Digital twin model performance mapping method and system based on point cloud and displacement perception

Abstract

The invention provides a digital twin model performance mapping method and system based on point cloud and displacement perception, and relates to the technical field of structural engineering digitization and health monitoring, and the method comprises the steps: obtaining three-dimensional point cloud data of an in-service device, generating a high-precision solid mesh model through reverse modeling, and importing the high-precision solid mesh model into finite element software to construct a digital twinborn model; synchronously acquiring local deformation images through a consumer-level camera, and extracting and filtering through a DIC algorithm to obtain a displacement field; and taking the displacement field as a load to drive a digital twin model for simulation, further extracting performance indexes such as stress, strain and energy dissipation for visualization, and triggering a risk alarm based on a threshold value. According to the invention, end-to-end closed loop from physical monitoring to digital simulation is realized, the technical problem that the in-service energy consumption device is difficult to carry out high-fidelity geometric reduction, real-time deformation driving and online performance early warning is solved, and an efficient and accurate tool is provided for structural health monitoring and intelligent operation and maintenance.

Description

Technical Field

[0001] This invention relates to the field of digital structural engineering and health monitoring technology, and in particular to a performance mapping method and system for digital twin models based on point cloud and displacement sensing. Background Technology

[0002] Structural energy dissipation devices (such as dampers and buckling-restrained braces) are key components in the seismic resistance and vibration reduction systems of buildings, bridges, and other engineering structures. Their service condition (such as geometric deformation, localized damage, and material performance degradation) directly affects the overall safety and service life of the structure. Accurate and real-time performance evaluation and health monitoring of in-service energy dissipation devices are crucial for ensuring the safe operation and maintenance of critical infrastructure.

[0003] Traditional performance evaluation methods primarily rely on design parameters, periodic manual inspections, and monitoring using contact sensors (such as strain gauges and displacement meters). Manual inspections are highly subjective, have limited coverage, and are difficult to implement continuously. While contact sensors offer high accuracy, they are costly to deploy, complex to maintain, and struggle to fully cover critical areas in complex geometries or harsh environments. In recent years, digital technologies such as 3D laser scanning and photogrammetry have provided efficient means to acquire structural surface geometry information. High-precision solid models can be reconstructed through point cloud reverse modeling, and combined with finite element analysis, digital simulation of component behavior can be achieved. However, current technologies lack an end-to-end solution that can deeply integrate on-site measured geometry with real-time deformation and displacement data, and directly drive high-fidelity digital twin models for online performance mapping and early warning. This makes it difficult to achieve a closed loop from physical monitoring to virtual simulation and then to intelligent decision-making, failing to meet the engineering requirements for real-time, accurate, and visualized characterization and evolution prediction of the service performance of in-service energy-consuming devices. Summary of the Invention

[0004] To address the technical problem of the lack of end-to-end technology in the aforementioned background technology for real-time and precise driving and fusion of on-site measured deformation displacement and high-fidelity digital twin models, this invention introduces the core technical features of high-precision geometric reconstruction based on 3D point cloud reverse modeling, vision-based non-contact displacement sensing, and mapping the sensed displacement to load-driven digital twin simulation. This provides a performance mapping method and system for digital twin models based on point cloud and displacement sensing, enabling real-time, visual characterization and intelligent early warning of the in-service performance of structural energy-consuming devices.

[0005] To achieve the above objectives, the first aspect of the present invention provides a method for performance mapping of digital twin models based on point clouds and displacement sensing, comprising: Acquire three-dimensional point cloud data of in-service structural energy dissipation devices; The three-dimensional point cloud data is reverse-engineered and processed to generate a high-precision solid model suitable for finite element analysis. The solid model is imported into the finite element solver, and a physical digital twin model is constructed by defining geometry, material constitutive properties, contact relationships and boundary conditions. Local deformation images of the energy-consuming device are acquired, the displacement field is extracted, and filtering is performed. The filtered displacement field is input as a load into the physical digital twin model to drive the model to perform simulation; Service performance indicators are extracted from simulation results, visualized, and risk alarms are triggered based on preset thresholds.

[0006] Furthermore, the acquisition of the three-dimensional point cloud data is achieved using three-dimensional laser scanning or reconstruction technology based on multi-view two-dimensional images, and the point cloud is then processed by registration, denoising, and hole filling.

[0007] Furthermore, the reverse modeling and processing includes topology optimization of the point cloud to meet the mesh quality requirements of finite element analysis and to support the identification and preservation of fine features.

[0008] Furthermore, the material constitutive model includes at least one of an elastic model, an elastoplastic model, and a damage evolution model.

[0009] Furthermore, the acquisition of locally deformed images is achieved using a consumer-grade camera, and the displacement field is extracted using a digital image correlation algorithm. The filtering process employs a bandpass filter. The digital image correlation algorithm tracks the pixel coordinate changes of feature points in the image sequence and, in conjunction with camera parameters, calculates the three-dimensional displacement field of the object's surface. The coordinate transformation relationship includes: transforming points in the world coordinate system... By rotation matrix Translation vector Points transformed to the camera coordinate system ,Right now Subsequently, the camera model was used to... Projecting onto the image plane yields pixel coordinates (X, Y), and the projection relationship is as follows: , ,in, For the camera focal length, ( , , )for The coordinate components.

[0010] Furthermore, the service performance indicators include at least one of stress, strain, stiffness, energy dissipation, cumulative plastic deformation, and damage factor; the risk alarm is triggered when the performance indicators exceed the corresponding thresholds, and the alarm event and related data are recorded.

[0011] A second aspect of the present invention provides a performance mapping system for a digital twin model based on point cloud and displacement sensing, comprising: The point cloud data acquisition module is used to acquire three-dimensional point cloud data of in-service structural energy-consuming devices; The reverse modeling module is used to reverse model and process the three-dimensional point cloud data to generate a high-precision solid model suitable for finite element analysis. The physical digital twin model simulation module is used to import the solid model into the finite element solver and construct the physical digital twin model by defining geometry, material constitutive model, contact relationship and boundary conditions; The image acquisition and processing module is used to acquire local deformation images of the energy-consuming device, extract the displacement field, and perform filtering processing. The driving and mapping module is used to input the filtered displacement field as a load into the physical digital twin model and drive the model to perform simulation. The visualization and alarm module is used to extract service performance indicators from simulation results, visualize them, and trigger risk alarms based on preset thresholds.

[0012] A third aspect of the present invention provides an electronic device comprising: Data acquisition and processing device, used to acquire three-dimensional point cloud data of structural energy-consuming device; A displacement sensing device is used to acquire local deformation images of the energy-consuming device; The input device receives the point cloud data and displacement data; A processor for performing the steps of the point cloud and displacement-aware digital twin model performance mapping method as described in the first aspect of the present invention; The memory is used to store point cloud data, displacement data, simulation models, and performance indicators. Output device for visualizing performance mapping results; An alarm device is used to issue an alarm when a performance indicator exceeds a preset threshold.

[0013] A fourth aspect of the present invention provides a computer-readable storage medium having a program stored thereon that, when executed by a processor, implements the steps of the performance mapping method for a digital twin model based on point cloud and displacement sensing as described in the first aspect of the present invention.

[0014] A fifth aspect of the present invention provides a computer program product comprising software code, wherein the program in the software code performs the steps of the performance mapping method for a digital twin model based on point cloud and displacement sensing as described in the first aspect of the present invention.

[0015] Compared with existing technologies, the performance mapping method and system for digital twin models based on point cloud and displacement sensing provided by this invention has the following beneficial effects: (1) This invention obtains real point clouds through on-site 3D scanning, and automatically generates a high-precision solid mesh model that meets the requirements of finite element analysis after processing such as registration, denoising, hole filling and topology optimization. This process realizes high-fidelity geometric restoration from the physical world to the digital space, ensures the accuracy of the geometric foundation of the digital twin model, and solves the problem that traditional simulation models rely on design drawings and cannot reflect the actual geometric state of the in-service structure, laying the foundation for subsequent accurate simulation.

[0016] (2) This invention uses a consumer-grade camera for non-contact image acquisition and utilizes the digital image correlation (DIC) algorithm to extract local high-precision displacement fields. This method enables low-cost and easy-to-implement continuous monitoring of deformation in key areas, overcoming the shortcomings of traditional sensors in terms of spatial coverage and long-term deployment.

[0017] (3) This invention uses the filtered real displacement field as a displacement-type load, accurately maps it, and inputs it into the constructed physical digital twin model. A direct bridge between "physical perception and digital driving" is established, enabling the digital twin model to reflect the real mechanical response of the structure in real time, and realizing data-driven dynamic simulation.

[0018] (4) This invention automatically extracts key performance indicators such as stress, strain, and energy dissipation from the driving simulation results, displays them in a visual cloud map, and automatically triggers risk alarms based on preset thresholds. This function realizes the intuitiveness of service performance and the automation of risk warning, providing maintenance personnel with a clear decision support interface and timely handling prompts. Attached Figure Description

[0019] The accompanying drawings, which form part of this disclosure, are used to provide a further understanding of this disclosure. The illustrative embodiments of this disclosure and their descriptions are used to explain this disclosure and do not constitute an undue limitation of this disclosure.

[0020] Figure 1 This is a flowchart illustrating the performance mapping method for a digital twin model based on point cloud and displacement sensing provided in Embodiment 1 of the present invention.

[0021] Figure 2 This is a schematic diagram of the overall process provided in Embodiment 1 of the present invention.

[0022] Figure 3 This is a schematic diagram of the module framework provided in Embodiment 1 of the present invention.

[0023] Figure 4 This is a framework diagram of the digital twin model performance mapping system based on point cloud and displacement sensing provided in Embodiment 2 of the present invention.

[0024] Figure 5 This is a schematic diagram of the hardware structure of the electronic device provided in Embodiment 3 of the present invention. Detailed Implementation

[0025] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0026] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form as well. Furthermore, it should be understood that the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0027] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0028] All data acquisition in this embodiment is carried out in accordance with laws and regulations and with user consent, and the data is used legally.

[0029] Example 1 like Figure 1 This embodiment provides a performance mapping method for digital twin models based on point clouds and displacement sensing, including: Acquire three-dimensional point cloud data of in-service structural energy dissipation devices; The three-dimensional point cloud data is reverse-engineered and processed to generate a high-precision solid model suitable for finite element analysis. The solid model is imported into the finite element solver, and a physical digital twin model is constructed by defining geometry, material constitutive properties, contact relationships and boundary conditions. Local deformation images of the energy-consuming device are acquired, the displacement field is extracted, and filtering is performed. The filtered displacement field is input as a load into the physical digital twin model to drive the model to perform simulation; Service performance indicators are extracted from simulation results, visualized, and risk alarms are triggered based on preset thresholds.

[0030] This invention aims to overcome a long-standing technical obstacle in this field: how to deeply integrate high-fidelity geometric models obtained through point cloud reverse modeling with real-time deformation data obtained through visual displacement sensing technology and directly drive digital twin simulation. While existing technologies can achieve geometric reconstruction and displacement measurement separately, they face the following key integration difficulties: First, the problem of data heterogeneity: point clouds are static three-dimensional discrete point sets, while displacement fields are time-series dynamic vectors. The two differ fundamentally in data structure, coordinate system, and sampling density, lacking a unified mapping interface. Second, the lack of a driving mechanism: even with both geometric models and displacement data, there is a lack of standardized methods and driving logic for directly and accurately inputting measured displacement as a load into the finite element model, leading to a disconnect between "sensing" and "simulation." Third, the contradiction between real-time performance and fidelity: high-fidelity finite element simulation involves large computational loads, making it difficult to meet the real-time monitoring needs of engineering projects, while simplified models cannot accurately reflect the true mechanical behavior of the structure.

[0031] To address the aforementioned challenges, this invention first designs an automated reverse modeling process from point clouds to solveable finite element meshes, ensuring that the geometric model is both highly faithful and meets the mesh quality requirements for simulation calculations. Secondly, it proposes a displacement-mesh precise mapping method based on coordinate system one and spatial interpolation algorithms, establishing a bridge for the transformation from visual displacement fields to finite element nodal loads. Then, it constructs an efficient displacement-driven solution strategy and fault-tolerant mechanism, ensuring simulation accuracy while meeting the real-time requirements of engineering. Finally, it forms a complete closed loop from physical perception to digital driving and then to intelligent early warning, achieving a synergistic technical effect of "1+1>2".

[0032] Specifically, the acquisition of 3D point cloud data is achieved using 3D laser scanning or reconstruction techniques based on multi-view 2D images, followed by registration, denoising, and hole filling processing. A massive number of 3D coordinate points on the object's surface are obtained by actively emitting lasers or calculating from multi-view images. Registration unifies multi-station or multi-view data into the same coordinate system; denoising removes outliers caused by measurement errors; and hole filling repairs data gaps caused by occlusion. This step solves the problem of traditional modeling relying on idealized design drawings that do not match the actual object, providing a precise geometric foundation for constructing high-fidelity digital twin models.

[0033] Specifically, the reverse modeling and processing includes topology optimization of the point cloud to meet the mesh quality requirements of finite element analysis and to support the identification and preservation of fine features. Its core lies in generating a continuous surface from the discrete point cloud and meshing it using algorithms (such as Poisson reconstruction), while automatically repairing mesh defects (such as holes and self-intersections) and identifying and preserving fine geometric features (such as welds and micro-dimples) that have a critical impact on the mechanical response. This step solves the problem of poor mesh quality generated directly from coarse scan data, making it unsuitable for high-precision finite element calculations.

[0034] Specifically, the material constitutive model includes at least one of an elastic model, an elastoplastic model, and a damage evolution model. Users can select or customize the corresponding constitutive model and parameters in the finite element software based on the actual material properties of the energy dissipation device (such as the elastoplasticity of steel and the damage of concrete). This feature enables the digital twin model to more realistically simulate the nonlinear behavior and performance degradation process of materials under load.

[0035] Specifically, the acquisition of locally deformed images is achieved using a consumer-grade camera, and the displacement field is extracted using a digital image correlation (DIC) algorithm. The filtering process employs a bandpass filter. The DIC algorithm calculates the three-dimensional displacement field of the object surface by tracking the pixel coordinate changes of feature points in the image sequence and combining camera parameters. The coordinate transformation relationship includes: transforming points in the world coordinate system... By rotation matrix Translation vector Points transformed to the camera coordinate system ,Right now Subsequently, the camera model was used to... Projecting onto the image plane yields pixel coordinates (X, Y), and the projection relationship is as follows: , ,in, For the camera focal length, ( , , )for The coordinate components. The DIC algorithm calculates the full-field displacement data by tracking the movement of natural or artificial speckle patterns on the object's surface in an image sequence. Consumer-grade cameras significantly reduce the cost of monitoring hardware. Bandpass filtering effectively removes high-frequency noise caused by sudden changes in ambient light and slight camera vibrations, as well as low-frequency interference caused by temperature drift. This step enables non-contact, low-cost, and high-precision continuous measurement of deformation in key areas, overcoming the problems of difficult deployment and limited coverage of traditional sensors.

[0036] Specifically, the service performance indicators include at least one of stress, strain, stiffness, energy dissipation, cumulative plastic deformation, and damage factor. The risk alarm is triggered when the performance indicator exceeds a corresponding threshold, and the alarm event and related data are recorded. Under displacement load, the finite element solver can calculate the detailed mechanical response within the model. Indicators that intuitively reflect the health status of the device (such as cumulative plastic deformation characterizing the degree of damage) can be extracted from these results. The system continuously compares the calculated indicator values ​​with preset safety thresholds (such as yield stress and allowable plastic strain). Once a limit is exceeded, an alarm is automatically triggered and a snapshot is recorded. This step solves the problems of reliance on professional personnel for interpreting simulation results and untimely risk warnings, achieving automation and intelligence in performance evaluation and safety warning.

[0037] In one specific embodiment, see Figure 2 , Figure 3 ,in, Figure 2 (Flowchart) This diagram illustrates the four main steps of this embodiment: point cloud acquisition (p1), reverse modeling (p2), construction of a physical digital twin (p3), displacement sensing-driven operation and service mapping (p4). The arrows in the diagram indicate the closed-loop flow of data from field acquisition to model-driven operation. Figure 3 (Module Framework) This module demonstrates the functional chain of point cloud data acquisition, reverse modeling, physical digital twin model construction and simulation, and driving and mapping. All modules are interconnected via standard data format interfaces (PLY / OBJ / INP / CSV).

[0038] This embodiment is implemented according to the following steps: Step S1: Point cloud acquisition and recording (corresponding to...) Figure 2 p1, Figure 3 (point cloud data acquisition module) Equipment: Laser scanners (e.g., Faro / Leica), structured light sensors (e.g., Kinect Azure), or photogrammetry-based reconstruction systems using multi-view 2D images. A combination of these can be used depending on site conditions.

[0039] Data Acquisition Process: Coverage sampling is performed on the target energy-consuming device to ensure sufficient point density for key details (such as thin ribs, holes, and slots). Calibration parameters (external parameters) during acquisition are recorded synchronously and saved in standard point cloud format (PLY / PCD / PTS).

[0040] Example of acquisition accuracy and parameters: Laser scanning accuracy can be selected from 0.5–2 mm, photogrammetry pixel resolution is controlled at 0.1–1 mm / pixel equivalent scale, and an overlap rate of ≥60% is recommended to ensure reconstruction quality.

[0041] Step S2: Point cloud preprocessing and registration ( Figure 3 (The initial stage of the reverse modeling module) Rigid body registration: Feature-based (SIFT / ORB) pairing combined with RANSAC is used to eliminate mismatches, or fast registration is performed based on manually set calibration boards. Subsequently, global ICP (Iterative Closest Point) optimization is performed to reduce registration residuals. ICP termination conditions include RMSE < 0.5 mm or number of iterations > 50.

[0042] Outlier removal: Statistical outlier filtering is used (based on the average distance of the neighborhood, the threshold is usually the mean ± kσ, where k is 2–3). For areas with excessively high point density, voxel filtering is used for partitioned sampling. The sample voxel size is set according to the component size (0.5 mm–1 mm for small components, 5–10 mm for large components).

[0043] Step S3: Surface Reconstruction and Topology Repair Normal estimation and Poisson reconstruction: Estimate the point normals of the preprocessed point cloud (with the neighborhood radius set to 2–5 times the point cloud density), and use Poisson surface reconstruction to generate closed patches. The reconstruction parameters are such as depth (typically 8–12) to balance detail and computational cost.

[0044] Multi-resolution processing: Use higher depth or local reconstruction on key small features to preserve details.

[0045] Hole and self-intersection repair: Automatically detects opening boundaries and performs hole filling, using Boolean operations to preserve key geometric features such as thin walls and slots; performs trimming on self-intersecting or overlapping patches.

[0046] Surface subdivision: Perform necessary subdivision and smoothing of the surface mesh to ensure the boundary continuity of the subsequent volume mesh generation.

[0047] Step S4: Meshing and Finite Element Derivation (Constructing a Solvable Mesh) Surface to volume mesh: Generate volume mesh (tetrahedron or hexahedron) based on surface patches. Tetrahedrons are suitable for automated processes, while hexahedrons can be manually divided in critical areas to improve accuracy.

[0048] Mesh quality inspection: Check cell shape quality (minimum interior angle, aspect ratio, Jacobian). Example threshold settings: minimum angle > 15° (preferably 20°), aspect ratio < 10, Jacobian > 0.3. Cells that do not meet the requirements are locally refined or re-meshed.

[0049] Export format: Generates the format required by Abaqus to read the file or the equivalent solver.

[0050] Step S5: Numerical Model Establishment ( Figure 3 (Physical digital twin simulation module) Model import: Import the mesh into Abaqus, perform geometry compatibility checks, and divide the material domains and contact surfaces.

[0051] Constitutive model selection: Select a suitable constitutive model based on the structure of the energy-consuming device, such as elastic or plastic damage evolution models, or UMAT or material subroutines.

[0052] Contact and Boundary Settings: Define tangential frictional contact and normal hard contact, constraints (hinged or fixed), and set the boundary surfaces or node sets that receive displacement loads.

[0053] Step S6: Camera Deployment and Displacement Sensing (corresponding to...) Figure 2 p4 and Figure 3 Image acquisition module) Camera Selection and Deployment: Select an industrial camera, structured light camera, or RGB-D camera based on the local monitoring area. The camera sampling frequency should be set according to the dynamic characteristics being measured; for example, quasi-static conditions can be sampled once a day, while vibration events require 100–1000 Hz. The camera position should ensure that the visualization area covers the target deformation area and that redundant viewing angles are reserved.

[0054] 2D to 3D mapping and displacement extraction: If a depth camera is used, depth information is directly obtained using RGB-D, and pixel displacement is mapped to the target's 3D displacement in the physical coordinate system (using camera intrinsic and extrinsic parameters and depth data). This can be expressed by the formula: ;in, For real coordinates in the three-dimensional world, The coordinates are the origin of the camera's optical center.

[0055] The core coordinate transformation relationship can be expressed as follows: First, transform the points in the world coordinate system... By rotation matrix Translation vector Points transformed to the camera coordinate system ,Right now Subsequently, the camera model was used to... Projecting onto the image plane yields pixel coordinates (X, Y), and the projection relationship is as follows: , ,in, For the camera focal length, ( , , )for The coordinate components are given by (X,Y), which is the coordinate system of the image plane. The Digital Image Correlation (DIC) algorithm calculates the three-dimensional displacement field of the object surface by tracking the pixel coordinate changes of feature points in a sequence of images and combining the above transformation relationship.

[0056] Filtering and uncertainty assessment: Bandpass filtering is applied to the original displacement data; the uncertainty of the displacement values ​​is estimated (by calculating the sample covariance) and a confidence interval is attached.

[0057] Step S7: Displacement Mapping and Numerical Driving Mapping method: The processed displacement field is mapped to the boundary surface or set of nodes of the finite element model that receive displacement loads.

[0058] Solution strategy: Invoke Abaqus's static or dynamic nonlinear solver (Implicit / Explicit depending on the working condition); to ensure convergence, the Newton-Raphson method and arc length method can be enabled. Example convergence criteria: residual 2-norm less than 1e-6 or displacement increment ratio less than 1e-5. Parallel solution and hardware acceleration are supported (GPU acceleration if the solver allows).

[0059] Step S8: Indicator Extraction, Visualization, and Alerts ( Figure 3 Visualization and Alarm Module Index calculation: Extract Von Mises stress, principal strain, local stiffness (by fitting the loading / unloading curve), energy dissipation (energy absorption curve of the energy dissipator), and cumulative plastic deformation from the solution results.

[0060] Visualization: Displays performance cloud maps and profile diagrams on a 3D interactive interface, and supports animation playback and export (PDF / Excel / CSV).

[0061] Alarm strategy: Alarms are triggered based on preset thresholds (e.g., local Von Mises stress exceeds 1.1 times the material yield strength, local stiffness attenuation > 20%, cumulative plastic deformation exceeds a predetermined limit, etc.); the alarm content includes time, trigger index, relevant original observations (images and point cloud fragments) and snapshots of solution results, and supports logging and remote notification (email, SMS, mobile push).

[0062] Through the synergistic effect of the above steps, this invention achieves a deep integration of point cloud geometric reconstruction technology and visual displacement sensing technology, achieving a "1+1>2" technical effect that surpasses the simple superposition of single technologies: First, it realizes synchronous high-fidelity mapping between "geometric state" and "mechanical state." The digital twin model can not only reflect the true geometric shape of the structure, but also reveal its internal stress, strain, damage evolution, and other mechanical states in real time and accurately, solving the problem of the disconnect between geometric and mechanical analysis in traditional methods. Second, it expands the dimensions and depth of monitoring, combining the full-field coverage advantage of point clouds with the local high-precision advantage of DIC, realizing full-scale performance evaluation from macroscopic geometry to microscopic mechanical response. Third, it constructs an intelligent closed loop of "perception-drive-early warning," transforming physical world monitoring data into loads driving digital simulation, and automatically converting simulation results into operational and maintenance decision-making basis, forming a complete link from data acquisition to intelligent handling, significantly improving the automation, intelligence level, and engineering practical value of structural health monitoring.

[0063] Periodic monitoring and online updates Deployment mode: The edge (on-site) is responsible for image acquisition, preliminary DIC displacement extraction and bandpass filtering; the local server is responsible for point cloud reverse modeling, meshing, numerical simulation, visualization and database management.

[0064] Online updates: Acquire images and drive simulations during periodic events or when triggered by events (such as strong earthquakes or tests).

[0065] Multi-source fusion: Data from traditional sensors such as accelerometers and strain gauges are used as redundant inputs and fused with visual displacement data. Modal recognition is used to assist in verifying displacement measurement results.

[0066] Automation Strategies for Grid and Quality Control Automated inspection process: The reverse modeling module automatically calculates the mesh distortion rate, minimum or average interior angle, and element aspect ratio for the generated mesh and generates a quality report. If distortion is found or the results are substandard, manual intervention is prompted.

[0067] Example parameters: When the minimum angle of the mesh is less than 12°, the module returns a re-meshing request and records version control.

[0068] Displacement mapping details and fault tolerance Mapping interpolation fault tolerance: To prevent missing observations or instantaneous outliers from affecting the simulation, a threshold check strategy is introduced: If the proportion of missing observations is >30% at a certain moment, the previous prediction interpolation is used for this mapping; if the residual between the observation and the model prediction exceeds three times the standard deviation, the driving is paused and the abnormal event is recorded.

[0069] Spatial resolution matching: The visual displacement field is usually a sparse or local set of points. When interpolating to the finite element mesh, distance weighted (IDW) is preferred.

[0070] Alarm thresholds and handling procedures examples Threshold setting examples: Yield threshold alarm: Local Von Mises stress ≥ σyield × 1.0 triggers "Attention", ≥ σyield × 1.2 triggers "Warning"; Stiffness degradation threshold: A decrease in stiffness ≥ 10% relative to the initial stiffness triggers "Concern", ≥ 20% triggers "Handling". Thresholds can be determined by design values, experimental data, or historical evolution statistics and can be adjusted by engineers.

[0071] Handling Procedure: When the "Warning" level is triggered, the system automatically generates a report and sends a notification to the maintenance personnel, while also suggesting on-site inspection or loading of further high-resolution point cloud or image acquisition for confirmation.

[0072] Optional implementation variations: If it is not convenient to cover the point cloud completely on site, a historical CAD model can be used as a priori. The CAD model can be used to identify deviations in the point cloud and only the deviation areas can be locally reconstructed and meshed to speed up the engineering process.

[0073] Example 2 like Figure 4This embodiment provides a performance mapping system for digital twin models based on point clouds and displacement sensing, including: The point cloud data acquisition module is used to acquire three-dimensional point cloud data of in-service structural energy-consuming devices; The reverse modeling module is used to reverse model and process the three-dimensional point cloud data to generate a high-precision solid model suitable for finite element analysis. The physical digital twin model simulation module is used to import the solid model into the finite element solver and construct the physical digital twin model by defining geometry, material constitutive model, contact relationship and boundary conditions; The image acquisition and processing module is used to acquire local deformation images of the energy-consuming device, extract the displacement field, and perform filtering processing. The driving and mapping module is used to input the filtered displacement field as a load into the physical digital twin model and drive the model to perform simulation. The visualization and alarm module is used to extract service performance indicators from simulation results, visualize them, and trigger risk alarms based on preset thresholds.

[0074] Example 3 Embodiment 3 of the present invention provides an electronic device, comprising: Data acquisition and processing device, used to acquire three-dimensional point cloud data of structural energy-consuming device; A displacement sensing device is used to acquire local deformation images of the energy-consuming device; The input device receives the point cloud data and displacement data; A processor for performing the steps of the point cloud and displacement-aware digital twin model performance mapping method as described in the first aspect of the present invention; The memory is used to store point cloud data, displacement data, simulation models, and performance indicators. Output device for visualizing performance mapping results; An alarm device is used to issue an alarm when a performance indicator exceeds a preset threshold.

[0075] In one specific embodiment, such as Figure 5 The hardware structure shown illustrates the relationship between the input device (3D point cloud and displacement sensing device), processor, memory, output device, and alarm device: the data acquisition and processing device sends data to the input device, the processor completes the data acquisition and processing and stores it in the memory, the displacement sensing device sends data to the input device, and the output device compares the mapping result with the threshold. If the threshold is exceeded, the alarm device is activated.

[0076] The detailed steps are the same as those of the digital twin model performance mapping method based on point cloud and displacement sensing provided in Example 1, and will not be repeated here.

[0077] Example 4 Embodiment 4 of the present invention provides a computer-readable storage medium.

[0078] A computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps in the digital twin model performance mapping method based on point cloud and displacement sensing as described in Embodiment 1 of the present invention.

[0079] The detailed steps are the same as those of the digital twin model performance mapping method based on point cloud and displacement sensing provided in Example 1, and will not be repeated here.

[0080] Example 5 Embodiment 5 of the present invention provides a computer program product.

[0081] A computer program product includes software code, wherein the program in the software code performs the steps of the performance mapping method for digital twin models based on point cloud and displacement sensing as described in Embodiment 1 of the present invention.

[0082] The detailed steps are the same as those of the digital twin model performance mapping method based on point cloud and displacement sensing provided in Example 1, and will not be repeated here.

[0083] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0084] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0085] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0086] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0087] The above description is merely a preferred embodiment of this practice and is not intended to limit the scope of this practice. Various modifications and variations can be made to this practice by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this practice should be included within the protection scope of this practice.

Claims

1. A point cloud and displacement perception based digital twin model performance mapping method, characterized in that, The method comprises the following steps: acquiring three-dimensional point cloud data of an energy dissipation device of a structure in service; performing reverse modeling and processing on the three-dimensional point cloud data to generate a high-precision solid model suitable for finite element analysis; importing the solid model into a finite element solver, defining geometry, material constitutive, contact relationship and boundary conditions, and constructing a physical digital twin model; collecting local deformation images of the energy dissipation device, extracting displacement fields and performing filtering processing; inputting the filtered displacement field into the physical digital twin model as a load to drive the model to perform simulation; extracting service performance indicators from the simulation results, visualizing and displaying them, and triggering a risk warning based on a preset threshold.

2. The method of claim 1, wherein, The three-dimensional point cloud data is acquired by three-dimensional laser scanning or reconstruction technology based on multi-view two-dimensional images, and the point cloud is registered, denoised and hole-filled.

3. The method of claim 1, wherein, The reverse modeling and processing include topology optimization of the point cloud to meet the grid quality requirements of finite element analysis, and support identification and retention of small features.

4. The method of claim 1, wherein, The material constitutive includes at least one of an elastic model, an elastoplastic model and a damage evolution model.

5. The method of claim 1, wherein, The local deformation image is collected by using a consumer camera, and a displacement field is extracted by a digital image correlation algorithm, and the filtering processing adopts a band-pass filtering mode; the digital image correlation algorithm solves a three-dimensional displacement field of an object surface by tracking pixel coordinate changes of feature points in sequence images and combining camera parameters, and a coordinate transformation relationship includes: transforming a point in a world coordinate system to a point in a camera coordinate system by a rotation matrix and a translation vector , that is ; subsequently, the point is projected to an image plane by a camera model to obtain pixel coordinates (X, Y), and a projection relationship is , wherein, f is a camera focal length, and (x, y, z) are coordinate components of the point . , , .​ 6. The method of claim 1, wherein, The service performance indicators include at least one of stress, strain, stiffness, energy dissipation, cumulative plastic deformation and damage factor; the risk warning is triggered when the performance indicators exceed the corresponding threshold, and the warning events and related data are recorded.

7. A point cloud and displacement aware based digital twin model performance mapping system, comprising: The method comprises the following steps: a point cloud data acquisition module for acquiring three-dimensional point cloud data of an energy dissipation device of a structure in service; a reverse modeling module for performing reverse modeling and processing on the three-dimensional point cloud data to generate a high-precision solid model suitable for finite element analysis; a physical digital twin model simulation module for importing the solid model into a finite element solver, defining geometry, material constitutive, contact relationship and boundary conditions, and constructing a physical digital twin model; an image acquisition and processing module for collecting local deformation images of the energy dissipation device, extracting displacement fields and performing filtering processing; a driving and mapping module for inputting the filtered displacement field into the physical digital twin model as a load to drive the model to perform simulation; a visualization and warning module for extracting service performance indicators from the simulation results, visualizing and displaying them, and triggering a risk warning based on a preset threshold.

8. An electronic device, comprising: The method comprises the following steps: a data acquisition and processing device for acquiring three-dimensional point cloud data of an energy dissipation device of a structure; a displacement sensing device for collecting local deformation images of the energy dissipation device; an input device for receiving the point cloud data and displacement data; a processor for executing the steps of the point cloud and displacement sensing-based digital twin model performance mapping method according to any one of claims 1 to 6; a memory for storing point cloud data, displacement data, simulation models and performance indicators; an output device for visualizing and displaying performance mapping results; a warning device for issuing a warning when the performance indicators exceed the preset threshold.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the steps of the point cloud and displacement sensing-based digital twin model performance mapping method according to any one of claims 1 to 6.

10. A computer program product comprising software code, characterized in that, The programs in the software code perform the steps of the point cloud and displacement aware digital twin model performance mapping method as claimed in any one of claims 1 to 6.

Images