Building structure simulation result visualized interactive system

Abstract

The application relates to the technical field of structural simulation interaction, in particular to a building structure simulation result visual interaction system, which comprises the following modules: a data analysis module, which is used for acquiring a geometric model and a finite element analysis result file in parallel and extracting a coordinate sequence; a grid mapping module, which is used for calling a shape function interpolation algorithm based on physical field data flow to calculate local coordinate weights and projecting data to rendering grid vertices; an interactive detection module, which is used for capturing virtual probe tip world coordinates in real time and logically searching for surrounding elements based on voxel indexes; and a dynamic rendering module, which is used for rendering an over-limit area in a highlighted flickering mode in an enhanced three-dimensional rendering model in response to a judgment signal. In the application, local grid subdivision is performed on a high stress area according to a stress gradient change rate, an enhanced three-dimensional rendering model is constructed, a Boolean operation cutting is performed along a principal stress direction, and a dynamic section view and a local contour map are generated.

Description

Technical Field

[0001] This invention relates to the field of structural simulation interaction technology, and in particular to a visualization and interactive system for building structure simulation results. Background Technology

[0002] The field of structural simulation interactive technology encompasses a general technical scope that utilizes computer graphics and numerical calculation methods to perform three-dimensional visualization and dynamic analysis of the physical performance data of various building entities. Traditional building structure simulation result visualization interactive systems refer to systems that calculate stress, strain, and displacement data of building structures using finite element analysis software, import these discrete numerical results into a post-processing module, map them to different color levels based on their magnitude, and ultimately generate static color cloud maps or deformable mesh maps, which are then presented on a two-dimensional display screen for engineers to review and verify.

[0003] Existing technologies typically use static images generated directly from finite element calculation meshes for visualization. The mesh density of these images often differs from that of the geometric model mesh used for rendering. This leads to data interpolation distortion or jagged edges when scaling across different levels or viewing local details. Furthermore, this static mapping method cannot support users in conducting real-time stress data exploration of internal structures at arbitrary cross-sections or specific depths in three-dimensional space. This results in the negative consequences of unintuitive stress analysis of complex nodes and low efficiency in identifying potential structural hazards. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a visualization and interactive system for building structure simulation results.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a visualization and interactive system for building structure simulation results includes: The data parsing module is used to acquire the geometric model file and finite element analysis result file of the building project in parallel, parse and extract the spatial coordinate sequence of geometric vertices and the stress tensor and element topology index of the finite element nodes, perform normalization and cleaning operations on the stress tensor of the finite element nodes, and generate a physical field data stream. The mesh mapping module is used to calculate the local coordinate weight of each geometric rendering vertex in the finite element unit by calling the shape function interpolation algorithm through the physical field data stream and the spatial coordinate sequence, project the physical field data onto the rendering mesh vertex, perform local mesh subdivision on the high stress area according to the stress gradient change rate, and construct an enhanced 3D rendering model. The interactive detection module is used to capture the world coordinates of the tip of the virtual probe in the enhanced 3D rendering model in real time, retrieve the bounding cell based on the voxel index logic, perform trilinear interpolation on the node attribute values ​​of the bounding cell to obtain the precise physical value of the current probe position, compare the precise physical value with a preset safety threshold, and generate a judgment signal. The dynamic rendering module is used to render the over-limit area in the enhanced 3D rendering model in a high-brightness flashing manner in response to the judgment signal, perform Boolean operation cutting along the principal stress direction, and generate dynamic section view and local contour map.

[0006] As a further aspect of the present invention, the specific function of the data parsing module is as follows: The geometric topology alignment submodule is used to read the vertex index list in the geometric model file, load the node number mapping table in the finite element analysis result file, establish the spatial correspondence between geometric vertices and finite element nodes through a hash matching algorithm, remove redundant geometric primitives that have not participated in the calculation, and output the aligned topology index. The stress tensor extraction submodule is used to traverse the aligned finite element nodes, extract six-dimensional stress data including normal stress components and shear stress components from the binary data block of the finite element analysis result file, assemble them into the corresponding second-order tensor matrix, and synchronously associate the displacement vector and temperature scalar of the node to generate the original node attribute set. The data cleaning and normalization submodule is used to receive the original set of node attributes, impute missing values ​​and remove singular values, map the numerical range of multiple physical quantities to a standardized dimensionless interval, and package and serialize the processed node attributes with the topology index to generate a physical field data stream.

[0007] As a further aspect of the present invention, the specific function of the mesh mapping module is as follows: The local weight calculation submodule is used to parse the unit topology information in the physical field data stream. For each geometric rendering vertex, it uses an inverse mapping algorithm to determine its natural coordinate position inside the corresponding finite element unit, and constructs a shape function matrix describing the degree of influence of multiple nodes of the unit on the vertex, generating local coordinate weights. The attribute projection mapping submodule is used to perform weighted average calculation of the physical properties of the finite element nodes through the local coordinate weights, accurately transfer stress, displacement and temperature data from the sparse finite element mesh to the high-density rendering mesh vertices, and establish a dynamic binding relationship between rendering vertices and physical property values ​​to generate a basic rendering model. The mesh subdivision and reconstruction submodule is used to traverse all triangular faces of the basic rendering model, calculate the stress gradient change rate between adjacent vertices, identify regions where the gradient value exceeds a preset subdivision threshold, perform a center split operation on the triangular faces within the region and update the vertex connection relationship, increase the mesh density in the high stress gradient region, and construct the enhanced 3D rendering model.

[0008] As a further aspect of the present invention, the specific function of the interactive detection module is as follows: The probe coordinate tracking submodule is used to monitor the spatial displacement data of the user input device in real time through the graphical interface, update the position and attitude matrix of the virtual probe in the 3D scene, extract the 3D vector of the probe tip in the world coordinate system, and convert it into normalized coordinates relative to the model bounding box to generate the tip world coordinates. The unit index retrieval submodule is used to pre-construct the voxel index logic of the enhanced 3D rendering model using the octree space segmentation algorithm, quickly traverse the octree nodes according to the tip world coordinates, locate the smallest leaf node where the probe is currently located, and filter the target tetrahedron or hexahedron that actually includes the probe coordinates from the candidate unit list included in the leaf node to generate the enclosing unit. The attribute interpolation determination submodule is used to obtain the physical attribute values ​​of multiple vertices surrounding the unit and the relative position coefficient of the probe within the unit, perform trilinear interpolation based on volume coordinates to solve the instantaneous physical quantity of the probe point, compare the instantaneous physical quantity with the preset safety threshold of the material strength, and if the difference exceeds the threshold, trigger an alarm state and generate a determination signal.

[0009] As a further aspect of the present invention, the specific function of the dynamic rendering module is as follows: The abnormal highlight warning submodule is used to receive the judgment signal. If the signal indicates that there is an over-limit area, the pulse function is activated in the fragment shader to modify the material emission color of the rendering mesh vertex corresponding to the over-limit area to a high-saturation warning color, and the transparency channel is dynamically adjusted according to the sine wave law to render the over-limit area in the enhanced 3D rendering model in a highlight flashing manner. The dynamic section generation submodule is used to calculate the direction of the principal stress vector based on the stress tensor eigenvalue at the probe position, construct the equation of the section plane passing through the probe point and with the normal vector parallel to the principal stress direction, perform real-time Boolean subtraction on the model using template buffer technology, remove geometry on one side of the section plane, and generate a dynamic section view. The contour plotting submodule is used to map a continuous physical scalar field into discrete color textures on the cross section of the generated dynamic sectional view, extract pixels of the target numerical range through a fragment discarding algorithm to draw contour lines, and superimpose and project the contour lines onto the sectional surface to generate a local contour plot.

[0010] As a further aspect of the present invention, the data cleaning process of the data cleaning and normalization submodule includes: Obtain all stress tensor components in the original node attribute set, calculate the Mises equivalent yield stress value for each node, and calculate the mean and standard deviation of the stress values ​​across the entire field. The Raida criterion was used to identify and remove outlier data nodes that deviated from the mean by more than three standard deviations, and the weighted average of the nearest neighbor nodes was used to fill in the removal positions. Based on the material ultimate strength parameter corresponding to the preset safety threshold, the repaired stress value is linearly mapped to a floating-point range of zero to one, while retaining the metadata marker of the original value, thus generating a physical field data stream.

[0011] As a further aspect of the present invention, the calculation process of the local weight calculation submodule includes: Obtain the node coordinate matrix of the finite element, including the target rendering vertex, and establish the Jacobian transformation matrix from the global Cartesian coordinate system to the element natural coordinate system; The Newton-Raphson iteration method is used to solve the nonlinear equation system and calculate the parameter coordinates of the target rendering vertex in the natural coordinate system, ensuring that the iteration residual is less than the preset convergence tolerance. Substitute the obtained parameter coordinates into the Lagrange function expression of the element type, calculate the contribution coefficients of multiple finite element nodes to the rendered vertex, establish a weight vector including the weights of all nodes, and generate local coordinate weights.

[0012] As a further aspect of the present invention, the calculation process of the stress gradient change rate based on the mesh subdivision reconstruction submodule includes: Obtain the physical field values ​​of the three vertices of the current triangular facet and the geometric feature scale of the facet. Based on the ratio of the difference in physical quantities of neighboring vertices to their spatial distance, calculate the stress gradient change rate of the facet using the following gradient evaluation formula. : ; in, Represents the rate of change of the stress gradient. This represents the surface area value of the current triangular facet. Representative noodle piece The normalized stress values ​​at each vertex, This represents the average stress value at the three vertices of the facet. Representing the The length of the edge opposite each vertex. This represents the geometry factor used to correct the weight of the elongated triangle; The calculated stress gradient change rate is compared with a preset subdivision threshold to filter the set of target patches that need to be split.

[0013] As a further aspect of the present invention, the interpolation calculation process of the attribute interpolation determination submodule includes: Obtain the baseline attribute values ​​of the eight corner points of the enclosing cell in the natural coordinate system, as well as the offset vector of the probe point relative to the cell center; Linear interpolation is performed sequentially along the three principal axes of the natural coordinate system. First, the projection point attributes are calculated on the four edges parallel to the X-axis. Then, the secondary projection point attributes are calculated on the two Y-axis lines formed by the aforementioned projection points. Finally, the attribute value of the probe position is determined on the line segment connected in the Z-axis direction. The scalar and tensor properties obtained by interpolation are stored separately, and the interpolation results are corrected by combining the anisotropy coefficient of the material to eliminate the influence of mesh distortion on the interpolation accuracy and generate accurate physical values.

[0014] As a further aspect of the present invention, the process of constructing the cutting surface of the dynamic cutting generation submodule includes: Obtain the stress tensor matrix at the current position of the probe, call the Jacobi rotation algorithm to solve for the eigenvalues ​​and eigenvectors of the real symmetric matrix, and determine the eigenvector corresponding to the eigenvalue with the largest absolute value as the first principal stress direction; Using the spatial coordinates of the probe tip as the origin and the direction of the first principal stress as the normal vector, a point normal plane equation is established, and the orientation of the plane's normal is adaptively adjusted according to the viewpoint position to ensure that the cutting surface is facing the user's line of sight. A four-dimensional clipping vector is generated to describe the cutting plane, and it is passed to the rendering pipeline of the graphics processing unit. During the rasterization stage, all fragments in the negative half of the plane are discarded to generate a dynamic cut view.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, the geometric model of the building project and the finite element analysis results are acquired in parallel. The coordinates of the geometric vertices and the stress tensors of the finite element nodes are extracted by analysis. The local coordinate weights of the geometric rendering vertices are calculated using the shape function interpolation algorithm. The physical field data is projected onto the vertices of the rendering mesh. The high-stress area is subdivided locally according to the stress gradient change rate. The world coordinates of the virtual probe are captured and the bounding cells are retrieved based on the voxel index logic. Boolean operations are performed along the principal stress direction to cut and generate a dynamic section view. This eliminates the distortion of data when scaling across levels and enables real-time and accurate exploration of the internal structure in three-dimensional space. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the interactive visualization system for building structure simulation results of the present invention. Figure 2 This is a flowchart illustrating the internal logic interaction of the data parsing module in this invention. Figure 3 This is a flowchart of the data processing of the grid mapping module of the present invention; Figure 4 This is a flowchart illustrating the operational logic of the interactive detection module of the present invention. Figure 5 This is a flowchart of the rendering process of the dynamic rendering module of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the software-based technical solution is described in detail below with reference to system architecture diagrams and embodiments. It should be understood that the specific embodiments described herein are only for explaining the technical solutions of this invention and do not constitute a limitation on the scope of protection.

[0018] In the description of this invention, the system architecture relationships or data processing flows indicated by terms such as "layer," "module," "interface," "data flow," "client," and "server" are all defined based on the architecture diagram or flowchart corresponding to the embodiments. This way of describing is only used to clearly illustrate the logical relationships between the elements in the technical solution, and not to limit the physical deployment form. The term "multiple" includes two or more technical units, including but not limited to multiple data nodes, processing threads, service instances, or functional components and other scalable elements. The specific number is determined according to the actual business scenario and needs to be specifically specified.

[0019] Please see Figure 1 and Figure 2 This invention provides a technical solution: a visual interactive system for building structure simulation results, comprising: The data parsing module is used to acquire geometric model files and finite element analysis result files of building engineering in parallel, parse and extract the spatial coordinate sequence of geometric vertices and the stress tensor and element topology index of finite element nodes, perform normalization and cleaning operations on the stress tensor of finite element nodes, and generate physical field data stream.

[0020] The specific functions of the data parsing module are as follows: The geometric topology alignment submodule is used to read the vertex index list in the geometric model file, load the node number mapping table in the finite element analysis result file, establish the spatial correspondence between geometric vertices and finite element nodes through a hash matching algorithm, remove redundant geometric primitives that have not participated in the calculation, and output the aligned topology index. The stress tensor extraction submodule is used to traverse the aligned finite element nodes, extract six-dimensional stress data including normal stress components and shear stress components from the binary data block of the finite element analysis result file, assemble them into the corresponding second-order tensor matrix, and synchronously associate the displacement vector and temperature scalar of the node to generate the original node attribute set. The data cleaning and normalization submodule receives the original set of node attributes, imputes missing values ​​and removes singular values, maps the numerical ranges of multiple physical quantities to a standardized dimensionless interval, and packages and serializes the processed node attributes and topology index to generate a physical field data stream.

[0021] The data cleaning process in the data cleaning and normalization submodule includes: Obtain all stress tensor components in the original node attribute set, calculate the Mises equivalent yield stress value for each node, and calculate the mean and standard deviation of the stress values ​​across the entire field. The Raida criterion was used to identify and remove outlier data nodes that deviated from the mean by more than three standard deviations, and the weighted average of the nearest neighbor nodes was used to fill in the removal positions. Based on the material ultimate strength parameter corresponding to the preset safety threshold, the repaired stress value is linearly mapped to the floating-point range of zero to one, while retaining the metadata tag of the original value to generate a physical field data stream.

[0022] The data parsing module is configured to run in the memory space of a high-performance workstation, serving as the foundational data access point for the entire visualization and interactive system. The specific functionalities of this module include a geometric topology alignment submodule, a stress tensor extraction submodule, and a data cleaning and normalization submodule.

[0023] The geometric topology alignment submodule is specifically responsible for reconstructing spatial consistency from heterogeneous data sources. In its implementation, this submodule reads the geometric model file of the building project through the standard interface of the industrial basic class, parses and stores the Cartesian coordinates of the geometric vertices. The result is loaded into a double-precision floating-point array. Simultaneously, this submodule loads the finite element analysis result file in parallel and extracts the node number mapping table. To address the minor spatial discrepancies between the geometric model and the finite element model, this submodule incorporates spatial hash matching logic. This logic first defines a spatial bounding box, dividing the three-dimensional space into sections with sides of length... A cubic mesh. In this embodiment, The value is set to one-tenth of the minimum feature size of the model, specifically 5.0 mm. The submodule maps all finite element nodes to a hash table, with the key being the mesh index and the value being a list of node IDs. Subsequently, the submodule iterates through each geometric vertex, calculates its hash bucket index, and performs Euclidean distance calculations between it and the finite element nodes within that bucket and its 26 neighboring buckets. If the minimum distance is less than a preset tolerance threshold... If the two are determined to be at the same physical location, a mapping key-value pair between the geometric vertex ID and the finite element node ID is established. In this embodiment, the tolerance threshold... Set as For geometric primitives that fail to match, the submodule performs a culling operation, retaining only the topological indexes that are involved in the mechanical calculations, and outputs the aligned topological mapping table.

[0024] The stress tensor extraction submodule focuses on reconstructing high-dimensional mechanical data from a binary stream. This submodule is configured with a binary decoder for specific CAE software result files, enabling direct file pointer location to node solution data blocks. This submodule is configured to traverse the finite element node index aligned by the preceding module, and for each node, read its stored six-dimensional stress components: normal stress... With shear stress This submodule is responsible for constructing the second-order stress tensor matrix. For each node Instantiate a submodule in memory The real symmetric matrix object is used to fill the six extracted components into their corresponding positions according to the tensor definition. Simultaneously, this submodule extracts the displacement vector of the node under the current load step. and temperature scalar This submodule constructs an array of structures indexed by node IDs in memory. Each structure contains a tensor matrix, a displacement vector, and a temperature value, forming the original set of node attributes. During this process, the submodule verifies the floating-point precision of the data, forcibly converting single-precision data to double-precision format. All extracted data is tagged with metadata, including load step index and substep index.

[0025] The data cleaning and normalization submodule integrates a statistical analysis engine and an outlier filter. After receiving the original set of node attributes, this submodule first performs scalarization calculations of the physical quantities. For the stress tensor of each node, the submodule calls the Mises yield criterion algorithm to calculate the equivalent nodal stress value. During the data cleaning phase, this submodule is configured to statistically analyze all nodes across the entire dataset. arithmetic mean of values with standard deviation To eliminate non-physical singularities caused by mesh distortion or computational divergence, the submodule strictly adheres to the Laida criterion. The specific decision logic is as follows: if the stress value of a node... satisfy If the node is not found to be an anomaly, the submodule automatically searches for the nearest neighbor in the topological neighborhood of that anomaly. For each normal node, an inverse distance-weighted interpolation method is used to calculate the interpolation value, replacing the abnormal values. In this embodiment, The value is set to 5. Subsequently, this submodule performs normalization mapping. The system presets a safety threshold. This value is set based on the yield strength of the building material. In this embodiment, it is based on the Q345 steel standard. The pressure is set to 345 MPa. The submodule uses a linear mapping function to map the repaired stress value to a dimensionless range of zero to one. The calculation formula is as follows: This operation transforms physical quantities into texture coordinates or color factors suitable for processing by the graphics rendering pipeline, and finally packages and serializes the processed attribute data with the topology index to generate a standardized physical field data stream.

[0026] Please see Figure 1 and Figure 3 The mesh mapping module is used to calculate the local coordinate weight of each geometric rendering vertex in the finite element unit by calling the shape function interpolation algorithm through the physical field data stream and spatial coordinate sequence, project the physical field data onto the rendering mesh vertex, perform local mesh subdivision in the high stress area according to the stress gradient change rate, and build an enhanced 3D rendering model.

[0027] The specific functions of the mesh mapping module are as follows: The local weight calculation submodule is used to parse the unit topology information in the physical field data stream. For each geometrically rendered vertex, it uses an inverse mapping algorithm to determine its natural coordinate position inside the corresponding finite element unit, and constructs a shape function matrix describing the degree of influence of multiple nodes of the unit on the vertex, generating local coordinate weights.

[0028] The calculation process of the local weight calculation submodule includes: Obtain the node coordinate matrix of the finite element, including the target rendering vertex, and establish the Jacobian transformation matrix from the global Cartesian coordinate system to the element natural coordinate system; The Newton-Raphson iteration method is used to solve the nonlinear equation system and calculate the parameter coordinates of the target rendering vertex in the natural coordinate system, ensuring that the iteration residual is less than the preset convergence tolerance. Substitute the obtained parameter coordinates into the Lagrange function expression of the element type, calculate the contribution coefficients of multiple finite element nodes to the rendered vertex, establish a weight vector including the weights of all nodes, and generate local coordinate weights.

[0029] The attribute projection mapping submodule is used to perform weighted average calculation of the physical properties of finite element nodes through local coordinate weights, accurately transfer stress, displacement and temperature data from sparse finite element mesh to high-density rendering mesh vertices, and establish a dynamic binding relationship between rendering vertices and physical property values ​​to generate the basic rendering model. The Mesh Subdivision Reconstruction Submodule is used to traverse all triangles of the base rendering model, calculate the rate of change of stress gradient between adjacent vertices, identify regions where the gradient value exceeds the preset subdivision threshold, perform a center split operation on the triangles within the region and update the vertex connection relationship, increase the mesh density in high stress gradient regions, and construct an enhanced 3D rendering model.

[0030] The calculation process for the stress gradient change rate, which is the basis of the mesh subdivision and reconstruction submodule, includes: Obtain the physical field values ​​of the three vertices of the current triangular facet and the geometric feature scale of the facet. Based on the ratio of the difference in physical quantities of neighboring vertices to their spatial distance, calculate the stress gradient change rate of the facet using the following gradient evaluation formula. : ; in, Represents the rate of change of stress gradient. This represents the surface area value of the current triangular facet. Representative noodle piece The normalized stress values ​​at each vertex, This represents the average stress value at the three vertices of the facet. Representing the The length of the edge opposite each vertex. This represents the geometry factor used to correct the weight of the elongated triangle; The calculated stress gradient change rate is compared with the preset subdivision threshold to select the set of target patches that need to be split.

[0031] The mesh mapping module is used to accurately attach physical data to the rendering model. Its specific functions include a local weight calculation submodule, an attribute projection mapping submodule, and a mesh subdivision and reconstruction submodule.

[0032] The local weight calculation submodule has a built-in shape function library and is configured to find the element to which each geometric vertex belongs in the finite element mesh for each vertex to be rendered. In its implementation, this submodule first uses a spatial octree index to quickly locate the geometric vertices. The finite element unit Subsequently, the submodule acquires the unit. Construct a global Cartesian coordinate system from the global coordinates of all corner nodes. To the unit natural coordinate system Jacobian transformation matrix Since the mapping is nonlinear for irregular hexahedral elements, a Newton-Raphson iterative solver is integrated within the submodule. The solver uses the element center as the initial value and iteratively corrects the parametric coordinates. until the calculated spatial coordinates are consistent with The Euclidean distance residual is less than the convergence tolerance. In this embodiment, the convergence tolerance... Set as Once convergent natural coordinates are obtained... The submodule substitutes it into the Lagrange function expression of the corresponding unit type. The submodule stores this set of weights as a weight vector. And ensure that the sum of all weights equals 1.0, and generate local coordinate weight data.

[0033] The attribute projection mapping submodule is configured to receive the physics data stream and local coordinate weights in parallel, and perform batch weighting operations using the parallel computing capabilities of the graphics processing unit. For each rendered vertex, the submodule reads its corresponding finite element node index list and local weight vector. The operational logic executed by this submodule is as follows: Suppose a certain rendering vertex is affected by m finite element nodes, with weights respectively... The physical properties of the corresponding node are The submodule calculates the attribute values ​​of the rendered vertex. This submodule handles displacement vectors. Performing the same weighted composition, the deformation of the rendered vertices is obtained. After the calculation is complete, this submodule establishes a dynamic binding relationship between the rendered vertex indices and physical property values. The submodule then calculates the normalized stress values. Mapped to one-dimensional texture coordinates, the RGB color value of the vertex is determined by consulting a preset pseudo-color spectrum, and a basic rendering model is generated.

[0034] The mesh subdivision and reconstruction submodule is configured to dynamically optimize the geometric refinement of the rendered model. This submodule iterates through all triangles of the base rendered model and calculates the rate of change of stress gradient. This submodule calculates the stress gradient change rate of the facet based on the following gradient evaluation formula. : ; in, This represents the rate of change of the stress gradient, used to quantify the degree of drastic change in physical quantities in the current patch region; This represents the surface area value of the current triangular facet. Representative noodle piece Normalized stress values ​​at each vertex; This represents the average stress value at the three vertices of the facet. Representing the The numerical value of the length of the edge opposite each vertex; This represents the geometry factor, used to correct for the weighting effect of elongated triangles.

[0035] In this embodiment, to verify the calculation logic of the formula, a local mesh of a high-stress zone at a steel structure connection node is selected as an example. Through measurement and acquisition, the relevant parameters are assigned the following values: current triangular patch area. for Normalized stress values ​​at the three vertices , , The values ​​are 0.9, 0.5, and 0.55 respectively. The average stress can be calculated. It is 0.65; the length of the opposite side corresponding to the three vertices. , , The values ​​are 0.2m, 0.25m, and 0.22m respectively; geometric shape factor. Based on the aspect ratio of the triangle being set to 1.05, the above parameters are substituted into the formula for calculation: First, calculate the components of each term in the summation term: First item: ; Second item: ; Third item: ; The summation result is ; The final calculation yielded .

[0036] This submodule compares the calculated result 59.81 with the preset subdivision judgment threshold. A comparison is made. In this embodiment, Set to 30.0. Because... The result indicates that the stress change rate in the current region exceeds the allowable smooth transition range, belonging to a high gradient region. The submodule then marks the triangular facet as the target facet, performs a center split operation, and constructs an enhanced 3D rendering model.

[0037] Please see Figure 1 and Figure 4 The interactive detection module is used to capture the world coordinates of the virtual probe's tip in the enhanced 3D rendering model in real time. It retrieves the bounding cell based on voxel index logic, performs trilinear interpolation on the node attribute values ​​of the bounding cell, obtains the precise physical value of the current probe position, compares the precise physical value with a preset safety threshold, and generates a judgment signal.

[0038] The specific functions of the interactive detection module are as follows: The probe coordinate tracking submodule is used to monitor the spatial displacement data of the user input device in real time through the graphical interface, update the position and attitude matrix of the virtual probe in the 3D scene, extract the 3D vector of the probe tip in the world coordinate system, and convert it into normalized coordinates relative to the model bounding box to generate the tip world coordinates. The cell index retrieval submodule is used to pre-build the voxel index logic of the enhanced 3D rendering model using the octree spatial segmentation algorithm. It quickly traverses the octree nodes according to the world coordinates of the tip, locates the smallest leaf node where the probe is currently located, and filters the target tetrahedron or hexahedron that actually includes the probe coordinates from the candidate cell list included in the leaf node to generate the enclosing cell. The attribute interpolation judgment submodule is used to obtain the physical attribute values ​​of multiple vertices surrounding the unit and the relative position coefficient of the probe within the unit. It performs trilinear interpolation based on volume coordinates to solve for the instantaneous physical quantity of the probe point. The instantaneous physical quantity is compared with the preset safety threshold of the material strength. If the difference exceeds the threshold, an alarm state is triggered and a judgment signal is generated.

[0039] The interpolation calculation process of the attribute interpolation determination submodule includes: Obtain the baseline attribute values ​​of the eight corner points of the enclosing cell in the natural coordinate system, as well as the offset vector of the probe point relative to the cell center; Linear interpolation is performed sequentially along the three principal axes of the natural coordinate system. First, the projection point attributes are calculated on the four edges parallel to the X-axis. Then, the secondary projection point attributes are calculated on the two Y-axis lines formed by the aforementioned projection points. Finally, the attribute value of the probe position is determined on the line segment connected in the Z-axis direction. The scalar and tensor properties obtained by interpolation are stored separately, and the interpolation results are corrected by combining the anisotropy coefficient of the material to eliminate the influence of mesh distortion on the interpolation accuracy and generate accurate physical values.

[0040] The interactive detection module is used to enable real-time interaction between the user and the three-dimensional physical field. Its specific functions include a probe coordinate tracking submodule, a unit index retrieval submodule, and an attribute interpolation determination submodule.

[0041] The probe coordinate tracking submodule is responsible for mapping user actions to a virtual 3D space. This submodule continuously monitors the signal flow from the input device and maintains a position and orientation matrix for a virtual probe object. When the user moves the device, the submodule updates this matrix and extracts the vector of the probe's tip in the world coordinate system. To adapt to architectural models of different scales, the submodule needs to convert these coordinates to normalized coordinates relative to the model's bounding box. Let the minimum point of the model's bounding box be... The maximum point is Model size The submodule calculates normalized coordinates. During this process, the submodule applies the Kalman filter algorithm to smooth the input displacement data in order to eliminate coordinate noise caused by hand tremors.

[0042] The cell index retrieval submodule pre-constructs an octree spatial index structure for the enhanced 3D rendering model. The depth of this octree is dynamically set according to the total number of model cells, ensuring that the number of cells contained in each leaf node does not exceed a preset upper limit. In this embodiment, this upper limit is set to 50 cells. When the tip world coordinates are received, the submodule starts from the root node of the octree and determines the child node quadrant in which the coordinate point falls layer by layer until the smallest leaf node is reached. This leaf node stores a list of candidate cells. Subsequently, the submodule traverses all tetrahedral or hexahedral cells in the list and executes an algorithm to determine if a point is within a polyhedron. Once a target is locked, the submodule immediately outputs the globally unique identifier and topological connectivity of the cell, generating enclosing cell data.

[0043] The attribute interpolation determination submodule is responsible for calculating the precise physical values ​​of the probe position. This submodule obtains the corner attribute values ​​of the enclosing element and the relative position of the probe within the element. For hexahedral elements, this submodule performs trilinear interpolation based on isoparametric coordinates. The interpolation process is as follows: First, the submodule reads the physical property values ​​of the eight corner points of the enclosing unit. to And calculate its scaling factor in the local coordinate system of the element based on the probe coordinates. ; Secondly, perform fourth-order linear interpolation along the X-axis and calculate... Similarly, calculate ; Next, perform two linear interpolations along the Y-axis and calculate... and ; Finally, a linear interpolation is performed along the Z-axis to obtain... .

[0044] This submodule will calculate the instantaneous physical quantities The difference is compared with the system's preset material safety threshold. In this embodiment, the material safety threshold is set to 345 MPa. If When the pressure exceeds 345 MPa, the submodule immediately generates a high-level judgment signal. In addition, the submodule also incorporates the anisotropy coefficient matrix of the material to correct the interpolation results, preventing interpolation distortion caused by mesh non-orthogonality and generating accurate physical values.

[0045] Please see Figure 1 and Figure 5The dynamic rendering module is used to render the over-limit area in a highlighted flashing manner in response to the judgment signal in the enhanced 3D rendering model, perform Boolean operation cutting along the principal stress direction, and generate dynamic section view and local contour map.

[0046] The specific functions of the dynamic rendering module are as follows: The abnormal highlight warning submodule is used to receive judgment signals. If the signal indicates that there is an over-limit area, the pulse function is activated in the fragment shader to modify the material emission color of the rendering mesh vertex corresponding to the over-limit area to a high-saturation warning color, and the transparency channel is dynamically adjusted according to the sine wave law to render the over-limit area in a highlight flashing mode in the enhanced 3D rendering model. The dynamic section generation submodule is used to calculate the principal stress vector direction based on the stress tensor eigenvalue at the probe position, construct the section plane equation passing through the probe point and with the normal vector parallel to the principal stress direction, perform real-time Boolean subtraction on the model using template buffer technology, remove geometry on one side of the section plane, and generate a dynamic section view.

[0047] The process of constructing the cutting surface of the dynamic section generation submodule includes: Obtain the stress tensor matrix at the current position of the probe, call the Jacobi rotation algorithm to solve for the eigenvalues ​​and eigenvectors of the real symmetric matrix, and determine the eigenvector corresponding to the eigenvalue with the largest absolute value as the first principal stress direction; Using the spatial coordinates of the probe tip as the origin and the direction of the first principal stress as the normal vector, a point-normal plane equation is established, and the orientation of the plane's normal is adaptively adjusted according to the viewpoint position to ensure that the cutting surface is facing the user's line of sight. A four-dimensional clipping vector is generated to describe the cutting plane, and it is passed to the rendering pipeline of the graphics processing unit. During the rasterization stage, all fragments in the negative half of the plane are discarded to generate a dynamic cut view.

[0048] The contour plotting submodule is used to map a continuous physical scalar field into discrete color textures on the cross section of the generated dynamic sectional view. It extracts pixels of the target numerical range through a fragment discarding algorithm to draw contour lines, and then superimposes and projects the contour lines onto the sectional surface to generate a local contour plot.

[0049] The dynamic rendering module is used to respond to judgment signals and provide a visual analysis view. Its specific functions include an anomaly highlighting warning submodule, a dynamic section generation submodule, and a contour map drawing submodule.

[0050] The abnormal highlight warning submodule is configured to directly manipulate the fragment shader of the graphics processing unit. Upon receiving an over-limit alarm signal, this submodule immediately activates the pulse function logic. This submodule locks the vertex index corresponding to the over-limit region in video memory and sets the material's emissive color channel to a highly saturated warning color. In this embodiment, the warning color is set to pure red RGB(1.0,0.0,0.0). Simultaneously, the submodule introduces a time variable. According to the sine wave law Dynamically adjust transparency, including blink frequency The refresh rate is set to 2Hz. This causes the out-of-limit areas to appear as a flickering effect on the 3D model, ensuring that users can locate potential risk points immediately.

[0051] The dynamic section generation submodule adaptively generates the section surface based on the stress state at the probe position. This submodule first obtains the stress tensor matrix of the probe point. In this example, the obtained stress tensor matrix is: ; The submodule calls the Jacobian rotation algorithm to solve for the eigenvalues ​​of the real symmetric matrix. Through calculation, one eigenvalue of the matrix is ​​first determined. Regarding the remaining Submatrix blocks, solving the characteristic equation The other two eigenvalues ​​are obtained as follows: and Corresponding The eigenvector is calculated as follows .

[0052] The submodule selects the feature value with the largest absolute value. corresponding feature vector This is taken as the direction of the first principal stress. Subsequently, the coordinates of the probe tip are used. With the origin, For the normal vector, a point-normal plane equation is established. The submodule calculates the dot product of the normal and the view vector; if the dot product is less than zero, the normal direction is reversed to ensure the cutting plane faces the user. Finally, the submodule generates a four-dimensional clipping plane vector, which is passed to the rendering pipeline. During rasterization, all fragments in the negative half of the plane are discarded to generate a dynamic section view. The results show that the system-generated cutting plane is precisely perpendicular to the direction of maximum stress on the component.

[0053] The contour plotting submodule is responsible for overlaying quantitative analysis charts onto dynamically generated cross-sectional surfaces.

[0054] This submodule utilizes texture mapping technology to map a continuous physical scalar field into discrete color textures. To draw clear contour lines, the submodule employs a fragment discarding algorithm. In the fragment shader, the normalized physical value of the current pixel is obtained. Set the contour interval to . In this embodiment, Set to 0.1. Submodule calculation. Divide by The remainder. If the remainder falls within the preset line width range. If the pixel is within the specified range, it is rendered as the contour color; otherwise, it is rendered as the stress cloud map color based on the chromatographic mapping. In this embodiment, The value is set to 0.005. Through this logic, the submodule generates a dynamically updated local contour map on the section surface in real time.

[0055] Table 1. System Key Thresholds and Parameter Settings: ; As shown in Table 1, the key parameters in this system are all set based on clear physical criteria or statistical data, ensuring the stability of system operation and the engineering reliability of the results.

[0056] Table 2 Comparison of Mesh Subdivision Reconstruction Effects: ; As shown in Table 2, by implementing the gradient adaptive subdivision strategy of the present invention, the display error of the high-stress area is significantly reduced when the total number of patches increases by only about 21%, and the rendering frame rate is maintained at 95 FPS. The memory usage is much lower than that of the global subdivision scheme, which verifies that the mesh subdivision reconstruction submodule effectively optimizes the consumption of computing resources while ensuring accuracy.

[0057] The above embodiments illustrate preferred embodiments of the present invention. Any equivalent adjustments to the technical solution based on software engineering methods are within the scope of protection, including but not limited to: implementing algorithm logic using different programming languages, refactoring functional modules into services, adjusting data interaction protocols, and optimizing resource scheduling strategies. Any implementation scheme derived from reasonable modifications to the data processing flow, service call chain, or system architecture layer without departing from the core technology of the present invention should be considered within the protection scope defined by the technical solution of the present invention.

Claims

1. A visualization and interactive system for building structure simulation results, characterized in that, The system includes: The data parsing module is used to acquire the geometric model file and finite element analysis result file of the building project in parallel, parse and extract the spatial coordinate sequence of geometric vertices and the stress tensor and element topology index of the finite element nodes, perform normalization and cleaning operations on the stress tensor of the finite element nodes, and generate a physical field data stream. The mesh mapping module is used to calculate the local coordinate weight of each geometric rendering vertex in the finite element unit by calling the shape function interpolation algorithm through the physical field data stream and the spatial coordinate sequence, project the physical field data onto the rendering mesh vertex, perform local mesh subdivision on the high stress area according to the stress gradient change rate, and construct an enhanced 3D rendering model. The interactive detection module is used to capture the world coordinates of the tip of the virtual probe in the enhanced 3D rendering model in real time, retrieve the bounding cell based on the voxel index logic, perform trilinear interpolation on the node attribute values ​​of the bounding cell to obtain the precise physical value of the current probe position, compare the precise physical value with a preset safety threshold, and generate a judgment signal. The dynamic rendering module is used to render the over-limit area in the enhanced 3D rendering model in a high-brightness flashing manner in response to the judgment signal, perform Boolean operation cutting along the principal stress direction, and generate dynamic section view and local contour map.

2. The building structure simulation result visualization and interactive system according to claim 1, characterized in that, The specific functions of the data parsing module are as follows: The geometric topology alignment submodule is used to read the vertex index list in the geometric model file, load the node number mapping table in the finite element analysis result file, establish the spatial correspondence between geometric vertices and finite element nodes through a hash matching algorithm, remove redundant geometric primitives that have not participated in the calculation, and output the aligned topology index. The stress tensor extraction submodule is used to traverse the aligned finite element nodes, extract six-dimensional stress data including normal stress components and shear stress components from the binary data block of the finite element analysis result file, assemble them into the corresponding second-order tensor matrix, and synchronously associate the displacement vector and temperature scalar of the node to generate the original node attribute set. The data cleaning and normalization submodule is used to receive the original set of node attributes, impute missing values ​​and remove singular values, map the numerical range of multiple physical quantities to a standardized dimensionless interval, and package and serialize the processed node attributes with the topology index to generate a physical field data stream.

3. The building structure simulation result visualization and interactive system according to claim 1, characterized in that, The specific functions of the mesh mapping module are as follows: The local weight calculation submodule is used to parse the unit topology information in the physical field data stream. For each geometric rendering vertex, it uses an inverse mapping algorithm to determine its natural coordinate position inside the corresponding finite element unit, and constructs a shape function matrix describing the degree of influence of multiple nodes of the unit on the vertex, generating local coordinate weights. The attribute projection mapping submodule is used to perform weighted average calculation of the physical properties of the finite element nodes through the local coordinate weights, accurately transfer stress, displacement and temperature data from the sparse finite element mesh to the high-density rendering mesh vertices, and establish a dynamic binding relationship between rendering vertices and physical property values ​​to generate a basic rendering model. The mesh subdivision and reconstruction submodule is used to traverse all triangular faces of the basic rendering model, calculate the stress gradient change rate between adjacent vertices, identify regions where the gradient value exceeds a preset subdivision threshold, perform a center split operation on the triangular faces within the region and update the vertex connection relationship, increase the mesh density in the high stress gradient region, and construct the enhanced 3D rendering model.

4. The interactive visualization system for building structure simulation results according to claim 1, characterized in that, The specific functions of the interactive detection module are as follows: The probe coordinate tracking submodule is used to monitor the spatial displacement data of the user input device in real time through the graphical interface, update the position and attitude matrix of the virtual probe in the 3D scene, extract the 3D vector of the probe tip in the world coordinate system, and convert it into normalized coordinates relative to the model bounding box to generate the tip world coordinates. The unit index retrieval submodule is used to pre-construct the voxel index logic of the enhanced 3D rendering model using the octree space segmentation algorithm, quickly traverse the octree nodes according to the tip world coordinates, locate the smallest leaf node where the probe is currently located, and filter the target tetrahedron or hexahedron that actually includes the probe coordinates from the candidate unit list included in the leaf node to generate the enclosing unit. The attribute interpolation determination submodule is used to obtain the physical attribute values ​​of multiple vertices surrounding the unit and the relative position coefficient of the probe within the unit, perform trilinear interpolation based on volume coordinates to solve the instantaneous physical quantity of the probe point, compare the instantaneous physical quantity with the preset safety threshold of the material strength, and if the difference exceeds the threshold, trigger an alarm state and generate a determination signal.

5. The interactive visualization system for building structure simulation results according to claim 1, characterized in that, The specific functions of the dynamic rendering module are as follows: The abnormal highlight warning submodule is used to receive the judgment signal. If the signal indicates that there is an over-limit area, the pulse function is activated in the fragment shader to modify the material emission color of the rendering mesh vertex corresponding to the over-limit area to a high-saturation warning color, and the transparency channel is dynamically adjusted according to the sine wave law to render the over-limit area in the enhanced 3D rendering model in a highlight flashing manner. The dynamic section generation submodule is used to calculate the direction of the principal stress vector based on the stress tensor eigenvalue at the probe position, construct the equation of the section plane passing through the probe point and with the normal vector parallel to the principal stress direction, perform real-time Boolean subtraction on the model using template buffer technology, remove geometry on one side of the section plane, and generate a dynamic section view. The contour plotting submodule is used to map a continuous physical scalar field into discrete color textures on the cross section of the generated dynamic sectional view, extract pixels of the target numerical range through a fragment discarding algorithm to draw contour lines, and superimpose and project the contour lines onto the sectional surface to generate a local contour plot.

6. The building structure simulation result visualization and interactive system according to claim 2, characterized in that, The data cleaning process of the data cleaning and normalization submodule includes: Obtain all stress tensor components in the original node attribute set, calculate the Mises equivalent yield stress value for each node, and calculate the mean and standard deviation of the stress values ​​across the entire field. The Raida criterion was used to identify and remove outlier data nodes that deviated from the mean by more than three standard deviations, and the weighted average of the nearest neighbor nodes was used to fill in the removal positions. Based on the material ultimate strength parameter corresponding to the preset safety threshold, the repaired stress value is linearly mapped to a floating-point range of zero to one, while retaining the metadata marker of the original value, thus generating a physical field data stream.

7. The building structure simulation result visualization and interactive system according to claim 3, characterized in that, The calculation process of the local weight calculation submodule includes: Obtain the node coordinate matrix of the finite element, including the target rendering vertex, and establish the Jacobian transformation matrix from the global Cartesian coordinate system to the element natural coordinate system; The Newton-Raphson iteration method is used to solve the nonlinear equation system and calculate the parameter coordinates of the target rendering vertex in the natural coordinate system, ensuring that the iteration residual is less than the preset convergence tolerance. Substitute the obtained parameter coordinates into the Lagrange function expression of the element type, calculate the contribution coefficients of multiple finite element nodes to the rendered vertex, establish a weight vector including the weights of all nodes, and generate local coordinate weights.

8. The interactive visualization system for building structure simulation results according to claim 3, characterized in that, The calculation process of the stress gradient change rate, which is based on the mesh subdivision and reconstruction submodule, includes: Obtain the physical field values ​​of the three vertices of the current triangular facet and the geometric feature scale of the facet. Based on the ratio of the difference in physical quantities of neighboring vertices to their spatial distance, calculate the stress gradient change rate of the facet using the following gradient evaluation formula. : ； in, Represents the rate of change of the stress gradient. This represents the surface area value of the current triangular facet. Representative noodle piece The normalized stress values ​​at each vertex, This represents the average stress value at the three vertices of the facet. Representing the The length of the edge opposite each vertex. This represents the geometry factor used to correct the weight of the elongated triangle; The calculated stress gradient change rate is compared with a preset subdivision threshold to filter the set of target patches that need to be split.

9. The interactive visualization system for building structure simulation results according to claim 4, characterized in that, The interpolation calculation process of the attribute interpolation determination submodule includes: Obtain the baseline attribute values ​​of the eight corner points of the enclosing cell in the natural coordinate system, as well as the offset vector of the probe point relative to the cell center; Linear interpolation is performed sequentially along the three principal axes of the natural coordinate system. First, the projection point attributes are calculated on the four edges parallel to the X-axis. Then, the secondary projection point attributes are calculated on the two Y-axis lines formed by the aforementioned projection points. Finally, the attribute value of the probe position is determined on the line segment connected in the Z-axis direction. The scalar and tensor properties obtained by interpolation are stored separately, and the interpolation results are corrected by combining the anisotropy coefficient of the material to eliminate the influence of mesh distortion on the interpolation accuracy and generate accurate physical values.

10. The building structure simulation result visualization and interactive system according to claim 5, characterized in that, The process of constructing the cutting surface of the dynamic section generation submodule includes: Obtain the stress tensor matrix at the current position of the probe, call the Jacobi rotation algorithm to solve for the eigenvalues ​​and eigenvectors of the real symmetric matrix, and determine the eigenvector corresponding to the eigenvalue with the largest absolute value as the first principal stress direction; Using the spatial coordinates of the probe tip as the origin and the direction of the first principal stress as the normal vector, a point normal plane equation is established, and the orientation of the plane's normal is adaptively adjusted according to the viewpoint position to ensure that the cutting surface is facing the user's line of sight. A four-dimensional clipping vector is generated to describe the cutting plane, and it is passed to the rendering pipeline of the graphics processing unit. During the rasterization stage, all fragments in the negative half of the plane are discarded to generate a dynamic cut view.

Images