the computation time of different algorithms typically varies, and the computation time t and minimum node spacing also play a role. Furthermore, the sorting δ based on the compatibility of the cross-section processing method is generally a floating-point number or a discrete value with clearly distinguishable values; therefore, identical scores will not occur. This embodiment determines the unique optimal component conversion algorithm by selecting the algorithm with the highest score. If identical scores occur in special cases, it indicates that the two algorithms have comparable performance, and either one can be chosen.

[0128] Further, optionally, step S2 further includes:

[0129] If all candidate component conversion algorithms are removed after step S22 or step S23, then the preset backup component conversion algorithm in this component type is adopted as the optimal component conversion algorithm for all components with the same component type and component geometric feature type as the current component.

[0130] If, after step S23, only one candidate component transformation algorithm passes the preliminary convergence check, then that candidate component transformation algorithm is taken as the optimal component transformation algorithm, and is used as the optimal component transformation algorithm for subsequent components with the same component type and geometric feature type as that component.

[0131] Taking the bifurcated single-column vase-shaped pier component as an example, the specific operation method of step S2 is as follows:

[0132] 1) For the bifurcated single-column vase-shaped pier component, all component transformation algorithms in the sub-library of bifurcated single-column vase-shaped pier component transformation algorithms are selected and added to the candidate component transformation algorithm library, and the single-point hard-connection component transformation algorithm is selected as the backup algorithm. The candidate component transformation algorithm library includes the bifurcated single-column vase-shaped pier component transformation algorithm (0.5m minimum spacing), the bifurcated single-column vase-shaped pier component transformation algorithm (1m minimum spacing), and the equivalent variable cross-section replacement component transformation algorithm (the complex cross-section in the middle is replaced by a simple cross-section, 1m minimum spacing).

[0133] 2) The calculation manager retrieves the candidate component conversion algorithms exported from the algorithm manager. Since there are three algorithms, three threads are started to calculate them separately, without the need for batch calculation. This batch calculation is executed twice, and the average time of the three algorithms is recorded as 13.1s, 5.9s, and 0.2s. No algorithm takes more than 60s, meaning there are no algorithms that need to be eliminated.

[0134] 3) A preliminary convergence check was performed on the transformed finite metadata. All three algorithms passed the check. Further algorithm scoring will then be conducted.

[0135] 4) Determine the suitability of the candidate algorithm's cross-section processing method and the minimum spacing of finite element nodes. For example, the algorithm ranking for the transformation of a bifurcated single-column vase-shaped pier component is set to 1, and the algorithm ranking for the transformation of an equivalent variable cross-section replacement component is set to 2, and these algorithms will participate in subsequent scoring calculations. The minimum node spacing for the transformation algorithm of the bifurcated single-column vase-shaped pier component (minimum spacing of 0.5m) and the transformation algorithm of the bifurcated single-column vase-shaped pier component (minimum spacing of 1m) are 0.5m and 1m, respectively; the minimum node spacing for the transformation algorithm of the equivalent variable cross-section replacement component (using a simple cross-section to replace the complex cross-section in the middle, minimum spacing of 1m) is 1m.

[0136] 5) The three algorithms are scored, and the one with the highest score is selected as the optimal component conversion algorithm, namely the vase pier component conversion algorithm with bifurcation (minimum spacing of 0.5m).

[0137] 6) In the subsequent processing of this single-column vase pier, the optimal component transformation algorithm is directly selected as the component transformation algorithm for subsequent components.

[0138] Further, step S2 may also include: temporarily caching the global data in operation during the parallel transformation of multiple components.

[0139] Preferably, the global data in the temporary cache includes the first and last nodes of the finite element components and the sequential queue of each finite element element stored during the algorithm's execution.

[0140] The caching mechanism includes: 1) recording the coordinates of the first and last nodes of each component after its transformation, for use when connecting subsequent components; 2) maintaining a real-time index table of the generation and insertion order of finite element elements to quickly find the insertion position of new elements, avoiding repeated traversal and improving the transformation efficiency of finite element components. Through this caching mechanism, the operational efficiency of finite element transformation can be improved under high concurrency conditions.

[0141] Step S3: Connect the finite element components, establish the anchorage connection between the foundation structure and the ground, generate a complete bridge finite element model, and add load information.

[0142] 1) Add link relationships (such as common nodes, elastic links or rigid connections) between finite element components, and establish anchorage connections between pile foundation components and the ground to generate a complete bridge finite element model;

[0143] This completes the rapid conversion from component-level finite element elements to a bridge finite element skeleton model, such as... Figure 7 As shown.

[0144] 2) Add load information to the finite element model.

[0145] After the finite element skeleton model is built, loads need to be added to the model. In this embodiment, gravity loads, traffic loads, wind loads, and construction loads are added to the model according to standard construction conditions.

[0146] Understandably, the auxiliary structures themselves hardly participate in mechanical calculations and are usually only provided as gravity loads, which are then added to the finite element model as gravity load information.

[0147] The method described in this embodiment completes the construction of a general finite element model, which belongs to the general format data of finite element software. If it is necessary to import it into a specific finite element software, further format fine-tuning can be performed to generate the data format text of the specific finite element software, such as the MCT file for Midas software, the TCL file for OpenSEES, and the CDB file for ANSYS.

[0148] like Figure 8 As shown, the second embodiment of this disclosure proposes a bridge finite element model conversion system 100 based on multi-algorithm matching, used to execute the bridge finite element model conversion method based on multi-algorithm matching as described in the above embodiments. The system includes:

[0149] The structured data generation module 101 is used to extract component numbers, three-dimensional geometric information of components, and mechanically related attachment attributes based on the bridge BIM model, and to calculate the mechanical properties of the cross sections; using the component number as the primary key, it associates the three-dimensional geometric information, mechanically related attachment attributes, and cross section mechanical properties to obtain structured data;

[0150] Finite element component conversion module 102 is used to obtain finite element components based on structured data through a component conversion algorithm;

[0151] The finite element model conversion module 103 is used to connect finite element components, establish the anchorage connection between the foundation structure and the ground, generate a complete bridge finite element model, and add load information.

[0152] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0153] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0154] It should be understood that the above embodiments are only used to illustrate the technical solutions of this disclosure, and not to limit them; although this disclosure has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this disclosure.

Claims

1. A method for transforming a bridge finite element model based on multi-algorithm matching, characterized in that, include: S1, based on the bridge BIM model, extract the component number, the three-dimensional geometric information of the component and the mechanical connection attributes, and calculate the mechanical properties of the section; Using the component number as the primary key, and associating it with three-dimensional geometric information, mechanically related attachment attributes, and cross-sectional mechanical properties, structured data is obtained; S2, through a component transformation algorithm, finite element components are obtained based on structured data transformation; S3 connects the finite element components, establishes the anchorage connection between the foundation structure and the ground, generates a complete bridge finite element model, and adds load information. The step S2 comprises: S21. For components in the superstructure, substructure, and foundation, identify the current component type and geometric characteristics, and select candidate component conversion algorithms. S22. Execute each candidate component conversion algorithm and monitor the execution efficiency, and remove candidate component conversion algorithms that have timed out. S23. After step S22, if there is at least one candidate component conversion algorithm that has not been removed, then perform a preliminary convergence check on the candidate component conversion algorithm that has completed the conversion, and remove the candidate component conversion algorithm that has not passed the preliminary convergence check. S24. After step S23, if there are at least two candidate component conversion algorithms that have not been removed, then determine the suitability of the candidate component conversion algorithms; and calculate the score for the candidate component conversion algorithms that have passed the preliminary convergence check according to the following formula: ； Where α is the final score; t is the calculation time of the candidate component conversion algorithm in seconds; δ is the ranking of the degree of fit between the cross-section processing method and the corresponding algorithm, with the complete variable cross-section modeling processing method ranking as 1, the middle complex cross-section replaced by simple cross-section modeling processing method ranking as 2, and the equal cross-section equivalent replacement modeling processing method ranking as 3. The minimum spacing between finite element nodes in the transformation algorithm for this component is given in meters. S25. The algorithm with the highest score is selected as the optimal component conversion algorithm. The processing result of the optimal component conversion algorithm is selected as the current finite element component conversion result. The optimal component conversion algorithm is also used as the optimal component conversion algorithm for subsequent components with the same component type and geometric feature type as the current component. S26. For all components in the support structure, a unique preset component conversion algorithm is used to convert the structured data into finite element components.

2. The bridge finite element model transformation method based on multi-algorithm matching according to claim 1, characterized in that, The algorithm fit includes the degree of fit of the cross-section processing method and the minimum spacing of finite element nodes.

3. The bridge finite element model transformation method based on multi-algorithm matching according to claim 2, characterized in that, The degree of fit of the cross-section processing methods, from high to low, is as follows: complete variable cross-section modeling, complex cross-section in the middle replaced by simple cross-section modeling, and equivalent replacement modeling with equal cross-section.

4. The bridge finite element model conversion method based on multi-algorithm matching according to claim 1, characterized in that, In step S23, the preliminary convergence check of the conversion algorithm for the candidate components after the conversion is completed includes: Preliminary convergence verification is performed based on the constraints of the spacing between adjacent finite element nodes and the cross-sectional consistency constraint of the finite element component. If all constraints are met, the verification is passed; if any constraint is not met, the verification is failed.

5. The bridge finite element model conversion method based on multi-algorithm matching according to claim 1, characterized in that, Step S2 further includes: If all alternative component conversion algorithms are removed after step S22 or step S23, then the preset backup component conversion algorithm in this component type is adopted as the optimal component conversion algorithm for all components with the same component type and component geometric feature type as the current component.

6. The bridge finite element model transformation method based on multi-algorithm matching according to claim 1, characterized in that, Step S2 further includes: If, after step S23, only one candidate component transformation algorithm passes the preliminary convergence check, then that candidate component transformation algorithm is taken as the optimal component transformation algorithm, and the optimal component transformation algorithm is taken as the optimal component transformation algorithm for subsequent components with the same component type and component geometric feature type as that component.

7. The bridge finite element model transformation method based on multi-algorithm matching according to claim 1, characterized in that, The alternative component conversion algorithm includes a component conversion algorithm for the complete variable cross-section modeling processing method corresponding to the component geometric feature type under the current component type, and also includes algorithm variants of the component conversion algorithm for the complete variable cross-section modeling processing method in terms of cross-section processing method and / or finite element node minimum spacing refinement.

8. The bridge finite element model conversion method based on multi-algorithm matching according to claim 1, characterized in that, After further format fine-tuning, the finite element model generates data format text for specific finite element software, including MCT files for Midas software, TCL files for OpenSEES, and CDB files for ANSYS.

9. A bridge finite element model conversion system based on multi-algorithm matching, used to execute the bridge finite element model conversion method based on multi-algorithm matching as described in any one of claims 1-8, characterized in that, The system includes: The structured data generation module is used to extract component numbers, three-dimensional geometric information of components, and mechanically related attachment attributes based on the bridge BIM model, and to calculate the mechanical properties of the cross sections; using the component number as the primary key, it associates the three-dimensional geometric information, mechanically related attachment attributes, and cross-sectional mechanical properties to obtain structured data; The finite element component conversion module is used to convert structured data into finite element components through a component conversion algorithm. The finite element model conversion module is used to connect finite element components, establish anchorage between the foundation structure and the ground, generate a complete bridge finite element model, and add load information.

Images