A method for testing fatigue strength reliability of a vehicle rearview mirror

Abstract

This invention relates to the field of fatigue testing technology for vehicle components, and more particularly to a method for testing the fatigue strength and reliability of a vehicle rearview mirror. The method includes: acquiring the basic characteristic parameters of the rearview mirror under test to construct a digital three-dimensional model, and generating a time-domain load spectrum for the installation points. An improved genetic algorithm is used to optimize the virtual strain monitoring point network based on the strain energy density distribution and the trend of principal stress direction changes. The time-domain load spectrum is applied to the model installation points for dynamic response analysis, and the equivalent stress and principal stress direction time history data are extracted. The cumulative fatigue damage value is calculated by combining the material's S-N curve, and multi-axis correction is performed by the frequency of principal stress direction changes. The corrected data is then integrated to evaluate the fatigue life and reliability of the rearview mirror. This method can optimize the layout of monitoring points, match the actual stress characteristics of the structure, reduce calculation deviations caused by multi-axis alternating loads, and improve the accuracy of fatigue strength and reliability assessment.

Description

Technical Field

[0001] This invention belongs to the field of fatigue testing technology for vehicle components, specifically a method for testing the fatigue strength reliability of a vehicle rearview mirror. Background Technology

[0002] Current vehicle rearview mirror fatigue strength testing employs a combination of physical bench testing and conventional finite element simulation. Virtual strain monitoring points are either uniformly distributed or selected based on manual experience, with stress amplitude being the sole criterion for point optimization. A conventional time-domain load spectrum is applied to the model's installation points, and linear dynamic response analysis is performed. Equivalent stress data is extracted and combined with the material's S / N curves to calculate uniaxial fatigue damage accumulation, thereby assessing the structure's fatigue life and reliability.

[0003] The virtual strain monitoring point layout did not take into account the structural strain energy density distribution and the changing patterns of principal stress directions. The point distribution did not match the actual stress characteristics, and data collection was missing in key stress areas, while monitoring was redundant in non-critical areas. When the rearview mirror was subjected to alternating loads from a real vehicle, the principal stress direction continuously and dynamically changed. The uniaxial fatigue calculation did not consider the cumulative effect of the principal stress direction changes on damage accumulation, resulting in deviations between the damage values ​​and the actual multiaxial failure state. Consequently, the accuracy of fatigue life and reliability assessments was low.

[0004] The virtual strain monitoring point network needs to be optimized by combining the strain energy density distribution and the trend of principal stress direction changes. Multi-axis correction of the cumulative fatigue damage value needs to be performed based on the frequency of principal stress direction changes to reduce the difference between simulation calculations and actual stress conditions. Summary of the Invention

[0005] This invention aims to solve at least one of the technical problems existing in the prior art; Therefore, this invention proposes a method for testing the fatigue strength reliability of vehicle rearview mirrors, comprising: The basic feature parameters of the rearview mirror of the vehicle under test are obtained, a digital three-dimensional model of the rearview mirror of the vehicle under test is constructed based on the basic feature parameters, and a time-domain load spectrum for the installation point of the rearview mirror of the vehicle under test is generated. An improved genetic algorithm is used to optimize the network of virtual strain monitoring points configured on the digital three-dimensional model of the rearview mirror of the vehicle under test. The improved genetic algorithm is optimized based on the variation trend of strain energy density distribution and principal stress direction. The time-domain load spectrum is applied to the mounting points of the digital three-dimensional model of the rearview mirror of the vehicle under test, driving the digital three-dimensional model to perform dynamic response analysis based on the virtual strain monitoring point network. Extract the equivalent stress time history data and principal stress direction time history data of the virtual strain monitoring point network during the dynamic response analysis process; Based on the equivalent stress time history data and the material SN curve of the rearview mirror of the vehicle under test, the cumulative fatigue damage value of each of the virtual strain monitoring point networks is calculated. Based on the principal stress direction time history data, the frequency of principal stress direction changes at key critical points is identified, and the cumulative fatigue damage value is corrected in a multi-axis manner. By integrating the cumulative fatigue damage values ​​after network correction of all virtual strain monitoring points, the overall fatigue life and reliability of the rearview mirror of the vehicle under test are evaluated.

[0006] Further, the process of acquiring the basic feature parameters of the rearview mirror of the vehicle under test, constructing a digital three-dimensional model of the rearview mirror based on the basic feature parameters, and generating a time-domain load spectrum for the mounting point of the rearview mirror includes: Obtain initial data on the geometric dimensions, material parameters, and connection structure of the rearview mirror of the vehicle under test; Based on the initial data of the geometric dimensions, material parameters and connection structure, a digital three-dimensional model of the rearview mirror of the vehicle under test is constructed. Based on the preset actual road spectrum excitation signal and vehicle vibration transfer function, a time-domain load spectrum is generated for the mounting point of the rearview mirror of the vehicle under test, specifically including: Vibration data of the target vehicle model on various typical road surfaces are collected, filtered and statistically analyzed, and then synthesized to obtain the preset actual road spectrum excitation signal. The actual road spectrum excitation signal includes vertical, longitudinal and lateral acceleration components. The vehicle vibration transfer function from the vehicle suspension hardpoint to the rearview mirror mounting base is established, and the vehicle vibration transfer function is obtained by experimental modal analysis and frequency response function fitting. Each acceleration component of the preset actual road spectrum excitation signal is input into the corresponding vehicle vibration transfer function; The frequency domain response of the triaxial force and torque at the rearview mirror mounting base is calculated by convolution operation. An inverse Fourier transform is performed on the frequency domain response of the three forces and torques to generate a time-domain load spectrum acting on the rearview mirror mounting base. The time-domain load spectrum includes force components and torque components in three directions that vary with time.

[0007] Furthermore, the virtual strain monitoring point network optimized on the digital three-dimensional model of the rearview mirror of the vehicle under test using an improved genetic algorithm includes: Initialize the genetic algorithm population, where each individual represents a spatial distribution scheme of a virtual strain monitoring point network on the surface of the digital three-dimensional model of the rearview mirror of the vehicle under test. Perform the following fitness assessment process for each individual: Based on the static unit load condition, finite element analysis was performed on the digital three-dimensional model of the rearview mirror of the vehicle under test to calculate the strain energy density of each node on the model surface. At the monitoring point location corresponding to the distribution scheme represented by the individual, extract the strain energy density value and the principal stress direction of the monitoring point; Based on the optimization criteria of the improved genetic algorithm, the fitness value of the spatial distribution scheme is calculated. The optimization criteria require that the monitoring points should cover areas with high strain energy density and be able to capture areas where the principal stress direction changes drastically. Based on fitness values, selection, crossover, and mutation operations are performed on individuals in the population to generate a new generation of population; Repeatedly perform fitness assessment and genetic operations until the preset number of generations or fitness value convergence is reached, and output the set of spatial coordinates of the virtual strain monitoring point network represented by the optimal individual.

[0008] Furthermore, the improved genetic algorithm is optimized based on the variation trend of strain energy density distribution and principal stress direction, and its working principle includes: During the fitness evaluation process, two optimization indices are calculated for each spatial distribution scheme: the first index is the average strain energy density at the selected monitoring point, and the second index is the standard deviation of the principal stress direction angle at the selected monitoring point. The average strain energy density and the standard deviation of the principal stress direction angle are weighted and summed, and the sum is used as the initial fitness of the spatial distribution scheme. A penalty term is introduced: if the distance between any two monitoring points is less than a preset minimum distance, a penalty value inversely proportional to the distance is subtracted from the initial fitness. The final fitness value is the value adjusted for the penalty term; During the crossover process of genetic manipulation, monitoring point groups located in different high strain energy density regions from high-fit individuals are preferentially exchanged. During the mutation process, the monitoring point is allowed to move in a small range with a certain probability toward the direction of the current maximum stress gradient in order to better track the changing trend of the principal stress direction.

[0009] Furthermore, the time-domain load spectrum is applied to the mounting points of the digital three-dimensional model of the rearview mirror of the vehicle under test, driving the digital three-dimensional model to perform dynamic response analysis based on the virtual strain monitoring point network, including: The force and torque components in the time-domain load spectrum are applied as nodal forces to the set of nodes simulating the rearview mirror mounting base in the digital three-dimensional model. Set the material damping coefficient and structural damping coefficient of the digital 3D model, and activate the geometric nonlinearity option to account for large deformation effects; The explicit dynamic integration method is used to solve the time integral of the digital three-dimensional model with the applied time-domain load spectrum. During the solution process, the stress tensor at each monitoring point in the virtual strain monitoring point network is recorded in real time as it changes over time. From the recorded stress tensor data, the equivalent stress time history data and the principal stress direction time history data of each monitoring point are separated and output.

[0010] Furthermore, the step of employing an explicit dynamic integration method to solve the time integral of the digitized three-dimensional model to which the time-domain load spectrum has been applied includes: The time step for explicit dynamic integration is set, which is determined based on the smallest finite element size in the digital three-dimensional model and the sound velocity in the material, to ensure the stability of the numerical solution. At the beginning of each time step, the nodal force vectors applied to the set of nodes of the simulated rearview mirror mounting base are calculated based on the force and torque components corresponding to the time-domain load spectrum. Based on the displacement, velocity, and acceleration states of all nodes in the digital 3D model at the current moment, and combined with the material damping coefficient and structural damping coefficient, calculate the internal force vector of all finite element elements. Based on the explicit central difference method, the acceleration vectors of all nodes at the next time step are solved using the nodal force vectors and the internal force vectors. Integrate the acceleration vector over time and update the velocity and displacement vectors of all nodes; Using the updated displacement vector as boundary conditions, the strain and stress states of all finite element elements are recalculated to complete the solution for one time step. The steps are repeated until the dynamic response solution for the entire time period covered by the time-domain load spectrum is completed, thus obtaining the dynamic response data of the digital three-dimensional model throughout the entire loading process.

[0011] Further, based on the equivalent stress time history data and the material SN curve of the rearview mirror of the vehicle under test, the cumulative fatigue damage value of each of the virtual strain monitoring point networks is calculated, including: Rainflow counting was performed on the equivalent stress time history data of each monitoring point to count all complete stress cycles, and the stress amplitude and average stress of each stress cycle were extracted. Using the Goodman formula or Gerber formula, a stress cycle containing average stress can be equivalently converted into a symmetrical cyclic stress amplitude with the same damage. Based on the equivalent stress amplitude, the number of cycles at which the material fails under the stress amplitude can be obtained by looking up a table or by interpolating a formula on the material SN curve of the rearview mirror of the vehicle under test. According to the linear cumulative damage criterion, the damage caused by stress cycling is calculated as the reciprocal of the ratio of the actual number of cycles to the number of failure cycles. By summing up the damage caused by all stress cycles in the equivalent stress time history data, the preliminary cumulative fatigue damage value of the corresponding monitoring point is obtained.

[0012] Furthermore, based on the principal stress direction time history data, the frequency of principal stress direction changes at key critical points is identified, and the cumulative fatigue damage value is corrected using multi-axis methods, including: From all monitoring points, those with a preliminary cumulative fatigue damage value higher than a preset threshold are selected and defined as critical danger points. Extract the time history data of the principal stress direction for each critical hazard point, calculate the sequence of principal stress direction angle changes over time, and record it as the principal stress direction angle sequence; The principal stress direction angle sequence is statistically analyzed, and the number of times the principal stress direction angle changes by more than a specific angle per unit time is calculated as the frequency of principal stress direction changes. Based on the frequency of principal stress direction changes, the corresponding multiaxial fatigue correction coefficient is obtained from a preset lookup table. The multiaxial fatigue correction coefficient is positively correlated with the frequency of direction changes. Multiply the initial cumulative fatigue damage value of the critical hazard point by the corresponding multiaxial fatigue correction coefficient to obtain the corrected cumulative fatigue damage value of the critical hazard point. For non-critical hazard points, the corrected cumulative fatigue damage value is equal to the initial cumulative fatigue damage value.

[0013] Furthermore, the evaluation of the overall fatigue life and reliability of the rearview mirror of the vehicle under test by integrating the cumulative fatigue damage value after network correction of all virtual strain monitoring points includes: Find the maximum value among the corrected cumulative fatigue damage values ​​of all monitoring points and record it as the maximum point damage value; The reciprocal of the maximum point damage value is used as the predicted characteristic lifetime of the rearview mirror of the vehicle under test under the loading conditions represented by the time-domain load spectrum. The number of monitoring points whose cumulative fatigue damage value, after statistical correction, exceeds the expected damage threshold of the design life, and their spatial distribution; Based on the predicted characteristic lifespan and the distribution of monitoring points exceeding the damage threshold, and according to a predefined reliability evaluation matrix, the fatigue reliability level of the rearview mirror of the vehicle under test is determined.

[0014] Furthermore, the method also includes testing and verification and model calibration steps: Select a sample produced in the same batch as the rearview mirror of the vehicle under test and install it on the vibration test bench; Actual strain gauges are attached to the connection between the back of the sample's mirror shell and the support to form a physical strain monitoring network. The time-domain load spectrum was reproduced on the vibration test bench, and the sample was subjected to accelerated fatigue testing. The physical strain monitoring network is used to collect real-time strain time history data at key locations during the experiment. The actual strain time history data is compared with the strain time history data obtained from the corresponding virtual strain monitoring point network in dynamic response analysis; If the comparison error exceeds the allowable range, the boundary conditions or material parameters of the digital three-dimensional model are corrected based on the actual data, and some analysis steps are repeated until the simulation results and experimental results meet the consistency requirements.

[0015] Compared with the prior art, the beneficial effects of the present invention are: Based on the strain energy density distribution and the trend of principal stress direction changes, optimization objectives are set, and the spatial location and density of the virtual strain monitoring point network are iteratively adjusted through an improved genetic algorithm. Monitoring points can adaptively concentrate in areas of high strain energy density and areas with abrupt changes in principal stress direction, reducing the redundancy of monitoring points in low-stress areas. The distribution pattern of monitoring points matches the strain distribution and stress direction characteristics within the structure, avoiding the problem of missing data in key areas that exists with manual or uniform point placement. During dynamic response analysis, each monitoring point can stably acquire continuous stress change data at its corresponding location, and the stress acquisition range highly matches the actual stress area of ​​the structure, improving the targeting and completeness of data acquisition.

[0016] Time history data of principal stress directions at key critical points are extracted, and the cumulative uniaxial fatigue damage value is corrected for multiaxial damage by analyzing the frequency of principal stress direction changes. This correction process reflects the superimposed effect of principal stress vector variations under alternating loads on material damage, aligning with the multiaxial fatigue damage formation mechanism. It mitigates the calculation bias caused by the assumption of constant stress direction in uniaxial fatigue calculations, ensuring that the cumulative fatigue damage value remains consistent with the actual material degradation process under multiaxial stress. The corrected damage data objectively reflects the damage distribution within the structure under complex loads. After integrating damage values ​​from all monitoring points, the fatigue life and reliability calculation results closely match the complex stress conditions of real vehicles, reducing the discrepancy between simulation evaluation and actual failure states, and improving the accuracy of fatigue strength reliability assessment.

[0017] The above and other objects, advantages and features of the present invention will become more apparent to those skilled in the art from the following detailed description of specific embodiments of the invention in conjunction with the accompanying drawings. Attached Figure Description

[0018] The following sections will describe some specific embodiments of the invention in detail by way of example and not limitation, with reference to the accompanying drawings. The same reference numerals in the drawings denote the same or similar parts or portions. Those skilled in the art should understand that these drawings are not necessarily drawn to scale. In the drawings: Figure 1 This is a flowchart illustrating the steps of a vehicle rearview mirror fatigue strength reliability test method according to the present invention. Figure 2 A flowchart illustrating the optimization of the virtual strain monitoring point network using an improved genetic algorithm; Figure 3 This is a flowchart for dynamic response analysis. Detailed Implementation

[0019] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] See Figure 1 This invention provides a method for testing the fatigue strength reliability of vehicle rearview mirrors, the overall implementation of which is as follows: The basic characteristic parameters of the rearview mirror of the vehicle under test are obtained, including geometric dimensions, material properties, and connection structure information. Based on this, a digital 3D model of the rearview mirror is constructed. A time-domain load spectrum for the rearview mirror mounting base is generated according to a preset actual road spectrum excitation signal and vehicle vibration transfer function. An improved genetic algorithm is used to optimize the configuration of the virtual strain monitoring point network. This algorithm optimizes the spatial distribution of monitoring points based on the variation trend of strain energy density distribution and principal stress direction on the model surface. The generated time-domain load spectrum is applied to the mounting points of the digital 3D model to drive explicit dynamic solution and obtain dynamic response data. During the solution process, the equivalent stress time history data and principal stress direction time history data of the virtual strain monitoring point network are extracted. Based on the equivalent stress time history data and the material's SN curve, the preliminary fatigue damage accumulation value of each monitoring point is calculated. Based on the principal stress direction time history data, the frequency of principal stress direction changes at key critical points is identified, and the preliminary damage value is corrected multi-axis. The corrected fatigue damage accumulation values ​​of all monitoring points are integrated to evaluate the overall fatigue life and reliability level of the rearview mirror of the vehicle under test.

[0021] In one embodiment of the present invention, initial data on the geometric dimensions, material parameters, and connection structure of the rearview mirror of the vehicle under test are acquired. Based on the initial data on the geometric dimensions, material parameters, and connection structure, a digital three-dimensional model of the rearview mirror of the vehicle under test is constructed. Vibration data of the target vehicle model on various typical road surfaces are collected. After filtering and statistical analysis, a preset actual road spectrum excitation signal is synthesized. The actual road spectrum excitation signal includes vertical, longitudinal, and lateral acceleration components. A vehicle vibration transfer function is established from the vehicle suspension hardpoint to the rearview mirror mounting base. The vehicle vibration transfer function is obtained through experimental modal analysis and frequency response function fitting. Each acceleration component of the preset actual road spectrum excitation signal is input into the corresponding vehicle vibration transfer function. Through convolution operations, the frequency domain response of the three-dimensional force and torque at the rearview mirror mounting base is calculated. An inverse Fourier transform is performed on the frequency domain response of the three-dimensional force and torque to generate a time-domain load spectrum acting on the rearview mirror mounting base. The time-domain load spectrum includes force components and torque components in three directions that vary with time.

[0022] In specific implementation, initial data on the geometric dimensions, material parameters, and connection structure of the rearview mirror of the vehicle under test are acquired. Based on the initial data on the geometric dimensions, material parameters, and connection structure, a digital three-dimensional model of the rearview mirror of the vehicle under test is constructed. In some embodiments, the geometric dimensions are obtained by measuring with a three-dimensional laser scanner, the material parameters include the elastic modulus and Poisson's ratio, and the initial data on the connection structure includes the bolt preload and the coefficient of friction of the contact surface. The digital three-dimensional model is established in finite element preprocessing software, containing solid elements and shell elements to characterize the mirror shell and internal support structure.

[0023] In practice, vibration data of the target vehicle model on various typical road surfaces are collected. After filtering and statistical analysis, the preset actual road spectrum excitation signal is synthesized. The actual road spectrum excitation signal includes vertical, longitudinal, and lateral acceleration components. Typical road surfaces include asphalt paved roads, gravel roads, and Belgian cobblestone roads. The vibration data is recorded by a triaxial accelerometer mounted on the vehicle body, with a sampling frequency set to 1000 Hz. The filtering process uses a bandpass filter to retain frequency components from 0.5 to 200 Hz. Statistical analysis extracts the envelope of the data from each road surface and then splices them together to form a continuous actual road spectrum excitation signal.

[0024] In specific implementation, a vehicle vibration transfer function is established from the vehicle suspension hardpoint to the rearview mirror mounting base. This vehicle vibration transfer function is obtained through experimental modal analysis and frequency response function fitting. In some embodiments, experimental modal analysis obtains the frequency response from the suspension hardpoint to the vehicle body mounting point using the hammer impact method, and then combines this with the transmission characteristics from the vehicle body to the rearview mirror base to fit a six-degree-of-freedom vehicle vibration transfer function matrix. The vehicle vibration transfer function represents the ratio of force to torque output under unit acceleration input.

[0025] In specific implementation, each acceleration component of the preset actual road spectrum excitation signal is input into the corresponding vehicle vibration transfer function; through convolution operation, the frequency domain response of the triaxial force and torque at the rearview mirror mounting base is calculated. It can be understood that for any acceleration component... With the corresponding transfer function Frequency domain response The calculation formula is: in: acceleration components Fourier transform, subscript Indicates the direction of force or torque components, using subscripts. Indicate the direction of acceleration input. The frequency variable is used. Optionally, an inverse Fourier transform is performed on the frequency domain response of the three-dimensional force and torque to generate a time-domain load spectrum acting on the rearview mirror mounting base. The time-domain load spectrum includes force components and torque components in three directions that vary with time. Optionally, the inverse Fourier transform uses a fast algorithm to process the discrete frequency domain data, outputting a time-domain load sequence with a time interval of 0.001 seconds. The force component unit is Newton, and the torque component unit is Newton-meter.

[0026] In one embodiment of the present invention, see [reference] Figure 2The genetic algorithm population is initialized, with each individual representing a spatial distribution scheme of virtual strain monitoring point networks on the surface of a digital 3D model of the rearview mirror of the vehicle under test. A fitness evaluation process is performed on each individual: based on a static unit load condition, finite element analysis is conducted on the digital 3D model of the rearview mirror to calculate the strain energy density at each node on the model surface. At the monitoring point locations corresponding to the distribution scheme represented by the individual, the strain energy density value and the principal stress direction of the monitoring point are extracted. Based on the optimization criteria of the improved genetic algorithm, the fitness value of the spatial distribution scheme is calculated. The optimization criteria require that the monitoring points should cover areas with high strain energy density and be able to capture areas with drastic changes in principal stress direction. Based on the fitness value, selection, crossover, and mutation operations are performed on the individuals in the population to generate a new generation. The fitness evaluation and genetic operations are repeated until the preset number of generations or the fitness value converges, outputting the spatial coordinate set of the virtual strain monitoring point network represented by the optimal individual. During the fitness evaluation process, two optimization indices are calculated for each spatial distribution scheme: the first index is the average strain energy density at the selected monitoring points, and the second index is the standard deviation of the principal stress direction angle at the selected monitoring points. The average strain energy density and the standard deviation of the principal stress direction angles are weighted and summed; this sum is used as the initial fitness of the spatial distribution scheme. A penalty term is introduced: for any two monitoring points whose distance is less than a preset minimum distance, a penalty value inversely proportional to the distance is subtracted from the initial fitness. The final fitness value is the value corrected for the penalty term. During the crossover process in the genetic operation, monitoring point groups located in different high strain energy density regions from high-fitness individuals are preferentially exchanged. During the mutation process, monitoring points are allowed to move within a small range with a certain probability towards the direction of the current maximum stress gradient to better track the changing trend of the principal stress direction.

[0027] In the specific implementation, a genetic algorithm population is initialized, with each individual representing a spatial distribution scheme of a virtual strain monitoring point network on the surface of the digital 3D model of the rearview mirror of the vehicle under test. A fitness evaluation process is then performed on each individual. The population size is set to 100 individuals, each encoded as a set of 3D coordinates for 30 monitoring points. Fitness evaluation first requires finite element analysis of the digital 3D model of the rearview mirror under test based on a static unit load condition, calculating the strain energy density at each node on the model surface. At the monitoring point locations corresponding to the distribution scheme represented by the individual, the strain energy density value and the principal stress direction of the monitoring point are extracted. Based on the optimization criteria of the improved genetic algorithm, the fitness value of the spatial distribution scheme is calculated. The optimization criteria require that the monitoring points should cover areas with high strain energy density and be able to capture areas with drastic changes in the principal stress direction. Based on the fitness values, selection, crossover, and mutation operations are performed on the individuals in the population to generate a new generation. The fitness evaluation and genetic operations are repeated until a preset number of generations or convergence of the fitness value is reached, outputting the spatial coordinate set of the virtual strain monitoring point network represented by the optimal individual.

[0028] In practical implementation, during the fitness evaluation process, two optimization indices are calculated for each spatial distribution scheme: the first index is the average strain energy density at the selected monitoring points, and the second index is the standard deviation of the principal stress direction angles at the selected monitoring points. After normalizing the average strain energy density and the standard deviation of the principal stress direction angles to dimensionless quantities, a weighted sum is performed, and the sum is used as the initial fitness of the spatial distribution scheme. A penalty term is introduced: for any two monitoring points whose distance is less than a preset minimum distance, a penalty value inversely proportional to the distance is subtracted from the initial fitness. In some embodiments, the average strain energy density index... Take the arithmetic mean of the strain energy density at the monitoring points, and the standard deviation of the principal stress direction angles. The standard deviation of the principal stress direction angle sequence at the monitoring points is used as the initial fitness after normalization. The calculation is as follows: in: and These are the weighting coefficients of the two indicators, both of which are dimensionless. The maximum value of the strain energy density at the nodal points on the model surface is taken as the reference strain energy density. The reference angle is 90 degrees, or 1.57 radians; both terms on the right side of the formula are dimensionless, therefore the initial fit on the left side is... It is also dimensionless, consistent with the dimension of the penalty term. Penalty value. Distance from monitoring points The relationship is ,in It is a dimensionless constant factor. The minimum allowable distance is preset; the final fitness value is the value after adjustment by the penalty term.

[0029] Optionally, during the crossover process in genetic operations, priority is given to exchanging monitoring point groups from high-fitness individuals located in different high-strain energy density regions; during the mutation process, monitoring points are allowed to move within a small range with a certain probability towards the direction of the current maximum stress gradient, in order to better track the changing trend of the principal stress direction. Optionally, the crossover operation selects the top 20% of individuals in fitness as parents with an 80% probability, exchanging the coordinate blocks of monitoring points in their codes located in different high-strain energy density partitions; the mutation operation is triggered with a 5% probability, and the selected monitoring points move no more than 2 mm along the direction of the maximum stress gradient.

[0030] In one embodiment of the present invention, see [reference] Figure 3 The force and moment components of the time-domain load spectrum are applied as nodal forces to the node set simulating the rearview mirror mounting base in a digital 3D model. The material damping coefficient and structural damping coefficient of the digital 3D model are set, and the geometric nonlinearity option is activated to account for large deformation effects. An explicit dynamic integration method is used to solve the digital 3D model with the applied time-domain load spectrum using time integration. During the solution process, the stress tensor at each monitoring point in the virtual strain monitoring point network is recorded in real time as a function of time. From the recorded stress tensor data, the equivalent stress time history data and principal stress direction time history data for each monitoring point are separated and output. The time step of the explicit dynamic integration is set, and the time step is determined based on the smallest finite element size in the digital 3D model and the sound velocity in the material to ensure the stability of the numerical solution. At the beginning of each time step, the nodal force vector applied to the node set of the simulated rearview mirror mounting base is calculated according to the corresponding force and moment components in the time-domain load spectrum. Based on the displacement, velocity, and acceleration states of all nodes in the current digital 3D model, and combining the material damping coefficients and structural damping coefficients, the internal force vectors of all finite element elements are calculated. Using the explicit central difference method, the acceleration vectors of all nodes for the next time step are solved using the nodal force vectors and internal force vectors. The acceleration vectors are then integrated over time to update the velocity and displacement vectors of all nodes. Using the updated displacement vectors as boundary conditions, the strain and stress states of all finite element elements are recalculated, completing the solution for one time step. This process is repeated until the dynamic response solution for the entire time-domain load spectrum is completed, obtaining the dynamic response data of the digital 3D model throughout the entire loading process.

[0031] In specific implementation, the force and moment components in the time-domain load spectrum are applied as nodal forces to the node set simulating the rearview mirror mounting base in the digital 3D model. The material damping coefficient and structural damping coefficient of the digital 3D model are set, and the geometric nonlinearity option is activated to consider large deformation effects. An explicit dynamic integration method is used to solve the digital 3D model with the applied time-domain load spectrum through time integration. During the solution process, the stress tensor at each monitoring point in the virtual strain monitoring point network is recorded in real time as a function of time. From the recorded stress tensor data, the equivalent stress time history data and principal stress direction time history data for each monitoring point are separated and output. In some embodiments, the material damping coefficient is set to 0.02, the structural damping coefficient to 0.01, and the geometric nonlinearity option is enabled to account for the stiffness change of the rearview mirror bracket under large amplitude vibrations. The nodal forces are applied using a concentrated force and moment coupling method, distributed to each node of the mounting base node set.

[0032] In the specific implementation, a time step for explicit dynamic integration is set. The time step is determined based on the smallest finite element size in the digital 3D model and the sound velocity in the material to ensure the stability of the numerical solution. At the beginning of each time step, the nodal force vectors applied to the node set of the simulated rearview mirror mounting base are calculated according to the corresponding force and torque components in the time-domain load spectrum. Based on the displacement, velocity, and acceleration states of all nodes in the digital 3D model at the current moment, combined with the material damping coefficient and structural damping coefficient, the internal force vectors of all finite element elements are calculated. Using the explicit central difference method, the acceleration vectors of all nodes in the next time step are solved using the nodal force vectors and internal force vectors. The acceleration vectors are integrated over time to update the velocity and displacement vectors of all nodes. The updated displacement vectors are used as boundary conditions to recalculate the strain and stress states of all finite element elements, completing the solution for one time step. The steps are repeated until the dynamic response solution for the entire time period covered by the time-domain load spectrum is completed, obtaining the dynamic response data of the digital 3D model throughout the entire loading process. The time step can be understood as... The calculation uses the Courant-Friedrichs-Lewy conditions, expressed as follows: in: This represents the smallest feature size of the finite element in the model. The speed of sound is the speed at which longitudinal waves propagate in a material. The stability safety factor is set between 0.8 and 0.9. Refer to Table 1, which shows the minimum element size and corresponding settling time step for different mesh refinement levels: Table 1: Explicit dynamic settling time steps for different mesh sizes Optionally, the nodal force vector is distributed to the mounting base node set based on the triaxial force and triaxial moment components in the time-domain load spectrum using the static equivalence principle; the force magnitude of each node is related to its distance from the geometric center of the mounting base, and the moment components are converted into equivalent couples and applied to the corresponding node pairs. Optionally, the internal force vector calculation incorporates material nonlinear effects, employing a bilinear kinematic hardening model to describe the stress-strain behavior of the rearview mirror plastic material under cyclic loading.

[0033] In one embodiment of the present invention, rainflow counting is performed on the equivalent stress time history data of each monitoring point to statistically determine all complete stress cycles, and the stress amplitude and mean stress of each stress cycle are extracted. Using the Goodman formula or Gerber formula, the stress cycle containing the mean stress is equivalently converted into a symmetrical cyclic stress amplitude with the same damage. Based on the equivalent converted stress amplitude, the number of cycles at which the material fails under the stress amplitude is obtained by looking up a table or interpolating using a formula on the material SN curve of the rearview mirror of the vehicle under test. According to the linear cumulative damage criterion, the damage caused by the stress cycle is calculated as the reciprocal of the ratio of the actual number of cycles to the number of failure cycles. The damage caused by all stress cycles in the equivalent stress time history data is accumulated to obtain the preliminary cumulative fatigue damage value of the corresponding monitoring point. From all monitoring points, those with a preliminary cumulative fatigue damage value higher than a preset threshold are selected and defined as critical danger points. The principal stress direction time history data of each critical danger point is extracted, and the sequence of principal stress direction angle changes over time is calculated and recorded as the principal stress direction angle sequence. The principal stress direction angle sequence is statistically analyzed, and the number of times the principal stress direction angle changes by more than a specific angle per unit time is calculated as the principal stress direction change frequency. Based on the principal stress direction change frequency, the corresponding multiaxial fatigue correction coefficient is obtained from a pre-set lookup table. The multiaxial fatigue correction coefficient is positively correlated with the direction change frequency. The preliminary cumulative fatigue damage value of the critical hazard point is multiplied by the corresponding multiaxial fatigue correction coefficient to obtain the corrected cumulative fatigue damage value of the critical hazard point. For non-critical hazard points, the corrected cumulative fatigue damage value is equal to their preliminary cumulative fatigue damage value.

[0034] In practice, rainflow counting is performed on the equivalent stress time history data of each monitoring point to count all complete stress cycles and extract the stress amplitude and mean stress of each stress cycle. Using the Goodman or Gerber formula, the stress cycle containing the mean stress is equivalently converted into a symmetrical cyclic stress amplitude with the same damage. Based on the equivalent stress amplitude, the number of cycles at which material failure occurs under the stress amplitude is obtained from the material's SN curve on the rearview mirror of the vehicle under test, either by referring to a table or through formula interpolation. According to the linear cumulative damage criterion, the damage caused by the stress cycle is calculated as the reciprocal of the ratio of the actual number of cycles to the number of failure cycles. The damage caused by all stress cycles in the equivalent stress time history data is summed to obtain the preliminary cumulative fatigue damage value for the corresponding monitoring point. In some embodiments, the rainflow counting process uses a four-peak valley detection method to identify closed stress hysteresis loops, and the output parameters of each loop include the stress amplitude. With average stress For cycles with non-zero mean stress, the Goodman formula is used to convert them into equivalent symmetrical cyclic stress amplitudes. The conversion formula is: in: Indicates the ultimate tensile strength of a material. The stress amplitude of the original stress cycle. This represents the average stress of the original stress cycles. The material's SN curve is in power-law form. ,in This represents the number of failure cycles. and Material constant; damage caused by a single cycle Calculated as ,in The number of times this cycle occurs in the actual load spectrum, and the preliminary cumulative fatigue damage value at the monitoring point. This is the sum of all cyclic damage.

[0035] In practice, monitoring points with preliminary cumulative fatigue damage values ​​exceeding a preset threshold are selected from all monitoring points and defined as critical hazard points. The time history data of the principal stress directions for each critical hazard point are extracted, and the sequence of principal stress direction angle changes over time is calculated and recorded as the principal stress direction angle sequence. The principal stress direction angle sequence is statistically analyzed to calculate the number of times the principal stress direction angle changes by more than a specific angle per unit time, which is taken as the principal stress direction change frequency. Based on the principal stress direction change frequency, the corresponding multiaxial fatigue correction coefficient is obtained from a preset lookup table. The multiaxial fatigue correction coefficient is positively correlated with the direction change frequency. The preliminary cumulative fatigue damage value of the critical hazard point is multiplied by the corresponding multiaxial fatigue correction coefficient to obtain the corrected cumulative fatigue damage value for the critical hazard point. For non-critical hazard points, the corrected cumulative fatigue damage value is equal to their preliminary cumulative fatigue damage value. Refer to Table 2, which shows the multiaxial fatigue correction coefficients corresponding to a typical critical hazard point in different principal stress direction change frequency ranges. Table 2: Relationship between frequency of principal stress direction changes and multiaxial fatigue correction coefficient Optionally, the preset threshold is set to 10% of the expected total damage over the design life. That is, when the initial cumulative fatigue damage at a monitoring point exceeds one-tenth of the total damage, that point is considered a critical danger point. The principal stress direction angle sequence is obtained by sampling the projection angle of the principal stress vector on the reference plane, with the sampling frequency consistent with the output frequency of the dynamic response analysis. Optionally, one second is taken as the unit of time, and 30 degrees is taken as the specific angle. That is, the cumulative number of times the principal stress direction angle changes by more than 30 degrees per second is counted as the change frequency. The lookup table is calibrated based on multiaxial fatigue test data, reflecting the weakening effect of frequent changes in principal stress direction on the material's fatigue limit.

[0036] In one embodiment of the present invention, the maximum value is found from the corrected cumulative fatigue damage values ​​of all monitoring points and recorded as the maximum point damage value. The reciprocal of the maximum point damage value is used as the predicted characteristic life of the rearview mirror of the vehicle under test under the loading conditions represented by the time-domain load spectrum. The number of monitoring points whose corrected cumulative fatigue damage values ​​exceed the expected damage threshold of the design life and their spatial distribution are statistically analyzed. Combining the predicted characteristic life and the distribution of monitoring points exceeding the damage threshold, the fatigue reliability level of the rearview mirror of the vehicle under test is determined according to a predefined reliability evaluation matrix. Samples produced in the same batch as the rearview mirror of the vehicle under test are selected and installed on a vibration test bench. Actual strain gauges are attached to the connection between the back of the mirror shell and the bracket of the sample to form a physical strain monitoring network. On the vibration test bench, the time-domain load spectrum is reproduced, and accelerated fatigue testing is performed on the sample. The actual strain time history data of key locations during the test is collected in real time through the physical strain monitoring network. The actual strain time history data is compared with the strain time history data obtained from the corresponding virtual strain monitoring point network in dynamic response analysis. If the comparison error exceeds the allowable range, the boundary conditions or material parameters of the digital 3D model are corrected based on the actual data, and some analysis steps are repeated until the simulation results and experimental results meet the consistency requirements.

[0037] In practical implementation, the maximum value is identified from the corrected cumulative fatigue damage values ​​of all monitoring points and recorded as the maximum point damage value. This maximum point damage value is then substituted into a preset life calculation formula to obtain the predicted characteristic life of the rearview mirror of the vehicle under test under the loading conditions represented by the time-domain load spectrum. The number and spatial distribution of monitoring points whose corrected cumulative fatigue damage values ​​exceed the expected damage threshold of the design life are statistically analyzed. Combining the predicted characteristic life with the distribution of monitoring points exceeding the damage threshold, and based on a predefined reliability evaluation matrix, the fatigue reliability level of the rearview mirror of the vehicle under test is determined. It can be understood that if the maximum point damage value is recorded as... Then predict the characteristic lifetime. The calculation formula is: in: This represents the total loading time corresponding to the time-domain load spectrum. The maximum point damage value is dimensionless, therefore the characteristic lifetime is predicted. The dimension of the matrix is ​​time, consistent with the dimension definition of engineering life; the rows of the reliability evaluation matrix correspond to the interval classification of the predicted characteristic life, the columns correspond to the interval of the proportion of monitoring points exceeding the standard, and the matrix elements store qualitative level labels from "low reliability" to "high reliability".

[0038] In the specific implementation, samples produced in the same batch as the rearview mirror of the vehicle under test are selected and installed on a vibration test bench. Actual strain gauges are attached to the back of the mirror housing and the connection between the mirror housing and the bracket, forming a physical strain monitoring network. On the vibration test bench, the time-domain load spectrum is reproduced, and accelerated fatigue testing is performed on the samples. The actual strain time history data at key locations during the test is collected in real time through the physical strain monitoring network. The actual strain time history data is compared with the strain time history data obtained from the corresponding virtual strain monitoring point network in dynamic response analysis. If the comparison error exceeds the allowable range, the boundary conditions or material parameters of the digital three-dimensional model are corrected based on the actual data, and some analysis steps are repeated until the simulation results and test results meet the consistency requirements. In some embodiments, the physical strain monitoring network is arranged with twelve uniaxial strain gauges, eight of which are attached to the stress concentration area on the back of the mirror housing, and four are attached to the connection flange between the bracket and the vehicle body. The accelerated fatigue test is achieved by amplifying the amplitude of the time-domain load spectrum or increasing the cycle rate, while keeping the temporal and spectral characteristics of the load unchanged.

[0039] Optionally, the expected damage threshold for the design life is set to 1.0, meaning that when the corrected cumulative fatigue damage value of a monitoring point exceeds 1.0, the point is determined to be damaged within the design life. The spatial distribution of monitoring points exceeding the standard is clustered statistically according to the component areas of the rearview mirror, such as the upper cover area, the lower shell area, and the pivot connection area. Optionally, the consistency requirement is defined as the root mean square error of strain between the virtual strain monitoring point network and the physical strain monitoring network at the same time point not exceeding 15%. If the error limit is exceeded, the connection stiffness of the rubber bushing or the damping parameters of the plastic material in the digital 3D model are adjusted first, and then the dynamic response analysis and fatigue damage calculation process is rerun.

[0040] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A method for testing the fatigue strength reliability of a vehicle rearview mirror, characterized in that, include: The basic feature parameters of the rearview mirror of the vehicle under test are obtained, a digital three-dimensional model of the rearview mirror of the vehicle under test is constructed based on the basic feature parameters, and a time-domain load spectrum for the installation point of the rearview mirror of the vehicle under test is generated. An improved genetic algorithm is used to optimize the network of virtual strain monitoring points configured on the digital three-dimensional model of the rearview mirror of the vehicle under test. The improved genetic algorithm is optimized based on the variation trend of strain energy density distribution and principal stress direction. The time-domain load spectrum is applied to the mounting points of the digital three-dimensional model of the rearview mirror of the vehicle under test, driving the digital three-dimensional model to perform dynamic response analysis based on the virtual strain monitoring point network. Extract the equivalent stress time history data and principal stress direction time history data of the virtual strain monitoring point network during the dynamic response analysis process; Based on the equivalent stress time history data and the material SN curve of the rearview mirror of the vehicle under test, the cumulative fatigue damage value of each of the virtual strain monitoring point networks is calculated. Based on the principal stress direction time history data, the frequency of principal stress direction changes at key critical points is identified, and the cumulative fatigue damage value is corrected in a multi-axis manner. By integrating the cumulative fatigue damage values ​​after network correction of all virtual strain monitoring points, the overall fatigue life and reliability of the rearview mirror of the vehicle under test are evaluated.

2. The method for testing the fatigue strength reliability of a vehicle rearview mirror according to claim 1, characterized in that, The process of acquiring the basic feature parameters of the rearview mirror of the vehicle under test, constructing a digital three-dimensional model of the rearview mirror based on the basic feature parameters, and generating a time-domain load spectrum for the mounting point of the rearview mirror includes: Obtain initial data on the geometric dimensions, material parameters, and connection structure of the rearview mirror of the vehicle under test; Based on the initial data of the geometric dimensions, material parameters and connection structure, a digital three-dimensional model of the rearview mirror of the vehicle under test is constructed. Based on the preset actual road spectrum excitation signal and vehicle vibration transfer function, a time-domain load spectrum is generated for the mounting point of the rearview mirror of the vehicle under test, specifically including: Vibration data of the target vehicle model on various typical road surfaces are collected, filtered and statistically analyzed, and then synthesized to obtain the preset actual road spectrum excitation signal. The actual road spectrum excitation signal includes vertical, longitudinal and lateral acceleration components. The vehicle vibration transfer function from the vehicle suspension hardpoint to the rearview mirror mounting base is established, and the vehicle vibration transfer function is obtained by experimental modal analysis and frequency response function fitting. Each acceleration component of the preset actual road spectrum excitation signal is input into the corresponding vehicle vibration transfer function; The frequency domain response of the triaxial force and torque at the rearview mirror mounting base is calculated by convolution operation. An inverse Fourier transform is performed on the frequency domain response of the three forces and torques to generate a time-domain load spectrum acting on the rearview mirror mounting base. The time-domain load spectrum includes force components and torque components in three directions that vary with time.

3. The method for testing the fatigue strength reliability of a vehicle rearview mirror according to claim 1, characterized in that, The virtual strain monitoring point network optimized on the digital 3D model of the rearview mirror of the vehicle under test using an improved genetic algorithm includes: Initialize the genetic algorithm population, where each individual represents a spatial distribution scheme of a virtual strain monitoring point network on the surface of the digital three-dimensional model of the rearview mirror of the vehicle under test. Perform the following fitness assessment process for each individual: Based on the static unit load condition, finite element analysis was performed on the digital three-dimensional model of the rearview mirror of the vehicle under test to calculate the strain energy density of each node on the model surface. At the monitoring point location corresponding to the distribution scheme represented by the individual, extract the strain energy density value and the principal stress direction of the monitoring point; Based on the optimization criteria of the improved genetic algorithm, the fitness value of the spatial distribution scheme is calculated. The optimization criteria require that the monitoring points should cover areas with high strain energy density and be able to capture areas where the principal stress direction changes drastically. Based on fitness values, selection, crossover, and mutation operations are performed on individuals in the population to generate a new generation of population; Repeatedly perform fitness assessment and genetic operations until the preset number of generations or fitness value convergence is reached, and output the set of spatial coordinates of the virtual strain monitoring point network represented by the optimal individual.

4. The method for testing the fatigue strength reliability of a vehicle rearview mirror according to claim 3, characterized in that, The improved genetic algorithm is optimized based on the variation trend of strain energy density distribution and principal stress direction. Its working principle includes: During the fitness evaluation process, two optimization indices are calculated for each spatial distribution scheme: the first index is the average strain energy density at the selected monitoring point, and the second index is the standard deviation of the principal stress direction angle at the selected monitoring point. The average strain energy density and the standard deviation of the principal stress direction angle are weighted and summed, and the sum is used as the initial fitness of the spatial distribution scheme. A penalty term is introduced: if the distance between any two monitoring points is less than a preset minimum distance, a penalty value inversely proportional to the distance is subtracted from the initial fitness. The final fitness value is the value adjusted for the penalty term; During the crossover process of genetic manipulation, monitoring point groups located in different high strain energy density regions from high-fit individuals are preferentially exchanged. During the mutation process, the monitoring point is allowed to move in a small range with a certain probability toward the direction of the current maximum stress gradient in order to better track the changing trend of the principal stress direction.

5. The method for testing the fatigue strength reliability of a vehicle rearview mirror according to claim 1, characterized in that, The time-domain load spectrum is applied to the mounting points of the digital 3D model of the rearview mirror of the vehicle under test, driving the digital 3D model to perform dynamic response analysis based on the virtual strain monitoring point network, including: The force and torque components in the time-domain load spectrum are applied as nodal forces to the set of nodes simulating the rearview mirror mounting base in the digital three-dimensional model. Set the material damping coefficient and structural damping coefficient of the digital 3D model, and activate the geometric nonlinearity option to account for large deformation effects; The explicit dynamic integration method is used to solve the time integral of the digital three-dimensional model with the applied time-domain load spectrum. During the solution process, the stress tensor at each monitoring point in the virtual strain monitoring point network is recorded in real time as it changes over time. From the recorded stress tensor data, the equivalent stress time history data and the principal stress direction time history data of each monitoring point are separated and output.

6. The method for testing the fatigue strength reliability of a vehicle rearview mirror according to claim 5, characterized in that, The explicit dynamic integration method is used to solve the time integral problem on the digitized three-dimensional model with the applied time-domain load spectrum, including: The time step for explicit dynamic integration is set, which is determined based on the smallest finite element size in the digital three-dimensional model and the sound velocity in the material, to ensure the stability of the numerical solution. At the beginning of each time step, the nodal force vectors applied to the set of nodes of the simulated rearview mirror mounting base are calculated based on the force and torque components corresponding to the time-domain load spectrum. Based on the displacement, velocity, and acceleration states of all nodes in the digital 3D model at the current moment, and combined with the material damping coefficient and structural damping coefficient, calculate the internal force vector of all finite element elements. Based on the explicit central difference method, the acceleration vectors of all nodes at the next time step are solved using the nodal force vectors and the internal force vectors. Integrate the acceleration vector over time and update the velocity and displacement vectors of all nodes; Using the updated displacement vector as boundary conditions, the strain and stress states of all finite element elements are recalculated to complete the solution for one time step. The steps are repeated until the dynamic response solution for the entire time period covered by the time-domain load spectrum is completed, thus obtaining the dynamic response data of the digital three-dimensional model throughout the entire loading process.

7. The method for testing the fatigue strength reliability of a vehicle rearview mirror according to claim 1, characterized in that, Based on the equivalent stress time history data and the material SN curve of the rearview mirror of the vehicle under test, the cumulative fatigue damage value of each of the virtual strain monitoring point networks is calculated, including: Rainflow counting was performed on the equivalent stress time history data of each monitoring point to count all complete stress cycles, and the stress amplitude and average stress of each stress cycle were extracted. Using the Goodman formula or Gerber formula, a stress cycle containing average stress can be equivalently converted into a symmetrical cyclic stress amplitude with the same damage. Based on the equivalent stress amplitude, the number of cycles at which the material fails under the stress amplitude can be obtained by looking up a table or by interpolating a formula on the material SN curve of the rearview mirror of the vehicle under test. According to the linear cumulative damage criterion, the damage caused by stress cycling is calculated as the reciprocal of the ratio of the actual number of cycles to the number of failure cycles. By summing up the damage caused by all stress cycles in the equivalent stress time history data, the preliminary cumulative fatigue damage value of the corresponding monitoring point is obtained.

8. The method for testing the fatigue strength reliability of a vehicle rearview mirror according to claim 7, characterized in that, Based on the principal stress direction time history data, the frequency of principal stress direction changes at key critical points is identified, and the cumulative fatigue damage value is corrected using multi-axis methods, including: From all monitoring points, those with a preliminary cumulative fatigue damage value higher than a preset threshold are selected and defined as critical danger points. Extract the time history data of the principal stress direction for each critical hazard point, calculate the sequence of principal stress direction angle changes over time, and record it as the principal stress direction angle sequence; The principal stress direction angle sequence is statistically analyzed, and the number of times the principal stress direction angle changes by more than a specific angle per unit time is calculated as the frequency of principal stress direction changes. Based on the frequency of principal stress direction changes, the corresponding multiaxial fatigue correction coefficient is obtained from a preset lookup table. The multiaxial fatigue correction coefficient is positively correlated with the frequency of direction changes. Multiply the initial cumulative fatigue damage value of the critical hazard point by the corresponding multiaxial fatigue correction coefficient to obtain the corrected cumulative fatigue damage value of the critical hazard point. For non-critical hazard points, the corrected cumulative fatigue damage value is equal to the initial cumulative fatigue damage value.

9. The method for testing the fatigue strength reliability of a vehicle rearview mirror according to claim 1, characterized in that, The fatigue damage accumulation value, after integrating all virtual strain monitoring point network corrections, is used to evaluate the overall fatigue life and reliability of the rearview mirror of the vehicle under test, including: Find the maximum value among the corrected cumulative fatigue damage values ​​of all monitoring points and record it as the maximum point damage value; The reciprocal of the maximum point damage value is used as the predicted characteristic lifetime of the rearview mirror of the vehicle under test under the loading conditions represented by the time-domain load spectrum. The number of monitoring points whose cumulative fatigue damage value, after statistical correction, exceeds the expected damage threshold of the design life, and their spatial distribution; Based on the predicted characteristic lifespan and the distribution of monitoring points exceeding the damage threshold, and according to a predefined reliability evaluation matrix, the fatigue reliability level of the rearview mirror of the vehicle under test is determined.

10. A method for testing the fatigue strength reliability of a vehicle rearview mirror according to claim 9, characterized in that, The method also includes testing and verification and model calibration steps: Select a sample produced in the same batch as the rearview mirror of the vehicle under test and install it on the vibration test bench; Actual strain gauges are attached to the connection between the back of the sample's mirror shell and the support to form a physical strain monitoring network. The time-domain load spectrum was reproduced on the vibration test bench, and the sample was subjected to accelerated fatigue testing. The physical strain monitoring network is used to collect real-time strain time history data at key locations during the experiment. The actual strain time history data is compared with the strain time history data obtained from the corresponding virtual strain monitoring point network in dynamic response analysis; If the comparison error exceeds the allowable range, the boundary conditions or material parameters of the digital three-dimensional model are corrected based on the actual data, and some analysis steps are repeated until the simulation results and experimental results meet the consistency requirements.

Images