A method and system for analyzing the deformation of an automobile chassis

Abstract

This application belongs to the field of new energy vehicles, specifically disclosing a method and system for analyzing the deformation of an automobile chassis. The method includes the following steps: S1. Establishing a three-dimensional finite element model of the automobile chassis and obtaining the strain components of the detection points; S2. Determining whether the detection area corresponding to the detection point is a potential deformation area; S3. Obtaining the strain components of the Gaussian integral point corresponding to the detection point; S4. Obtaining the displacement field of the three-dimensional finite element model; S5. Obtaining the geometric feature parameters corresponding to the potential deformation area, and based on the geometric feature parameters, determining whether deformation has occurred in the potential deformation area and the type of deformation. This application, by obtaining the detection points in the detection area, obtaining the strain components of the detection points, and analyzing them, can determine whether deformation exists in the detection area and the type of deformation; thus, by arranging fiber optic sensors in the automobile chassis area, deformation early warning can be provided, making it easier to find the cause of deformation.

Description

Technical Field

[0001] This application pertains to the field of new energy vehicles, and more specifically, relates to a method and system for analyzing the deformation of an automobile chassis. Background Technology

[0002] New energy vehicles have become an essential mode of transportation for many people's daily lives, travel, and business. If the chassis is impacted and deformed during driving, the battery pack located in the chassis will also be affected, potentially leading to damage or displacement, thus compromising driving safety. Therefore, it is necessary to provide timely warnings to the driver when chassis deformation occurs.

[0003] Patent application CN113060068A discloses an electric vehicle chassis collision protection and alarm system that uses a micro-bent fiber optic sensor to acquire optical signals to detect whether the chassis and battery box have deformed. However, it has the following drawbacks and shortcomings: the light transmittance of the micro-bent fiber optic used in this technology changes due to pressure variations, thus converting it into an electrical signal containing deformation information; however, temperature fluctuations also affect the fiber transmittance, interfering with the reading of deformation values. Although a sleeve-type micro-bent fiber optic sensor can be used to isolate the influence of temperature fluctuations, the vehicle chassis is located in an environment with poor heat dissipation and is very close to ground radiation sources, resulting in high temperature fluctuations. The temperature variable cannot be simply ignored using a sleeve-type fiber optic sensor; a method should be found to distinguish or compensate for the influence of temperature on deformation values. Furthermore, the rate of change of the electrical signal converted by the photoelectric converter changes abruptly with the dynamic deformation of the vehicle. Different levels of warning are issued by comparing this abrupt change with the threshold of the warning system. However, this kind of early warning system can only reflect the deformation range at the location of the deformation of the car chassis. It does not convert the electrical signal containing strain information into a deformation signal, output intuitive deformation value or deformation image, nor does it distinguish the type of deformation. Therefore, it is impossible to find the cause of the deformation.

[0004] From an application perspective, patent application CN113060068A primarily focuses on early warning of deformation in the chassis of electric vehicles. Since the battery pack is the most crucial structure in the entire electric vehicle chassis, the focus of deformation warning is on the vicinity of the battery pack installation area. However, it suffers from drawbacks such as insufficient monitoring scope and inapplicability to chassis of other types of vehicles besides electric vehicles. While the monitoring technology mentioned in patent CN113060068A focuses on the vicinity of the battery pack installation area, other important locations on the chassis, such as suspension mounting points and longitudinal beams, also require key monitoring for the specific type of vehicle used. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this application is to perform deformation analysis on the automobile chassis in real time by acquiring the strain components of different regions of the chassis.

[0006] To achieve the above objectives, in a first aspect, this application provides a method for analyzing the deformation of an automobile chassis, comprising the following steps: S1. Divide the vehicle chassis into multiple testing areas, and use the geometric center of each testing area as a testing point to obtain the strain components at the testing points; the strain components include axial strain ε. xx Transverse shear strain γ xy and torsional shear strain γ θ ; S2. Based on the detection point and the corresponding strain component, determine whether the detection area corresponding to the detection point is a possible deformation area, and at the same time establish a three-dimensional finite element model of the automobile chassis; S3. Using the three-dimensional finite element model, obtain the strain components of the Gaussian integral point corresponding to the detection point; S4. Obtain the displacement field corresponding to the Gaussian integration point based on the strain components of the Gaussian integration point. U ; S5. Based on the displacement field of the Gaussian integral point corresponding to the possible deformation region, obtain the geometric feature parameters corresponding to the possible deformation region, and based on the geometric feature parameters, determine whether the possible deformation region has undergone deformation and the type of deformation.

[0007] Preferably, in step S1, the strain components are obtained by: obtaining the strain value of the detection point and calculating and classifying it into strain components.

[0008] Preferably, in step S2, the method for determining whether the detection area corresponding to the detection point is a possible deformation area is as follows: determining the axial strain ε corresponding to the detection point. xx Transverse shear strain γ xy and torsional shear strain γ θ Whether the thresholds exceed the first, second, and third thresholds simultaneously.

[0009] Preferably, step S3 includes the following sub-steps: S31. In the three-dimensional finite element model, for each component of the stress at each detection point, a variation function model between detection combinations is established; the detection combination includes the detection point and other detection points, as well as the detection point and the corresponding Gaussian integral point in the three-dimensional finite element model; S32. Based on the variogram model, establish the Kriging equations to obtain the strain components at the Gaussian integral point.

[0010] As a further preferred embodiment, in step S31, a variogram model is fitted according to the parts of the vehicle chassis; wherein, the battery pack part is fitted with a Gaussian model, the longitudinal beam part is fitted with an exponential model, the suspension part is fitted with a spherical model, the welded joint part is fitted with a linear model, the large-area bottom plate part is fitted with an exponential model, and the periodic structure part is fitted with a hole effect model.

[0011] Preferably, in step S4, the displacement fields are established respectively. U The displacement field is obtained by using the equilibrium iterative equations of the strain components. U .

[0012] Preferably, in step S5, the deformation type is bending deformation, denting deformation, torsional deformation, or breakage and cracking.

[0013] As a further preferred embodiment, in step S5, the geometric characteristic parameters used to determine bending deformation are the maximum principal curvature, displacement gradient norm, and shear strain; the geometric characteristic parameters used to determine indentation deformation are Gaussian curvature, biaxial compressive strain, and displacement field Laplace; the geometric characteristic parameters used to determine torsional deformation are torsional angle gradient, shear strain to normal strain ratio, and cross-sectional torsional angle; and the geometric characteristic parameters used to determine breakage cracks are displacement gradient Frobenius norm, strain eigenvalue ratio, and displacement step.

[0014] Preferably, after step S5, step S6 is further included: visually rendering the possible deformable regions that have undergone deformation according to the type of deformation of the possible deformable regions.

[0015] Secondly, this application provides a system for performing deformation analysis using the above-described method for automobile chassis deformation analysis.

[0016] Preferably, the system includes a strain component acquisition module, a first deformation judgment module, a three-dimensional finite element module, a displacement field module, and a second deformation judgment module; The three-dimensional finite element module is used to establish a three-dimensional finite element model of the automobile chassis; The strain component acquisition module is used to acquire the strain components at the detection point; The first deformation judgment module is used to determine whether the area corresponding to the detection point is a possible deformation area based on the strain component; The displacement field module is used to obtain the displacement field of the three-dimensional finite element model based on the strain components of the three-dimensional finite element model and the detection points. The second deformation determination module is used to determine whether the possible deformation region has undergone deformation and the type of deformation based on the displacement field corresponding to the possible deformation region.

[0017] As a further preferred embodiment, the system further includes a visualization module; the visualization module is used to perform visual rendering of the deformable region based on the type of deformation of the deformable region.

[0018] Overall, the technical solutions conceived in this application have the following beneficial effects compared with the prior art: 1. This application can determine whether deformation exists in the detection area and the type of deformation by obtaining the detection points of the detection points and analyzing them; thus, deformation early warning can be performed by arranging fiber optic sensors in the vehicle chassis area, making it easier to find the cause of deformation. 2. When a certain position or area of ​​the car chassis deforms, the strain component at the detection point will shift in frequency. By analyzing and processing this frequency shift, the deformation signal can be obtained, and the type of deformation can be distinguished, including bending deformation, torsional deformation, dent deformation, and crack damage. 3. The preferred method is to establish a set of Kriging equations to map the strain components of the detection points to the corresponding Gaussian integral points in the three-dimensional finite element model, and calculate the displacement field data; by analyzing the geometric characteristic parameters of the displacement field, the deformation type of the automobile chassis can be identified and distinguished. 4. The method and system for analyzing vehicle chassis deformation disclosed in this application can meet the chassis deformation monitoring needs of various types of vehicles, including electric vehicles, with different structures. By defining the finite element mesh and CAD 3D model of the vehicle chassis, the data detected by the sensors can be adaptively mapped onto the corresponding model to form the required deformation image, which is convenient for users to view. After obtaining the possible deformation areas where deformation occurs, a visualized dynamic vehicle chassis deformation image can be obtained by using different color spectrum rendering and dynamics, which provides convenience for users to find the cause of deformation and facilitates users to repair and avoid road conditions that cause chassis deformation. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the method for analyzing the deformation of an automobile chassis provided in this application; Figure 2 This is a schematic diagram of the system for analyzing the deformation of the automobile chassis provided in this application; Figure 3 This is a top view of a simplified model of an automobile chassis used in the embodiments of this application; Figure 4 This is a schematic diagram of an automobile chassis divided into regions according to the location of components, according to an embodiment of this application; Figure 5 This is a visual rendering of the bending deformation of an embodiment of this application; Figure 6This is a visual rendering of the concave deformation in an embodiment of this application; Figure 7 This is a visual rendering diagram of the distortion and deformation of an embodiment of this application; Figure 8 This is a visual rendering of the cracks and damage in an embodiment of this application; Figure 9 This is a two-dimensional contour map of the deformation of a car chassis according to an embodiment of this application; Figure 10 This is a three-dimensional deformation surface diagram of an automobile chassis according to an embodiment of this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0021] To achieve the above objectives, firstly, this application provides a method for analyzing the deformation of an automobile chassis, such as... Figure 1 As shown, it includes the following steps: S1. Divide the car chassis into multiple inspection areas and establish a three-dimensional finite element model of the car chassis; in some embodiments, CAD model import and automatic mesh generation method are used, the main structure is generated using second-order tetrahedral mesh, and the key structure is generated using hexahedral mesh, then the material properties are defined, including elastic modulus, Poisson's ratio and coefficient of thermal expansion; finally, the boundary conditions and connection relationships are determined to complete the creation of the three-dimensional finite element model; Using the geometric center of each detection area as the detection point, distributed fiber optic sensors are set up to detect and differentiate the strain components at the detection points, including axial strain ε. xx Transverse shear strain γ xy and torsional shear strain γ θ The strain components mentioned above are converted into electrical signals by a photoelectric converter in an OFDR (Optical Frequency Domain Reflectometry) system. S2. Based on the detection point and its corresponding strain component, determine whether the detection area corresponding to the detection point is a potential deformation region. The usual method for determining whether a detection area is a potential deformation region is to determine the axial strain ε corresponding to the detection point. xx Transverse shear strain γ xy and torsional shear strain γ θ Does it simultaneously exceed the corresponding threshold? Taking DP780 high-strength steel, the chassis material, as an example, ε xx γ xy γ θThe thresholds are 0.0008, 0.0003, and 0.0002, respectively; if no detection area is a possible deformation area, the subsequent steps can be omitted; S3. Obtain the strain components at the Gaussian integral points corresponding to the detection points using a three-dimensional finite element model; this includes the following sub-steps: S31. In the three-dimensional finite element model, for each stress component of each detection point, a variogram model between detection combinations is established; the detection combination includes the detection point and other detection points, as well as the detection point and the corresponding Gaussian integral point in the three-dimensional finite element model; in the process of establishing the variogram model, the variogram model is fitted according to the parts of the car chassis, as shown in Table 1; among them, the battery pack part is fitted with a Gaussian model, the longitudinal beam part is fitted with an exponential model, the suspension part is fitted with a spherical model, the welded joint part is fitted with a linear model, the large area bottom plate part is fitted with an exponential model, and the periodic structure part is fitted with a hole effect model; Table 1. Selection of Fitting Models for Automobile Chassis Parts

[0022] S32. Based on the variogram model, establish the Kriging equations to obtain the strain components at the Gaussian integral point.

[0023] The Kriging equations are a method for unbiased estimation of strain components based on the spatial relationship established by the variogram. The equations are as follows:

[0024] λ The weighting coefficients to be determined are: μ For Lagrange multipliers, Γ For two known detection points p and q The matrix composed of the variogram values: , γ 0 is a vector composed of the variogram values ​​between the Gaussian integral points to be interpolated and the corresponding detection points. p and q These are the serial numbers used to mark the detection points. n This represents the total number of detection points. r p and r q These represent two known detection points. p and q Spatial location:

[0025] r0 represents the spatial location of the Gaussian integral point.

[0026] Solving for weight coefficients λ Finally, the three strain components of the Gaussian integral point corresponding to each detection point are obtained.

[0027] S4. Obtain the displacement field of the three-dimensional finite element model based on the strain components at the Gaussian integral points. U ; First solve the displacement field U The strain components in any direction can be represented by a set of equilibrium iterative equations:

[0028] superscript i For the first i iteration i =1~ m The number of equations m That is, the number of iterations. m ; Let be the tangent stiffness matrix for the current iteration. The external force vector is known and is represented by the strain components at the Gaussian integral point obtained in step S32. For internal force vectors, Let be the displacement increment to be determined in the i-th iteration.

[0029] Discretized equilibrium equations Both sides of the equation represent the internal and external force vectors, and the difference between the internal and external forces is the residual force. R i= The iterative solution ultimately needs to minimize the residual force. R i Approaching 0, that is To satisfy the equilibrium condition of the discretized equilibrium equation (i.e. (The left and right sides of the equation tend to be equal), where κ To achieve convergence tolerance, based on the displacement field U The accuracy requirements are set accordingly. When the accuracy requirement of the displacement field is... U When the convergence tolerance is ±0.5mm, it can be set to... κ =1×10 -6 N.

[0030] After obtaining the displacement field U After determining the strain components in the three directions, the displacement field can be further obtained. U .

[0031] S5. Based on the displacement field of the three-dimensional finite element model UThe geometric feature parameters corresponding to the possible deformation regions can be obtained respectively, and based on the geometric feature parameters, it can be determined whether the possible deformation regions have undergone deformation and the type of deformation.

[0032] The deformation type is bending deformation, denting deformation, torsional deformation, or fracture crack; the geometric characteristic parameters used to determine bending deformation are the maximum principal curvature, displacement gradient norm, and shear strain; in one embodiment, the criterion for determining bending deformation is: i. maximum principal curvature ≥ 0.02 mm -1 ii. Displacement gradient norm ≥ 0.15 mm / mm; iii. Shear strain ≤ 0.0008; The geometric characteristic parameters used to determine indentation deformation are Gaussian curvature, biaxial compressive strain, and displacement field Laplace; in one embodiment, the criterion for determining indentation deformation is: i. Gaussian curvature ≤ -10 -4 mm -2 ii. Biaxial compressive strain ≤ -0.001; iii. Displacement field Laplace ≥ 0.02 mm -1 ; The geometric characteristic parameters used to determine torsional deformation are the torsional angle gradient, the ratio of shear strain to normal strain, and the torsional angle of the cross section. In one embodiment, the criteria for determining torsional deformation are: i. torsional angle gradient ≥ 0.04 rad / mm; ii. shear strain to normal strain ratio ≥ 3.5; iii. cross section torsional angle ≥ 0.008 rad. The geometric characteristic parameters used to determine a broken crack are the Frobenius norm of the displacement gradient, the ratio of strain eigenvalues, and the displacement step. In one embodiment, the criterion for determining a broken crack is: i. the Frobenius norm of the displacement gradient ≥ 0.25 mm. -1 ii. Strain characteristic value ratio ≥ 50; iii. Displacement step ≥ 0.3 mm.

[0033] S6. Based on the type of deformation of the possible deformation region, perform visual rendering of the possible deformation region that has undergone deformation.

[0034] This application also provides a system for performing deformation analysis using the above-described method for automobile chassis deformation analysis, such as... Figure 2 As shown; the system includes a strain component acquisition module, a first deformation judgment module, a three-dimensional finite element module, a displacement field module, a second deformation judgment module, and a visualization module; The three-dimensional finite element module is used to establish a three-dimensional finite element model of the automobile chassis; The strain component acquisition module is used to acquire the strain components at the detection point; The first deformation judgment module is used to determine whether the area corresponding to the detection point is a possible deformation area based on the strain component; The displacement field module is used to obtain the displacement field of the three-dimensional finite element model based on the strain components of the three-dimensional finite element model and the detection points. The second deformation determination module is used to determine whether the possible deformation region has deformed and the type of deformation based on the displacement field corresponding to the possible deformation region; The visualization module is used to visualize and render the possible deformation areas based on the type of deformation.

[0035] The following are examples, which will be described below with reference to the accompanying drawings.

[0036] Figure 3 This is a top view of the simplified car chassis model used in this embodiment, which is a B-class sedan platform of a certain T brand electric vehicle. The areas marked with mesh lines in the figure are distributed fiber optic sensors. As shown in the figure, the entire chassis plane is covered with a mesh density of 1cm×1cm, and the mesh density is further increased at the battery pack outline edge, the center lines of the front and rear longitudinal beams, and the four suspension tower tops (because they are more likely to deform than other areas). The density of the increased density area is 0.5cm×0.5cm. The center of the mesh is the setting point of the distributed fiber optic sensor, i.e., the detection point.

[0037] The 3D finite element module uses CAD model import and automatic mesh generation to establish a high-precision 3D finite element model of the automobile chassis. The main structure is meshed using second-order tetrahedral meshes, while key structures are meshed using hexahedral meshes. Material properties are then defined, including elastic modulus, Poisson's ratio, and coefficient of thermal expansion. Finally, boundary conditions and connection relationships are determined to create a 3D finite element model with a displacement accuracy of ±0.5mm. In this embodiment, the material properties are high-strength steel DP780, with an elastic modulus of 210 GPa, a Poisson's ratio of 0.3, and a coefficient of thermal expansion of 1.2 × 10⁻⁶. -5 ℃ -1 The boundary conditions are as follows: fixed constraints (fixed constraints are applied at the bolt hole positions connecting the chassis to the body or subframe, restricting all degrees of freedom), displacement constraints (at the suspension mounting points, corresponding displacement constraints are set according to the actual motion relationship, such as restricting Z-direction displacement to simulate suspension support), symmetry boundary conditions (for the symmetrical chassis structure in this embodiment, symmetry constraints are applied on the symmetry plane to simplify the three-dimensional finite element model and improve computational efficiency), and load boundaries (external loads are applied according to actual working conditions, such as gravity, road reaction force, etc., to simulate the stress state of the chassis during operation). The connection relationship is a rigid connection with the subframe through 4 M12 bolts.

[0038] S1. The strain component acquisition module detects and calculates strain components through multiple detection points of a distributed fiber optic sensor, dividing the strain components into strain components; the strain components include axial strain ε. xx Transverse shear strain γ xy and torsional shear strain γ θ The signal is then converted into an electrical signal by a photoelectric converter in an OFDR (Optical Frequency Domain Reflectometry) system. S2. Here, the concept of a strain anomaly point is introduced. When the strain component detection value of the sensor at a certain location on the chassis is abnormal, this detection point is called a strain anomaly point. For the high-strength steel DP780 chassis material in this embodiment, when any three strain components ε xx γ xy γ θ When the detected values ​​are greater than 0.0008, 0.0003 and 0.0002 respectively, the first deformation judgment module identifies the detection point as a strain anomaly point and the area corresponding to the detection point as a possible deformation area.

[0039] In subsequent step S4, a continuous displacement field is established for the entire chassis based on a three-dimensional finite element model. After analyzing the higher-order geometric characteristics of the displacement field in step S5, the actual deformation areas can be found throughout the chassis. The geometric center of the deformation area may correspond to the location of the strain anomaly point. However, some strain anomaly points may not show the four types of deformation in S5 after calculation. Therefore, it is assumed that the areas corresponding to these strain anomaly points have not actually undergone deformation.

[0040] S3. In the previously established three-dimensional finite element model, each detection point has a corresponding Gaussian integral point. Kriging interpolation is used to accurately map the detection points to the Gaussian integral points of the three-dimensional finite element model, ensuring that the data can drive the update of the three-dimensional finite element model. Specifically, in the three-dimensional finite element model, for each strain component of each detection point, a variogram model is established between detection points and between the detection point and its corresponding Gaussian integral point, and the strain components of the Gaussian integral point are obtained. This includes: S31. By calculating the spatial distance vector and strain difference between all detection combinations, a correspondence between strain and model spatial position is established, and an experimental variogram point set is obtained. A detection combination refers to the combination of a detection point with any other detection point, or the combination of a detection point with the corresponding Gaussian integral point. Therefore, there are two types of variograms, which respectively constitute the matrix on the left side and the vector on the right side of the Kriging equation system established in step S32. like Figure 4As shown, the car chassis is divided into regions based on component locations, including sub-domains such as the battery pack, longitudinal beams, and suspension. Zonal Kriging is then performed on each sub-domain, and the most suitable fitting model is selected for the experimental points to obtain more accurate mapping results. Specifically, the battery pack is fitted using a Gaussian model, the longitudinal beams using an exponential model, the suspension using a spherical model, the welded joint area using a linear model, large-area floor plates using an exponential model, and periodic structural areas such as corrugated skid plates using a hole effect model. Taking a test point in a battery pack area as an example, a Gaussian model is selected to fit the experimental points. A variogram is independently established for the strain component of each test point. The variogram refers to the function applied to each strain component (axial strain ε). xx Transverse shear strain γ xy Torsional shear strain γ θ Establish variogram models γ( ) independently. h ), defined as:

[0041] That is, the variance of the strain components of the detection combination. Half of, of which, It is in space r The strain component value at any detection point. h If the spatial distance vector between combinations is detected, then in the formula... This refers to the strain difference of the detection assembly. Then, based on the characteristics of the battery pack region, a Gaussian model is selected to fit the variogram function:

[0042] Where c0 is the nugget constant, reflecting the measurement error of the nugget effect, which is taken as 0.001 in this embodiment. 2 c is the sill value, which reflects the intensity of spatial variability. In this embodiment, it is taken as 0.000005. a is the range, which is taken as 0.4m in this embodiment. e is the natural constant.

[0043] The formula is:

[0044] (with axial strain ε) xx For example, h (Unit: meters) S32. Using the fitted variogram model, for each Gaussian integration point of the three-dimensional finite element model, a Kriging equation system is established to solve for the corresponding unknown strain value. This Kriging equation system is used to predict the deformation at the Gaussian integration point. This is achieved by solving the Kriging weights to obtain the strain components at each Gaussian integration point. The Kriging equation system is a method for unbiased estimation of strain components based on the spatial relationship established by the variogram function. The equation system is as follows:

[0045] λ The weighting coefficients to be determined are: μ For Lagrange multipliers, Γ For two known detection points p and q The matrix composed of the variogram values: , γ 0 is a vector composed of the variogram values ​​between the Gaussian integral points to be interpolated and the corresponding detection points. p and q These are the serial numbers used to mark the detection points. n This represents the total number of detection points. r p and r q These represent two known detection points. p and q Spatial location:

[0046] r 0 represents the spatial location of the Gaussian integration point. The subsequent steps involve solving the weighting coefficients in the Kriging equations. λ The strain component values ​​at discrete points can be interpolated onto the Gaussian integral points of the finite element model to form a continuous strain field.

[0047] Solving for weight coefficients λ Finally, the strain component of the Gaussian integral point corresponding to each detection point is obtained, and this strain component also has values ​​in three different directions.

[0048] S4. The displacement field module obtains the displacement field of the three-dimensional finite element model based on the variogram model. U The detailed steps are as follows: After obtaining the strain components at each Gaussian integration point, the deformation value is calculated. A new objective function is introduced, and the conjugate gradient method is used to optimize the solution of the nonlinear finite element problem. Substituting the strain components into the solution yields a displacement field that approximates the actual deformation. UThe accuracy of this displacement field depends on the number of equations. By decomposing the displacement field into strain components in three directions, the exact deformation values ​​can be obtained and output. The strain components in any one direction can be expressed as:

[0049] superscript i For the first i iteration i =1~ m The number of equations m That is, the number of iterations. m ; Let be the tangent stiffness matrix for the current iteration. The external force vector is known and can be represented by the strain components obtained in step S32. For internal force vectors, Let be the displacement increment to be determined in the i-th iteration.

[0050] This set of equations is for the displacement field. U The equilibrium iterative equations for a certain strain component are used to solve the discretized equilibrium equations. Both sides of the equation represent the internal and external force vectors, and the difference between the internal and external forces is the residual force. R i ,Right now R i= The iterative solution ultimately needs to minimize the residual force. R i The equilibrium value tends to 0 to satisfy the equilibrium condition of the discretized equilibrium equation (i.e., ... (When both sides of the equation tend to be equal); the most effective method for solving this type of nonlinear algebraic problem is the equilibrium iteration method (Newton-Raphson method), the core idea of ​​which is to linearize the solution near the estimated value of the current solution, solve the linearized equation to obtain the increment of the solution, and iterate until convergence.

[0051] Initialization is required before iterative solution, including setting the initial displacement vector. (In this embodiment, the initial displacement vector is a zero vector 0), maximum number of iterations m (In this embodiment, it is set to 1.56 million) and the tangent stiffness matrix is ​​calculated for each iteration. The tangent stiffness matrix is ​​the internal force vector. For the displacement vector The derivative; before solving, the displacement after each iteration is expressed as:

[0052] This process is repeated iteratively until convergence, which yields the desired displacement field. U's A certain strain component can cause residual forceR i The equilibrium equations tend to zero. During the iteration process, whether the solution to the system's equilibrium equations converges depends primarily on the residual force vector. R i Whether the norm is lower than the preset tolerance. Specifically, the convergence condition is set as follows:

[0053] in κ To minimize convergence tolerance, the displacement field is adjusted according to this embodiment. U Given the accuracy requirement of ±0.5mm and the material properties of the chassis, all convergence tolerances are set to... κ =1×10 -6 N.

[0054] In this embodiment, all equations converge after approximately 1.56 million iterations (m≈156×10). 4 Thus, the displacement field of the three-dimensional finite element model is obtained. U The strain components in the three directions are then used to further obtain the displacement field. U .

[0055] S5. Based on the fitted displacement field analysis, the high-order geometric features and local deformation features of the possible deformation region are used to determine the deformation type. The methods for distinguishing deformation types are integrated into a second deformation judgment module. The displacement gradient tensor is calculated and obtained. Curvature analysis, Gaussian curvature analysis, torsion angle analysis, and unique discontinuity (singularity) analysis are used to determine whether the chassis exhibits bending deformation, denting deformation, torsional deformation, or damage / cracks. Curvature analysis is used to obtain parameters for bending deformation, including the maximum principal curvature; Gaussian curvature analysis is used to obtain parameters for denting deformation, including Gaussian curvature and the displacement field Laplace; torsion angle analysis is used to obtain parameters for torsional deformation, including the cross-sectional torsion angle and torsion angle gradient; and unique discontinuity (singularity) is used to obtain parameters corresponding to damage or cracks, including the displacement step. Identification layer: The calculated displacement field... U High-order geometric feature analysis is performed to obtain all the parameters required for deformation type criteria; correspondingly, displacement field gradient tensor calculation is used to obtain the above-mentioned parameters, thereby identifying deformation features based on these parameters.

[0056] Bending deformation: The identification criteria are that the following conditions must be met simultaneously: i. Maximum principal curvature ≥ 0.02 mm -1 ii. Displacement gradient norm ≥ 0.15 mm / mm; iii. Shear strain ≤ 0.0008, such as Figure 8 The figure shows a deformation point of bending deformation, whose maximum principal curvature value is calculated to be 0.025 mm. -1 The calculated shear strain is 0.0005, which satisfies the bending deformation criterion. Concave deformation: The identification criteria are that it simultaneously meets the following conditions: i. Gaussian curvature ≤ -10 -4 mm -2 ii. Biaxial compressive strain ≤ -0.001; iii. Displacement field Laplace ≥ 0.02 mm -1 ,like Figure 9 The image shows a deformation point of a concave deformation, whose Gaussian curvature is calculated to be -1.5 × 10⁻⁶. -4 mm -2 The biaxial compressive strain was calculated to be -0.0012, and the contour curve is a closed ellipse; the Laplace displacement field is 0.025 mm. -1 ; Satisfies the indentation deformation criterion; Torsional deformation: The identification criteria are that the following conditions must be met simultaneously: i. Torsional angle gradient ≥ 0.04 rad / mm; ii. Shear strain to normal strain ratio ≥ 3.5; iii. Cross-sectional torsional angle ≥ 0.008 rad. Figure 10 The figure shows a deformation point of torsional deformation, with a calculated torsional angle gradient of 0.05 rad / mm, a shear strain to normal strain ratio of 5.5, and a displacement field showing obvious cross-sectional rotation; thus satisfying the torsional deformation criterion. Damage and cracks: The identification criteria are that both of the following must be met simultaneously: i. Displacement gradient Frobenius norm ≥ 0.25 mm -1 ii. Strain eigenvalue ratio ≥ 50; iii. Displacement step ≥ 0.3 mm, a deformation point of failure and crack, where the axial strain is calculated to abruptly change from 0.001 to -0.005 within a 1 mm distance, the strain eigenvalue ratio reaches 60, and the displacement gradient Frobenius norm is calculated to be 0.35 mm. -1 The displacement step was calculated to be 0.4 mm, which meets the criteria for damage and cracking.

[0057] After obtaining the deformation type, the area corresponding to the Gaussian integral point of that deformation type can be used to determine the area where the final car chassis deformation occurs. If two adjacent deformation areas with the same deformation type exist, they are merged into one area by default. The deformation recognition module will label the areas that meet a certain type criterion, so this deformation area is composed of displacement field data and deformation type labels. Visualization in S6 can generate an observable image of the chassis deformation area.

[0058] S6. A visualization image is generated by combining displacement field data and deformation type. This process takes place in the visualization module. The visualization module's workflow involves creating a 3D CAD model of the car chassis, mapping the deformation data to the visualization model, using isosurfaces to delineate the deformation magnitude, and employing different colors for visual rendering for different deformation types. Specifically: (1) Bending deformation: such as Figure 5As shown, a red gradient color spectrum is used, and the dynamic effect is an animation of the color spectrum undulating along the axis of the component. Parallel contour lines are added to facilitate the reading of deformation values, and bidirectional arrows indicate the bending direction.

[0059] (2) Depression: such as Figure 6 As shown, a blue gradient chromatogram is used. The dynamic effect is that the chromatogram flows inward and contracts. Closed ring contour lines are added to indicate the deformation value, and arrows indicate the center of the depression and its depth.

[0060] (3) Twisting and deformation: such as Figure 7 As shown, a green gradient color scheme is used to render the twist angle, with gray representing no twisting. A diagonal line is introduced to indicate the degree of axis twist after the twist. The dynamic effect is an animation of the rotational twist.

[0061] (4) Damage and cracks: such as Figure 8 As shown, the damaged areas are marked with highlighted jagged lines, and the direction of crack propagation is indicated by arrows.

[0062] For areas exhibiting complex deformation, color mixing and geometric overlay are used for display. For example, the overlay of bending and depression is presented as a purple mixture of red and blue, while the overlay of damage and depression is presented as a coexistence of contour lines and jagged lines. Other areas without obvious deformation are displayed in the same color as the background. The final generated 3D display image file is in MP4 format, allowing users to observe the dynamic process from deformation to expansion.

[0063] By operating the visualization software, the deformation pattern can be viewed. The deformation type is dent accompanied by slight bending, and the location is on the impact force transmission path of the suspension. After inspection and analysis, the cause of the deformation is that it was caused by a violent impact with a deep pothole on the road surface during driving.

[0064] The system supports historical data retrospective analysis. By comparing deformation patterns under different working conditions (full load / no load, flat road / off-road), it assesses the fatigue life of the chassis structure and recommends maintenance strategies.

[0065] Under off-road conditions, the strain amplitude Δε of the depression region (i.e., the average value of the strain tensor norm (the length of the vector) of the region, which is obtained by averaging the strain tensor norm under different road conditions) increased by about 220% compared to the flat road conditions, reaching 0.00272, while it was 0.00085 under the flat road conditions.

[0066] The strain amplitudes for both off-road and flat road conditions are obtained through actual load tests under these two conditions. In this embodiment, the flat road condition adopts a Class A road surface as specified in ISO 8608, and the off-road condition adopts a Class E road surface as specified in the standard. The road surface characteristics are a gravel road with a large amount of loose gravel. The acquisition of the strain amplitude essentially involves real-time acquisition of strain data from various detection points on the chassis using distributed fiber optic sensors, followed by integration of these strain data into a strain tensor (usually a 3×3 matrix describing the strain state at a point; it is the integration of all strain components at a point, which can be represented as...). The relationship between the component and the component is as follows:

[0067] The strain tensor norm, as the strain amplitude, is obtained by calculating the Frobenius norm of the tensor (i.e., the square root of the sum of the squares of the components), and represents the average strain intensity of the entire chassis area of ​​the vehicle.

[0068] Actual load test setup: Using a four-column vibration table or road spectrum test track, road surface roughness spectrum excitation conforming to specific standards (such as ISO 8608) is applied to the chassis to reproduce multi-dimensional dynamic impact loads to simulate off-road conditions. Quasi-static loads caused by self-weight (curb weight) and stable aerodynamic loads are applied to simulate flat road conditions. Finally, by configuring standard counterweights, the overall vehicle mass and its distribution are precisely changed to evaluate the impact of load size (full load or no load) on chassis deformation characteristics. The execution standards set above are as follows: (1) Simulate the vehicle state: "Applying quasi-static load caused by self-weight (curb mass) and stable aerodynamic load" is to simulate the full-load state of the actual vehicle on the road. This is based on the provisions of the automotive industry standard QC / T 1096-2018 that "the installation state of the test bench should be the same as the installation state of the actual vehicle (full load)"; (2) "Configure standard counterweights" refers to the common test method in the industry, that is, calculate the number and weight ratio of counterweights according to the curb mass and bearing area of ​​the vehicle to ensure that the load is evenly distributed; (3) "Apply road surface roughness spectrum excitation that conforms to specific standards (such as ISO 8608)" directly adopts the road surface classification method specified in the international standard ISO 8606:2016. The four-column vibration table test generates different levels of road surface spectrum according to this standard to simulate the actual road conditions. (4) Standards for road condition types: Flat road condition: corresponding to Class A (good road surface) and Class B (general road surface) as defined in ISO 8608 standard. This type of road surface spectrum can represent typical paved roads such as expressways and Class I highways; Off-road condition: corresponding to higher grade road surfaces (such as Class E and above) in ISO 8608 standard. Its typical characteristics include unpaved road surfaces such as soil, gravel, pebbles, and mud, as well as characteristic road surfaces such as stone roads and twisted roads.

[0069] The fatigue life of the chassis structure is improved by introducing fatigue damage degree.D To describe.

[0070] First, calculate the number of cycles required for fatigue failure of the material under a certain working condition based on the material's SN curve. :

[0071] Then, based on Miner's linear cumulative damage theory, the fatigue damage degree is calculated. D :

[0072] in In the first h The actual number of load cycles under various load conditions can be obtained from the vehicle management system and is a known value. This refers to the number of cycles required for the material to undergo fatigue failure under this operating condition.

[0073] The SN curve is a curve that describes the relationship between stress level and the number of cycles leading to fatigue failure. Let ρ be the stress amplitude, and ρ be the slope of the curve. For the DP780 high-strength steel chassis material in this embodiment, ρ=4 and the material constant C=10 are selected. 12 MPa 4 Assuming that flat road conditions account for 90% and off-road conditions account for 10%, substituting these values ​​into the formula, we can finally calculate the number of cycles that lead to failure. Finally, substituting the values ​​into the damage degree formula, the fatigue life of the chassis structure is obtained as the damage degree. D =0.15, with a remaining lifespan of approximately 350,000 kilometers. Therefore, the recommended maintenance strategy is to perform maintenance at the next maintenance cycle (approximately 10,000 kilometers later).

[0074] Based on the implementation method in the technical solution, we set up a rectangular area as a simplified chassis model of the car and conducted a simple simulation according to the experimental steps in the instruction manual. We randomly set the three deformations at different positions on the chassis and obtained the two-dimensional and three-dimensional simulation results, which are shown below. Figure 9 , Figure 10 The simulation results do not represent the final visualization interface effect, but they can clearly show the location and extent of the deformation of the chassis.

[0075] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for analyzing the deformation of an automobile chassis, characterized in that, Includes the following steps: S1. Divide the vehicle chassis into multiple inspection areas, and establish a three-dimensional finite element model of the vehicle chassis with the geometric center of each inspection area as the inspection point, and obtain the strain components of the inspection point; the strain components include axial strain, transverse shear strain and torsional shear strain; S2. Based on the detection point and the corresponding strain component, determine whether the detection area corresponding to the detection point is a possible deformation area; S3. Using the three-dimensional finite element model, obtain the strain components of the Gaussian integral point corresponding to the detection point; S4. Obtain the displacement field of the three-dimensional finite element model based on the strain components at the Gaussian integral point; S5. Based on the displacement field, obtain the geometric feature parameters corresponding to the possible deformation region, and based on the geometric feature parameters, determine whether the possible deformation region has undergone deformation and the type of deformation.

2. The method as described in claim 1, characterized in that, In step S1, the strain components are obtained by obtaining the strain value at the detection point and calculating and dividing it into strain components.

3. The method as claimed in claim 1, characterized in that, The method for determining whether the detection area corresponding to the detection point is a possible deformation area is as follows: determine whether the axial strain, transverse shear strain and torsional shear strain corresponding to the detection point exceed the first threshold, the second threshold and the third threshold at the same time.

4. The method as claimed in claim 1, characterized in that, Step S3 includes the following sub-steps: S31. In the three-dimensional finite element model, for each component of the stress at each detection point, a variation function model between detection combinations is established; the detection combination includes the detection point and other detection points, as well as the detection point and the corresponding Gaussian integral point in the three-dimensional finite element model; S32. Based on the variogram model, establish the Kriging equations to obtain the strain components at the Gaussian integral point.

5. The method as claimed in claim 4, characterized in that, In step S31, the variogram model is fitted according to the parts of the car chassis; the battery pack part is fitted with a Gaussian model, the longitudinal beam part is fitted with an exponential model, the suspension part is fitted with a spherical model, the welded joint part is fitted with a linear model, the large area bottom plate part is fitted with an exponential model, and the periodic structure part is fitted with a hole effect model.

6. The method as claimed in claim 1, characterized in that, In step S4, the displacement field is obtained by establishing a set of equilibrium iterative equations for the strain components of the displacement field.

7. The method as claimed in claim 1, characterized in that, In step S5, the deformation type is bending deformation, denting deformation, torsional deformation, or breakage and cracking.

8. The method as claimed in claim 7, characterized in that, In step S5, the geometric characteristic parameters used to determine bending deformation are the maximum principal curvature, displacement gradient norm, and shear strain; the geometric characteristic parameters used to determine indentation deformation are Gaussian curvature, biaxial compressive strain, and displacement field Laplace; the geometric characteristic parameters used to determine torsional deformation are torsional angle gradient, shear strain to normal strain ratio, and cross-sectional torsional angle; and the geometric characteristic parameters used to determine fracture cracks are displacement gradient Frobenius norm, strain eigenvalue ratio, and displacement step.

9. A system for performing deformation analysis using the method described in any one of claims 1-8.

10. The system as described in claim 9, characterized in that, It includes a strain component acquisition module, a first deformation judgment module, a three-dimensional finite element module, a displacement field module, and a second deformation judgment module; The three-dimensional finite element module is used to establish a three-dimensional finite element model of the automobile chassis; The strain component acquisition module is used to acquire the strain components at the detection point; The first deformation judgment module is used to determine whether the area corresponding to the detection point is a possible deformation area based on the strain component; The displacement field module is used to obtain the displacement field of the three-dimensional finite element model based on the strain components of the three-dimensional finite element model and the detection points. The second deformation determination module is used to determine whether the possible deformation region has undergone deformation and the type of deformation based on the displacement field corresponding to the possible deformation region.

Images