A method and device for evaluating bionic performance of automobile collision dummy neck

By using the neck calibration device to collect data and build an optimized performance curve model in a car collision dummy, the evaluation index of the neck calibration test is solved, and the problems of low accuracy and poor stability of the neck calibration test evaluation method in the prior art are solved. , achieving more accurate evaluation of neck bionic performance.

CN118520262BActive Publication Date: 2024-11-05CHINA AUTOMOTIVE TECH & RES CENT CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202410995464.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-24
Publication Date
2024-11-05
Estimated Expiration
2044-07-24

AI Technical Summary

Technical Problem

The evaluation method of the existing automobile collision dummy neck calibration test is low in accuracy, unable to effectively display dynamic performance, and the initial conditions are different, resulting in unstable results.

Method used

The neck calibration device is used to collect standard time series data, divide it into training set, verification set and test set, build and optimize the neck calibration test performance curve model, calculate evaluation indicators through the D plane angle model and the occipital moment model, and determine the false car collision Bionic performance of human neck.

Benefits of technology

It improves the accuracy and stability of the evaluation of the neck bionic performance of the car collision dummy, can more accurately reflect the dynamic performance of the neck, and reduces the error caused by differences in initial conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118520262B_ABST
    Figure CN118520262B_ABST
Patent Text Reader

Abstract

The present application discloses a method and device for evaluating the bionic performance of the neck of an automobile collision dummy. The method comprises the following steps: collecting standard time series data of a standard automobile collision dummy through experiments and dividing multiple groups of standard time series data into a training set, a validation set and a test set, so as to build, train and optimize a neck calibration test performance curve model and calculate a standard evaluation index. The time series data to be tested of the automobile collision dummy to be tested is also collected through experiments, and the evaluation index to be tested of the automobile collision dummy to be tested is calculated in combination with the neck calibration test performance curve model. The bionic performance of the neck of the automobile collision dummy to be tested is determined according to the evaluation index to be tested and the standard evaluation index. The model is built using the standard automobile collision dummy to improve the stability and accuracy of the standard evaluation index, thereby improving the evaluation accuracy of the bionic performance of the neck of the automobile collision dummy.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of automobile collision dummy neck bionics, and in particular to a method and device for evaluating the bionic performance of an automobile collision dummy neck. Background Art

[0002] The automobile crash dummy is the main tool for testing automobile safety. In order to ensure that the automobile crash dummy can restore the movement posture and injury of the real human body to the greatest extent during the collision test, it is necessary to calibrate each part of the automobile crash dummy according to the regulations to determine the bionic performance of the dummy. The existing automobile crash dummy neck calibration tests mainly include neck bending, neck extension, neck twisting, and neck lateral bending.

[0003] The existing evaluation method is to compare the difference between the average mechanical response curve of the car crash dummy and the average mechanical response curve of the real human body, and then judge the calibration result of the car crash dummy. However, the representative curve constructed by this method only retains the average change relationship of the mechanical response index over time, with low accuracy, and cannot well show the dynamic performance of the curve. In actual operation, the initial conditions of each calibration test are often different. Therefore, a method is needed to accurately evaluate the bionic performance of the neck of the car crash dummy. Summary of the invention

[0004] In order to solve the above technical problems, the present application is proposed. The embodiments of the present application provide a method and device for evaluating the bionic performance of the neck of a car collision dummy.

[0005] According to one aspect of the present application, a method for evaluating the bionic performance of the neck of an automobile collision dummy is provided, comprising: using a neck calibration device to experiment on the head and neck assembly of a standard automobile collision dummy to collect multiple groups of standard time series data corresponding to the standard automobile collision dummy; dividing the multiple groups of standard time series data into a training set, a validation set and a test set; building and training a neck calibration test performance curve model based on the training set; optimizing the parameters of the neck calibration test performance curve model based on the validation set to obtain a final neck calibration test performance curve model; inputting the test set into the final neck calibration test performance curve model to calculate a standard evaluation index; using the neck calibration device to experiment on the head and neck assembly of a to-be-tested automobile collision dummy to collect multiple groups of to-be-tested time series data corresponding to the to-be-tested automobile collision dummy; inputting the to-be-tested time series data into the final neck calibration test performance curve model to calculate a to-be-tested evaluation index; and determining the bionic performance of the neck of the to-be-tested automobile collision dummy based on the to-be-tested evaluation index and the standard evaluation index.

[0006] In one embodiment, the final neck calibration test performance curve model includes a D-plane angle model and an occipital moment model; wherein, inputting the test set into the final neck calibration test performance curve model to calculate the standard evaluation index includes: inputting the test set into the D-plane angle model and the occipital moment model, and respectively calculating the standard D-plane angle residual and the standard occipital moment residual; calculating the standard evaluation index based on the standard D-plane angle residual and the standard occipital moment residual.

[0007] In one embodiment, the standard evaluation index is calculated based on the standard D plane angle residual and the standard occipital moment residual, and includes: the standard evaluation index H The calculation formula for 1 is:

[0008] ;

[0009] in, , , is the statistic of the i-th sample point, is the weight of the historical residual to the current statistic, A is the overall covariance matrix of the residual sequence, is the population mean vector of residuals, , It is a two-dimensional residual vector composed of the standard D-plane angle residual and the standard occipital moment residual.

[0010] In one embodiment, the step of inputting the time series data to be measured into the final neck calibration test performance curve model to calculate the evaluation index to be measured comprises: inputting the time series data to be measured into the D-plane angle model and the occipital moment model to respectively calculate the D-plane angle residual to be measured and the occipital moment residual to be measured; and calculating the evaluation index to be measured based on the D-plane angle residual to be measured and the occipital moment residual to be measured.

[0011] In one embodiment, the evaluation index to be measured is calculated based on the D plane angle residual to be measured and the occipital moment residual to be measured, and includes: the evaluation index to be measured H The calculation formula for 2 is:

[0012] ;

[0013] in, , , is the statistic of the i-th sample point, is the weight of the historical residual to the current statistic, A is the overall covariance matrix of the residual sequence, is the population mean vector of residuals, , It is a two-dimensional residual vector composed of the D-plane angle residual to be measured and the occipital moment residual to be measured.

[0014] In one embodiment, the standard automobile collision dummies include multiple ones; wherein, determining the neck bionic performance of the automobile collision dummy to be tested based on the evaluation index to be tested and the standard evaluation index includes: calculating the average value of the standard evaluation indexes corresponding to the multiple standard automobile collision dummies to obtain an average standard evaluation index; and determining the neck bionic performance of the automobile collision dummy to be tested based on the evaluation index to be tested and the average standard evaluation index.

[0015] In one embodiment, the standard time series data includes time, speed characteristics, a first acceleration characteristic, and a second acceleration characteristic; wherein, collecting multiple sets of standard time series data corresponding to the standard automobile collision dummy includes: collecting multiple sets of time, speed characteristics, a first acceleration characteristic, and a second acceleration characteristic corresponding to the standard automobile collision dummy; wherein, the calculation formula of the speed characteristic is:

[0016] ;

[0017] The calculation formula of the first acceleration characteristic is:

[0018] ;

[0019] The calculation formula of the second acceleration characteristic is:

[0020] ;

[0021] Among them, the vector It is all the data collected by the calibration test at a certain moment. It is A variable in a vector used to store speed information, This is the first The speed at which the swing arm grating hole passes through the photoelectric gate, is the number of swing arm gratings, is the first acceleration feature, is the second acceleration characteristic, is the time when the pendulum acceleration curve first reaches its peak value, is the time corresponding to the pendulum acceleration returning to 0. The pendulum acceleration curve is The acceleration corresponding to the moment.

[0022] In one embodiment, before dividing the multiple sets of standard time series data into a training set, a validation set, and a test set, the method for evaluating the bionic performance of the neck of a car collision dummy further includes: normalizing the multiple sets of standard time series data to obtain multiple sets of normalized standard time series data; wherein the normalization processing formula is:

[0023] ;

[0024] in, is the parameter value before normalization, is the maximum value of the parameter. is the minimum value of the parameter. is a normalized value; dividing the multiple groups of standard time series data into a training set, a validation set and a test set includes: dividing the multiple groups of normalized standard time series data into a training set, a validation set and a test set.

[0025] In one embodiment, the neck calibration test performance curve model is optimized based on the validation set to obtain the final neck calibration test performance curve model, including: inputting the validation set into the neck calibration test performance curve model to calculate the loss function; with the goal of minimizing the loss function, the neck calibration test performance curve model is optimized to obtain the final neck calibration test performance curve model.

[0026] According to another aspect of the present application, a device for evaluating the bionic performance of the neck of an automobile collision dummy is provided, comprising: a standard data acquisition module, used to use a neck calibration device to experiment on the head and neck assembly of a standard automobile collision dummy to collect multiple groups of standard time series data corresponding to the standard automobile collision dummy; a standard data division module, used to divide the multiple groups of standard time series data into a training set, a validation set and a test set; a curve model building module, used to build and train a neck calibration test performance curve model based on the training set; a curve model optimization module, used to optimize the parameters of the neck calibration test performance curve model based on the validation set to obtain a final neck calibration test performance curve model. a final neck calibration test performance curve model; a standard index calculation module, used for inputting the test set into the final neck calibration test performance curve model to calculate the standard evaluation index; a test data acquisition module, used for using the neck calibration device to experiment on the head and neck assembly of the automobile collision dummy to be tested, so as to collect multiple groups of time series data to be tested corresponding to the automobile collision dummy to be tested; a test index calculation module, used for inputting the time series data to be tested into the final neck calibration test performance curve model to calculate the evaluation index to be tested; and a test performance evaluation module, used for determining the neck bionic performance of the automobile collision dummy to be tested based on the evaluation index to be tested and the standard evaluation index.

[0027] The present application provides a method and device for evaluating the bionic performance of the neck of an automobile collision dummy. The method and device adopt a neck calibration device to experiment on the head and neck assembly of a standard automobile collision dummy to collect multiple groups of standard time series data corresponding to the standard automobile collision dummy; divide the multiple groups of standard time series data into a training set, a validation set and a test set; build and train a neck calibration test performance curve model based on the training set; optimize the parameters of the neck calibration test performance curve model based on the validation set to obtain a final neck calibration test performance curve model; input the test set into the final neck calibration test performance curve model to calculate and obtain standard evaluation indicators; adopt a neck calibration device to experiment on the head and neck assembly of a to-be-tested automobile collision dummy to collect multiple groups of to-be-tested time series data corresponding to the to-be-tested automobile collision dummy; input the to-be-tested time series data into the final neck calibration test performance curve model; energy curve model, and calculate the evaluation index to be tested; based on the evaluation index to be tested and the standard evaluation index, determine the neck bionic performance of the automobile collision dummy to be tested; that is, collect standard time series data of the standard automobile collision dummy through experiments and divide multiple groups of standard time series data into training set, verification set and test set to build, train and optimize the neck calibration test performance curve model and calculate the standard evaluation index; similarly, collect the time series data to be tested of the automobile collision dummy to be tested through experiments, and calculate the evaluation index to be tested of the automobile collision dummy to be tested in combination with the neck calibration test performance curve model; determine the neck bionic performance of the automobile collision dummy to be tested according to the evaluation index to be tested and the standard evaluation index; use the standard automobile collision dummy to build a model to improve the stability and accuracy of the standard evaluation index, thereby improving the evaluation accuracy of the neck bionic performance of the automobile collision dummy. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0029] Figure 1 It is a flow chart of a method for evaluating the bionic performance of a car collision dummy's neck provided by an exemplary embodiment of the present application.

[0030] Figure 2 It is a schematic diagram of a D-plane angle curve collected by a neck calibration device provided by an exemplary embodiment of the present application.

[0031] Figure 3 It is a schematic diagram of an occipital moment curve collected by a neck calibration device provided by an exemplary embodiment of the present application.

[0032] Figure 4 It is a schematic diagram of a prediction curve of a D-plane angle model provided by an exemplary embodiment of the present application.

[0033] Figure 5 It is a schematic diagram of a prediction curve of an occipital moment model provided by an exemplary embodiment of the present application.

[0034] Figure 6 It is a schematic diagram of the prediction curve residual of the D-plane angle model provided by an exemplary embodiment of the present application.

[0035] Figure 7 It is a schematic diagram of the residual error of the prediction curve of the occipital moment model provided by an exemplary embodiment of the present application.

[0036] Figure 8 It is a structural schematic diagram of a pendulum acceleration curve of a neck calibration device provided by an exemplary embodiment of the present application.

[0037] Fig. 9 It is a schematic structural diagram of a device for evaluating the bionic performance of a car collision dummy neck provided by an exemplary embodiment of the present application.

[0038] Fig.10 It is a structural diagram of an electronic device provided by an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0039] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0040] Figure 1 FIG. 1 is a flow chart of a method for evaluating the bionic performance of a car collision dummy neck provided by an exemplary embodiment of the present application. Figure 1 As shown, the method for evaluating the bionic performance of the neck of a car collision dummy comprises the following steps:

[0041] Step 110: Use a neck calibration device to conduct an experiment on the head and neck assembly of a standard automobile collision dummy to collect multiple groups of standard time series data corresponding to the standard automobile collision dummy.

[0042] Specifically, the neck calibration device in this application is a pendulum mounted on a test bench that strikes the honeycomb aluminum, and uses the pendulum to give an acceleration to the head and neck assembly of the standard automobile collision dummy, and measures the head and neck related data of the standard automobile collision dummy through a sensor to obtain the standard time series data corresponding to the standard automobile collision dummy (head and neck assembly). This application uses the neck calibration device to conduct multiple repeated experiments on the standard automobile collision dummy, and collects multiple groups of time series data including time, pendulum acceleration, pendulum speed, neck D plane angle, neck force, and occipital moment.

[0043] Step 120: Divide multiple sets of standard time series data into a training set, a validation set, and a test set.

[0044] After obtaining multiple sets of standard time series data, the multiple sets of standard time series data are divided into training sets, validation sets and test sets to realize the training, validation and testing functions of the model.

[0045] Step 130: Build and train a neck calibration test performance curve model based on the training set.

[0046] Specifically, the present application uses a PSO-XGBoost neural network to build a neck calibration test performance curve model. XGBoost is a machine learning algorithm based on decision trees and integration ideas. It determines the candidate set of split nodes through a distributed QuantileSketch, defines the scoring function of the tree structure, and can dynamically select the division of the tree split nodes through structural scoring. The present application uses a PSO-XGBoost neural network to build a neck calibration test performance curve model, and uses a training set to train the neck calibration test performance curve model.

[0047] Step 140: Optimize the parameters of the neck calibration test performance curve model based on the validation set to obtain a final neck calibration test performance curve model.

[0048] In one embodiment, the specific implementation method of the above step 140 may be: inputting the verification set into the neck calibration test performance curve model to calculate the loss function; with the goal of minimizing the loss function, optimizing the parameters of the neck calibration test performance curve model to obtain the final neck calibration test performance curve model.

[0049] After building and training the neck calibration test performance curve model, this application uses a validation set to verify the neck calibration test performance curve model and optimizes the parameters of the neck calibration test performance curve model based on the validation results to obtain the final neck calibration test performance curve model. Specifically, the XGBoost model predicts the sample results as follows:

[0050] ;

[0051] in, is the output of the model, For the The feature vector corresponding to the samples is is the leaf weight (indicating the Tree pair samples ), is the sum of all trees.

[0052] XGBoost is a greedy algorithm, and the new tree generated in each iteration is used to fit the residual of the previous tree. The predicted value at the iteration ,but:

[0053] ;

[0054] The loss function at this time is:

[0055] ;

[0056] in, is the complexity of the tree, is a constant.

[0057] This application uses the PSO particle swarm optimization algorithm to optimize the parameters of the XGBoost model. The PSO particle swarm optimization algorithm abstracts the feasible solution of each function to be optimized into particles. Each particle has the following elements: speed, position, and the optimal position experienced by itself. The movement of particles is constrained by both speed and position. This application uses an adaptive PSO algorithm that adaptively changes weights according to the number of iterations to achieve parameter optimization. The parameters of the XGBoost model optimized by the PSO algorithm include at least the following 6 types: learning_rate learning rate, n_estimators number of decision trees, max_depth maximum depth of decision trees, min_child_weight minimum leaf weight, reg_alpha L1 regularization term, reg_lambda L2 regularization term.

[0058] The objective function of the parameter optimization is the loss function of the XGBoost model fitting:

[0059] ;

[0060] in, is the model output, vector is the normalized model input feature, is a vector The true value of the model output variable contained in is the number of sample points.

[0061] The evaluation of the bionic performance of automobile collision dummies mainly focuses on the dynamic performance of the model output, such as the maximum value and the time corresponding to the maximum value. Therefore, in order to ensure the fitting performance of the XGBoost model in dynamic performance, this application designs the loss function in an exponential form to measure the difference between the predicted value and the actual value, so that the XGBoost model parameters can be optimized to focus more on the fitting effect of outliers, thereby improving the model accuracy.

[0062] Step 150: Input the test set into the final neck calibration test performance curve model to calculate the standard evaluation index.

[0063] Specifically, the fitting residual of the test set of this application is used as an evaluation index for evaluating the neck bionic performance of the automobile collision dummy. After obtaining the trained neck calibration test performance curve model, this application uses the test set to calculate the standard evaluation index, that is, the fitting residual between the true value of the test set and the output (predicted value) of the neck calibration test performance curve model, as a reference value for measuring the neck bionic performance of the automobile collision dummy to be tested.

[0064] Step 160: using a neck calibration device to conduct experiments on the head and neck assembly of the automobile collision dummy to be tested, so as to collect multiple groups of time series data corresponding to the automobile collision dummy to be tested.

[0065] Specifically, the present application uses the same neck calibration device to collect multiple groups of time series data to be tested corresponding to the automobile collision dummy (head and neck assembly) to be tested, wherein the time series data to be tested includes time, pendulum acceleration, pendulum velocity, neck D-plane angle, neck force, and occipital moment.

[0066] Step 170: Input the time series data to be tested into the final neck calibration test performance curve model to calculate the evaluation index to be tested.

[0067] After collecting multiple groups of time series data to be tested of the automobile collision dummy to be tested, the application inputs the time series data to be tested into the trained neck calibration test performance curve model to calculate the evaluation index to be tested corresponding to the time series data to be tested.

[0068] Step 180: Determine the neck bionic performance of the automobile collision dummy to be tested based on the evaluation index to be tested and the standard evaluation index.

[0069] After the evaluation indicators to be tested of the automobile collision dummy to be tested are calculated, the standard evaluation indicators and the evaluation indicators to be tested are compared to determine the bionic performance of the neck of the automobile collision dummy to be tested.

[0070] The present application provides a method for evaluating the bionic performance of the neck of an automobile collision dummy. The method comprises the following steps: using a neck calibration device to experiment on the head and neck assembly of a standard automobile collision dummy to collect multiple groups of standard time series data corresponding to the standard automobile collision dummy; dividing the multiple groups of standard time series data into a training set, a validation set, and a test set; building and training a neck calibration test performance curve model based on the training set; optimizing the parameters of the neck calibration test performance curve model based on the validation set to obtain a final neck calibration test performance curve model; inputting the test set into the final neck calibration test performance curve model to calculate and obtain standard evaluation indicators; using a neck calibration device to experiment on the head and neck assembly of a to-be-tested automobile collision dummy to collect multiple groups of to-be-tested time series data corresponding to the to-be-tested automobile collision dummy; inputting the to-be-tested time series data into the final neck calibration test performance curve model; and The method comprises the following steps: a linear model is used to calculate the evaluation index to be tested; based on the evaluation index to be tested and the standard evaluation index, the neck bionic performance of the automobile collision dummy to be tested is determined; that is, standard time series data of the standard automobile collision dummy are collected through experiments, and multiple groups of standard time series data are divided into a training set, a validation set and a test set to build, train and optimize the neck calibration test performance curve model and calculate the standard evaluation index; the time series data to be tested of the automobile collision dummy to be tested are collected through experiments, and the evaluation index to be tested of the automobile collision dummy to be tested is calculated in combination with the neck calibration test performance curve model; the neck bionic performance of the automobile collision dummy to be tested is determined according to the evaluation index to be tested and the standard evaluation index; the model is built using the standard automobile collision dummy to improve the stability and accuracy of the standard evaluation index, thereby improving the evaluation accuracy of the neck bionic performance of the automobile collision dummy.

[0071] In one embodiment, the final neck calibration test performance curve model includes a D-plane angle model and an occipital moment model; wherein, the specific implementation method of the above step 150 may be: inputting the test set into the D-plane angle model and the occipital moment model, and respectively calculating the standard D-plane angle residual and the standard occipital moment residual; based on the standard D-plane angle residual and the standard occipital moment residual, calculating the standard evaluation index.

[0072] In the neck bending test, the present application determines whether the neck of the automobile collision dummy is qualified based on two variables: the D-plane angle and the occipital moment. Therefore, the present application establishes two regression models with the D-plane angle and the occipital moment as outputs, respectively, to describe the performance of the neck of the automobile collision dummy. Among them, each regression model corresponds to an output variable and several input variables, and the input variables include initial characteristics, time, and output variables corresponding to another regression model. Exemplarily, in the D-plane angle model of the neck bending test, the input variables include velocity characteristics, first acceleration characteristics, second acceleration characteristics, time characteristics, and occipital moment characteristics (the output of the occipital moment model), and the output variable is the D-plane angle curve (such as Figure 2The input variables of the occipital moment model include velocity feature, first acceleration feature, second acceleration feature, time feature and D plane angle feature, and the output variable is the occipital moment curve (as shown in Figure 3 As shown). This application inputs the test set into the D-plane angle model and the occipital moment model, and the models output the D-plane angles corresponding to the test set (as shown in Figure 4 ) and the occipital moment curve (as Figure 5 The standard D-plane angle residual (as shown in Figure 2) is calculated based on the difference between the D-plane angle output by the model and the actual D-plane angle, and the difference between the occipital moment output by the model and the actual occipital moment. Figure 6 ) and the standardized occipital moment residual (as Figure 7 As shown in Figure 2), the standard evaluation index is then calculated based on the standard D-plane angle residual and the standard occipital moment residual. Specifically, the calculation formula for the fitting residual of each test set data is as follows:

[0073] ;

[0074] in, is the two-dimensional residual vector, i is the sample point number, is the true value of the D plane angle variable in the test set data, is the predicted value of the D plane angle given by the performance curve model, is the true value of the occipital moment variable in the test set data, Predicted values ​​of occipital moment given by the performance curve model.

[0075] In one embodiment, the specific calculation method of the standard evaluation index can be: H The calculation formula for 1 is:

[0076] ;

[0077] in, , , is the statistic of the i-th sample point, is the weight of the historical residual to the current statistic, A is the overall covariance matrix of the residual sequence, is the population mean vector of residuals, , is a two-dimensional residual vector consisting of the standard D-plane angle residual and the standard occipital moment residual.

[0078] The residual sequence in the present application includes multiple sample points, each sample point includes D-plane angle residual and occipital moment residual data, and the residuals of the sample points are judged by statistics to comprehensively measure the difference between the predicted value and the actual value of the D-plane angle model and the occipital moment model, that is, to measure the evaluation index of the neck bionic performance of the automobile collision dummy.

[0079] In one embodiment, the specific implementation method of the above step 170 can be: input the time series data to be measured into the D-plane angle model and the occipital moment model, and calculate the D-plane angle residual to be measured and the occipital moment residual to be measured respectively; according to the D-plane angle residual to be measured and the occipital moment residual to be measured, calculate the evaluation index to be measured.

[0080] In this application, the time series data to be tested is input into the D-plane angle model and the occipital moment model, and the models respectively output the D-plane angle residual to be tested and the occipital moment residual to be tested, and then the evaluation index to be tested is calculated based on the D-plane angle residual to be tested and the occipital moment residual to be tested.

[0081] Specifically, the calculation formula for the fitting residual of each set of test set data is as follows:

[0082] ;

[0083] in, is the two-dimensional residual vector, i is the sample point number, is the true value of the D plane angle variable in the test set data, is the predicted value of the D plane angle given by the performance curve model, is the true value of the occipital moment variable in the test set data, Predicted values ​​of occipital moment given by the performance curve model.

[0084] In one embodiment, the specific calculation method of the evaluation index to be measured can be: H The calculation formula for 2 is:

[0085] ;

[0086] in, , , is the statistic of the i-th sample point, is the weight of the historical residual to the current statistic, A is the overall covariance matrix of the residual sequence, is the population mean vector of residuals, , It is a two-dimensional residual vector composed of the D-plane angle residual to be measured and the occipital moment residual to be measured.

[0087] In one embodiment, the standard automobile collision dummies include multiple ones; wherein, the specific implementation method of the above step 180 may be: calculating the average value of the standard evaluation indicators corresponding to the multiple standard automobile collision dummies to obtain the average standard evaluation indicator; and determining the neck bionic performance of the automobile collision dummy to be tested based on the evaluation indicator to be tested and the average standard evaluation indicator.

[0088] Specifically, this application uses 5 standard car crash dummies, and each standard car crash dummy performs at least 5 tests to obtain sufficient data. The standard evaluation indicators of the 5 standard car crash dummies are calculated respectively, and the average value is calculated based on the 5 standard evaluation indicators to obtain the average standard evaluation index , if the evaluation index to be measured is calculated to be If the value is within the range, the neck bionic performance of the corresponding automobile collision dummy to be tested is considered qualified, otherwise it is considered unqualified.

[0089] In one embodiment, the standard time series data includes time, velocity characteristics, first acceleration characteristics and second acceleration characteristics; wherein, the specific implementation of the above step 110 may be: collecting multiple groups of time, velocity characteristics, first acceleration characteristics and second acceleration characteristics corresponding to the standard automobile collision dummy; wherein, the velocity characteristic represents the total energy when the pendulum hits, which is mainly related to the state of the calibration device, and the calculation formula of the velocity characteristic is:

[0090] ;

[0091] The first acceleration characteristic (the front part of the acceleration, such as Figure 8 As shown in Figure 2, it indicates the acceleration of the calibration device acting on the lower end of the neck of the car collision dummy, which is related to the state of the honeycomb aluminum. The calculation formula of the first acceleration characteristic is:

[0092] ;

[0093] The second acceleration characteristic (the latter part of the acceleration, such as Figure 8 The figure shows the average acceleration of the pendulum after the pendulum contacts the honeycomb aluminum. The calculation formula of the second acceleration characteristic is:

[0094] ;

[0095] Among them, the vector It is all the data collected by the calibration test at a certain moment. It is A variable in a vector used to store speed information, This is the first The speed at which the swing arm grating hole passes through the photoelectric gate, is the number of swing arm gratings, is the first acceleration characteristic, is the second acceleration characteristic, is the time when the pendulum acceleration curve first reaches its peak value, is the time corresponding to the pendulum acceleration returning to 0. The pendulum acceleration curve is The acceleration corresponding to the moment.

[0096] In one embodiment, before step 120, the above-mentioned automobile collision dummy neck bionic performance evaluation method may further include: normalizing multiple groups of standard time series data to obtain multiple groups of normalized standard time series data; wherein the normalization processing formula is:

[0097] ;

[0098] in, is the parameter value before normalization, is the maximum value of the parameter. The minimum value of the parameter. is the normalized value.

[0099] Correspondingly, a specific implementation of the above step 120 may be: dividing multiple groups of normalized standard time series data into a training set, a validation set, and a test set.

[0100] Fig. 9 FIG. 1 is a schematic diagram of the structure of a vehicle collision dummy neck bionic performance evaluation device provided by an exemplary embodiment of the present application. Fig. 9As shown, the vehicle collision dummy neck bionic performance evaluation device 90 includes: a standard data acquisition module 91, which is used to use a neck calibration device to experiment on the head and neck assembly of a standard vehicle collision dummy to collect multiple groups of standard time series data corresponding to the standard vehicle collision dummy; a standard data division module 92, which is used to divide the multiple groups of standard time series data into a training set, a verification set and a test set; a curve model building module 93, which is used to build and train a neck calibration test performance curve model based on the training set; a curve model optimization module 94, which is used to optimize the parameters of the neck calibration test performance curve model based on the verification set to obtain a final neck calibration test performance curve model. a test performance curve model; a standard index calculation module 95, which is used to input the test set into the final neck calibration test performance curve model to calculate the standard evaluation index; a test data acquisition module 96, which is used to use the neck calibration device to experiment on the head and neck assembly of the automobile collision dummy to be tested, so as to collect multiple groups of time series data to be tested corresponding to the automobile collision dummy to be tested; a test index calculation module 97, which is used to input the time series data to be tested into the final neck calibration test performance curve model to calculate the evaluation index to be tested; a test performance evaluation module 98, which is used to determine the neck bionic performance of the automobile collision dummy to be tested based on the evaluation index to be tested and the standard evaluation index.

[0101] The present application provides a device for evaluating the bionic performance of the neck of an automobile collision dummy. The device uses a neck calibration device to perform experiments on the head and neck assembly of a standard automobile collision dummy through a standard data acquisition module 91 to collect multiple groups of standard time series data corresponding to the standard automobile collision dummy; a standard data division module 92 divides the multiple groups of standard time series data into a training set, a verification set and a test set; a curve model building module 93 builds and trains a neck calibration test performance curve model based on the training set; a curve model optimization module 94 optimizes the parameters of the neck calibration test performance curve model based on the verification set to obtain a final neck calibration test performance curve model; a standard index calculation module 95 inputs the test set into the final neck calibration test performance curve model to calculate the standard evaluation index; a test data acquisition module 96 uses a neck calibration device to perform experiments on the head and neck assembly of the automobile collision dummy to be tested to collect multiple groups of test time series data corresponding to the automobile collision dummy to be tested; The index calculation module 97 inputs the time series data to be tested into the final neck calibration test performance curve model to calculate the evaluation index to be tested; the performance evaluation module 98 determines the neck bionic performance of the automobile collision dummy to be tested based on the evaluation index to be tested and the standard evaluation index; that is, the standard time series data of the standard automobile collision dummy is collected through experiments and multiple groups of standard time series data are divided into training sets, verification sets and test sets to build, train and optimize the neck calibration test performance curve model and calculate the standard evaluation index. The time series data to be tested of the automobile collision dummy to be tested is also collected through experiments, and the evaluation index to be tested of the automobile collision dummy to be tested is calculated in combination with the neck calibration test performance curve model. The neck bionic performance of the automobile collision dummy to be tested is determined according to the evaluation index to be tested and the standard evaluation index. The model is built using the standard automobile collision dummy to improve the stability and accuracy of the standard evaluation index, thereby improving the evaluation accuracy of the neck bionic performance of the automobile collision dummy.

[0102] In one embodiment, the curve model optimization module 94 can be further configured as follows: inputting the validation set into the neck calibration test performance curve model to calculate the loss function; optimizing the parameters of the neck calibration test performance curve model with the goal of minimizing the loss function to obtain the final neck calibration test performance curve model.

[0103] In one embodiment, the final neck calibration test performance curve model includes a D-plane angle model and an occipital moment model; wherein, the above-mentioned standard index calculation module 95 can be further configured as: inputting the test set into the D-plane angle model and the occipital moment model, and respectively calculating the standard D-plane angle residual and the standard occipital moment residual; according to the standard D-plane angle residual and the standard occipital moment residual, calculating the standard evaluation index.

[0104] In one embodiment, the specific calculation method of the standard evaluation index can be:H The calculation formula for 1 is:

[0105] ;

[0106] in, , , is the statistic of the i-th sample point, is the weight of the historical residual to the current statistic, A is the overall covariance matrix of the residual sequence, is the population mean vector of residuals, , is a two-dimensional residual vector consisting of the standard D-plane angle residual and the standard occipital moment residual.

[0107] In one embodiment, the above-mentioned indicator calculation module 97 can be further configured as: inputting the time series data to be measured into the D-plane angle model and the occipital moment model, and respectively calculating the D-plane angle residual to be measured and the occipital moment residual to be measured; and calculating the evaluation indicator to be measured based on the D-plane angle residual to be measured and the occipital moment residual to be measured.

[0108] In one embodiment, the specific calculation method of the evaluation index to be measured can be: H The calculation formula for 2 is:

[0109] ;

[0110] in, , , is the statistic of the i-th sample point, is the weight of the historical residual to the current statistic, A is the overall covariance matrix of the residual sequence, is the population mean vector of residuals, , It is a two-dimensional residual vector composed of the D-plane angle residual to be measured and the occipital moment residual to be measured.

[0111] In one embodiment, the standard automobile collision dummies include multiple ones; wherein the above-mentioned performance evaluation module 98 to be tested can be further configured to: calculate the average value of the standard evaluation indicators corresponding to the multiple standard automobile collision dummies to obtain the average standard evaluation indicator; based on the evaluation indicator to be tested and the average standard evaluation indicator, determine the neck bionic performance of the automobile collision dummy to be tested.

[0112] In one embodiment, the standard time series data includes time, speed characteristics, first acceleration characteristics and second acceleration characteristics; wherein the standard data acquisition module 91 can be further configured to: acquire multiple groups of time, speed characteristics, first acceleration characteristics and second acceleration characteristics corresponding to the standard automobile collision dummy; wherein the calculation formula of the speed characteristic is:

[0113] ;

[0114] The calculation formula of the first acceleration characteristic is:

[0115] ;

[0116] The calculation formula of the second acceleration characteristic is:

[0117] ;

[0118] Among them, the vector It is all the data collected by the calibration test at a certain moment. It is A variable in a vector used to store speed information, This is the first The speed at which the swing arm grating hole passes through the photoelectric gate, is the number of swing arm gratings, is the first acceleration characteristic, is the second acceleration characteristic, is the time when the pendulum acceleration curve first reaches its peak value, is the time corresponding to the pendulum acceleration returning to 0. The pendulum acceleration curve is The acceleration corresponding to the moment.

[0119] In one embodiment, the above-mentioned automobile collision dummy neck bionic performance evaluation device 90 can be further configured to: perform normalization processing on multiple sets of standard time series data to obtain multiple sets of normalized standard time series data; wherein the normalization processing formula is:

[0120] ;

[0121] in, is the parameter value before normalization, is the maximum value of the parameter. The minimum value of the parameter. is the normalized value.

[0122] Correspondingly, the above-mentioned standard data division module 92 can be further configured to: divide the multiple groups of normalized standard time series data into a training set, a validation set and a test set.

[0123] Below, reference Fig.10 The electronic device according to the embodiment of the present application is described. The electronic device may be any one or both of the first device and the second device, or a stand-alone device independent of them, and the stand-alone device may communicate with the first device and the second device to receive the collected input signal from them.

[0124] Fig.10 A block diagram of an electronic device according to an embodiment of the present application is illustrated.

[0125] like Fig.10 As shown, the electronic device 10 includes one or more processors 11 and a memory 12 .

[0126] The processor 11 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.

[0127] The memory 12 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 11 may run the program instructions to implement the methods of the various embodiments of the present application described above and / or other desired functions. Various contents such as input signals, signal components, noise components, etc. may also be stored in the computer-readable storage medium.

[0128] In one example, the electronic device 10 may further include: an input device 13 and an output device 14 , and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0129] When the electronic device is a stand-alone device, the input device 13 may be a communication network connector, which is used to receive the collected input signals from the first device and the second device.

[0130] In addition, the input device 13 may also include, for example, a keyboard, a mouse, and the like.

[0131] The output device 14 can output various information to the outside, including the determined distance information, direction information, etc. The output device 14 can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, and the like.

[0132] Of course, to simplify, Fig.10 Only some of the components related to the present application in the electronic device 10 are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, according to specific application situations, the electronic device 10 may also include any other appropriate components.

[0133] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the method according to various embodiments of the present application described in the above-mentioned "Exemplary Method" section of this specification.

[0134] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0135] In addition, an embodiment of the present application may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the steps of the method according to various embodiments of the present application described in the above "Exemplary Method" section of this specification.

[0136] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0137] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, not for limitation, and the above details do not limit the present application to being implemented by adopting the above specific details.

[0138] The block diagrams of the devices, apparatuses, equipment, and systems involved in this application are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The words "such as" used here refer to the phrase "such as but not limited to", and can be used interchangeably with them.

[0139] It should also be noted that in the apparatus, device and method of the present application, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present application.

[0140] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0141] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A method for evaluating the bionic performance of a car collision dummy neck, characterized in that: include: Using a neck calibration device to test the head and neck assembly of a standard automobile collision dummy, so as to collect multiple groups of standard time series data corresponding to the standard automobile collision dummy; Dividing the multiple groups of standard time series data into a training set, a validation set and a test set; A neck calibration test performance curve model is constructed and trained based on the training set; Optimizing the parameters of the neck calibration test performance curve model based on the validation set to obtain a final neck calibration test performance curve model; Inputting the test set into the final neck calibration test performance curve model to calculate the standard evaluation index; The neck calibration device is used to conduct experiments on the head and neck assembly of the automobile collision dummy to be tested, so as to collect multiple groups of time series data to be tested corresponding to the automobile collision dummy to be tested; Inputting the time series data to be tested into the final neck calibration test performance curve model to calculate the evaluation index to be tested; Determining the neck bionic performance of the automobile collision dummy to be tested based on the evaluation index to be tested and the standard evaluation index; The final neck calibration test performance curve model includes a D-plane angle model and an occipital moment model; wherein the test set is input into the final neck calibration test performance curve model, and the standard evaluation index is calculated to include: Inputting the test set into the D-plane angle model and the occipital moment model, and respectively calculating the standard D-plane angle residual and the standard occipital moment residual; The standard evaluation index is calculated based on the standard D-plane angle residual and the standard occipital moment residual.

2. The method for evaluating the bionic performance of the neck of a car collision dummy according to claim 1, characterized in that: The standard evaluation index calculated according to the standard D-plane angle residual and the standard occipital moment residual includes: The standard evaluation index H The calculation formula for 1 is: ; in, , , is the statistic of the i-th sample point, n is the total number of sample points, is the weight of historical residuals on current statistics, is the historical residual, for The transpose of , A is the overall covariance matrix of the residual sequence, is the population mean vector of residuals, , It is a two-dimensional residual vector composed of the standard D-plane angle residual and the standard occipital moment residual.

3. The method for evaluating the bionic performance of the neck of a car collision dummy according to claim 1, characterized in that: The inputting of the time series data to be tested into the final neck calibration test performance curve model to calculate the evaluation index to be tested includes: Inputting the time series data to be measured into the D-plane angle model and the occipital moment model, and respectively calculating the D-plane angle residual to be measured and the occipital moment residual to be measured; The evaluation index to be measured is calculated according to the D-plane angle residual to be measured and the occipital moment residual to be measured.

4. The method for evaluating the bionic performance of the neck of a car collision dummy according to claim 3, characterized in that: The evaluation index to be measured is calculated according to the D-plane angle residual to be measured and the occipital moment residual to be measured, and includes: The evaluation index to be tested H The calculation formula for 2 is: ; in, , , is the statistic of the i-th sample point, n is the total number of sample points, is the weight of historical residuals on current statistics, is the historical residual, for The transpose of , A is the overall covariance matrix of the residual sequence, is the population mean vector of residuals, , It is a two-dimensional residual vector composed of the D-plane angle residual to be measured and the occipital moment residual to be measured.

5. The method for evaluating the bionic performance of the neck of a car collision dummy according to claim 1, characterized in that: The standard automobile collision dummies include a plurality of dummies; wherein, determining the neck bionic performance of the automobile collision dummy to be tested based on the evaluation index to be tested and the standard evaluation index includes: Calculating an average value of standard evaluation indicators corresponding to a plurality of the standard automobile collision dummies to obtain an average standard evaluation indicator; Based on the evaluation index to be tested and the average standard evaluation index, the neck bionic performance of the automobile collision dummy to be tested is determined.

6. The method for evaluating the bionic performance of the neck of a car collision dummy according to claim 1, characterized in that: The standard time series data includes time, speed characteristics, first acceleration characteristics and second acceleration characteristics; wherein the collecting of multiple groups of standard time series data corresponding to the standard automobile collision dummy includes: Collecting multiple groups of time, speed characteristics, first acceleration characteristics and second acceleration characteristics corresponding to the standard automobile collision dummy; The calculation formula of the speed characteristic is: ; The calculation formula of the first acceleration characteristic is: ; The calculation formula of the second acceleration characteristic is: ; in, It is a variable used to store speed information in all the data collected at a certain moment in the calibration test. This is the first The speed at which the swing arm grating hole passes through the photoelectric gate, is the number of swing arm gratings, is the first acceleration feature, is the second acceleration characteristic, is the time when the pendulum acceleration curve first reaches its peak value, is the time corresponding to the pendulum acceleration returning to 0. The pendulum acceleration curve is The acceleration corresponding to the moment.

7. The method for evaluating the bionic performance of the neck of a car collision dummy according to claim 1, characterized in that: Before dividing the plurality of groups of standard time series data into a training set, a validation set and a test set, the method for evaluating the bionic performance of the neck of a car collision dummy further comprises: The multiple sets of standard time series data are normalized to obtain multiple sets of normalized standard time series data; wherein the normalization formula is: ; in, is the parameter value before normalization, is the maximum value of the parameter. is the minimum value of the parameter. is the normalized value; The dividing the plurality of groups of standard time series data into a training set, a validation set and a test set comprises: The multiple groups of normalized standard time series data are divided into a training set, a validation set and a test set.

8. The method for evaluating the bionic performance of the neck of a car collision dummy according to claim 1, characterized in that: The step of optimizing the parameters of the neck calibration test performance curve model based on the validation set to obtain the final neck calibration test performance curve model comprises: Inputting the verification set into the neck calibration test performance curve model to calculate the loss function; With the goal of minimizing the loss function, the parameters of the neck calibration test performance curve model are optimized to obtain the final neck calibration test performance curve model.

9. A device for evaluating the bionic performance of a car collision dummy neck, characterized in that: include: A standard data acquisition module, used to perform an experiment on the head and neck assembly of a standard automobile collision dummy using a neck calibration device, so as to collect multiple groups of standard time series data corresponding to the standard automobile collision dummy; A standard data partitioning module, used to divide the multiple groups of standard time series data into a training set, a validation set and a test set; A curve model building module, used to build and train a neck calibration test performance curve model based on the training set; A curve model optimization module, used to optimize the parameters of the neck calibration test performance curve model based on the verification set to obtain a final neck calibration test performance curve model; A standard index calculation module, used for inputting the test set into the final neck calibration test performance curve model to calculate the standard evaluation index; A test data acquisition module, used to use the neck calibration device to test the head and neck assembly of the automobile collision dummy to be tested, so as to collect multiple groups of test time series data corresponding to the automobile collision dummy to be tested; A test index calculation module, used for inputting the test time series data into the final neck calibration test performance curve model to calculate the test evaluation index; A performance evaluation module to be tested, used for determining the neck bionic performance of the automobile collision dummy to be tested based on the evaluation index to be tested and the standard evaluation index; The final neck calibration test performance curve model includes a D-plane angle model and an occipital moment model; wherein the index calculation module to be measured is further configured as follows: Inputting the test set into the D-plane angle model and the occipital moment model, and respectively calculating the standard D-plane angle residual and the standard occipital moment residual; The standard evaluation index is calculated based on the standard D-plane angle residual and the standard occipital moment residual.

Citation Information

Patent Citations

  • Collision dummy neck structure reflecting human neck mechanical properties and design method thereof

    CN103926046A

  • Fast equivalent modeling method for a new structure of a crash dummy neck for automobile rear-end collision test

    CN109255166A

  • Driving posture determination method for reducing head and neck injuries of dummy in automobile collision and medium

    CN117268798A

  • Method for predicting evaluation index of dummy injury in automobile collision

    CN118296913A