Digital twin method integrating multi-modal data

Abstract

The present application relates to the technical field of data integration, and specifically relates to a digital twin method for integrating multi-modal data, including entity dynamic deformation field monitoring: real-time capture of three-dimensional deformation data of a physical entity to generate a dynamic deformation field; non-rigid reference frame construction; solving the motion vector of each point on the surface of the entity to generate a non-rigid reference frame grid that is updated in real time with deformation; twin reference frame binding: dynamically binding the virtual coordinate system of the digital twin with the non-rigid reference frame grid; multi-modal data remapping: spatial transformation of multi-modal sensing data collected by the physical entity to output a remapping data stream; entity deformation suppression closed loop: detecting abnormal deformation areas based on the remapping data stream to generate actuator control instructions to suppress entity deformation and collect post-suppression data. The present application not only improves the accuracy and response speed of abnormal identification, but also has predictive protection capability, improving the stability and safety margin of system operation.

Description

Technical Field

[0001] This invention relates to the field of data integration technology, and in particular to a digital twin method for integrating multimodal data. Background Technology

[0002] During the operation of complex structural equipment (such as wind turbine blades, aircraft wing surfaces, and large bridge structures), the entities often exhibit nonlinear, multi-scale, and multi-source coupled deformation behavior, and their spatiotemporal evolution is highly dynamic and locally sudden. Traditional digital twin modeling methods are mostly based on static coordinate mapping or rigid approximations, which make it difficult to accurately track the real state of the entity during non-rigid deformation processes. In addition, multimodal sensing data (such as fiber optic strain, thermal imaging, vibration, stress, etc.) are often distributed in different acquisition channels and spatiotemporal coordinate systems, resulting in inaccurate data fusion and insufficient collaborative interpretation, thus restricting the real-time accuracy and controllability of the twin.

[0003] Meanwhile, when early signs of abnormal physical operation appear, such as sudden increases in local strain or abrupt changes in curvature, traditional methods typically rely on post-processing analysis, failing to achieve real-time identification and proactive intervention, resulting in delays in risk response. Existing control strategies are mostly open-loop operations, lacking high-precision suppression capabilities based on twin feedback closed loops, making it difficult to guarantee system stability, especially when dealing with rapid dynamic disturbances and local deformations. Therefore, there is an urgent need to construct a digital twin method that integrates non-rigid reference frame binding, cross-modal data remapping, and actuator closed-loop control to achieve a fully intelligent closed loop from "precise modeling" to "active suppression." Summary of the Invention

[0004] This invention provides a digital twin method for integrating multimodal data.

[0005] A digital twin approach integrating multimodal data includes the following steps:

[0006] S1, Entity Dynamic Deformation Field Monitoring: Real-time capture of three-dimensional deformation data of physical entities through distributed fiber optic sensors and laser holographic scanning, generating dynamic deformation fields;

[0007] S2, Construction of non-rigid reference system: Based on the dynamic deformation field, the motion vector of each point on the solid surface is calculated to generate a non-rigid reference system mesh that is updated in real time with deformation;

[0008] S3, Twin Reference System Binding: Dynamically bind the virtual coordinate system of the digital twin to the non-rigid reference system mesh, and output the twin with the bound reference system;

[0009] S4, Multimodal Data Remapping: The multimodal sensing data collected by the physical entity is spatially transformed according to the reference grid coordinates at the binding time, and the remapped data stream is output.

[0010] S5, Entity deformation suppression closed loop: Based on the remapped data stream, detect abnormal deformation regions, generate actuator control commands to suppress entity deformation, and collect suppressed data to return to S1.

[0011] Optionally, S1 includes:

[0012] S11, Fiber optic strain data calculation: Wavelength offset is collected by distributed fiber optic sensors, and micro-strain distribution is calculated based on the fiber Bragg grating strain sensing model.

[0013] S12, Holographic Deformation Field Reconstruction: The phase difference of the interference fringes is obtained by laser holographic scanning, and the normal deformation is reconstructed based on the holographic interferometric model;

[0014] S13, Spatiotemporal data fusion: Input fiber microstrain data and holographic normal deformation into the spatiotemporal fusion model to generate a three-dimensional dynamic deformation field tensor.

[0015] Optionally, S13 includes:

[0016] S131, Mesh Discretization: Defines the spatial grid coordinates for discretization;

[0017] S132, the second-order partial derivative discretization in the x-direction is approximately calculated using a three-point central difference scheme;

[0018] S133, the second-order partial derivative discretization in the y-direction is calculated by central difference discretization;

[0019] S134, add the difference formulas in the x and y directions to obtain the total curvature operator;

[0020] S135, Boundary condition handling: Boundary points are handled using second-order forward / backward difference formulas.

[0021] Optionally, S2 includes:

[0022] S21, Construction of motion vector field: Based on the normal displacement component and strain component in the dynamic deformation field, calculate the three-dimensional motion vector at the grid point;

[0023] S22, Thermal Deformation Compensation: Considering the thermal expansion effect of the material, thermal field correction is applied to the motion vector;

[0024] S23, Reference Frame Mesh Generation: The discrete motion vector field is mapped to a continuous non-rigid reference frame mesh using the thin plate spline interpolation method.

[0025] Optionally, S21 includes:

[0026] S211, Displacement gradient and curvature extraction: Based on the normal deformation obtained in the dynamic deformation field, the displacement gradient vector of the grid points is calculated using the central difference method;

[0027] S212, Integrated Motion Vector Expression: Combining displacement gradient, curvature and micro-strain values ​​to form a three-dimensional motion vector at the grid point.

[0028] Optionally, S3 includes:

[0029] S31, Control point coordinate mapping: Construct a one-to-one mapping relationship between the virtual control point set and the physical control point set;

[0030] S32, Affine Transformation Solution: Solve for the optimal rigid transformation by minimizing the squared difference between the real and imaginary coordinates;

[0031] S33, Nonlinear Residual Compensation: Calculate the affine transformation residuals and construct a thin-plate spline interpolation function based on the residual points;

[0032] S34, Binding Function Composition: Based on the constructed thin-plate spline interpolation function, a dynamic binding transformation function is constructed;

[0033] S35, Full Model Binding: Apply the transformation function to all twin model vertices.

[0034] Optionally, S33 includes:

[0035] S331, Affine Residual Calculation: After the affine transformation is completed, calculate the corresponding transformation residual vector for each pair of control points.

[0036] S332, Residual Interpolation Function Construction: Based on the spatial distribution of all residual points, and using the distributed sensor index positions as interpolation control points, a thin plate spline interpolation function is constructed.

[0037] Optionally, S4 includes:

[0038] S41, Sensor Parameter Coordinate Extraction: Extract the parameter coordinates of all sensors on the physical entity to form a parameter coordinate set;

[0039] S42, Real-time spatial position calculation: Calculate the real-time three-dimensional spatial position of the sensor based on the current non-rigid reference frame mesh function;

[0040] S43, Binding Position Calculation: Call the grid snapshot function recorded at the binding time to calculate the binding reference position of each sensor;

[0041] S44, Spatial displacement vector generation: Calculate the spatial displacement of each sensor;

[0042] S45, Data remapping execution: Spatial translation of the original sensor data to eliminate the effects of physical deformation;

[0043] S46, Data Flow Reconstruction: Aggregate all remapped data and output an aligned remapped data flow.

[0044] Optionally, S5 includes:

[0045] S51, extract the deformation-related feature set from the remapped data stream;

[0046] S52, Abnormal Region Identification: Identify abnormal regions based on a multi-level threshold strategy;

[0047] S53, Calculation of the suppressive force field: Calculate the required reverse suppressive force based on the characteristic values ​​of the abnormal region;

[0048] S54, Actuator instruction generation: Generate a set of control instructions based on the calculated inhibition force;

[0049] S55, Closed-loop execution and feedback: After the control command is issued to the entity, the closed-loop feedback process is executed.

[0050] Optionally, S54 includes:

[0051] S541, Hydraulic compensation command calculation: Generate the drive area control command of the hydraulic system based on the suppression force field and the normal vector of the abnormal region;

[0052] S542, Piezoelectric micro-vibration command calculation: Utilizes a piezoelectric actuator to generate a vibration control signal that matches the resonant frequency of the structure;

[0053] S543, Active Cooling Control Command Calculation: Based on the integral of deformation and safety threshold, generate corresponding thermal control cooling power commands to reduce the source of thermal strain.

[0054] The beneficial effects of this invention are:

[0055] This invention significantly improves the adaptability of digital twins to the deformation process of physical entities by introducing a non-rigid reference system binding mechanism. Compared with traditional rigid or linear mapping methods, this invention constructs a non-rigid mesh that updates in real time with deformation based on a dynamic deformation field, and binds the virtual twin coordinate system to this mesh through affine transformation and residual compensation, thereby achieving high-precision synchronous mapping. This mechanism effectively solves the problem of twin model deviation under conditions such as large deformation and local distortion of the physical entity, ensuring that the twin is always in a state of accurate correspondence with the real physical scene.

[0056] This invention introduces a closed-loop variable suppression strategy driven by multi-source actuators into the digital twin framework. By extracting deformation features such as strain, curvature, and strain rate from the remapped data stream, a multi-dimensional anomaly detection index system is constructed. Furthermore, a spatially distributed suppression force field is generated for the anomaly region, which is further transformed into multi-mode execution commands such as hydraulic compensation, piezoelectric vibration, and active cooling, forming a highly responsive closed-loop control path. This mechanism not only improves the accuracy and response speed of anomaly identification but also has predictive protection capabilities, significantly improving the stability and safety margin of system operation. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a flowchart of a method according to an embodiment of the present invention;

[0059] Figure 2 This is a deformation suppression diagram according to an embodiment of the present invention. Detailed Implementation

[0060] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0061] It should be noted that the use of terms such as "an embodiment," "an embodiment," "an exemplary embodiment," and "some embodiments" in the specification indicates that the described embodiment may include a specific feature, structure, or characteristic, but not every embodiment necessarily includes that specific feature, structure, or characteristic. Furthermore, when a specific feature, structure, or characteristic is described in connection with an embodiment, implementing such a feature, structure, or characteristic in conjunction with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the art.

[0062] Generally, terms can be understood at least partly from their use in context. For example, depending at least partly on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in a singular sense, or a combination of features, structures, or characteristics in a plural sense. Additionally, the term "based on" can be understood not necessarily to convey an exclusive set of factors, but rather, alternatively, depending at least partly on the context, to allow for the presence of other factors that are not necessarily explicitly described.

[0063] like Figures 1-2 As shown, the digital twin method for integrating multimodal data includes the following steps:

[0064] S1, Entity Dynamic Deformation Field Monitoring: Real-time capture of three-dimensional deformation data of physical entities through distributed fiber optic sensors and laser holographic scanning, generating dynamic deformation fields;

[0065] S2, Construction of Non-rigid Reference System: Based on the dynamic deformation field, the motion vector of each point on the solid surface is calculated to generate a non-rigid reference system mesh that is updated in real time with deformation;

[0066] S3, Twin Reference System Binding: Dynamically binds the virtual coordinate system of the digital twin to a non-rigid reference system mesh, and outputs the twin with the bound reference system;

[0067] S4, Multimodal Data Remapping: The multimodal sensing data collected by the physical entity is spatially transformed according to the reference grid coordinates at the binding time, and the remapped data stream is output.

[0068] S5, Entity Deformation Suppression Closed Loop: Based on the remapped data stream, abnormal deformation regions are detected, actuator control commands are generated to suppress entity deformation, and suppressed data is collected and returned to S1.

[0069] S1 specifically includes:

[0070] S11, Fiber optic strain data calculation: Wavelength offset is acquired through distributed fiber optic sensors. The micro-strain distribution is calculated based on the fiber Bragg grating strain sensing model and is expressed as:

[0071] ;

[0072] in, It is the micro-strain value, in units of , This represents the wavelength offset of the fiber optic grating, in nm. The initial center wavelength of the fiber optic grating is in nm. It is the strain sensitivity coefficient, with a value of 0.78;

[0073] S12, Holographic Deformation Field Reconstruction: Obtaining the Phase Difference of Interference Fringes via Laser Holographic Scanning The normal deformation, reconstructed based on the holographic interferometric model, is expressed as:

[0074] ;

[0075] in, It is the normal deformation, with units of , It is the phase difference of the interference fringes, measured in rad. It is the laser wavelength, measured in nm. It is the laser incident angle, in units of ;

[0076] when When the light is incident perpendicularly, the formula is expressed as:

[0077] ;

[0078] S13, Spatiotemporal data fusion: integrating fiber optic microstrain data With holographic normal deformation Input the spatiotemporal fusion model to generate a three-dimensional dynamic deformation field tensor, represented as:

[0079] ;

[0080] in, It is a dynamic deformation field tensor. It is the three-dimensional spatial coordinate of the entity. It's a timestamp. It is the curvature operator of the normal deformation, representing the rate of change of the local surface shape. The curvature operator is defined as follows:

[0081] ;

[0082] In the discretization process, the central difference method can be used to approximate the solution of the above second-order partial derivatives, specifically including:

[0083] (1) Mesh discretization: Define the spatial grid coordinates of the discretized grid, expressed as:

[0084] ;

[0085] ;

[0086] in, These are the grid spacings in the x and y directions, respectively, in millimeters (mm). It represents grid points The normal deformation value at the location is expressed in micrometers. ;

[0087] (2) Discretization of the second-order partial derivative in the x-direction: Approximate calculation is performed using a three-point central difference scheme, expressed as:

[0088] ;

[0089] (3) Directional second-order partial derivative discretization: using central difference discretization, expressed as:

[0090] ;

[0091] (4) Curvature operator composition expression: Adding the difference formulas for directions yields the total curvature operator, expressed as:

[0092] ;

[0093] when hour, ;

[0094] (5) Boundary condition handling: For boundary points, the second-order forward / backward difference formula is used for handling, specifically including:

[0095] 1) Left boundary ( ):

[0096] ;

[0097] 2) Right boundary ( ):

[0098] ;

[0099] 3) Upper / lower boundaries or .

[0100] S2 specifically includes:

[0101] S21, Construction of Motion Vector Field: Based on Dynamic Deformation Field Normal displacement components in With strain components Calculate grid points The three-dimensional motion vector at that location is represented as:

[0102] ;

[0103] ;

[0104] ;

[0105] in, Grid points The motion vector at that location, The displacement gradient is calculated discretely using central difference. It is the micro-strain value, in units of (One part per million strain) It is the curvature value, in units of ;

[0106] S22, Thermal Deformation Compensation: Considering the thermal expansion effect of the material, the motion vector is corrected by thermal field adjustment to obtain the corrected vector, expressed as:

[0107] ;

[0108] in, It is the coefficient of thermal expansion of the material (unit: ), The temperature at the grid points obtained by infrared thermal imaging (unit: ), Reference temperature (unit: ), It is the motion vector after temperature correction;

[0109] S23, Reference Frame Mesh Generation: The discrete motion vector field is mapped to a continuous non-rigid reference frame mesh using a thin-plate spline interpolation method. (Mesh points...) The location is represented as:

[0110] ;

[0111] ;

[0112] in, Represents parameter points Corresponding three-dimensional spatial coordinates, control points The three-dimensional position is The interpolation weights are in vector form. , These are thin-plate spline basis functions. , For the location of control points, For spline interpolation weights, solve the linear system get, For elements of the basis functions of thin plate splines, , indicating the first The and the first Radial basis function response between control points , For affine terms, .

[0113] S3 specifically includes:

[0114] S31, Control Point Coordinate Mapping: Constructing a Virtual Control Point Set With physical control point set A one-to-one mapping relationship between them, where, These are the predefined anchor point coordinates in a digital twin. These are the spatial coordinates of the points corresponding to the physical entity. For distributed sensor grid indexing;

[0115] S32, Affine Transformation Solution: The optimal rigid transformation is found by minimizing the squared difference between the real and imaginary coordinates, expressed as:

[0116] ;

[0117] ;

[0118] in, It is a linear transformation matrix. It is a translation vector;

[0119] S33, Nonlinear Residual Compensation: Calculate the affine transformation residuals, and construct a thin-plate spline interpolation function based on the residual points, expressed as:

[0120] ;

[0121] ;

[0122] in, These are thin-plate spline basis functions, with the domain being... , Interpolation control weights;

[0123] S34, Binding Function Composition: Construct a dynamic binding transformation function, represented as:

[0124] ;

[0125] in, It is a dynamic binding transformation function. This indicates that three-dimensional points Projected onto the parametric plane of the non-rigid reference mesh, This represents the value of the residual field at that point.

[0126] S35, Full Model Binding: Binding all twin model vertices. Applying the transformation function, it can be expressed as:

[0127] ;

[0128] in, For the first after dynamic reference frame binding The vertex positions are used to generate a digital twin model that is dynamically consistent with the non-rigid reference frame of the physical entity.

[0129] S4 specifically includes:

[0130] S41, Sensor Parameter Coordinate Extraction: Extract the parameter coordinates of all sensors on the physical entity, forming a parameter coordinate set, represented as:

[0131] ;

[0132] in, It is the first The fixed coordinates of each sensor on the physical parameter plane This represents the total number of sensors;

[0133] S42, Real-time Spatial Position Calculation: Based on the current non-rigid reference frame mesh function Calculate the real-time three-dimensional spatial position of the sensor. , is represented as:

[0134] ;

[0135] S43, Binding Position Calculation: Call the mesh snapshot function recorded at the binding time. Calculate the reference positions for each sensor binding. , is represented as:

[0136] ;

[0137] S44, Spatial Displacement Vector Generation: Calculate the spatial displacement of each sensor. , is represented as:

[0138] ;

[0139] in, ;

[0140] S45, Data Remapping Execution: Remap raw sensor data Spatial translation to eliminate the effects of solid deformation is represented as:

[0141] ;

[0142] in, The sensor data is reproduced after replay;

[0143] S46, Data Flow Reconstruction: Aggregate all remapped data and output an aligned remapped data flow. , is represented as:

[0144] .

[0145] S5 specifically includes:

[0146] S51, extract the deformation-related feature set from the remapped data stream, represented as:

[0147] ;

[0148] in, Remapping microstrain (unit: ), Remapping curvature (unit: ), Strain change rate (unit: ), It is the Laplace gradient of curvature (unit: );

[0149] S52, Abnormal Region Identification: Identify abnormal regions based on a multi-level threshold strategy. , is represented as:

[0150] ;

[0151] in, It is the yield strain threshold of the material, and its value range is... , It is the safety curvature threshold, and its value range is... , It is the strain rate threshold, with a value range of 100%. , It is the curvature abrupt change spatial threshold, with a value range of [value missing]. ;

[0152] S53, Calculation of the suppressive force field: Based on the characteristic values ​​of the anomalous region, calculate the required reverse suppressive force. , is represented as:

[0153] ;

[0154] in, Proportional gain coefficient (unit: The value ranges from 0.1 to 1.0. Differential gain coefficient (unit: The value ranges from 0.001 to 0.05. Curvature compensation gain (unit: The value ranges from 0.5 to 5.0.

[0155] S54, Actuator Instruction Generation: Generates a control instruction set based on the calculated inhibition force. , is represented as:

[0156] ;

[0157] The specific definitions of each sub-instruction of the control instructions include:

[0158] (1) Hydraulic compensation command:

[0159] ;

[0160] (2) Piezoelectric micro-vibration command:

[0161] ;

[0162] (3) Active cooling command:

[0163] ;

[0164] in, The surface normal vector of the abnormal region. The system resonant frequency (unit: Hz). For conversion coefficients of various actuators, , , ;

[0165] S55, Closed-loop execution and feedback: After the control command is issued to the entity, the closed-loop feedback process is executed, which specifically includes:

[0166] (1) Execute control commands ;

[0167] (2) Waiting delay The condition is expressed as:

[0168] ;in, For hydraulic response time, For piezoelectric response time, Cooling response time;

[0169] (3) Collect deformation data again and feed it back to step S1.

[0170] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0171] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method of integrating multi-modal data for digital twin, characterized in that, The method comprises the following steps: S1, entity dynamic deformation field monitoring: capturing three-dimensional deformation data of a physical entity in real time through distributed optical fiber sensors and laser holographic scanning to generate a dynamic deformation field; S2, non-rigid reference system construction: calculating motion vectors of each point on the surface of the entity based on the dynamic deformation field to generate a non-rigid reference system grid that is updated in real time with the deformation; Specifically comprising: S21, motion vector field construction: calculating three-dimensional motion vectors at grid points based on normal displacement components and strain components in the dynamic deformation field; S22, thermal deformation compensation: considering the thermal expansion effect of the material, performing thermal field correction on the motion vectors; S23, reference system grid generation: mapping the discrete motion vector field to a continuous non-rigid reference system grid by using a thin plate spline interpolation method; S3, twin reference system binding: dynamically binding a virtual coordinate system of a digital twin to the non-rigid reference system grid to output a twin of the bound reference system; S4, multi-modal data remapping: performing spatial transformation on multi-modal sensing data collected by the physical entity according to the reference system grid coordinates at the binding time to output remapped data streams; specifically comprising: S41, sensor parameter coordinate extraction: extracting parameter coordinates of all sensors on the physical entity to form a parameter coordinate set; S42, real-time spatial position calculation: calculating real-time three-dimensional spatial positions of the sensors based on the current non-rigid reference system grid function; S43, bound position calculation: calculating bound reference positions of the sensors by calling the non-rigid reference system grid function recorded at the binding time; S44, spatial displacement vector generation: calculating spatial displacements of each sensor; S45, data remapping execution: performing spatial translation on the original sensor data to eliminate the influence of entity deformation; S46, data stream reconstruction: aggregating all remapped data to output aligned remapped data streams; S5, entity deformation suppression closed loop: detecting abnormal deformation areas based on the remapped data streams to generate actuator control instructions to suppress entity deformation, and collecting post-suppression data to return to S1; specifically comprising: S51, extracting a feature set related to deformation from the remapped data streams; S52, abnormal area identification: identifying abnormal areas according to a multi-level threshold strategy; S53, suppression force field calculation: calculating required reverse suppression forces according to feature values of the abnormal areas; S54, actuator instruction generation: generating control instruction sets according to the calculated suppression forces; S55, closed loop execution and feedback: after the control instructions are issued to the entity, performing a closed loop feedback process.

2. The digital twin method of integrating multi-modal data according to claim 1, wherein, The S1 comprises: S11, optical fiber strain data calculation: collecting wavelength shifts through distributed optical fiber sensors and calculating micro-strain distribution based on an optical fiber Bragg grating strain sensing model; S12, holographic deformation field reconstruction: obtaining interference fringe phase differences through laser holographic scanning and reconstructing normal deformation variables based on a holographic interferometry model; S13, spatio-temporal data fusion: inputting optical fiber micro-strain data and holographic normal deformation variables into a spatio-temporal fusion model to generate a three-dimensional dynamic deformation field tensor.

3. The digital twin method of integrating multi-modal data according to claim 2, wherein, The S13 comprises: S131, grid discretization: defining discretized spatial grid coordinates; S132, the three-point central difference format is used to approximate the calculation of the second-order partial derivative in the x direction; S133, the central difference is used to calculate the second-order partial derivative in the y direction; S134, the difference formulas in the x and y directions are added to obtain the total curvature operator; S135, boundary condition processing: using the second-order forward / backward difference formula to process the boundary points.

4. The digital twin method of integrating multi-modal data according to claim 1, wherein, The S21 includes: S211, displacement gradient and curvature extraction: based on the normal deformation obtained in the dynamic deformation field, the central difference method is used to calculate the displacement gradient vector of the grid point; S212, motion vector integration expression: combining the displacement gradient, curvature and micro-strain value to form a three-dimensional motion vector at the grid point.

5. The digital twin method of integrating multi-modal data according to claim 4, wherein, The S3 includes: S31, control point coordinate mapping: constructing a one-to-one mapping relationship between the virtual control point set and the physical control point set; S32, affine transformation solution: solving the optimal rigid transformation by minimizing the square difference between virtual and physical coordinates; S33, nonlinear residual compensation: calculating the affine transformation residual, and constructing a thin plate spline interpolation function based on the residual points; S34, binding function synthesis: constructing a dynamic binding transformation function based on the constructed thin plate spline interpolation function; S35, full model binding: applying the transformation function to all vertexes of the twin model.

6. The digital twin method of integrating multi-modal data according to claim 5, wherein, The S33 includes: S331, affine residual calculation: after the affine transformation is completed, the transformation residual vector of each control point pair is calculated; S332, residual interpolation function construction: based on the spatial distribution of all residual points, the distributed sensor index position is used as the interpolation control point to construct a thin plate spline interpolation function.

7. The digital twin method of integrating multi-modal data according to claim 1, wherein, The S54 includes: S541, hydraulic compensation instruction calculation: generating the driving area control instruction of the hydraulic system according to the inhibition force field and the normal vector of the abnormal area; S542, piezoelectric micro-vibration instruction calculation: using the piezoelectric actuator to generate a vibration control signal matched with the structural resonance frequency; S543, active cooling control instruction calculation: based on the integral quantity of the deformation and the safety threshold, the corresponding thermal control cooling power instruction is generated to reduce the thermal strain source.

Images