A method for converting grid structure BIM model and point cloud model to finite element model
The conversion method addresses inefficiencies in generating finite element models from BIM models by extracting and cleaning structural data, enabling accurate simulation and safety in large-scale grid structures through a digital twin approach.
Patent Information
- Application Number
- CN202310115289.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-02-13
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2043-02-13
AI Technical Summary
Existing methods for generating finite element models from BIM models of large-scale grid structures are inefficient and fail to accurately reflect the dynamic changes and deformations during construction, leading to safety risks and inefficiencies due to information islands between models.
A method to convert a grid structure BIM model and point cloud model into a finite element model by extracting nodes, elements, and material information, performing geometric cleaning, and dynamically updating coordinates using a digital twin approach.
The method efficiently breaks information islands, accurately reflects structural responses, and supports rapid, accurate simulation of construction stages, enhancing safety and efficiency in large-scale grid structure construction.
Smart Images

Figure CN116305418B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of digital model processing in grid structure engineering, and in particular to a method for converting a grid structure BIM model and a point cloud model into a finite element model. Background Art
[0002] In recent years, with the development of emerging technologies such as BIM, IOT, digital twins, and artificial intelligence, the traditional construction industry can no longer meet the development needs of the new era, and the transformation of the construction industry to digitalization is an inevitable development trend. Among them, the large-scale grid structure in the building is more likely to have safety risks during the construction stage due to its complex structure, multiple construction stages, and the time-varying load conditions, displacement states, and boundary conditions of the structure at different construction stages. Although finite element simulation is generally performed on the entire process of structural construction before construction, there are many rod nodes in large-scale grid structures, and conventional finite element analysis methods are usually based on GUI operations, which are less efficient in dynamic correction. At the same time, there are information islands between the BIM model and the finite element model during the entire construction process, and the finite element model often cannot reflect the actual response of the structure. The BIM model is a refined model with highly integrated information throughout the entire life cycle of the project. If it cannot be combined with the finite element, its full value cannot be brought into play.
[0003] In addition, large-scale grid structure construction errors and deformations are common problems in engineering. If the errors and deformations exceed the design allowable values, it will affect the normal construction of the project and even endanger the safety of the structure. The use of traditional methods such as total stations is not only inefficient, but also difficult to fully and accurately reflect the overall situation of large-scale grid structures.
[0004] In summary, it is necessary to develop a method to generate a finite element model from a BIM model, break the information islands between models, and realize dynamic interaction of models throughout the construction process. This can not only create new value for the BIM model, but also the generated finite element model can more efficiently and accurately reflect the correction of the BIM model and accept the real coordinates and size data of the point cloud model, thereby providing a guarantee for more rapid and accurate reflection of the real response of the construction stage of large-scale grid structures. Summary of the invention
[0005] In view of the deficiencies in the prior art, an object of the present invention is to provide a method for converting a grid structure BIM model and a point cloud model into a finite element model, which can solve the problems described in the background technology.
[0006] The technical solution to achieve the purpose of the present invention is: a method for converting a grid structure BIM model and a point cloud model to a finite element model, comprising the following steps:
[0007] Step 1: Obtain the BIM model file of the target grid structure, and export the centerline model file and the surface model file from the BIM model file;
[0008] Step 2: Extract the nodes, elements, sections and material information representing the structural parts from the surface model file, and generate an initial finite element model according to the nodes, elements, sections and material information in a preset data exchange format;
[0009] Step 3: Perform geometric cleaning on redundant line segments in the centerline model in the centerline model file to obtain a cleaned centerline model;
[0010] Step 4: Assign the node matching coordinate data of the cleaned centerline model to the nodes of the initial finite element model through the nearest distance matching criterion, and complete the establishment of the finite element command flow model;
[0011] Step 5: Based on the point cloud model of the target grid structure, the real coordinate information of each rod of the target grid structure is obtained, and the real coordinate information of each rod in the point cloud model is dynamically updated, and the dynamic establishment and correction of the finite element command flow model is realized based on the digital twin.
[0012] Furthermore, the target grid structure is a grid structure or a truss structure, the surface model file includes structural member information, and the structural member information includes node, unit, section and material information.
[0013] Furthermore, the structural component information includes rods and welding balls.
[0014] Furthermore, in step 1, when exporting the centerline model file from the BIM model file, the format of the centerline model file is selected as DXF format, and the output option is uncut.
[0015] Furthermore, the specific implementation process of generating the initial finite element model according to the node, unit, section and material information in a preset defined data exchange format includes the following steps:
[0016] Step 21: Traverse the surface model file and filter out all the bar units as unit libraries P according to the unit number or section name N ′,P N ′={P1,P2,…,P n ,…,P N}, P n It represents the unit data volume corresponding to the nth member, and N represents the total number of members;
[0017] Step 22: Extract the unit number, cross-sectional dimensions, material information, and coordinates of the nodes at both ends of the unit from the unit data body;
[0018] Step 23: Output the extracted unit number, section size, material information, and node coordinates at both ends of the unit into a finite element model format according to a predefined data exchange format, thereby generating an initial finite element model.
[0019] Furthermore, in step 21, the corresponding element representing the rod is searched by searching characters, and the element number and the section name are represented by characters.
[0020] Furthermore, the geometric cleaning of redundant line segments in the centerline model in the centerline model file to obtain a cleaned centerline model includes the following steps:
[0021] Step 31: Determine the redundant line segments that are exported from the BIM model and stored in the centerline model file. The redundant line segments include redundant free line segments, redundant cantilevered line segments that are longer than the actual members, and redundant missing line segments that are shorter than the actual members. When the target grid structure is a truss structure, if there is a diagonal brace on the main member, and no corresponding node is generated on the main member to connect with the diagonal brace, the line segment corresponding to the main member is a redundant line segment, and the redundant line segment is recorded as a redundant line segment of the diagonal brace;
[0022] Step 32: Clean up the redundant free line segments determined in step 31, specifically including:
[0023] Traverse all the bar elements in the centerline model file, calculate the number of times the start point and end point of each bar element appear in the element library, regard the bar element corresponding to the start point and end point that each appears only once as an abnormal bar element, delete the abnormal bar element from the centerline model, including deleting the start point and end point of the abnormal bar element and the abnormal bar element itself, thereby deleting redundant free line segments, and thus obtaining the first centerline model;
[0024] Step 33: clearing the first centerline model of the redundant cantilevered line segments that are longer than the actual rods and the redundant missing line segments that are shorter than the actual rods determined in step 31, specifically including:
[0025] If the current node has only one node number connected to it, the current node is considered as a marked node, and all nodes of the first centerline model are traversed, and each node is considered as the current node, so as to obtain all marked nodes.
[0026] Calculate the distance between each marked node and other nodes respectively, so as to obtain the node corresponding to the minimum distance with the current marked node among other nodes, record the node with the minimum distance ≤ threshold k1 as the minimum distance node, traverse each marked node, so as to obtain the minimum distance node corresponding to all marked nodes, extract the node number of the marked node and the node number of the minimum distance node, and assign the node number and coordinates of the minimum distance node to the marked node, so as to obtain the second centerline model;
[0027] Step 34: Cleaning the redundant line segments of the diagonal brace determined in step 31 from the second centerline model, specifically including:
[0028] Traverse all units in the first centerline model, generate the midpoint coordinates and number of each unit, calculate the distance between the midpoint of each unit and all nodes in the second centerline model, the distance between the midpoint and the node is calculated based on the midpoint coordinates and the node coordinates, the m nodes determined in the second centerline model determined in step 33, then the distance between the unit midpoint and all nodes is also m, that is, the distance set, the distance set is recorded as {Dis1, Dis2, ..., Dis m}, the distance set {Dis1,Dis2,…,Dis m The unit corresponding to the midpoint of the threshold value k2 in} is recorded as the first unit, the first unit and the node with the minimum distance corresponding to the first unit are extracted, and the first unit is deleted from the unit set corresponding to the distance set.
[0029] And generate the unit Et corresponding to the node number with the smallest distance between each unit starting point N1 and the unit midpoint m+1 , and generate the unit Et corresponding to the node number with the smallest distance between each unit starting point N2 and the midpoint of the unit m+2 , and assign the cross-sectional new type of the first unit in the unit set to the unit Et m+1 and unit Et m+2 , thus deleting the original truss unit with diagonal braces, and the original truss unit becomes two new units: unit Et m+1 and unit Et m+2 , thus generating a third centerline model.
[0030] Furthermore, after executing step 34, it also includes repeating step 34 to continuously update the third centerline model, so as to obtain the final third centerline model. The third centerline model properly processed by step 34 or the final third centerline model properly processed by repeating step 34 is the cleaned centerline model.
[0031] Furthermore, the threshold k1 is an empirical value, and the threshold k1 is set to be smaller than the minimum length of the truss brace in the target grid structure.
[0032] Furthermore, the threshold k2=0.1m.
[0033] The beneficial effects of the present invention are as follows: the present invention utilizes the characteristics of BIM model throughout the entire construction life cycle and highly integrated information to extract the parameter information required for the establishment of a large-scale grid structure finite element model to generate a finite element digital model. 1. It can not only break the information islands between models and reduce the workload of repeated model establishment, but also the model conversion method used for different grid structures (grids and trusses) is the same, which brings new value to the BIM model; 2. Large-scale grid structures have many rods and nodes. Compared with the model established by GUI method, the digital finite element model can more efficiently match the changes and corrections that occur in the BIM model during the construction process, and promote the digital transformation of the construction industry; 3. The digital finite element model is also more in line with the concept of digital twin technology, and has stronger interactivity with emerging technologies such as three-dimensional scanning. At the same time, considering the time-varying working conditions of large-scale grid structures at different construction stages, the dynamic interaction of the entire construction process of point cloud models, BIM models and finite element digital models can enable the model to more accurately simulate the real response of the structure. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] Figure 1 A flowchart of the method for model building;
[0035] Figure 2 It is a flow chart of the bridge structure parameter prediction method;
[0036] Figure 3 A schematic diagram of some redundant line segments existing in the centerline model of the target grid structure;
[0037] Figure 4 For Figure 3 Schematic diagram of the finite element command flow model established after the above steps;
[0038] Figure 5 An example diagram of the sphere center coordinates of the welding sphere nodes obtained after fitting the welding sphere based on the point cloud model;
[0039] Figure 6 It is a schematic diagram of a unit composed of four nodes. DETAILED DESCRIPTION
[0040] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments:
[0041] like Figure 1-Figure 6 As shown, a method for converting a grid structure BIM model and a point cloud model to a finite element model includes the following steps:
[0042] Step 1: Obtain the BIM model file of the target grid structure, and export the centerline model file and the surface model file from the BIM model file.
[0043] The grid structure is usually a grid structure or a truss structure or a mixture of the two. The surface model file includes the target grid structure's rods, welded balls and other structural member information. The structural member information includes node, unit, section and material information. When exporting the centerline model file from the BIM model file, the centerline model file format can be selected as DXF format, and the output option can be uncut.
[0044] Step 2: Extract the nodes, elements, sections and material information representing the structural member information from the surface model file, and generate an initial finite element model according to the nodes, elements, sections and material information in a preset defined data exchange format.
[0045] In this step, the relevant information of the rod or welded ball as a node or unit can be extracted from the surface model file. In the target mesh structure, multiple rods are usually included. The rods are usually round tubes, such as round steel tubes. Each rod is equivalent to a unit. The round tube has a cross section and its own material type. Therefore, a finite element model can be generated based on the information including nodes, units, cross sections and materials. Specifically, the specific implementation process of generating an initial finite element model according to the node, unit, cross section and material information in a preset defined data exchange format includes the following steps:
[0046] Step 21: Traverse the surface model file and filter out all the bar units as unit libraries P according to the unit number or section name N ′,P N ′={P1,P2,…,P n ,…,P N}, P n It represents the unit data volume corresponding to the nth member, that is, the unit data volume of the nth unit in the surface model file. N represents the total number of members, that is, the total number of members in the surface model file.
[0047] In the surface model file, the rod (usually a round steel pipe) is usually named pip. Therefore, the cross-section name of the rod is usually represented by the character pip. Therefore, whenever a pip character is searched, it means that a rod is searched, so all rods can be traversed to obtain the unit library P representing all rods. N '. Of course, in actual use, if the rod is a square tube or an I-shaped or other shape, its character is not pip, but has other corresponding characters, which can be searched out according to the corresponding characters.
[0048] Step 22: Extract the unit number, cross-sectional dimensions, material information, and coordinates of the nodes at both ends of the unit from the unit data body.
[0049] In the surface model file, the information length LEN of a bar as a unit is 8. The format of each unit data body is the same, and the information content is also the same. The unit number, section size, material information (usually material type) and the coordinates of the nodes at both ends of the unit are located in the 1st, 4th, 6th and 7th lines of the unit data body, respectively, so that the unit number, section size, material information and the coordinates of the nodes at both ends of the unit can be quickly extracted from the unit data body. Figure 2 , Figure 2 is an example of a unit data body, which includes the unit number, section size, material information, and the coordinates of the two end nodes of the unit. Among them, the unit number and section size are both in the first row, the material information is in the fourth row, and the coordinates of the two end nodes of the unit are in the sixth and seventh rows respectively. The material type in the figure refers to the material information, and the two end points are the unit nodes 1 and 2 in the figure. It can be seen from the figure that the coordinates of the two end nodes are three-dimensional coordinates.
[0050] Step 23: Output the extracted unit number, section size, material information, and node coordinates at both ends of the unit into a finite element model format according to a predefined data exchange format, thereby generating an initial finite element model.
[0051] In the process of converting the unit number, section size, material information, and the coordinates of the nodes at both ends of the unit in the surface model into the finite element model, the unit number, section size, material information, and the coordinates of the nodes at both ends of the unit usually correspond to different syntaxes. The syntax of the same parameter is the same, but the corresponding data is different.
[0052] Step 3: Perform geometric cleanup on redundant line segments in the centerline model in the centerline model file to obtain a cleaned centerline model.
[0053] In this step, whether it is a redundant line segment can be manually determined or features can be manually marked and then the software can automatically extract the features and then automatically determine whether it is a redundant line segment.
[0054] The geometric cleaning of redundant line segments in the centerline model in the centerline model file to obtain the cleaned centerline model includes the following steps:
[0055] Step 31: Determine the redundant line segments that are exported from the BIM model and retained in the centerline model file. The redundant line segments include redundant free line segments, redundant cantilevered line segments that are longer than the actual members, and redundant missing line segments that are shorter than the actual members. When the target grid structure is a truss structure, if there are diagonal braces on the main members, and no corresponding nodes and diagonal brace connections are generated on the main members, then the line segment corresponding to the main members is a redundant line segment, and this redundant line segment is recorded as a redundant diagonal brace line segment.
[0056] refer to Figure 3 , Figure 3 Schematic diagram of some redundant line segments existing in the centerline model of the target grid structure. The line segments in the area framed by the number marks (1-4) in the figure correspond to different redundant line segment situations.
[0057] Step 32: Clear the redundant free line segments determined in step 31.
[0058] In this step, all the bar elements in the centerline model file are traversed, and the number of times the start point and end point of each bar element appear in the element library is calculated respectively. The bar element corresponding to the start point and end point that each appears only once is regarded as an abnormal bar element, and the abnormal bar element is deleted from the centerline model, including deleting the start point and end point of the abnormal bar element and the abnormal bar element itself, that is, deleting the redundant free line segments, thereby obtaining the first centerline model.
[0059] Step 33: Clear the first centerline model of the redundant cantilevered line segments that are longer than the actual rods and the redundant missing line segments that are shorter than the actual rods determined in step 31.
[0060] In this step, if the current node has only one node number connected to the current node, the current node is regarded as a marked node, and all nodes of the first centerline model are traversed, and each node is regarded as the current node respectively, so as to obtain all marked nodes.
[0061] In step 22, it is obtained that both ends of each unit have a node, so one unit corresponds to two nodes.
[0062] Calculate the distance between each marked node and other nodes (including unmarked nodes and marked nodes) to obtain the node with the minimum distance to the current marked node among other nodes, and record the node with the minimum distance ≤ threshold k1 as the node with the minimum distance. Traverse each marked node to obtain the node with the minimum distance corresponding to all marked nodes. Of course, not every marked node has a corresponding node with the minimum distance. Extract the node number of the marked node and the node number of the node with the minimum distance, and assign the node number and coordinates of the node with the minimum distance to the marked node, that is, replace the original node number and node coordinates of the marked node with the node number and coordinates of the node with the minimum distance, so as to obtain the second centerline model.
[0063] The threshold k1 is an empirical value. In this embodiment, the value of the threshold k1 is determined according to the rod size of the target grid structure. Specifically, the threshold k1 can be set to be less than the minimum length of the truss brace in the target grid structure to prevent the truss brace rod from being cleared. For example, in this embodiment, the minimum length of the truss brace is 0.796m (meter), and the threshold k1 is 0.4m.
[0064] Step 34: Clear the redundant line segments of the diagonal brace determined in step 31 from the second centerline model.
[0065] In this step, all units in the second centerline model are traversed, the midpoint coordinates and numbers of each unit are generated, and the distances between the midpoints of each unit and all nodes in the second centerline model are calculated. The distances between the midpoints and nodes are calculated based on the midpoint coordinates and node coordinates. For the m nodes determined in the second centerline model in step 33, there are also m distances between the midpoints of the units and all nodes, i.e., a distance set. The distance set is recorded as {Dis1, Dis2, …, Dis m}, the distance set {Dis1,Dis2,…,Dis m} is recorded as the first unit, and the node with the minimum distance between the first unit and the first unit is extracted, and the node is selected from the distance set {Dis1, Dis2, …, Dis m} corresponds to the unit set {Et1,Et2,…,Et m} and delete the first unit.
[0066] refer to Figure 6 , Figure 6 This step also includes generating a unit Et corresponding to the node number with the smallest distance between each unit starting point N1 and the midpoint of the unit. m+1 , and generate the unit Et corresponding to the node number with the smallest distance between each unit starting point N2 and the midpoint of the unit m+2 , and the unit set {Et1,Et2,…,Et m The cross section of the first element Et1 in} is newly assigned to the element Et m+1 and unit Et m+2 , thereby deleting the original truss unit with diagonal braces, and the original truss unit becomes two new units (unit Et m+1 and unit Et m+2 ), thereby generating a third centerline model.
[0067] by Figure 6 For example, Figure 6There are 4 nodes in it, among which the line segment connecting node 2 and node 4 is a diagonal brace, and the line segment connecting node 1 and node 3 is a truss. For the finite element model, it requires 1-2 unit (i.e., a line segment corresponding to the line segment connecting node 1 and node 2), 2-3 unit, 3-4 unit, 4-1 unit and 4-2 unit. However, in the second centerline model, the corresponding generated units include 1-3 unit, 3-4 unit, 4-1 unit and 4-2 unit. Therefore, it is necessary to further generate the midpoint of each member. Assuming that the 1-3 unit generates the midpoint 5, the distance between the midpoint 5 and the node 2 in all other nodes will be less than the threshold, so it is necessary to delete the 1-3 unit (i.e., the truss unit with a diagonal brace) and replace (generate) the 1-2 unit and the 2-3 unit. Because one unit corresponds to one member, and the member includes two endpoints, one unit also includes two endpoints. It should be noted that the two newly generated units here are just to divide the original truss unit with diagonal braces into two (equivalent to deleting the original truss unit with diagonal braces).
[0068] The above example is a truss structure with only midpoint braces. For components with three-point and four-point braces, the midpoints generated in the above steps can be changed to three-point and four-point. The threshold k2 should theoretically be 0. The generated truss main bar midpoint and brace theoretically coincide. However, considering the slight errors caused by model conversion and data accuracy, in this embodiment, the threshold k2 can be 0.1m, which can be adjusted according to actual conditions.
[0069] Step 34 may be repeated to continuously update the third centerline model, thereby obtaining the final third centerline model.
[0070] The third centerline model processed properly in step 34 or the final third centerline model processed properly in step 34 is the cleaned centerline model.
[0071] In this embodiment, the truss structures on both sides of the target grid structure have diagonal braces only at the midpoints of the rods, and the upper trusses in the middle part have trusses at the quarter points in addition to the midpoints. Step 34 clears the node problems at the midpoints of all trusses. After repeating step 34, the node problems at the quarter points of all trusses can be further cleared. Similarly, if the structure has diagonal braces at the eighth points, repeat step 34, and so on, until all truss diagonal brace nodes are cleared.
[0072] Step 4: Assign the node matching coordinate data of the final third centerline model to the nodes of the initial finite element model through the closest distance matching criterion to complete the establishment of the finite element command flow model.
[0073] refer to Figure 4 , Figure 4 For Figure 3Schematic diagram of the finite element command flow model established after the above steps.
[0074] In this step, all nodes of the final third centerline model and each node of the initial finite element model are traversed to search for the node with the shortest distance from each node of the initial finite element model in the final third centerline model, and the coordinate value of the node with the shortest distance is assigned to the corresponding node of the initial finite element model, thereby establishing the finite element command flow model.
[0075] Step 5: Based on the point cloud model of the target grid structure, the real coordinate information of each bar in the current minimum construction stage of the target grid structure is obtained. The real coordinate information refers to the physical coordinates of the entity bars in the actual grid structure. The real coordinate information of each bar in the point cloud model is dynamically updated based on different construction stages, and the dynamic establishment and correction of the finite element command flow model is realized based on the digital twin.
[0076] At different construction stages, due to different boundary conditions, construction errors, and load conditions, there will be differences between the actual node coordinates and theoretical coordinates of the structure. In order to make the model more accurate, the three-dimensional scanning technology can be used to obtain the true coordinate values of the structural nodes, and then update the extracted theoretical coordinate data to complete the update and correction of the model coordinates. The model correction includes node coordinates, boundary conditions, material parameters, etc. Here, only the correction of coordinate data is considered. The update and correction here are also based on the generation of an accurate theoretical model, so that the model is more in line with the actual situation of the structure, so it is corrected.
[0077] In this step, the actual structural node coordinate information of the target mesh structure at different stages is obtained through the point cloud model of the target mesh structure. In this embodiment, the structural nodes adopt welding ball nodes, and the welding balls are fitted based on the point cloud model to obtain the sphere center coordinates of the welding ball nodes. The obtained sphere center coordinate data is dynamically updated to the finite element command flow model, thereby realizing the dynamic establishment and correction of a high-fidelity finite element command flow model based on digital twins.
[0078] refer to Figure 5 , Figure 5 This is an example diagram of the sphere center coordinates of the welding sphere nodes obtained after fitting the welding sphere based on the point cloud model.
[0079] Among them, the Boolean operation of the BIM model will cause the overlapping parts of the rod and the welded ball node to be cut off, and the endpoints of each rod will not converge on the center of the welded ball, but on the center of the cut sphere. However, for the finite element model, the coordinates of the welded ball node should be the coordinates of the center of the sphere. The nodes of the centerline model will not perform Boolean operations, and all rods at the same node will converge at the node. However, the centerline model has the problem of geometric cleaning. Therefore, the final centerline model obtained after geometric cleaning is the data required for the establishment of the finite element model under ideal conditions. What is missing is only the node coordinates and the nodes corresponding to the rods, so the node coordinates and the corresponding nodes need to be updated.
[0080] The reason why the minimum distance principle is used for matching is that the node coordinates of the centerline model after geometric cleaning are the ideal finite element node coordinates. Therefore, it is only necessary to correct the coordinates of the cut nodes in the finite element model to the corresponding nodes. By finding the node closest to all the nodes of the final centerline model for each node in the finite element model, because each cut node (endpoint) in the finite element model is closest to the node corresponding to the ideal state, this closest distance is the radius of the welding ball node. Therefore, the minimum distance principle can be used to correct the coordinates of the cut nodes in the initial finite element model to the theoretical and actual nodes.
[0081] The present invention utilizes the characteristics of BIM model throughout the entire construction life cycle and highly integrated information to extract the parameter information required for the establishment of a large-scale grid structure finite element model to generate a finite element digital model. 1. It can not only break the information islands between the models and reduce the workload of repeated model establishment, but also for different grid structures (grids and trusses), the model conversion method used is the same, which brings new value to the BIM model; 2. Large-scale grid structures have many rods and nodes. Compared with the model established by GUI method, the digital finite element model can more efficiently match the changes and corrections that occur in the BIM model during the construction process, and promote the digital transformation of the construction industry; 3. The digital finite element model is also more in line with the concept of digital twin technology, and has stronger interactivity with emerging technologies such as three-dimensional scanning. At the same time, considering the time-varying working conditions of large-scale grid structures at different construction stages, the dynamic interaction of the entire construction process of point cloud models, BIM models and finite element digital models can enable the model to more accurately simulate the real response of the structure.
[0082] The embodiment disclosed in this specification is only an example of a unilateral feature of the present invention, and the protection scope of the present invention is not limited to this embodiment, and any other functionally equivalent embodiments fall within the protection scope of the present invention. For those skilled in the art, various other corresponding changes and deformations can be made according to the technical solutions and concepts described above, and all of these changes and deformations should fall within the protection scope of the claims of the present invention.
Claims
1. A method for converting a grid structure BIM model and a point cloud model to a finite element model, characterized in that: The following steps are involved: Step 1: Obtain the BIM model file of the target grid structure, and export the centerline model file and the surface model file from the BIM model file; Step 2: Extract the nodes, elements, sections and material information representing the structural parts from the surface model file, and generate an initial finite element model according to the nodes, elements, sections and material information in a preset data exchange format; Step 3: Perform geometric cleanup on the redundant line segments in the centerline model in the centerline model file to obtain the cleaned centerline model. The geometric cleaning of redundant line segments in the centerline model in the centerline model file to obtain a cleaned centerline model includes the following steps: Step 31: Determine the redundant line segments that are exported from the BIM model and stored in the centerline model file. The redundant line segments include redundant free line segments, redundant cantilevered line segments that are longer than the actual members, and redundant missing line segments that are shorter than the actual members. When the target grid structure is a truss structure, if there is a diagonal brace on the main member, and no corresponding node is generated on the main member to connect with the diagonal brace, the line segment corresponding to the main member is a redundant line segment, and the redundant line segment is recorded as a redundant line segment of the diagonal brace; Step 32: Clean up the redundant free line segments determined in step 31, specifically including: Traverse all the bar elements in the centerline model file, calculate the number of times the start point and end point of each bar element appear in the element library, regard the bar element corresponding to the start point and end point that each appears only once as an abnormal bar element, delete the abnormal bar element from the centerline model, including deleting the start point and end point of the abnormal bar element and the abnormal bar element itself, thereby deleting redundant free line segments, and thus obtaining the first centerline model; Step 33: clearing the first centerline model of the redundant cantilevered line segments that are longer than the actual rods and the redundant missing line segments that are shorter than the actual rods determined in step 31, specifically including: If the current node has only one node number connected to it, the current node is considered as a marked node, and all nodes of the first centerline model are traversed, and each node is considered as the current node, so as to obtain all marked nodes. Calculate the distance between each marked node and other nodes respectively, so as to obtain the node corresponding to the minimum distance with the current marked node among other nodes, record the node with the minimum distance ≤ threshold k1 as the minimum distance node, traverse each marked node, so as to obtain the minimum distance node corresponding to all marked nodes, extract the node number of the marked node and the node number of the minimum distance node, and assign the node number and coordinates of the minimum distance node to the marked node, so as to obtain the second centerline model; Step 34: Cleaning the redundant line segments of the diagonal brace determined in step 31 from the second centerline model, specifically including: Traverse all units in the first centerline model, generate the midpoint coordinates and number of each unit, calculate the distance between the midpoint of each unit and all nodes in the second centerline model, the distance between the midpoint and the node is calculated based on the midpoint coordinates and the node coordinates, the m nodes determined in the second centerline model determined in step 33, then the distance between the unit midpoint and all nodes is also m, that is, the distance set, the distance set is recorded as , the distance set The unit corresponding to the midpoint of the value < threshold k2 is recorded as the first unit, the node with the minimum distance between the first unit and the first unit is extracted, and the first unit is deleted from the unit set corresponding to the distance set. And generate the unit corresponding to the node number with the smallest distance between each unit starting point N1 and the midpoint of the unit , and generate the unit corresponding to the node number with the smallest distance between each unit starting point N2 and the midpoint of the unit , and assign the cross-section information of the first element in the element set to the element and unit , thereby deleting the original truss unit with the diagonal brace, and the original truss unit becomes two new units: and unit , thereby generating a third centerline model; Step 4: Assign the node matching coordinate data of the cleaned centerline model to the nodes of the initial finite element model through the nearest distance matching criterion, and complete the establishment of the finite element command flow model; Step 5: Based on the point cloud model of the target grid structure, the real coordinate information of each rod of the target grid structure is obtained, and the real coordinate information of each rod in the point cloud model is dynamically updated, and the dynamic establishment and correction of the finite element command flow model is realized based on the digital twin.
2. The method for converting a grid structure BIM model and a point cloud model to a finite element model according to claim 1, characterized in that: The target grid structure is a grid structure or a truss structure, and the surface model file includes structural member information, and the structural member information includes node, unit, section and material information.
3. The method for converting a grid structure BIM model and a point cloud model to a finite element model according to claim 1, characterized in that: Structural parts information includes rods and welding balls.
4. The method for converting a grid structure BIM model and a point cloud model to a finite element model according to claim 1, characterized in that: In step 1, export the centerline model file from the BIM model file, select the DXF format for the centerline model file, and set the output option to uncut.
5. The method for converting a grid structure BIM model and a point cloud model to a finite element model according to claim 1, characterized in that: The specific implementation process of generating the initial finite element model according to the node, element, section and material information in a preset data exchange format includes the following steps: Step 21: Traverse the surface model file and filter out the unit library consisting of all rods as units according to the unit number or section name , , Represents the unit data volume corresponding to the nth member, and N represents the total number of members; Step 22: Extract the unit number, cross-sectional dimensions, material information, and coordinates of the nodes at both ends of the unit from the unit data body; Step 23: Output the extracted unit number, section size, material information, and node coordinates at both ends of the unit into a finite element model format according to a predefined data exchange format, thereby generating an initial finite element model.
6. The method for converting a grid structure BIM model and a point cloud model to a finite element model according to claim 5, characterized in that: In step 21, the corresponding element representing the rod is searched by searching characters, and the element number and section name are represented by characters.
7. The method for converting a grid structure BIM model and a point cloud model to a finite element model according to claim 1, characterized in that: After executing step 34, it also includes repeating step 34 to continuously update the third centerline model, so as to obtain the final third centerline model. The third centerline model properly processed by step 34 or the final third centerline model properly processed by repeating step 34 is the cleaned centerline model.
8. The method for converting a grid structure BIM model and a point cloud model to a finite element model according to claim 1, characterized in that: The threshold k1 is an empirical value, and the threshold k1 is set to be smaller than the minimum length of the truss brace in the target grid structure.
9. The method for converting a grid structure BIM model and a point cloud model to a finite element model according to claim 1, characterized in that: Threshold k2=0.1m.
