Automobile bumper parameter determination method and system based on AI modeling

Abstract

The invention discloses an AI modeling-based automobile bumper parameter determination method and system, relates to the technical field of AI modeling, is used for solving the problem of deviation in automobile bumper parameter determination, and selects a distance between an impact point and an automobile door according to an impact angle to calculate a side angle impact score by calling an AI modeling bumper parameter data packet and a confirmation mechanism. A simulated impact experiment is conducted, a diagonal deformation difference value and an extrusion force difference value of the front bumper and the rear bumper are obtained to generate a diagonal response score, a collision angle collision coefficient is obtained in combination with a side angle collision score and marked as side face collision or front face collision, a side face collision label is called, the number is counted, and whether a parameter supplement rule is introduced or not is judged; the maximum deformation and the residual deformation are collected to calculate the deformation recovery rate, a fatigue supplement coefficient is obtained by combining the number of the side impact labels, whether the deformation recovery rate is introduced or not is judged according to the fatigue supplement coefficient, a parameter confirmation mechanism is imported after screening, and intelligent distinguishing of impact modes at different impact angles is achieved.

Description

Technical Field

[0001] This invention relates to the field of AI modeling technology, and more specifically, to a method and system for determining automobile bumper parameters based on AI modeling. Background Technology

[0002] With the development of the automotive industry, the safety issues faced by vehicles during operation are becoming increasingly prominent. As an important component of energy-absorbing structures, car bumpers need to play a role in mitigating impact and reducing damage under different conditions such as frontal, side, and oblique collisions. Existing methods for determining bumper parameters mostly rely on finite element simulation or real vehicle tests, primarily evaluating bumper performance through frontal collision data. However, they often cannot accurately classify bumpers when facing complex collisions such as side and oblique impacts.

[0003] The existing technology has the following shortcomings: Currently, due to different collision angles, when the front bumper is hit from the left, the rear bumper often exhibits right-side compression. That is, the four corners of the front and rear bumpers exert diagonal compression on the other bumper during the collision, making it difficult to clearly distinguish between frontal and side collisions. During the collision, the frontal or side impact mode of the front and rear bumpers cannot be clearly marked. In the AI ​​modeling scenario, the introduction of metal fatigue parameters lacks dynamic adjustment, leading to deviations in subsequent parameter determination. Parameters are prone to lag or over-correction, further reducing the accuracy of automobile bumper parameter determination. Therefore, this paper proposes an AI-based method and system for determining automobile bumper parameters.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a method and system for determining automobile bumper parameters based on AI modeling. This method addresses the problems mentioned in the background art by employing multi-angle collision score calculation, diagonal response coefficient marking, and a dynamic introduction mechanism for fatigue supplementary parameters.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for determining automobile bumper parameters based on AI modeling, comprising the following steps: Step S1: Call the AI-modeled car bumper parameter data package and bumper parameter confirmation mechanism, set the collision angle, select the distance between each impact point and the door, and calculate the side angle impact score; Step S2: Conduct a simulated impact experiment based on the collision angle, obtain the difference in deformation and the difference in compressive force between the front and rear bumpers to obtain the diagonal response score, combine the side angle impact score to obtain the collision angle collision coefficient, and mark each collision angle as a side impact or a frontal impact. Step S3: Call the side impact tags and count the number. Select whether to enter the parameter supplementation rule. Calculate the deformation recovery rate of the bumper based on the maximum deformation of the bumper after each simulation and the residual deformation after unloading the collision object, and obtain the fatigue supplementation coefficient by combining the number of side impact tags. Step S4: Based on the fatigue compensation coefficient analysis, determine whether to include the deformation recovery rate as a parameter to be introduced. After processing and screening the deformation recovery rate, import it into the bumper parameter confirmation mechanism.

[0007] In a preferred embodiment, in step S1, the AI-modeled car bumper parameter data package and the bumper parameter confirmation mechanism are invoked through the control command interface. AI-modeled car bumper parameter data package refers to a multi-dimensional parameter set generated during the training phase based on finite element simulation data, collision test data, and vehicle geometry data. The bumper parameter confirmation mechanism refers to a set of calculation rules used to retrieve, call, and dynamically correct parameters in a data packet under different collision conditions.

[0008] In a preferred embodiment, in step S1, the corresponding impact points are determined by the collision angles preset in the AI-modeled car bumper parameter data package. The spatial coordinates of the impact point and the spatial coordinates of the door reference point are obtained in the vehicle coordinate system. The distance between each impact point and the door is calculated using the Euclidean distance formula. Subtracting the standardized distances between each impact point and the car door from 1 yields the side corner impact score, calculated using the following formula: ; In the formula, The distance between each impact point and the car door after standardization. Let i be the side impact score corresponding to the i-th collision angle, i = 1, 2, ..., n, where n is the total number of collision angles and i is the i-th collision angle.

[0009] In a preferred embodiment, in step S2, based on the multiple sets of collision angles preset in the car bumper parameter data package modeled by AI, a controlled simulation test is performed on the front end or side front end area of ​​the car, and the corresponding deformation difference and compressive force difference of the front and rear bumpers are generated. The diagonal response score is calculated by multiplying the difference in deformation and the difference in compressive force between the front and rear bumpers after standardization.

[0010] In a preferred embodiment, in step S2, the diagonal response score and the side-angle impact score are substituted into a polynomial regression calculation to obtain the collision angle collision coefficient. If the collision angle collision coefficient exceeds the side threshold, the current collision angle is marked as a side impact; If the collision angle collision coefficient is lower than the side threshold, the current collision angle is marked as a frontal impact and imported into the bumper parameter confirmation mechanism.

[0011] In a preferred embodiment, in step S3, the number of collisions marked as side impacts in the simulation experiment is counted and recorded as the number of side impact tags. If the number of side-impact tags exceeds the preset threshold, then the parameter supplementation rule will be applied. If the number of side-impact tags is lower than the preset threshold, the parameter supplementation rule will not be applied.

[0012] In a preferred embodiment, in step S3, when the parameter supplementation rule is passed, the maximum deformation of the front and rear bumpers during the collision and the residual deformation after the collision object is unloaded are collected according to the simulated impact experiment. The maximum deformation represents the maximum instantaneous displacement of the bumper at each vertex under the impact of a collision; The residual deformation after the collision object is unloaded represents the deformation value that the bumper still retains after the collision object is unloaded. The deformation recovery rate of the bumper is calculated based on the maximum deformation and the residual deformation after the collision object is unloaded. The fatigue compensation coefficient was calculated by combining the number of side impact tags and the deformation recovery rate.

[0013] In a preferred embodiment, in step S4, when the fatigue compensation coefficient exceeds the fatigue determination threshold, the deformation recovery rate is determined to be introduced as a parameter. When the fatigue compensation coefficient is lower than the fatigue judgment threshold, the deformation recovery rate will not be used as a parameter to be introduced. When determining whether to use deformation recovery rate as a parameter to be introduced, the deformation recovery rate is processed and filtered.

[0014] In a preferred embodiment, in step S4, the median absolute deviation algorithm is used to calculate the anomaly determination value of each deformation recovery rate; The deformation recovery rates of those with abnormal judgment values ​​lower than or equal to the preset deviation coefficient threshold are selected and their arithmetic average is calculated to obtain the deformation recovery rates after screening. The deformation recovery rate after screening is imported into the bumper parameter confirmation mechanism; The bumper parameter confirmation mechanism adds deformation recovery rate as an additional parameter input to the existing parameter system.

[0015] A car bumper parameter determination system based on AI modeling includes an impact scoring module, an angle evaluation module, a coefficient calculation module, and a filtering and import module. The functions of each module are as follows: The impact scoring module is used to call the AI-modeled car bumper parameter data package and the bumper parameter confirmation mechanism, set the collision angle, select the distance between each impact point and the car door, and calculate the side angle impact score. The angle evaluation module is used to simulate impact experiments based on the collision angle, obtain the difference in deformation and the difference in compressive force between the front and rear bumpers to obtain the diagonal response score, and combine the side angle impact score to obtain the collision angle collision coefficient and mark each collision angle as a side impact or a frontal impact. The coefficient calculation module is used to call the side impact tags and count the number to select whether to enter the parameter supplementation rules. Based on the maximum deformation of the bumper after each simulation and the residual deformation after the collision object is unloaded, the deformation recovery rate of the bumper is calculated, and the fatigue supplementation coefficient is obtained by combining the number of side impact tags. The filter and import module is used to analyze whether to include the deformation recovery rate as a parameter to be imported based on the fatigue compensation coefficient. After processing and filtering the deformation recovery rate, it is imported into the bumper parameter confirmation mechanism.

[0016] The technical effects and advantages of this invention are as follows: This invention utilizes AI-modeled car bumper parameter data packages and a bumper parameter confirmation mechanism. It sets the collision angle, selects the distance between each impact point and the car door, calculates the side-angle impact score, conducts simulated impact experiments based on the collision angle, obtains the difference in deformation and compressive force between the front and rear bumpers to obtain the diagonal response score, combines the side-angle impact score to obtain the collision angle collision coefficient, and marks each collision angle as either a side impact or a frontal impact. It calls the side impact tags, counts the number, and selects whether to include them in the parameter supplementation rules. Based on the maximum deformation of the rear bumper after each simulated impact experiment and the residual deformation after unloading the impacting object, it calculates the bumper's deformation recovery rate. Combining the number of side impact tags, it obtains the fatigue supplementation coefficient, analyzes whether to include the deformation recovery rate as a parameter to be introduced based on the fatigue supplementation coefficient, processes and filters the deformation recovery rate, and then imports it into the bumper parameter confirmation mechanism. This achieves intelligent differentiation of impact modes under different collision angles and improves the flexibility of introducing metal fatigue factors into the bumper. Attached Figure Description

[0017] Figure 1 This is a flowchart of a method for determining automobile bumper parameters based on AI modeling according to the present invention.

[0018] Figure 2 This is a schematic diagram of a system for determining the parameters of an automobile bumper based on AI modeling, according to the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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] This invention utilizes AI-modeled car bumper parameter data packages and a bumper parameter confirmation mechanism. It sets the collision angle, selects the distance between each impact point and the car door, calculates the side-angle impact score, conducts simulated impact experiments based on the collision angle, obtains the difference in deformation and compressive force between the front and rear bumpers to obtain the diagonal response score, combines the side-angle impact score to obtain the collision angle collision coefficient, and marks each collision angle as either a side impact or a frontal impact. It calls the side impact tags, counts the number, and selects whether to include them in the parameter supplementation rules. Based on the maximum deformation of the rear bumper and the residual deformation after unloading the impacting object, it calculates the bumper's deformation recovery rate. Combining the number of side impact tags, it obtains the fatigue supplementation coefficient, analyzes whether to include the deformation recovery rate as a parameter, processes and filters the deformation recovery rate, and then imports it into the bumper parameter confirmation mechanism, thus achieving intelligent differentiation of impact modes under different collision angles.

[0021] Example 1 Please see Figure 1 A method for determining car bumper parameters based on AI modeling, the specific operation process is as follows: Step S1: Call the AI-modeled car bumper parameter data package and bumper parameter confirmation mechanism, set the collision angle, select the distance between each impact point and the door, and calculate the side angle impact score; Step S2: Conduct a simulated impact experiment based on the collision angle, obtain the difference in deformation and the difference in compressive force between the front and rear bumpers to obtain the diagonal response score, combine the side angle impact score to obtain the collision angle collision coefficient, and mark each collision angle as a side impact or a frontal impact. Step S3: Call the side impact tags and count the number. Select whether to enter the parameter supplementation rule. Calculate the deformation recovery rate of the bumper based on the maximum deformation of the bumper after each simulation and the residual deformation after unloading the collision object, and obtain the fatigue supplementation coefficient by combining the number of side impact tags. Step S4: Based on the fatigue compensation coefficient analysis, determine whether to include the deformation recovery rate as a parameter to be introduced. After processing and screening the deformation recovery rate, import it into the bumper parameter confirmation mechanism.

[0022] The specific implementation process is as follows: In step S1, the AI-modeled car bumper parameter data package and bumper parameter confirmation mechanism are invoked through the control command interface; Among them, the control command interface refers to the data interaction channel provided by AI modeling, which is used for parameter input, result output and label calling between the modeling environment and the parameter confirmation mechanism; Furthermore, the AI-modeled car bumper parameter data package refers to a multi-dimensional parameter set generated during the training phase based on finite element simulation data, collision test data, and vehicle geometric structure data. The data package includes, but is not limited to, the multi-dimensional mapping relationship of bumper material density, elastic modulus, yield strength, ultimate tensile strength, Poisson's ratio, thickness, deformation threshold, and collision angle. Furthermore, the bumper parameter confirmation mechanism refers to a set of calculation rules used to retrieve, call, and dynamically correct parameters in the data packet under different collision conditions, and is used to determine the parameter changes of the car bumper modeled by AI. The corresponding impact points are determined by the pre-set collision angles in the car bumper parameter data package modeled by AI. The logic for obtaining the distance between each impact point and the car door is to establish a car coordinate system based on the geometric center of the car and set a car door reference point. The spatial coordinates of the impact point and the spatial coordinates of the car door reference point are obtained in the car coordinate system. The distance between each impact point and the car door is calculated using the Euclidean distance formula. Optionally, the specific implementation is as follows: Using the vehicle's geometric center as a reference, the vehicle is divided equally according to the number of doors. For example, for a two-door vehicle, the vehicle is divided equally using its geometric center, resulting in a right half and a left half. The right half corresponds to the right door and the left door, respectively. Furthermore, for a four-door vehicle, the vehicle is divided equally using both its geometric center and the number of doors, resulting in a front right quarter, a front left quarter, a rear right quarter, and a rear left quarter. Each quarter corresponds to the front right, front left, rear right, and rear left doors, etc. Therefore, it can be seen that the aforementioned collisions... The distance between the impact point and the car door refers to the distance between the spatial coordinates of the car door corresponding to each part after equal division. The impact points are also divided according to the equal division results. For example, if the equal division results are the right front quarter of the car, the left front quarter of the car, and the right rear quarter of the car, and the impact point is located in the left front quarter of the car, then the impact point is divided into the left front quarter of the car and its spatial coordinates are calculated. The Euclidean distance between the spatial coordinates of the impact point and the spatial coordinates of the reference point of the left front door is calculated to obtain the distance between each impact point and the car door, etc. The specific implementation method or division method is not limited. This example is only given as an implementation means and will not be elaborated here. Specifically, the door reference point is determined by the researchers based on the door area and door hinge installation coordinates in the AI-modeled car bumper parameter data package. The door geometric boundary parameters are used to limit the boundary range of the door in the vehicle coordinate system. The hinge installation coordinates are used to locate the reference point of the door under the geometric center. The spatial coordinates of the door reference point are obtained by combining the door area. Furthermore, the Euclidean distance formula is common knowledge to those skilled in the art, and the specific implementation method is not limited and will not be elaborated here; The distances between each impact point and the car door are standardized so that the values ​​of the distances between each impact point and the car door are expressed between 0 and 1. It should be noted that the standardization methods include, but are not limited to, standard linear transformation based on interval scaling, statistical Z-Score standardization method, or normalization method based on nonlinear mapping function. The application methods of standardization will not be elaborated here. The standardized distances between each impact point and the door are substituted into the side-angle impact rating calculation formula, as follows: The side-angle impact rating is obtained by subtracting the standardized distances between each impact point and the door from 1, as shown in the following formula: ; In the formula, The distance between each impact point and the car door after standardization. Let i be the side impact score corresponding to the i-th collision angle, i = 1, 2, ..., n, where n is the total number of collision angles and i is the i-th collision angle; Among them, the side impact score corresponds to each impact angle, so it can be seen that the number of impact angles is the same as the number of side impact scores. In step S2, the simulated impact experiment based on the collision angle refers to the controlled simulation test of the front or side front area of ​​the car based on multiple sets of collision angles preset in the car bumper parameter data package modeled by AI. Among them, the simulated impact experiment uses a virtual collision body (such as a standardized rigid barrier or a deformable movable barrier) to conduct multiple impact experiments at a fixed speed and with a fixed mass, only changing the angle (i.e. the collision angle) with the vehicle's geometric center line. This generates the corresponding difference in deformation and pressure between the front and rear bumpers under different angle conditions. It should be noted that simulated impact experiments were conducted for each collision angle, focusing on the four corner vertices of the front and rear bumpers, namely the left front vertex, right front vertex, left rear vertex, and right rear vertex. The four corner vertices were defined as two pairs of opposite corners, namely the left front and right rear vertex as one pair of opposite corners, and the right front and left rear vertex as the other pair of opposite corners. For the two pairs of opposite corners, the difference in deformation and the difference in compressive force of the front and rear bumper opposite corners at each collision angle were analyzed. The logic for obtaining the difference in deformation between the front and rear bumpers at the diagonal is as follows: under the collision angle, two preset diagonal combinations are selected respectively. For each diagonal combination, the maximum deformation of the corresponding corner point under the collision angle is obtained. The absolute difference of each diagonal combination is calculated to obtain the difference in deformation of each diagonal combination and then accumulated to obtain the difference in deformation between the front and rear bumpers at the diagonal. Specifically, the two sets of opposite corners have been described above and will not be repeated here; Furthermore, if the left front and right rear vertices and the diagonal points opposite the right front and left rear vertices are selected as diagonal combinations, then the formula for calculating the deformation difference of each diagonal combination is: ; In the formula, Let A be the difference in deformation between the diagonal combinations. Let B be the difference in deformation between the two diagonal combinations. In the collision angle The maximum deformation of the lower left front vertex In the collision angle The maximum deformation of the bottom right rear vertex In the collision angle The maximum deformation of the lower right front vertex. In the collision angle The maximum deformation of the bottom left rear vertex; The maximum deformation refers to the maximum value of the displacement vector magnitude of the vertex in the three-dimensional coordinate system during the entire collision process. That is, the initial coordinates of the vertex are recorded at the beginning of the collision simulation time and the deformed coordinates of the vertex are recorded after the collision simulation time ends. The maximum deformation is obtained by calculating the Euclidean distance between the deformed coordinates and the initial coordinates. Furthermore, the deformation difference between the diagonal combinations of A and B is accumulated to calculate the deformation difference between the front and rear bumpers. The logic for obtaining the difference in extrusion force between the front and rear bumpers diagonally is as follows: under the collision angle, two preset diagonal combinations are selected respectively. For each diagonal combination, the peak extrusion force of the corresponding corner point under the collision angle is obtained. The absolute difference of the peak extrusion force of each diagonal combination is calculated to obtain the extrusion force difference of each diagonal combination. The results are then accumulated to obtain the difference in extrusion force between the front and rear bumpers diagonally. Specifically, peak compressive force refers to the maximum value of the contact force transmitted along the normal direction at the apex during the entire collision process. In the AI ​​modeling scenario, the normal contact force sequence during the collision simulation process is sampled time by time, the inner product of the contact force vector and the unit normal vector at each time is calculated, and the maximum value of the inner product is selected over the entire collision simulation time. This maximum value is taken as the peak compressive force. The difference in deformation and the difference in compressive force between the front and rear bumpers diagonally are standardized to keep them under the same dimension and to express their values ​​between 0 and 1. Specifically, the standardization process has been described above and will not be repeated here. The diagonal response score is calculated by multiplying the difference in deformation and the difference in compressive force between the front and rear bumpers after standardization. It should be noted that since the difference in deformation and the difference in compressive force between the front and rear bumpers are standardized so that their values ​​are between 0 and 1, it can be determined that the diagonal response score is dimensionless and has a value between 0 and 1, with its maximum value being 1. For example, even if the difference in deformation and the difference in compressive force between the front and rear bumpers are both at the maximum value of 1, the product calculation result is still 1, which will not be elaborated here. Substituting the diagonal response score and the side-angle impact score into a multinomial regression calculation, the collision angle collision coefficient is obtained, and the specific formula is expressed as follows: ; In the formula, Let i be the collision coefficient at the i-th collision angle. The diagonal response score is given for the i-th collision angle. The side-angle impact score corresponds to the i-th collision angle. To adjust the parameters, as well as The weighting coefficients for the diagonal response score and the side impact score are i = 1, 2, ..., n, where n is the total number of collision angles and i is the i-th collision angle. It should be noted that the order and form of the polynomial regression calculation formula in this invention can be selected according to the actual application requirements. No specific order and function form are limited here. Those skilled in the art can determine them based on the characteristics of the AI ​​modeling scenario, which will not be elaborated here. The weight coefficients mentioned in the multinomial regression calculation formula were obtained by our researchers based on the principle of minimizing the loss function and corresponding optimization algorithms (such as ordinary least squares, gradient descent, etc.). This is common knowledge to those skilled in the art and will not be elaborated here. It should be noted that when the diagonal response score and the side impact score are larger, it indicates that the collision deformation and force characteristics are more significant, the collision risk is higher, the collision coefficient of the collision angle is larger, and it is more likely to be marked as a side impact. Conversely, when both the diagonal response score and the side impact score are smaller, it indicates that the vehicle mainly exhibits axial frontal force characteristics, the collision coefficient is smaller, and it is more likely to be marked as a frontal impact. The collision coefficients at each collision angle are compared with the preset side threshold. If the collision angle collision coefficient exceeds the side threshold, the current collision angle is marked as a side impact; If the collision angle collision coefficient is lower than the side threshold, the current collision angle is marked as a frontal collision and imported into the bumper parameter confirmation mechanism. It should be noted that the side impact threshold was set by our researchers based on the vehicle's structural geometry and historical crash test data, and will not be elaborated upon here. Furthermore, the collision angle marked as a frontal impact is imported into the bumper parameter confirmation mechanism to dynamically correct and confirm the bumper's performance parameters under frontal collision conditions, which will not be elaborated here. In step S3, the side impact tags are retrieved and their number is counted. Based on the side impact tags, it is selected whether to introduce parameter supplementation rules. After conducting simulated impact experiments at various collision angles, the force on the front and rear bumpers during each simulated collision is recorded based on the comparison between the collision angle collision coefficient and the preset side threshold, and a corresponding side impact label is generated to identify whether the collision angle belongs to a side collision. After obtaining the side impact labels, the number of collisions marked as side impacts in the simulation experiment is counted and recorded as the number of side impact labels, which is used to reflect the frequency of side force on the bumper under different collision angles.

[0023] The number of side-impact tags is compared with a preset threshold: If the number of side-impact tags exceeds the preset threshold, then the parameter supplementation rule will be applied. If the number of side-impact tags is lower than the preset threshold, the parameter supplementation rule will not be applied. It should be noted that the preset quantity threshold is a critical standard used to determine whether to introduce parameter supplementation rules. By collecting a large amount of simulation data from vehicle crash tests and statistically analyzing the data distribution of the number of side collisions for different vehicles, the number of side collisions at which the recovery ability of the vehicle bumper begins to show a decreasing trend is determined as the quantity threshold. For example, when the number of side collisions reaches 5, the average recovery ability of the bumper decreases by more than 10%, so the quantity threshold is preset to 5. That is, when the number of side impact tags in the simulation experiment exceeds 5, it is determined that the parameter supplementation rules should be introduced.

[0024] When supplementing the input parameters, the maximum deformation of the front and rear bumpers during the collision and the residual deformation after the collision object is unloaded are collected based on the simulated impact experiment. Among them, the maximum deformation represents the maximum instantaneous displacement value of the bumper at each vertex under the impact, reflecting the instantaneous deformation capability of the bumper; the residual deformation after the impact object is unloaded represents the deformation value that the bumper still retains after the impact object is unloaded, which is used to evaluate the shape recovery capability of the bumper. Based on the maximum deformation and the residual deformation after the impact object is unloaded, the deformation recovery rate of the bumper is calculated using the following formula: ; in, For deformation recovery rate, For the maximum deformation, This represents the residual deformation after the colliding object is unloaded.

[0025] Specifically, the maximum instantaneous displacement value of each vertex is recorded at each collision. After the simulated impact experiment is completed, the maximum value is selected as the maximum deformation based on the statistical maximum instantaneous displacement value of each vertex. Furthermore, when the maximum deformation is obtained, the deformation value that the bumper still retains after the collision object is unloaded is taken as the residual deformation after the collision object is unloaded. The deformation recovery rate reflects the fatigue characteristics of the bumper material and structure. The value ranges from 0 to 1. The closer the value is to 1, the stronger the bumper's recovery ability after the collision is unloaded. The closer the value is to 0, the greater the residual deformation of the bumper and the weaker its recovery ability.

[0026] The fatigue compensation coefficient is further calculated by combining the number of side impact tags and the deformation recovery rate. The specific calculation formula is as follows: ; in, For fatigue compensation coefficient, The number of tags for side impact. This is a threshold value used to normalize the number of side impact tags. For deformation recovery rate, and , which is a weighting coefficient used to adjust the influence of the number of side impact tags and deformation recovery rate on the fatigue compensation coefficient.

[0027] Specifically, the fatigue compensation coefficient has only one numerical expression. Since the deformation recovery rate is obtained by filtering each vertex in the deformation recovery rate calculation process (i.e., selecting the maximum value as the maximum deformation based on the maximum instantaneous displacement value of each vertex), the deformation recovery rate has only one vertex. Therefore, the number of numerical expressions of the fatigue compensation coefficient is the same as the number of numerical expressions of the deformation recovery rate, which will not be elaborated here. It should be noted that the weighting coefficients are set through regression analysis of historical crash test data. This compares the contributions of the number of side impacts and the deformation recovery rate to the cumulative residual deformation of the bumper and the material fatigue performance, thereby determining the range of values ​​for the weighting coefficients. For example, in a set of experimental data, if the increase in the number of side impacts accounts for 60% of the overall fatigue performance, while the decrease in the deformation recovery rate accounts for 40%, then the weighting coefficients will be determined accordingly. Set to 0.6. Set to 0.4.

[0028] The fatigue compensation coefficient reflects the fatigue effect that the bumper may experience due to repeated stress under multiple side impacts. The larger the number of side impact tags, the lower the deformation recovery rate and the larger the fatigue compensation coefficient, thus indicating that the bumper parameter confirmation mechanism needs to introduce fatigue influencing factors in subsequent parameter optimization. When the number of side impact tags is smaller, the deformation recovery rate is higher and the fatigue compensation coefficient is smaller, it means that no additional parameters need to be introduced.

[0029] In step S4, the deformation recovery rate is determined as a parameter to be introduced based on the fatigue compensation coefficient, and the deformation recovery rate is processed and screened before being imported into the bumper parameter confirmation mechanism. The fatigue compensation coefficient is determined based on a preset fatigue assessment threshold. When the fatigue compensation coefficient exceeds the fatigue judgment threshold, it indicates that the bumper has a fatigue accumulation effect under multiple collision conditions, and the deformation recovery rate is determined to be introduced as a parameter. When the fatigue compensation coefficient is lower than the fatigue judgment threshold, it indicates that the fatigue accumulation effect of the bumper is not significant, and the deformation recovery rate is not to be included as a parameter.

[0030] It should be noted that the fatigue judgment threshold is used as a benchmark to determine whether the fatigue compensation coefficient has reached a significant level. Its value is set by collecting a large amount of experimental data on bumpers under different collision numbers and deformation recovery rates. By comparing the cumulative residual deformation of the bumper structure under different coefficient ranges, the critical point at which the bumper performance begins to decline significantly is determined as the fatigue judgment threshold. For example, in a set of statistical experiments, when the fatigue compensation coefficient reaches 0.8, the residual deformation of the bumper exceeds 25% of the maximum deformation on average. Therefore, the fatigue judgment threshold is preset to 0.8. That is, when the fatigue compensation coefficient exceeds 0.8, the deformation recovery rate is determined to be the parameter to be introduced.

[0031] When determining whether to use deformation recovery rate as a parameter to be introduced, the deformation recovery rate is processed and filtered as follows: The deformation recovery rates from multiple simulated impact experiments were constructed into a deformation recovery rate dataset. Obtain the median of the deformation recovery rate dataset to get the median deformation recovery rate; Calculate the absolute value of the deviation between each deformation recovery rate and the median, and construct a set of deviations; Obtain the median of the deviation set to get the deviation median; The anomaly determination value for each deformation recovery rate is obtained by subtracting the median of each deformation recovery rate from the median of the deviations. The anomaly detection value is compared with the preset deviation coefficient threshold: If the abnormal judgment value is greater than the preset deviation coefficient threshold, the deformation recovery rate is determined to be an abnormal value and is removed. Conversely, if the deformation recovery rate is not an outlier, it is determined to be retained. It should be noted that the deviation coefficient threshold is a benchmark used to determine whether the deformation recovery rate is an outlier. According to statistical standards, the deviation coefficient threshold can be 2.5 or 3. For example, in a set of experimental data, the calculated outlier judgment value of the deformation recovery rate is 3.2. When the deviation coefficient threshold is preset to 3, the deformation recovery rate is judged as an outlier and removed; if the outlier judgment value is 2.1, it is retained.

[0032] After outlier removal, the arithmetic mean of the remaining deformation recovery rates is calculated to obtain the deformation recovery rate after screening. The deformation recovery rate after screening is imported into the bumper parameter confirmation mechanism. The bumper parameter confirmation mechanism adds the deformation recovery rate as an additional parameter input on the basis of the original parameter system, which is used to correct and optimize the performance parameter feedback of the bumper under low speed and side collision conditions.

[0033] By incorporating deformation recovery rate into the verification mechanism, the accuracy of bumper parameters under complex collision scenarios is improved, ensuring that the final parameter determination results are more in line with actual usage conditions.

[0034] Example 2 Please see Figure 2 A car bumper parameter determination system based on AI modeling includes an impact scoring module, an angle evaluation module, a coefficient calculation module, and a filtering and import module. The functions of each module are as follows: The impact scoring module is used to call the AI-modeled car bumper parameter data package and the bumper parameter confirmation mechanism, set the collision angle, select the distance between each impact point and the car door, and calculate the side angle impact score. The angle evaluation module is used to simulate impact experiments based on the collision angle, obtain the difference in deformation and the difference in compressive force between the front and rear bumpers to obtain the diagonal response score, and combine the side angle impact score to obtain the collision angle collision coefficient and mark each collision angle as a side impact or a frontal impact. The coefficient calculation module is used to call the side impact tags and count the number to select whether to enter the parameter supplementation rules. Based on the maximum deformation of the bumper after each simulation and the residual deformation after the collision object is unloaded, the deformation recovery rate of the bumper is calculated, and the fatigue supplementation coefficient is obtained by combining the number of side impact tags. The filter and import module is used to analyze whether to include the deformation recovery rate as a parameter to be imported based on the fatigue compensation coefficient. After processing and filtering the deformation recovery rate, it is imported into the bumper parameter confirmation mechanism.

[0035] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0036] Finally, it should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0037] Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0038] In this document, the singular forms “a,” “an,” and “the” may also include the plural forms unless the context clearly indicates otherwise. It should also be understood that terms such as “comprising / including” or “having” specify the presence of the stated features, integrals, steps, operations, components, parts, or combinations thereof, but do not preclude the possibility of the presence or addition of one or more other features, integrals, steps, operations, components, parts, or combinations thereof. Meanwhile, the term “and / or” as used in this specification includes any and all combinations of the associated listed items.

[0039] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0040] The above description of the disclosed embodiments will enable those skilled in the art to make or use various modifications to these embodiments. It will be readily apparent to those skilled in the art that the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for determining automobile bumper parameters based on AI modeling, characterized in that: Includes the following steps: Step S1: Call the AI-modeled car bumper parameter data package and bumper parameter confirmation mechanism, set the collision angle, select the distance between each impact point and the door, and calculate the side angle impact score; Step S2: Conduct a simulated impact experiment based on the collision angle, obtain the difference in deformation and the difference in compressive force between the front and rear bumpers to obtain the diagonal response score, combine the side angle impact score to obtain the collision angle collision coefficient, and mark each collision angle as a side impact or a frontal impact. Step S3: Call the side impact tags and count the number. Select whether to enter the parameter supplementation rule. Calculate the deformation recovery rate of the bumper based on the maximum deformation of the bumper after each simulation and the residual deformation after unloading the collision object, and obtain the fatigue supplementation coefficient by combining the number of side impact tags. Step S4: Based on the fatigue compensation coefficient analysis, determine whether to include the deformation recovery rate as a parameter to be introduced. After processing and screening the deformation recovery rate, import it into the bumper parameter confirmation mechanism.

2. The method for determining car bumper parameters based on AI modeling according to claim 1, characterized in that: In step S1, the AI-modeled car bumper parameter data package and bumper parameter confirmation mechanism are invoked through the control command interface; AI-modeled car bumper parameter data package refers to a multi-dimensional parameter set generated during the training phase based on finite element simulation data, collision test data, and vehicle geometry data. The bumper parameter confirmation mechanism refers to a set of calculation rules used to retrieve, call, and dynamically correct parameters in a data packet under different collision conditions.

3. The method for determining car bumper parameters based on AI modeling according to claim 2, characterized in that: In step S1, the corresponding impact points are determined by the pre-set collision angles in the car bumper parameter data package modeled by AI. The spatial coordinates of the impact point and the spatial coordinates of the door reference point are obtained in the vehicle coordinate system. The distance between each impact point and the door is calculated using the Euclidean distance formula. Subtracting the standardized distances between each impact point and the car door from 1 yields the side corner impact score, calculated using the following formula: ； In the formula, The distance between each impact point and the car door after standardization. Let i be the side impact score corresponding to the i-th collision angle, i = 1, 2, ..., n, where n is the total number of collision angles and i is the i-th collision angle.

4. The method for determining car bumper parameters based on AI modeling according to claim 1, characterized in that: In step S2, based on the multiple sets of collision angles preset in the car bumper parameter data package modeled by AI, a controlled simulation test is performed on the front or side front area of ​​the car, and the corresponding deformation difference and compressive force difference of the front and rear bumpers are generated. The diagonal response score is calculated by multiplying the difference in deformation and the difference in compressive force between the front and rear bumpers after standardization.

5. The method for determining car bumper parameters based on AI modeling according to claim 4, characterized in that: In step S2, the diagonal response score and the side-angle impact score are substituted into the polynomial regression calculation to obtain the collision angle collision coefficient. If the collision angle collision coefficient exceeds the side threshold, the current collision angle is marked as a side impact; If the collision angle collision coefficient is lower than the side threshold, the current collision angle is marked as a frontal impact and imported into the bumper parameter confirmation mechanism.

6. The method for determining automobile bumper parameters based on AI modeling according to claim 5, characterized in that: In step S3, the number of collisions marked as side impacts in the simulation experiment is counted and recorded as the number of side impact tags; If the number of side-impact tags exceeds the preset threshold, then the parameter supplementation rule will be applied. If the number of side-impact tags is lower than the preset threshold, the parameter supplementation rule will not be applied.

7. The method for determining automobile bumper parameters based on AI modeling according to claim 1, characterized in that: In step S3, when the parameter supplementation rule is passed, the maximum deformation of the front and rear bumpers during the collision and the residual deformation after the collision object is unloaded are collected according to the simulated impact experiment. The maximum deformation represents the maximum instantaneous displacement of the bumper at each vertex under the impact of a collision; The residual deformation after the collision object is unloaded represents the deformation value that the bumper still retains after the collision object is unloaded. The deformation recovery rate of the bumper is calculated based on the maximum deformation and the residual deformation after the collision object is unloaded. The fatigue compensation coefficient was calculated by combining the number of side impact tags and the deformation recovery rate.

8. The method for determining car bumper parameters based on AI modeling according to claim 7, characterized in that: In step S4, when the fatigue compensation coefficient exceeds the fatigue judgment threshold, the deformation recovery rate is determined to be introduced as a parameter. When the fatigue compensation coefficient is lower than the fatigue judgment threshold, the deformation recovery rate will not be used as a parameter to be introduced. When determining whether to use deformation recovery rate as a parameter to be introduced, the deformation recovery rate is processed and filtered.

9. The method for determining automobile bumper parameters based on AI modeling according to claim 8, characterized in that: In step S4, the median absolute deviation algorithm is used to calculate the anomaly determination value of each deformation recovery rate; The deformation recovery rates of those with abnormal judgment values ​​lower than or equal to the preset deviation coefficient threshold are selected and their arithmetic average is calculated to obtain the deformation recovery rates after screening. The deformation recovery rate after screening is imported into the bumper parameter confirmation mechanism; The bumper parameter confirmation mechanism adds deformation recovery rate as an additional parameter input to the existing parameter system.

10. A system for determining automobile bumper parameters based on AI modeling, used to implement the method for determining automobile bumper parameters based on AI modeling as described in any one of claims 1-9, characterized in that: It includes an impact rating module, an angle evaluation module, a coefficient calculation module, and a filtering and import module. The functions of each module are as follows: The impact scoring module is used to call the AI-modeled car bumper parameter data package and the bumper parameter confirmation mechanism, set the collision angle, select the distance between each impact point and the car door, and calculate the side angle impact score. The angle evaluation module is used to simulate impact experiments based on the collision angle, obtain the difference in deformation and the difference in compressive force between the front and rear bumpers to obtain the diagonal response score, and combine the side angle impact score to obtain the collision angle collision coefficient and mark each collision angle as a side impact or a frontal impact. The coefficient calculation module is used to call the side impact tags and count the number to select whether to enter the parameter supplementation rules. Based on the maximum deformation of the bumper after each simulation and the residual deformation after the collision object is unloaded, the deformation recovery rate of the bumper is calculated, and the fatigue supplementation coefficient is obtained by combining the number of side impact tags. The filter and import module is used to analyze whether to include the deformation recovery rate as a parameter to be imported based on the fatigue compensation coefficient. After processing and filtering the deformation recovery rate, it is imported into the bumper parameter confirmation mechanism.

Images