An automobile windshield glass optical performance simulation method and system based on digital twinning

Abstract

This invention discloses a method and system for simulating the optical performance of automotive windshields based on digital twins, belonging to the field of automotive windshield digital twin simulation technology. The method collects geometric surface parameters of the windshield, polyvinyl butyral adhesive layer material and viscoelastic characteristic parameters, and multiphysics service data to construct a digital twin geometric model and divide the double-layer glass and adhesive layer regions. A viscoelastic refractive index spatiotemporal coupling iterative algorithm is used to solve the spatial non-uniform gradient distribution of refractive index under temperature-induced aging and the transient changes in molecular chain orientation and local fluctuations of refractive index under alternating stress. This is mapped region by region to the adhesive layer region to obtain the real-time refractive index distribution. This invention simultaneously solves the spatial non-uniform gradient distribution of refractive index and local fluctuations under the dual effects of temperature-induced aging and alternating stress of the adhesive layer through the viscoelastic refractive index spatiotemporal coupling iterative algorithm, unifying gradient background distortion and dynamic local perturbations into ray tracing, making the simulation output approximate the real driving visual state.

Description

Technical Field

[0001] This invention relates to the field of digital twin simulation technology for automotive windshields, specifically to a method and system for simulating the optical performance of automotive windshields based on digital twins. Background Technology

[0002] Automotive head-up display (HUD) systems project driving information onto the driver's field of vision through the windshield. The optical quality of the windshield directly determines the clarity, positional accuracy, and visual comfort of the virtual image. As HUD systems evolve towards larger screen sizes and augmented reality, the need for evaluation and optimization of windshield optical performance continues to increase.

[0003] The windshield is a sandwich structure composed of double-layered glass and a polyvinyl butyral (PVB) adhesive layer. Its optical properties depend on the combined effect of the glass's curved surface geometry and the refractive index distribution of the adhesive layer. In the vehicle development process, to shorten the R&D cycle and reduce the cost of physical prototype manufacturing, the industry has introduced digital twin technology to virtually simulate the optical performance of windshields. This technology constructs geometric and optical property models of the windshield in digital space, simulates the propagation behavior of light within the glass structure, and outputs parameters such as optical distortion and line-of-sight offset in the main viewing area, providing simulation data support for the design of the glass's optical structure.

[0004] During its service life, the windshield is subjected to various factors such as changes in ambient temperature, solar radiation, and vehicle body structural loads, causing the polyvinyl butyral (PVB) adhesive layer inside the laminated structure to undergo physical state evolution. To characterize these effects in a digital twin model, the conventional approach is to assign optical equivalent values ​​to the adhesive layer material based on its reference refractive index parameters, establish a laminated optical model of the windshield, and then perform ray tracing calculations and evaluate relevant optical indicators.

[0005] The limitations of existing technologies include at least the following problems: In the digital twin optical simulation modeling of automotive windshields, the industry generally uses homogeneous equivalent modeling for the polyvinyl butyral adhesive layer inside the sandwich structure, ignoring the non-uniform spatial gradient distribution of the refractive index formed by the polyvinyl butyral adhesive layer under environmental temperature changes and long-term service aging. It also does not consider the transient changes in molecular chain orientation of the polyvinyl butyral adhesive layer due to viscoelastic characteristics in the alternating stress field of high-speed driving, and the resulting local fluctuations in refractive index. The modeling settings are doubly disconnected from the actual optical distribution state of the polyvinyl butyral adhesive layer. The simulated optical path characterization is limited to the geometric deformation of the outer glass panel, which is difficult to cover the combined effects of gradient background distortion and dynamic local disturbances on the propagating light. The optical distortion and line-of-sight offset parameters of the main viewing area output by this method deviate from the actual driving visual state, making it difficult for the glass optical structure design to match the actual field of vision after installation, and making it difficult to avoid the field of vision hazards in dynamic driving scenarios from the source of simulation. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a method and system for simulating the optical performance of automotive windshields based on digital twins. This solves the problem in existing technologies where the polyvinyl butyral adhesive layer of the windshield is simplified to a uniform medium, leading to a double disconnect between the digital twin optical simulation results and the actual service conditions.

[0007] To achieve the above objectives, this invention provides the following technical solution: a digital twin-based simulation method for the optical performance of automotive windshields, comprising the following steps: collecting geometric surface parameters of the double-layer glass, material and viscoelastic characteristics of the interlayer polyvinyl butyral adhesive layer, as well as environmental temperature changes, long-term service aging, and high-speed driving alternating stress field parameters corresponding to the service process; constructing a digital twin geometric model based on the geometric surface parameters of the double-layer glass, and dividing the double-layer glass and polyvinyl butyral adhesive layer structural regions; inputting the environmental temperature change and long-term service aging parameters into a viscoelastic refractive index spatiotemporal coupling iterative algorithm to calculate the non-uniform spatial gradient of the polyvinyl butyral adhesive layer under temperature-induced aging. The algorithm inputs the high-speed alternating stress field and the viscoelastic characteristic parameters of the polyvinyl butyral adhesive layer into a spatiotemporal coupling iterative algorithm for viscoelastic refractive index, calculating the transient changes in molecular chain orientation and local fluctuations in refractive index of the polyvinyl butyral adhesive layer under alternating stress. It then maps the non-uniform spatial gradient distribution of refractive index, the transient changes in molecular chain orientation, and the local fluctuations in refractive index region-by-region to the polyvinyl butyral adhesive layer structure region of the digital twin geometric model, obtaining the real-time distribution of the adhesive layer's refractive index. Finally, it imports the geometric deformation data of the outer glass panel, integrates it into the digital twin geometric model, and updates the surface parameters of the double-layered glass. A global ray tracing operation is then performed on the updated digital twin geometric model, outputting the optical distortion and line-of-sight offset data for the main viewing area.

[0008] Furthermore, the specific steps for solving the spatial non-uniform gradient distribution of refractive index are as follows: construct time-varying temperature field boundary conditions using environmental temperature variation parameters, and determine the aging time scale of the polyvinyl butyral adhesive layer using long-term service aging parameters; input the time-varying temperature field boundary conditions and aging time scale into the viscoelastic refractive index spatiotemporal coupling iterative algorithm, call the pre-calibrated temperature variation aging refractive index mapping relationship, and solve the refractive index offset point by point along each spatial coordinate of the polyvinyl butyral adhesive layer structure region; perform spatial interpolation and smoothing processing on the refractive index offset of each spatial coordinate to generate a spatial non-uniform gradient distribution of refractive index.

[0009] Further, the specific steps for calculating the transient changes in molecular chain orientation and the local fluctuations in refractive index are as follows: The parameters of the high-speed alternating stress field are expanded according to a time series, and the stress tensor components of each node within the polyvinyl butyral adhesive layer structure at each moment are extracted; the stress tensor components of each node and the viscoelastic characteristic parameters of the polyvinyl butyral adhesive layer are input into a spatiotemporal coupling iterative algorithm for viscoelastic refractive index, and the nonlinear constitutive relation of the stress optical coefficient as a function of strain rate is invoked to calculate the transient molecular chain orientation at each node; based on the transient molecular chain orientation, the anisotropic increment of refractive index at each node is calculated, and the principal component is taken as the local fluctuation of refractive index at that node; the transient changes in molecular chain orientation and the local fluctuations of refractive index at each node at each moment are summarized.

[0010] Furthermore, the specific steps for invoking the nonlinear constitutive relation are as follows: obtain experimental data of the stress optical coefficient of polyvinyl butyral material under different strain rate conditions; perform nonlinear fitting on the experimental data with the logarithm of strain rate as the independent variable and the stress optical coefficient as the dependent variable; and use the fitted polynomial function or piecewise function as the nonlinear constitutive relation of the stress optical coefficient as a function of strain rate.

[0011] Further, the specific steps to obtain the real-time refractive index distribution of the adhesive layer are as follows: the polyvinyl butyral adhesive layer structure region is divided into grid cells, and spatial index coordinates of each grid cell are established; using the spatial index coordinates as the mapping reference, the non-uniform gradient distribution of refractive index in space is used as the refractive index reference value of each grid cell and filled into each grid cell; the anisotropic refractive index increment associated with the transient change in molecular chain orientation of the corresponding spatial coordinates of the grid cell is superimposed on the refractive index reference value, and the local fluctuation of refractive index is superimposed, and the real-time spatial distribution of refractive index is calculated for each grid cell.

[0012] Furthermore, the specific steps for superimposing the refractive index anisotropy increment and the refractive index local fluctuation are as follows: extract the relaxation time spectrum from the viscoelastic characteristic parameters of the polyvinyl butyral adhesive layer; input the refractive index local fluctuation and the relaxation time spectrum into the viscoelastic refractive index spatiotemporal coupling iterative algorithm to perform time-domain phase correction on the refractive index local fluctuation; and superimpose the time-domain phase-corrected refractive index local fluctuation onto the refractive index reference value and the refractive index anisotropy increment of the corresponding grid cell.

[0013] Further, the specific steps for extracting the relaxation time spectrum are as follows: perform dynamic mechanical analysis frequency scanning tests on the polyvinyl butyral material to obtain test data on the changes in storage modulus and loss modulus with frequency; fit the test data with a generalized Maxwell model to obtain the relaxation time spectrum of the polyvinyl butyral material.

[0014] Furthermore, the specific steps for updating the surface parameters of the double-layer glass are as follows: collect measured point cloud data or finite element deformation simulation data of the outer glass panel of the automotive windshield in the actual vehicle assembly state as the geometric deformation data of the outer glass panel; use the initial node coordinates of the double-layer glass structure region in the digital twin geometric model as the reference, decompose the geometric deformation data of the outer glass panel into displacement vectors of each node; apply displacement vectors to update the surface parameters of the double-layer glass node by node.

[0015] Furthermore, the specific steps for outputting the optical distortion and line-of-sight offset data of the main viewing area are as follows: taking the driver's eye position as the starting point of ray tracing, a ray tracing beam is emitted towards the incident side of the updated digital twin geometric model; the ray tracing beam passes sequentially through the double-layer glass structure region after updating the surface parameters and the polyvinyl butyral adhesive layer structure region that completes the real-time refractive index distribution mapping, and the position coordinates and direction vectors of each ray on the exit side are recorded; the exit position coordinates and direction vectors of each ray are compared with the exit position coordinates and direction vectors of the theoretical interference-free reference ray, the optical distortion and line-of-sight offset data of the main viewing area are calculated and output.

[0016] A digital twin-based simulation system for the optical performance of automotive windshields includes: a parameter acquisition module for acquiring geometric surface parameters of the double-layer glass, material and viscoelastic characteristics of the interlayer polyvinyl butyral adhesive layer, and environmental temperature changes, long-term service aging, and high-speed driving alternating stress field parameters during service; a geometric modeling module for constructing a digital twin geometric model based on the geometric surface parameters of the double-layer glass and dividing the structural regions of the double-layer glass and the polyvinyl butyral adhesive layer; and a viscoelastic refractive index calculation module for calling the spatiotemporal calculation of the viscoelastic refractive index. The coupled iterative algorithm solves for the non-uniform gradient distribution of refractive index space, transient changes in molecular chain orientation, and local fluctuations in refractive index. The mapping and deformation fusion module maps the non-uniform gradient distribution of refractive index space, transient changes in molecular chain orientation, and local fluctuations in refractive index region by region to the polyvinyl butyral adhesive layer structure region, and imports the geometric deformation data of the outer glass plate to update the surface parameters of the double-layer glass. The ray tracing output module performs global ray tracing calculations on the updated digital twin geometric model and outputs optical distortion and line-of-sight offset data in the main viewing area.

[0017] The present invention has the following beneficial effects: (1) The simulation method of the optical performance of automotive windshield glass based on digital twins drives the spatiotemporal coupling iterative algorithm of viscoelastic refractive index by collecting environmental temperature change and long-term service aging parameters. The refractive index offset is calculated point by point along each spatial coordinate of the polyvinyl butyral adhesive layer structure region and a non-uniform gradient distribution of refractive index is generated. This makes the adhesive layer present the gradient refractive characteristics that evolve with the service process in the digital twin model. The deflection effect of the refractive index difference inside the adhesive layer on the propagation of light is incorporated into the simulation optical path calculation, thus eliminating the static disconnect between traditional modeling and the real optical distribution of the adhesive layer.

[0018] (2) The simulation method of automotive windshield optical performance based on digital twin expands the alternating stress field parameters of high-speed driving according to the time series, extracts the stress tensor components of each node in the adhesive layer structure region and solves the transient molecular chain orientation degree node by node, and quantitatively characterizes the instantaneous fluctuation of the refractive index inside the adhesive layer by the anisotropic increment of refractive index, so that the dynamic optical disturbances caused by service conditions such as road excitation and airflow pulsation are captured by the digital twin model, and the dynamic disconnect between traditional modeling and the real optical distribution of the adhesive layer is resolved.

[0019] (3) The simulation method for the optical performance of automotive windshield glass based on digital twins uses the non-uniform gradient distribution of refractive index space as the reference value, and the anisotropic increment of refractive index associated with the transient change of molecular chain orientation and the local fluctuation of refractive index as the superposition quantity. The real-time spatial distribution of the refractive index of the adhesive layer is constructed cell by cell, so that the digital twin model has the optical representation capability of both time and space. The tracing beam passes through the double-layer glass after updating the surface parameters and the adhesive layer structure region after completing the refractive index mapping. By comparing the outgoing light with the theoretical interference-free reference light, the optical distortion and line of sight offset data of the main viewing area are output, so that the simulation results are close to the real driving and riding visual state.

[0020] (4) The automotive windshield optical performance simulation system based on digital twins, through the collaborative architecture of parameter acquisition, geometric modeling, viscoelastic refractive index calculation, mapping and deformation fusion and ray tracing output modules, orderly drives the dispersed physical inputs such as environmental temperature change, long-term service aging, and high-speed driving alternating stress field into a complete simulation link. The viscoelastic refractive index calculation module undertakes the dual tasks of temperature change aging gradient calculation and alternating stress fluctuation calculation. The mapping and deformation fusion module integrates the calculation results and the geometric deformation data of the glass outer plate into the digital twin geometric model, realizing the physical coupling of geometric surface changes and medium refractive index changes within the same model framework.

[0021] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0022] Figure 1This is a flowchart of a digital twin-based simulation method for the optical performance of automotive windshields according to the present invention.

[0023] Figure 2 This is a flowchart illustrating the specific steps involved in solving the spatial non-uniform gradient distribution of refractive index in a digital twin-based simulation method for the optical performance of automotive windshields according to the present invention.

[0024] Figure 3 This is a block diagram of a digital twin-based simulation system for the optical performance of automotive windshields according to the present invention. Detailed Implementation

[0025] Please see Figure 1 This invention provides a technical solution: a method for simulating the optical performance of automotive windshields based on digital twins, comprising the following steps: collecting geometric surface parameters of the double-layer glass, material and viscoelastic characteristics of the interlayer polyvinyl butyral adhesive layer, as well as environmental temperature changes, long-term service aging, and high-speed driving alternating stress field parameters corresponding to the service process; constructing a digital twin geometric model based on the geometric surface parameters of the double-layer glass, and dividing the structural regions of the double-layer glass and the polyvinyl butyral adhesive layer; inputting the environmental temperature change and long-term service aging parameters into a viscoelastic refractive index spatiotemporal coupling iterative algorithm to calculate the non-uniform spatial gradient distribution of the refractive index of the polyvinyl butyral adhesive layer under temperature change aging; and then... The alternating stress field at high speed and the viscoelastic characteristic parameters of the polyvinyl butyral adhesive layer are input into a spatiotemporal coupled iterative algorithm for viscoelastic refractive index. The transient changes in molecular chain orientation and local fluctuations in refractive index of the polyvinyl butyral adhesive layer under alternating stress are calculated. The non-uniform gradient distribution of refractive index space, the transient changes in molecular chain orientation, and the local fluctuations in refractive index are mapped region by region to the polyvinyl butyral adhesive layer structure region of the digital twin geometric model to obtain the real-time distribution of the adhesive layer refractive index. The geometric deformation data of the outer glass panel are imported, integrated into the digital twin geometric model, and the surface parameters of the double-layer glass are updated. A global ray tracing operation is performed on the updated digital twin geometric model to output the optical distortion and line-of-sight offset data of the main viewing area.

[0026] Among them, the viscoelastic refractive index spatiotemporal coupling iterative algorithm uses the finite element discrete iterative method to realize the spatiotemporal correlation calculation of the temperature change aging cumulative effect and the alternating stress transient effect. The iteration step size can be flexibly adjusted according to the simulation accuracy requirements. In one implementation, the iteration step size is set to 0.01s-0.1s; All collected parameters need to be denoised to remove invalid data and outliers, and then converted to the same data format before being input into subsequent steps. Among them, the geometric surface parameters of the double-layer glass are controlled with an accuracy of ±0.01mm, the parameters of the polyvinyl butyral adhesive layer include density, glass transition temperature, and Poisson's ratio, the viscoelastic characteristic parameters include storage modulus, loss modulus, and relaxation time, the environmental temperature change parameters cover the temperature range of -40℃ to 80℃ commonly used in automotive service, the long-term service aging parameters are divided into 0-5 year service cycles, and the high-speed driving alternating stress field parameters cover the stress change range at vehicle speeds of 60-180km / h.

[0027] Specifically, such as Figure 2 As shown, the specific steps for solving the non-uniform spatial gradient distribution of refractive index are as follows: The time-varying temperature field boundary conditions are constructed using environmental temperature variation parameters, and the aging time scale of the polyvinyl butyral adhesive layer is determined using long-term service aging parameters. Specifically: The environmental temperature change parameters include real-time temperature data, temperature change rate and temperature duration under different service conditions. The time-varying temperature field boundary conditions are constructed according to the time series based on these parameters, covering the temperature differences in different areas of the windshield (main viewing area and edge area). In one embodiment, the temperature change rate of the main viewing area is set to 0.5℃ / min-2℃ / min, and the temperature change rate of the edge area is set to 0.3℃ / min-1.5℃ / min; The long-term service life parameters are divided by quarter, and the aging time scale corresponds to five gradients: 0.5 years, 1 year, 2 years, 3 years and 5 years. Each gradient corresponds to a different degree of aging. Among them, 0.5 years corresponds to mild aging, and 5 years corresponds to severe aging. The aging parameter can accurately characterize the aging and decay law of polyvinyl butyral adhesive layer with service time. The time-varying temperature field boundary conditions and aging timescale are input into the viscoelastic refractive index spatiotemporal coupling iterative algorithm. A pre-calibrated temperature-varying aging refractive index mapping relationship is invoked to calculate the refractive index offset point-by-point along each spatial coordinate of the polyvinyl butyral adhesive layer structure region. Specifically: The pre-calibrated temperature-dependent aging refractive index mapping relationship was obtained through a large number of experiments. The experiments used polyvinyl butyral adhesive layer samples identical to those in the actual vehicle, and were tested at different temperatures (-40℃, -20℃, 0℃, 25℃, 40℃, 60℃, 80℃). By establishing the correspondence between temperature, aging time and refractive index of the test samples under different aging time scales, a mapping table is formed (as shown in Table 1). The algorithm calls this mapping table and, in combination with the time-varying temperature field boundary conditions and aging time scale, calculates the difference between the refractive index and the reference refractive index (25℃, refractive index in the unaged state) at each spatial coordinate point by point, which is the refractive index offset.

[0028] Spatial interpolation and smoothing are performed on the refractive index offsets of each spatial coordinate to generate a non-uniform spatial gradient distribution of refractive index, specifically as follows: A linear interpolation method is used to interpolate the refractive index offset between adjacent spatial coordinates to fill in the offset data in the coordinate gaps. The smoothing process uses a Gaussian filtering method to filter out abnormal fluctuations in the offset data. The filter window size is set to 3×3. Without changing the overall gradient trend, the refractive index spatial distribution is made to better match the actual optical properties of the polyvinyl butyral adhesive layer. After processing, the non-uniform gradient distribution of the refractive index space is obtained by organizing according to the spatial coordinates, which clearly presents the refractive index differences in different regions of the adhesive layer.

[0029] In this implementation scheme, by using both environmental temperature change parameters and long-term service aging parameters as input drivers, the time-varying temperature field boundary conditions are constructed and the aging time scale of the polyvinyl butyral adhesive layer is determined. This allows the evolution of the adhesive layer's refractive index to correspond to the actual service history of the vehicle. The pre-calibrated temperature-varying aging refractive index mapping relationship solidifies the correlation between temperature, aging time, and refractive index into callable data support. After the algorithm calculates the refractive index offset point by point along each spatial coordinate of the adhesive layer structure region, it generates a continuous spatial non-uniform gradient distribution of refractive index through spatial interpolation and smoothing. This makes the adhesive layer in the digital twin model exhibit a gradually differentiating refractive characteristic with spatial position and service status. The deflection effect of the refractive index difference inside the adhesive layer on light propagation is incorporated into the simulation optical path calculation, bridging the gap between traditional homogeneous equivalent modeling and the real optical distribution of the adhesive layer.

[0030] Specifically, the steps for calculating the transient changes in molecular chain orientation and the local fluctuations in refractive index are as follows: The alternating stress field parameters during high-speed driving are expanded according to a time series, and the stress tensor components of each node in the polyvinyl butyral adhesive layer structure region at each time point are extracted, specifically as follows: The parameters of the high-speed driving alternating stress field include stress amplitude, stress frequency, stress direction and stress duration. They are unfolded into continuous stress data with one data point every 0.01s in time series, covering different high-speed driving conditions such as acceleration, constant speed and deceleration. The nodes within the polyvinyl butyral adhesive layer structure are divided into meshes, with each node corresponding to a spatial coordinate. The stress tensor components of each node in the x, y, and z directions at each time point are extracted. ; Where x-direction corresponds to the horizontal direction of the windshield, y-direction corresponds to the vertical direction, and z-direction corresponds to the thickness direction of the glass. In one implementation, during high-speed constant-speed travel (120 km / h), the node's The stress tensor components are 15-25 MPa. The tensile component is 10-20 MPa; The stress tensor components of each node and the viscoelastic characteristic parameters of the polyvinyl butyral adhesive layer are input into the spatiotemporal coupling iterative algorithm of viscoelastic refractive index. The nonlinear constitutive relation of stress optical coefficients as a function of strain rate is invoked to solve the transient molecular chain orientation degree node by node. Specifically: Storage modulus and loss modulus in the viscoelastic characteristic parameters are used to characterize the viscoelastic response characteristics of the polyvinyl butyral adhesive layer. The algorithm combines the stress tensor components of each node to calculate the strain rate of each node, calls the nonlinear constitutive relation, determines the corresponding stress optical coefficient based on the strain rate, and then solves the transient molecular chain orientation degree. The value of molecular chain orientation degree is in the range of 0-1. The closer the value is to 1, the more orderly the molecular chain orientation is. The closer the value is to 0, the more disordered the molecular chain orientation is. In one implementation, when the strain rate is 10... -3 s -1 At this time, the molecular chain orientation degree is 0.3-0.5, and when the strain rate is 10... -2 s -1 At that time, the degree of molecular chain orientation is 0.6-0.8; The anisotropic increment of refractive index at each node is calculated based on the transient molecular chain orientation. The principal component is taken as the local fluctuation of refractive index at that node. The transient change in molecular chain orientation and the local fluctuation of refractive index at each node at each time point are summarized as follows: The increase in refractive index anisotropy is positively correlated with the degree of molecular chain orientation; the higher the degree of molecular chain orientation, the greater the increase in refractive index anisotropy. This can be verified by the formula... calculate; in, denoted as the refractive index anisotropy increment, and k is the proportionality coefficient (determined by the properties of polyvinyl butyral materials, with a value range of 0.001-0.005). This is a molecular chain orientation function. The molecular chain orientation angle; The principal component of the anisotropic refractive index increment (i.e. the maximum increment value) is taken as the local refractive index fluctuation at that node to avoid calculation errors caused by the superposition of increments in multiple directions. By summarizing the molecular chain orientation change data (i.e., transient changes in molecular chain orientation) and the local fluctuations in refractive index at all times and at all nodes, a spatiotemporally continuous dataset is formed.

[0031] The specific steps for invoking nonlinear constitutive relations are as follows: Experimental data on the stress optical coefficient of polyvinyl butyral material under different strain rate conditions were obtained, specifically as follows: The polyvinyl butyral adhesive layer sample was tested using a dynamic mechanical analyzer, with the test strain rate range set to 10.-4 s -1 Up to 10 -1 s -1 The experiment covered the strain rate range that the adhesive layer might withstand during high-speed vehicle operation. Three parallel experiments were set up for each strain rate gradient to obtain the stress optical coefficient data at the corresponding strain rate. After removing abnormal experimental data, the average value of the three parallel experiments was taken as the final stress optical coefficient data at that strain rate. Some experimental data are shown in Table 2.

[0032] Using the logarithm of strain rate as the independent variable and the stress optical coefficient as the dependent variable, a nonlinear fitting was performed on the experimental data, specifically as follows: Take the natural logarithm of the strain rate ( Using the x-axis as the x-axis and the corresponding stress optical coefficient as the y-axis, a scatter plot is drawn. The scatter plot is then fitted nonlinearly using the least squares method, with a quadratic polynomial function chosen as the fitting function. The fitting formula is: ; Where C is the stress optical coefficient. Let a be the strain rate, and b and c be the fitting coefficients. In one embodiment, the fitted values ​​of a are 0.05-0.15, b are 0.2-0.4, and c are 2.5-3.0. The fitted polynomial or piecewise function is used as the nonlinear constitutive relation of the stress optical coefficients as a function of strain rate, specifically: When the strain rate range is 10 -4 s -1 Up to 10 -1 s -1 In this case, the aforementioned quadratic polynomial function is used as the nonlinear constitutive relation; If the strain rate exceeds this range, a piecewise function is used to compensate, i.e., when the strain rate is less than 10... -4 s -1 When the strain rate is greater than 10, the stress optical coefficient is supplemented according to a linear decreasing trend. -1 s -1 At that time, the stress optical coefficient is supplemented in a linear increasing trend to ensure that the corresponding stress optical coefficient can be called under different strain rate conditions.

[0033] In this implementation scheme, by expanding the alternating stress field parameters of high-speed driving according to the time series, the stress tensor components of each node in the adhesive layer structure region are extracted and input together with the viscoelastic characteristic parameters of the polyvinyl butyral adhesive layer into the spatiotemporal coupling iterative algorithm of viscoelastic refractive index. The nonlinear constitutive relation of stress optical coefficient with strain rate is called to solve the transient molecular chain orientation degree node by node, and then the anisotropic increment of refractive index is calculated and its principal component is taken as the local fluctuation of refractive index. This path incorporates the instantaneous disturbance of the internal refractive index of the adhesive layer caused by service conditions such as road excitation and airflow pulsation into the quantitative characterization, so that the digital twin model can capture the dynamic optical fluctuations that are completely ignored by traditional simulation, thereby resolving the dynamic disconnect between the model setting and the actual optical distribution of the adhesive layer.

[0034] Specifically, the steps to obtain the real-time refractive index distribution of the adhesive layer are as follows: The polyvinyl butyral adhesive layer structure is divided into mesh elements, and the spatial index coordinates of each mesh element are established, as follows: Quadrilateral grids were used to divide the polyvinyl butyral adhesive layer structure region, and the grid cell size was set according to the simulation accuracy requirements. In one implementation, the mesh cell size of the main viewing area is set to 1mm×1mm, and the mesh cell size of the edge area is set to 2mm×2mm. The main viewing area uses a finer mesh to ensure the calculation accuracy of the refractive index distribution in the main viewing area. Each grid cell is assigned a unique spatial index coordinate, with the coordinate format being (x, y, z). Where x and y correspond to the planar coordinates of the windshield surface, z corresponds to the coordinates in the direction of adhesive layer thickness, and the value of z ranges from 0.38mm to 2.0mm (covering the thickness of PVB adhesive layer in common automotive windshields). Using spatial index coordinates as the mapping reference, the non-uniform spatial gradient distribution of refractive index is used as the refractive index reference value for each grid cell and filled into each grid cell, specifically as follows: The calculated non-uniform spatial gradient distribution data of refractive index is filled with the refractive index reference value for each grid cell according to the correspondence between spatial index coordinates and grid cells, ensuring that the refractive index reference value of each grid cell is accurately matched with the spatial position of the grid cell. In one implementation, the refractive index reference value of the central grid cell in the main view area is 1.488-1.490, and the refractive index reference value of the edge grid cell is 1.486-1.488, reflecting the difference in refractive index gradient caused by temperature change aging. The anisotropic refractive index increment, associated with the transient changes in molecular chain orientation at the corresponding spatial coordinates of the grid cells, is superimposed on the refractive index reference value. Local refractive index fluctuations are then superimposed, and the real-time spatial distribution of the refractive index is calculated cell by cell. Specifically: Based on the spatial index coordinates of the grid cells, the corresponding transient changes in molecular chain orientation and the associated refractive index anisotropy increment and refractive index local fluctuation are found. The real-time refractive index data is calculated cell by cell according to the calculation method of real-time refractive index value = refractive index reference value + refractive index anisotropy increment + refractive index local fluctuation. In one implementation, the refractive index reference value of a certain main view area grid cell is 1.489, the refractive index anisotropy increment is 0.002, and the local refractive index fluctuation is 0.001, then the real-time refractive index value of the grid cell is 1.492. After the calculation is completed, the real-time refractive index values ​​of all grid cells are integrated to form the real-time spatial distribution of the refractive index of the polyvinyl butyral adhesive layer.

[0035] The specific steps for superimposing the anisotropic increment of refractive index and the local fluctuation of refractive index are as follows: The relaxation time spectrum was extracted from the viscoelastic characteristic parameters of the polyvinyl butyral adhesive layer, specifically as follows: The viscoelastic characteristic parameters of the polyvinyl butyral adhesive layer include storage modulus and loss modulus data at different frequencies. The relaxation time spectrum is extracted by fitting these data. The extracted relaxation time spectrum contains multiple relaxation time components, covering short relaxation times (10). -6 S-10 -3 s), medium relaxation time (10 -3 s-100s), long relaxation time (100s-10) 6 s) These correspond to different modes of motion of the molecular chain; The local refractive index fluctuations and the relaxation time spectrum are input into the viscoelastic refractive index spatiotemporal coupling iterative algorithm to perform time-domain phase correction on the local refractive index fluctuations, specifically as follows: The local fluctuations in refractive index change dynamically over time, and the relaxation time spectrum characterizes the time response characteristics of the polyvinyl butyral adhesive layer. The algorithm combines the relaxation time spectrum to perform phase correction on the local fluctuations in refractive index at different times. The core of the correction is to eliminate the lag deviation of fluctuations caused by the relaxation effect of molecular chains. In one implementation, the phase correction angle is set to 5°-10° for fluctuations corresponding to short relaxation times and 15°-25° for fluctuations corresponding to long relaxation times. The local refractive index fluctuations after time-domain phase correction are superimposed onto the refractive index reference value and the refractive index anisotropy increment of the corresponding grid cell. Specifically: The corrected local fluctuation of refractive index is superimposed with the anisotropic increment of refractive index of the grid cell to obtain the dynamic optical increment value. The dynamic optical increment value is superimposed on the refractive index reference value. During the superposition process, all data of the same grid cell correspond to the same spatial index coordinates to avoid data confusion between different coordinates. After the superposition is completed, the final real-time value of the refractive index of the grid cell is obtained, which is used for subsequent updates of the digital twin geometric model.

[0036] The specific steps for extracting the relaxation time spectrum are as follows: Dynamic mechanical analysis frequency scanning tests were performed on polyvinyl butyral materials to obtain test data on the changes in storage modulus and loss modulus with frequency. Specifically: A dynamic mechanical analyzer was used to perform frequency scanning tests on the polyvinyl butyral adhesive layer samples. The test frequency range was set to 0.01Hz-1000Hz, and the test temperature was set to 25℃ (normal service conditions). During the test, the samples were kept under tension, and the tensile stress was set to 5-10MPa to avoid sample damage. A set of energy storage modulus (E') and loss modulus (E'') data was recorded every 0.1 Hz. After the test was completed, continuous data curves of energy storage modulus and loss modulus changing with frequency were obtained. Some test data are shown in Table 3:

[0037] The relaxation time spectrum of polyvinyl butyral material was obtained by fitting the test data to the generalized Maxwell model, as follows: The generalized Maxwell model consists of multiple Maxwell elements connected in parallel, with each Maxwell element corresponding to a relaxation time. The model expression is as follows: ; in, For energy storage modulus, For loss modulus, Angular frequency, Let be the elastic modulus of the i-th Maxwell element. Let n be the relaxation time of the i-th Maxwell cell, and n be the number of Maxwell cells (in one implementation, n = 5 - 8). The least squares method was used to fit the test data to determine the performance of each Maxwell cell. Integrate all The relaxation time spectrum of polyvinyl butyral material was obtained, clearly showing the molecular chain motion characteristics corresponding to different relaxation times.

[0038] In this implementation scheme, the polyvinyl butyral adhesive layer structure is discretized into grid cells and spatial index coordinates are established. The non-uniform spatial gradient distribution of refractive index is used as a reference value to accurately match each grid cell. The anisotropic increment of refractive index associated with the transient change of molecular chain orientation and the local fluctuation of refractive index are superimposed cell by cell to construct a real-time distribution of adhesive layer refractive index that evolves synchronously with spatial position and time. During the superposition process, the relaxation time spectrum is extracted from the viscoelastic characteristic parameters and the local fluctuation of refractive index is corrected in the time domain. This eliminates the lag deviation of fluctuation caused by molecular chain relaxation effect and keeps the dynamic optical correction amount synchronized with the loading sequence of alternating stress field.

[0039] Specifically, the steps for updating the double-glazed surface parameters are as follows: Collect measured point cloud data or finite element deformation simulation data of the outer glass panel of the automotive windshield in its actual vehicle assembly state as geometric deformation data of the outer glass panel. Specifically: The measured point cloud data was collected using a 3D laser scanner. The scanning range covered the entire glass outer panel, and the scanning accuracy was controlled within ±0.01mm. The collection density was 1 point cloud data per square millimeter. The collection conditions included common real vehicle conditions such as empty vehicle, fully loaded vehicle, high-speed driving, low temperature environment, and high temperature environment. Three sets of point cloud data were collected for each condition, and the average value was taken as the measured point cloud data for that condition. The finite element deformation simulation data is calculated using a finite element model of the car body. It simulates the effects of bolt preload, body load and ambient temperature on the deformation of the glass outer panel during actual vehicle assembly. The output deformation data is consistent with the measured point cloud data format and can be used as a supplement to the measured data. Using the initial node coordinates of the double-glazed structure region in the digital twin geometric model as a reference, the geometric deformation data of the outer glass panel is decomposed into displacement vectors of each node, specifically as follows: In the digital twin geometric model, the initial node coordinates of the double-glazed structure region are the standard coordinates before deformation. Using these coordinates as a reference, the geometric deformation data (point cloud data or finite element data) of the outer glass panel are compared with the initial coordinates to calculate the displacement of each node in the x, y, and z directions, forming the displacement vector of each node. ); in, This is the horizontal displacement. This is the vertical displacement. This represents displacement in the thickness direction. In one implementation, under high-speed driving conditions, the main view node's The displacement is 0.05mm-0.1mm. The displacement is 0.03mm-0.08mm; The displacement vector is applied node by node to update the surface parameters of the double-layer glass, specifically as follows: According to the displacement vector of each node, the node coordinates of the double-glass structure region in the digital twin geometric model are adjusted node by node. The initial node coordinates are superimposed with the displacement vector to obtain the deformed node coordinates, that is, the deformed coordinates = initial coordinates + displacement vector. After the adjustment is completed, the surface of the double-layer glass is smoothed to ensure that the surface is continuous without any breaks and to avoid surface distortion caused by node displacement. The updated surface parameters of the double-layer glass accurately represent the deformation characteristics of the glass outer panel in the actual vehicle assembly state.

[0040] In this implementation scheme, by collecting measured point cloud data or finite element deformation simulation data of the glass outer panel under the actual vehicle assembly state, the geometric deformation of the glass outer panel is decomposed into nodal displacement vectors and applied to the digital twin geometric model node by node, updating the double-layer glass surface parameters. This path enables the glass surface morphology in the digital twin model to realistically reflect the micro-deformation caused by actual vehicle working conditions such as assembly preload, vehicle body load, and ambient temperature. The comparison and decomposition of deformation data with the initial nodal coordinates ensures the spatial correspondence accuracy of the displacement vectors. The updated double-layer glass surface parameters provide a geometric input basis that matches the actual vehicle assembly state for subsequent ray tracing calculations, thereby reducing the deviation between the simulation output and the actual driving and riding visual state.

[0041] Specifically, the steps for outputting the optical distortion and line-of-sight offset data of the main view area are as follows: Using the driver's eye position as the starting point for ray tracing, a ray tracing beam is emitted towards the incident side of the updated digital twin geometry model, specifically as follows: The driver's eye position coordinates are set with reference to automotive industry standards, and are set as the eye point coordinates of the standard position of the driver's seat; in one embodiment, the eye position coordinates are set as (X=500mm, Y=1200mm, Z=700mm) (a coordinate system is established with the lower left corner of the windshield as the origin); The tracking beam uses a parallel beam with a wavelength of 550nm (the center wavelength of visible light, which matches the visual characteristics of the human eye). The beam covers the main viewing area of ​​the windshield (a 100mm x 150mm area directly in front of the driver), and the beam spacing is set to 0.5mm to ensure that the beam can fully cover the main viewing area and capture the optical characteristics of all positions in the main viewing area. The tracing ray beam sequentially passes through the double-layered glass structure region after the surface parameters have been updated, and the polyvinyl butyral adhesive layer structure region after real-time refractive index distribution mapping is completed. The position coordinates and direction vectors of each ray on the exit side are recorded, specifically as follows: During ray tracing, following the law of refraction, and combining the refractive index of the double-layer glass (the refractive index of conventional automotive glass is 1.52) with the real-time distribution of the refractive index of the polyvinyl butyral adhesive layer, the refraction angle of each ray at the glass-adhesive layer interface and the adhesive layer-glass interface is calculated to trace the complete propagation path of the ray. On the side where the light rays emanate (inside the windshield, facing the driver), record the coordinates of the emanation position of each ray. ) and direction vector (dx, dy, dz), the direction vector is used to characterize the propagation direction of light after it is emitted; The emission position coordinates and direction vectors of each ray are compared with those of the theoretical interference-free reference ray. The optical distortion and line-of-sight offset data of the main field of view are calculated and output, as follows: The theoretically interference-free reference ray's exit position coordinates and direction vector are the exit data of the ray after passing through a digital twin geometric model with no geometric deformation and the refractive index of the polyvinyl butyral layer taking the spatial-temporal average value. The position coordinates obtained from the actual tracking are compared with the ideal coordinates, and the position deviation value is calculated, which is the optical distortion value. The actual direction vector is compared with the ideal direction vector, and the direction deviation angle is calculated, which is the line-of-sight offset. In one embodiment, the optical distortion value at the center of the main viewing area is controlled within 0.01mm-0.03mm, and the line-of-sight offset is controlled within 0.1°-0.3°; After the calculation is completed, the optical distortion values ​​and line-of-sight offsets at different positions in the main viewing area are compiled into a data table, the corresponding service conditions are marked, and simulation data for windshield optical structure design is output.

[0042] In this implementation scheme, a ray tracing beam is emitted from the driver's eye position as the starting point of ray tracing to a digital twin geometric model that has updated surface parameters and completed real-time refractive index distribution mapping. The ray passes sequentially through a double-layer glass structure region with actual vehicle deformation characteristics and a polyvinyl butyral adhesive layer structure region with gradient and fluctuation superposition characteristics. After recording the emission position coordinates and direction vectors of each ray, the data is compared with the emission data of the theoretical interference-free reference ray. The optical distortion and line-of-sight offset data of the main viewing area are calculated and output. This path integrates the combined effects of geometric surface changes and medium refractive index changes on light propagation into a quantitative comparison framework, making the simulation output results close to the real driving visual state.

[0043] Please see Figure 3This invention provides a technical solution: a digital twin-based simulation system for the optical performance of automotive windshields, comprising: a parameter acquisition module for acquiring geometric surface parameters of the double-layer glass, material and viscoelastic characteristics of the interlayer polyvinyl butyral adhesive layer, and environmental temperature changes, long-term service aging, and high-speed driving alternating stress field parameters corresponding to the service process; a geometric modeling module for constructing a digital twin geometric model based on the geometric surface parameters of the double-layer glass, and dividing the double-layer glass and polyvinyl butyral adhesive layer structural regions; and a viscoelastic refractive index calculation module for adjusting... A viscoelastic refractive index spatiotemporal coupling iterative algorithm is used to solve for the non-uniform gradient distribution of refractive index space, transient changes in molecular chain orientation, and local fluctuations in refractive index. The mapping and deformation fusion module is used to map the non-uniform gradient distribution of refractive index space, transient changes in molecular chain orientation, and local fluctuations in refractive index region by region to the polyvinyl butyral adhesive layer structure region, and import the geometric deformation data of the outer glass plate to update the surface parameters of the double-layer glass. The ray tracing output module is used to perform global ray tracing calculations on the updated digital twin geometric model and output the optical distortion and line-of-sight offset data of the main view area.

[0044] The parameter acquisition module includes a 3D laser scanner, a material performance tester, a stress sensor, and a temperature sensor. Among them, the 3D laser scanner is used to collect the geometric surface parameters of the double-layer glass, the material performance tester is used to collect the material and viscoelastic characteristic parameters of the polyvinyl butyral adhesive layer, the stress sensor is used to collect the parameters of the alternating stress field during high-speed driving, and the temperature sensor is used to collect the parameters of environmental temperature change. The geometric modeling module uses 3D modeling software and supports functions such as mesh generation, region identification, and surface verification, enabling the rapid construction and modification of digital twin geometric models; The viscoelastic refractive index calculation module has a built-in viscoelastic refractive index spatiotemporal coupling iterative algorithm, which has functions such as data calculation, iterative optimization, and abnormal data processing. The calculation accuracy can be adjusted according to the requirements. The mapping and deformation fusion module supports functions such as mesh coordinate mapping, data overlay, and surface update to ensure accurate fusion of refractive index data and geometric model; The ray tracing output module has a built-in ray tracing algorithm that supports global ray tracing, data comparison, distortion and offset calculation, and can output simulation results in various forms such as data tables and curves.

[0045] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0046] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for simulating the optical performance of automotive windshields based on digital twins, characterized in that, Includes the following steps: The geometric surface parameters of the double-layer glass of the automotive windshield, the material and viscoelastic characteristics of the interlayer polyvinyl butyral adhesive layer, as well as the environmental temperature change, long-term service aging and high-speed driving alternating stress field parameters during service are collected. A digital twin geometric model was constructed based on the geometric surface parameters of double-layer glass, and the structural regions of double-layer glass and polyvinyl butyral adhesive layer were divided. By inputting environmental temperature change and long-term service aging parameters into the viscoelastic refractive index spatiotemporal coupling iterative algorithm, the non-uniform spatial gradient distribution of refractive index of polyvinyl butyral adhesive layer under temperature change aging is obtained. The high-speed alternating stress field and the viscoelastic characteristic parameters of the polyvinyl butyral adhesive layer are input into the spatiotemporal coupling iterative algorithm of viscoelastic refractive index to calculate the transient change of molecular chain orientation and the local fluctuation of refractive index of the polyvinyl butyral adhesive layer under alternating stress. The non-uniform gradient distribution of refractive index space, transient changes in molecular chain orientation, and local fluctuations of refractive index are mapped region by region to the polyvinyl butyral adhesive layer structure region of the digital twin geometric model to obtain the real-time distribution of the adhesive layer refractive index. Import the geometric deformation data of the outer glass panel, integrate it into the digital twin geometric model, and update the surface parameters of the double-layer glass. Perform global ray tracing on the updated digital twin geometric model and output optical distortion and line-of-sight offset data for the main view area.

2. The method for simulating the optical performance of automotive windshields based on digital twins according to claim 1, characterized in that, The specific steps for solving the spatial non-uniform gradient distribution of refractive index are as follows: The time-varying temperature field boundary conditions were constructed using environmental temperature change parameters, and the aging time scale of the polyvinyl butyral adhesive layer was determined using long-term service aging parameters. The time-varying temperature field boundary conditions and aging time scale are input into the viscoelastic refractive index spatiotemporal coupling iterative algorithm. The pre-calibrated temperature-varying aging refractive index mapping relationship is called to calculate the refractive index offset point by point along each spatial coordinate of the polyvinyl butyral adhesive layer structure region. Spatial interpolation and smoothing are performed on the refractive index offsets of each spatial coordinate to generate a non-uniform spatial gradient distribution of refractive index.

3. The method for simulating the optical performance of automotive windshields based on digital twins according to claim 1, characterized in that, The specific steps for calculating the transient changes in molecular chain orientation and the local fluctuations in refractive index are as follows: The parameters of the high-speed alternating stress field are expanded according to the time series, and the stress tensor components of each node in the polyvinyl butyral adhesive layer structure region at each time are extracted. The stress tensor components of each node and the viscoelastic characteristic parameters of the polyvinyl butyral adhesive layer are input into the spatiotemporal coupling iterative algorithm of viscoelastic refractive index. The nonlinear constitutive relation of stress optical coefficient as a function of strain rate is called to solve the transient molecular chain orientation degree node by node. The anisotropic increment of refractive index at each node is calculated based on the transient molecular chain orientation. The principal component is taken as the local fluctuation of refractive index at that node. The transient change of molecular chain orientation and the local fluctuation of refractive index at each node at each time point are summarized.

4. The method for simulating the optical performance of automotive windshields based on digital twins according to claim 3, characterized in that, The specific steps for invoking nonlinear constitutive relations are as follows: Experimental data on the stress optical coefficient of polyvinyl butyral material under different strain rate conditions were obtained. The experimental data were nonlinearly fitted using the logarithm of strain rate as the independent variable and the stress optical coefficient as the dependent variable. The fitted polynomial or piecewise function is used as the nonlinear constitutive relation of the stress optical coefficient as a function of strain rate.

5. The method for simulating the optical performance of automotive windshields based on digital twins according to claim 1, characterized in that, The specific steps to obtain the real-time refractive index distribution of the adhesive layer are as follows: The polyvinyl butyral adhesive layer structure region is divided into grid cells, and the spatial index coordinates of each grid cell are established. Using spatial index coordinates as the mapping reference, the non-uniform spatial gradient distribution of refractive index is used as the refractive index reference value for each grid cell and filled into each grid cell. The anisotropic refractive index increment, which is associated with the transient change in molecular chain orientation of the corresponding spatial coordinates of the grid cell, is superimposed on the refractive index reference value. Then, the local fluctuation of refractive index is superimposed, and the real-time spatial distribution of refractive index is calculated on a grid cell-by-grid basis.

6. The method for simulating the optical performance of automotive windshields based on digital twins according to claim 5, characterized in that, The specific steps for superimposing the anisotropic increment of refractive index and the local fluctuation of refractive index are as follows: Relaxation time spectrum extracted from viscoelastic characteristic parameters of polyvinyl butyral adhesive layer; The local refractive index fluctuations and the relaxation time spectrum are input into the viscoelastic refractive index spatiotemporal coupling iterative algorithm to perform time-domain phase correction on the local refractive index fluctuations. The local fluctuations in refractive index after phase correction in the time domain are superimposed on the refractive index reference value and the refractive index anisotropy increment of the corresponding grid cell.

7. The method for simulating the optical performance of automotive windshields based on digital twins according to claim 6, characterized in that, The specific steps for extracting the relaxation time spectrum are as follows: Dynamic mechanical analysis frequency scanning tests were performed on polyvinyl butyral materials to obtain test data on the changes in storage modulus and loss modulus with frequency. The relaxation time spectrum of polyvinyl butyral material was obtained by fitting the test data with a generalized Maxwell model.

8. The method for simulating the optical performance of automotive windshields based on digital twins according to claim 1, characterized in that, The specific steps for updating the double-glazed surface parameters are as follows: Collect measured point cloud data or finite element deformation simulation data of the outer glass panel of the automotive windshield in the actual vehicle assembly state, as the geometric deformation data of the outer glass panel. Using the initial node coordinates of the double-glass structure region in the digital twin geometric model as a reference, the geometric deformation data of the outer glass panel is decomposed into displacement vectors of each node; Apply displacement vectors to update the surface parameters of the double-layer glass on a node-by-node basis.

9. The method for simulating the optical performance of automotive windshields based on digital twins according to claim 1, characterized in that, The specific steps for outputting the optical distortion and line-of-sight offset data of the main field of view are as follows: Starting from the driver's eye position, a tracing ray beam is emitted towards the incident side of the updated digital twin geometry model; The tracing beam sequentially passes through the double-glass structure region after updating the surface parameters and the polyvinyl butyral adhesive layer structure region that completes the real-time refractive index distribution mapping, recording the position coordinates and direction vectors of each ray on the emission side. The emission position coordinates and direction vectors of each ray are compared with the emission position coordinates and direction vectors of the theoretical interference-free reference ray. The optical distortion and line-of-sight offset data of the main field of view are calculated and output.

10. A digital twin-based simulation system for the optical performance of automotive windshields, employing the digital twin-based simulation method for the optical performance of automotive windshields as described in any one of claims 1-9, characterized in that, include: The parameter acquisition module is used to collect the geometric surface parameters of the double-layer glass of the automotive windshield, the material and viscoelastic characteristics of the interlayer polyvinyl butyral adhesive layer, as well as the environmental temperature change, long-term service aging and high-speed driving alternating stress field parameters corresponding to the service process. The geometric modeling module is used to construct a digital twin geometric model based on the geometric surface parameters of double-layer glass, and to divide the structural regions of double-layer glass and polyvinyl butyral adhesive layer. The viscoelastic refractive index calculation module is used to call the viscoelastic refractive index spatiotemporal coupling iterative algorithm to calculate the non-uniform gradient distribution of refractive index space, transient changes in molecular chain orientation, and local fluctuations in refractive index. The mapping and deformation fusion module is used to map the non-uniform gradient distribution of refractive index space, the transient change of molecular chain orientation and the local fluctuation of refractive index to the polyvinyl butyral adhesive layer structure region region by region, and import the geometric deformation data of the outer glass plate to update the double glass surface parameters. The ray tracing output module is used to perform global ray tracing calculations on the updated digital twin geometric model and output optical distortion and line-of-sight offset data for the main view area.

Images