Model training method, performance prediction method, device, equipment, vehicle and medium

Abstract

The invention relates to a model training method and device, a performance prediction method and device, equipment, a vehicle and a medium, and the method comprises the steps: obtaining a test data set which comprises a plurality of groups of test sample data, sequentially inputting each group of test sample data into a preset initial model, so as to obtain a first prediction performance parameter corresponding to each group of test sample data, acquiring simulation performance parameters corresponding to each group of test sample data, determining prediction precision of a preset initial model according to the first prediction performance parameters and the simulation performance parameters, and determining a target vehicle performance prediction model according to the prediction precision and the preset initial model, the target vehicle performance prediction model with high vehicle performance prediction precision can be effectively obtained, the vehicle performance corresponding to the vehicle design parameters can be predicted through the target vehicle performance prediction model, the determination efficiency of the vehicle design parameters can be effectively improved, the time cost required by the vehicle from design to machining can be effectively shortened, and the vehicle performance prediction efficiency is improved. And the design cost of a high-performance vehicle can be reduced.

Description

Technical Field

[0001] The present disclosure relates to the field of vehicle technology, and in particular to a model training method, a performance prediction method, a device, a equipment, a vehicle and a medium. Background Art

[0002] With the advancement of technology and the improvement of consumer awareness, the proportion of vehicle performance references when purchasing vehicles has gradually increased. Therefore, vehicle manufacturers need to further optimize vehicle performance. Vehicle performance is usually determined by vehicle design parameters. During the vehicle design process, it is usually necessary to simulate the vehicle design parameters. When the obtained simulation performance parameters meet the requirements, the vehicle is produced according to the vehicle design parameters; however, the existing vehicle simulation models usually cannot further improve the efficiency of vehicle design, which is not conducive to reducing the cost of high-performance vehicle design. Summary of the invention

[0003] In order to overcome the problems existing in the related art, the present disclosure provides a vehicle performance prediction model training method, a vehicle performance prediction method, a vehicle performance prediction model training device, an electronic device, a vehicle and a storage medium.

[0004] According to a first aspect of an embodiment of the present disclosure, a vehicle performance prediction model training method is provided, comprising:

[0005] Acquire a test data set, wherein the test data set includes multiple groups of test sample data, each group of test sample data includes one or more vehicle design parameters;

[0006] Inputting each group of the test sample data into a preset initial model in sequence to obtain a first predicted performance parameter corresponding to each group of the test sample data output by the preset initial model, wherein the preset initial model is a machine learning model for predicting a target performance parameter of a vehicle;

[0007] Obtaining simulation performance parameters corresponding to each group of test sample data;

[0008] Determining the prediction accuracy of the preset initial model according to the first prediction performance parameter and the simulation performance parameter corresponding to each group of the test sample data in the multiple groups of test sample data;

[0009] A target vehicle performance prediction model is determined according to the prediction accuracy and the preset initial model.

[0010] Optionally, the preset initial model is a machine learning model obtained by training a preset initial training data set, and determining the target vehicle performance prediction model according to the prediction accuracy and the preset initial model includes:

[0011] When it is determined that the prediction accuracy is greater than or equal to a preset accuracy threshold, determining target training data according to the preset initial training data set;

[0012] Training the preset initial model with the target training data to obtain an updated pending performance prediction model;

[0013] The target vehicle performance prediction model is determined according to the pending performance prediction model.

[0014] Optionally, the preset initial training data set includes multiple groups of training sample data, and determining the target training data according to the preset initial training data set includes:

[0015] Inputting each set of training sample data in the plurality of sets of training sample data into the preset initial model in sequence to obtain a first standby performance parameter corresponding to each set of training sample data output by the preset initial model;

[0016] In the case where it is determined that the first standby performance parameter does not belong to the preset performance parameter range, taking the training sample data corresponding to the first standby performance parameter as abnormal training data;

[0017] The abnormal training data is deleted from the preset initial training data set to obtain the target training data.

[0018] Optionally, determining target training data according to the preset initial training data set includes:

[0019] determining a sensitive characteristic design parameter of the one or more vehicle design parameters;

[0020] Obtaining a supplementary training data set corresponding to the sensitive feature design parameters;

[0021] The sensitive feature design parameters and the supplementary training data set in the preset initial training data set are used as the target training data.

[0022] Optionally, determining target training data according to the preset initial training data set includes:

[0023] Acquiring preset constraint condition data, wherein the preset constraint condition data includes a value range of each vehicle design parameter;

[0024] Inputting the preset initial model and the preset constraint conditions into a preset multi-objective optimization algorithm to obtain an optimal solution set of vehicle design parameters output by the preset multi-objective optimization algorithm, wherein the optimal solution set includes multiple groups of optimal solutions;

[0025] Inputting each set of optimal solutions into a preset simulation model to obtain a second standby performance parameter output by the preset simulation model;

[0026] Obtaining a second prediction performance parameter corresponding to the optimal solution;

[0027] If the difference between the second standby performance parameter and the second predicted performance parameter is greater than a preset deviation threshold, using the optimal solution as pending training data;

[0028] The preset initial training data set and the pending training data are used as the target training data.

[0029] Optionally, determining the prediction accuracy of the preset initial model according to the first prediction performance parameter and the simulation performance parameter corresponding to each group of the test sample data in the multiple groups of test sample data includes:

[0030] Determine, for each set of test sample data, a difference between the first predicted performance parameter and the simulated performance parameter;

[0031] The prediction accuracy of the preset initial model is determined according to the difference.

[0032] Optionally, determining the target vehicle performance prediction model according to the prediction accuracy and the preset initial model includes:

[0033] When it is determined that the prediction accuracy is less than a preset accuracy threshold, the preset initial model is used as the target vehicle performance prediction model.

[0034] Optionally, obtaining the simulation performance parameters corresponding to each group of the test sample data includes:

[0035] Each group of the test sample data is input into a preset simulation model in turn to obtain the simulation performance parameters corresponding to the test sample data output by the preset simulation model.

[0036] According to a second aspect of an embodiment of the present disclosure, a vehicle performance prediction method is provided, comprising:

[0037] Obtain target design parameters;

[0038] The target design parameters are input into the target vehicle performance prediction model described in the first aspect to obtain target performance parameters corresponding to the target design parameters output by the target vehicle performance prediction model.

[0039] According to a third aspect of an embodiment of the present disclosure, a vehicle performance prediction model training device is provided, comprising:

[0040] A first acquisition module is configured to acquire a test data set, wherein the test data set includes multiple groups of test sample data, each group of test sample data includes one or more vehicle design parameters;

[0041] A second acquisition module is configured to input each group of the test sample data into a preset initial model in sequence to obtain a first predicted performance parameter corresponding to each group of the test sample data output by the preset initial model, wherein the preset initial model is a machine learning model for predicting a target performance parameter of a vehicle;

[0042] A third acquisition module is configured to acquire simulation performance parameters corresponding to each group of test sample data;

[0043] A first determination module is configured to determine the prediction accuracy of the preset initial model according to the first prediction performance parameter and the simulation performance parameter corresponding to each group of the test sample data in the multiple groups of test sample data;

[0044] The second determination module is configured to determine a target vehicle performance prediction model according to the prediction accuracy and the preset initial model.

[0045] Optionally, the preset initial model is a machine learning model obtained by training a preset initial training data set, and the second determination module is configured to:

[0046] When it is determined that the prediction accuracy is greater than or equal to a preset accuracy threshold, determining target training data according to the preset initial training data set;

[0047] Training the preset initial model with the target training data to obtain an updated pending performance prediction model;

[0048] The target vehicle performance prediction model is determined according to the pending performance prediction model.

[0049] Optionally, the preset initial training data set includes multiple sets of training sample data, and the second determining module is configured to:

[0050] Inputting each set of training sample data in the plurality of sets of training sample data into the preset initial model in sequence to obtain a first standby performance parameter corresponding to each set of training sample data output by the preset initial model;

[0051] In the case where it is determined that the first standby performance parameter does not belong to the preset performance parameter range, taking the training sample data corresponding to the first standby performance parameter as abnormal training data;

[0052] The abnormal training data is deleted from the preset initial training data set to obtain the target training data.

[0053] Optionally, the second determining module is configured to:

[0054] determining a sensitive characteristic design parameter of the one or more vehicle design parameters;

[0055] Obtaining a supplementary training data set corresponding to the sensitive feature design parameters;

[0056] The sensitive feature design parameters and the supplementary training data set in the preset initial training data set are used as the target training data.

[0057] Optionally, the second determining module is configured to:

[0058] Acquiring preset constraint condition data, wherein the preset constraint condition data includes a value range of each vehicle design parameter;

[0059] Inputting the preset initial model and the preset constraint conditions into a preset multi-objective optimization algorithm to obtain an optimal solution set of vehicle design parameters output by the preset multi-objective optimization algorithm, wherein the optimal solution set includes multiple groups of optimal solutions;

[0060] Inputting each set of optimal solutions into a preset simulation model to obtain a second standby performance parameter output by the preset simulation model;

[0061] Obtaining a second prediction performance parameter corresponding to the optimal solution;

[0062] If the difference between the second standby performance parameter and the second predicted performance parameter is greater than a preset deviation threshold, using the optimal solution as pending training data;

[0063] The preset initial training data set and the pending training data are used as the target training data.

[0064] Optionally, the first determining module is configured to:

[0065] Determine, for each set of test sample data, a difference between the first predicted performance parameter and the simulated performance parameter;

[0066] The prediction accuracy of the preset initial model is determined according to the difference.

[0067] Optionally, the second determining module is configured to:

[0068] When it is determined that the prediction accuracy is less than a preset accuracy threshold, the preset initial model is used as the target vehicle performance prediction model.

[0069] Optionally, the third acquisition module is configured to:

[0070] Each group of the test sample data is input into a preset simulation model in turn to obtain the simulation performance parameters corresponding to the test sample data output by the preset simulation model.

[0071] According to a fourth aspect of an embodiment of the present disclosure, an electronic device is provided, including:

[0072] processor;

[0073] a memory for storing processor-executable instructions;

[0074] The processor is configured to execute the steps of the method described in the first aspect, or to execute the steps of the method described in the second aspect.

[0075] According to a fifth aspect of an embodiment of the present disclosure, a vehicle is provided, comprising: the electronic device as described in the fourth aspect of an embodiment of the present disclosure.

[0076] According to the sixth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored. When the program instructions are executed by a processor, the steps of the vehicle performance prediction model training method provided in the first aspect of the present disclosure, or the steps of the vehicle performance prediction method provided in the second aspect, are implemented.

[0077] The technical solution provided by the embodiments of the present disclosure may have the following beneficial effects:

[0078] By acquiring a test data set, the test data set includes multiple groups of test sample data, and inputting each group of test sample data into a preset initial model in turn to obtain the first prediction performance parameters corresponding to each group of test sample data output by the preset initial model, and obtaining the simulation performance parameters corresponding to each group of test sample data, the prediction accuracy of the preset initial model is determined according to the first prediction performance parameters and simulation performance parameters corresponding to each group of test sample data in the multiple groups of test sample data, and the target vehicle performance prediction model is determined according to the prediction accuracy and the preset initial model, so as to effectively obtain a target vehicle performance prediction model with high vehicle performance prediction accuracy, and predict the performance of the vehicle corresponding to the vehicle design parameters through the target vehicle performance prediction model, so as to effectively improve the efficiency of determining the vehicle design parameters, and then effectively shorten the time cost required for the vehicle from design to processing, which is conducive to reducing the design cost of high-performance vehicles.

[0079] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0081] Figure 1 is a flow chart of a vehicle performance prediction model training method according to an exemplary embodiment;

[0082] Figure 2 is based on Figure 1 A flow chart of a vehicle performance prediction model training method shown in the illustrated embodiment;

[0083] Figure 3 is a block diagram of a vehicle performance prediction model training device according to an exemplary embodiment;

[0084] Figure 4 is a block diagram of a vehicle performance prediction model training device according to an exemplary embodiment;

[0085] Figure 5 is a block diagram of a vehicle performance prediction model training device according to an exemplary embodiment;

[0086] Figure 6 is a block diagram of a vehicle according to an exemplary embodiment. DETAILED DESCRIPTION

[0087] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0088] It should be noted that all actions of acquiring signals, information or data in this application are carried out in compliance with the relevant data protection laws and policies of the country where they are located, and with the authorization given by the owner of the corresponding device.

[0089] Figure 1 is a flow chart of a vehicle performance prediction model training method according to an exemplary embodiment. Figure 1 As shown, the vehicle performance prediction model training method can be applied to electronic devices, which can be devices with processors such as servers and electronic terminals, such as computers, tablet computers, smart phones, etc. The vehicle performance prediction model training method includes:

[0090] In step 110 , a test data set is acquired, wherein the test data set may include multiple groups of test sample data, and each group of test sample data may include one or more vehicle design parameters.

[0091] Among them, the vehicle design parameters can be the size of the vehicle body, the parameters of the materials used for the vehicle body, and the position, size, and shape parameters of the vehicle components (for example, the thickness and mechanical parameters of the metal plates used for the vehicle body, the spatial coordinate parameters of the installation position of the vehicle power unit, etc.), or the size of the vehicle parts themselves and the parameters of the materials used (for example, the mechanical parameters of the materials used for the vehicle engine body, the size of the air spring and the mechanical parameters of the materials used, etc.); the mechanical parameters can be strength (the ability to resist destruction under the action of external forces, which may include yield limit and tensile strength), plasticity (the ability to resist permanent deformation under the action of external forces), strength (the ability to resist local deformation) and fatigue strength (the maximum stress value that can withstand an infinite number of cycles under cyclic stress without breaking).

[0092] It should be noted that the design parameter set corresponding to the vehicle performance can be collected through a hierarchical collection method such as Latin hypercube to obtain a small number of design parameter sets, and the design parameter set is divided into a training data set and a test data set. The division ratio can be 4:1 or 5:1, etc. The training data set is used to train the model, and the test data set is used to test the accuracy of the training model.

[0093] In step 120, each group of test sample data is input into the preset initial model in turn to obtain a first prediction performance parameter corresponding to each group of test sample data output by the preset initial model.

[0094] The preset initial model is a machine learning model that can be used to predict the target performance parameters of the vehicle. The preset initial model can be a machine learning model trained by a preset initial training data set. The first predicted performance parameter can be a vehicle performance parameter, including the strength, stiffness, fatigue durability of the vehicle body or parts, and the NVH (NoiseVibrationHarshness) of the vehicle. It should be noted that the vehicle performance parameters can include the above performance parameters, but are not limited to the above performance parameters.

[0095] In step 130, the simulation performance parameters corresponding to each set of test sample data are obtained.

[0096] In this step, each group of test sample data may be input into the preset simulation model in turn to obtain simulation performance parameters corresponding to the test sample data output by the preset simulation model.

[0097] Among them, the preset simulation model can be a finite element model, a boundary element model, a statistical energy model, a computational fluid dynamics model, a hardware-in-the-loop simulation model and other models. The simulation performance parameters can be simulation performance parameters corresponding to the vehicle design parameters obtained through the vehicle finite element model, or can be simulation performance parameters corresponding to the vehicle design parameters obtained through other simulation models. For example, the simulation model can include existing simulation models such as multi-body dynamics models, computational fluid dynamics models and simplified structure models, and the present disclosure does not make a limiting description.

[0098] In step 140, the prediction accuracy of the preset initial model is determined according to the first prediction performance parameter and the simulation performance parameter corresponding to each group of test sample data in the plurality of groups of test sample data.

[0099] Among them, the first predicted performance parameter can be a set of test sample data input into a preset initial model, and the predicted vehicle performance parameters (for example, the strength, stiffness, fatigue durability of the vehicle body or components and the NVH (Noise Vibration Harshness) of the vehicle) output by the preset initial model.

[0100] An implementation method of this step may be to determine the difference between the first prediction performance parameter and the simulation performance parameter for each set of test sample data; and determine the prediction accuracy of the preset initial model based on the difference.

[0101] It should be noted that the implementation method of determining the prediction accuracy of the preset initial model based on the difference described above may be to calculate the difference between the first prediction performance parameter and the simulation performance parameter corresponding to each group of test sample data to obtain multiple differences corresponding to multiple groups of test sample data, and take the maximum absolute value of the ratio of each difference to the simulation performance parameter as the prediction accuracy of the preset initial model, which can be expressed by the formula:

[0102] A=(x i -y i ) / y i ;

[0103] Among them, x i It can be the first prediction performance parameter corresponding to the i-th group of test sample data, y i It can be the simulation performance parameter corresponding to the i-th group of test sample data.

[0104] The average value of multiple differences corresponding to multiple groups of test sample data can also be calculated, and the ratio of the average value to the average value of the simulation performance parameter is used as the prediction accuracy of the preset initial model, which can be expressed by the formula:

[0105] Among them, x iIt can be the first prediction performance parameter corresponding to the i-th group of test sample data, y i It can be the simulation performance parameter corresponding to the i-th group of test sample data.

[0106] Another implementation of this step may be to determine the variance or standard deviation between the first prediction performance parameter and the simulation performance parameter for each set of test sample data; and determine the prediction accuracy of the preset initial model based on the variance or the standard deviation.

[0107] In step 150, a target vehicle performance prediction model is determined according to the prediction accuracy and a preset initial model.

[0108] This step can be done by Figure 2 The embodiment shown is implemented, Figure 2 is based on Figure 1 The embodiment shown is a flow chart of a vehicle performance prediction model training method, such as Figure 2 As shown, Figure 1 Step 150 may include the following steps:

[0109] In step 151, when it is determined that the prediction accuracy is greater than or equal to a preset accuracy threshold, target training data is determined according to a preset initial training data set.

[0110] The preset initial training data set includes multiple sets of training sample data.

[0111] It should be noted that when it is determined that the prediction accuracy is greater than or equal to the preset accuracy threshold, there is a deviation between the vehicle prediction performance parameters output by the preset initial model and the vehicle simulation performance parameters. The preset initial model cannot be used as a performance prediction model in the vehicle production process. The preset initial model needs to be optimized, which can be achieved by cleaning or updating the preset initial training data set to obtain cleaned or updated training data (i.e., target training data) so that the preset initial model can be optimized.

[0112] In this step, the implementation method of determining the target training data according to the preset initial training data set may include:

[0113] Implementation method one, input each group of training sample data in multiple groups of training sample data into a preset initial model in turn to obtain the first standby performance parameter corresponding to each group of training sample data output by the preset initial model; when it is determined that the first standby performance parameter does not belong to the preset performance parameter range, the training sample data corresponding to the first standby performance parameter is used as abnormal training data; and the abnormal training data is deleted from the preset initial training data set to obtain the target training data.

[0114] It should be noted that the preset performance parameter range can be a vehicle performance parameter range preset according to actual conditions. For example, in order to improve the ability of the vehicle body to resist damage from external forces, the preset yield strength range of the vehicle can be a strength value above 1000Mpa, or it can be (1300, 1600), that is, the preset yield strength of the vehicle is greater than 1300Mpa and less than 1600Mpa.

[0115] In this implementation, abnormal training data in a preset training data set may be determined by an outlier detection method, which may include a K-nearest neighbor method, a Z-score (standard score), a box plot, an isolation forest, and the like.

[0116] The implementation method of the K nearest neighbor method may be: when the difference between each first standby performance parameter and the first standby performance parameter adjacent to the first standby performance parameter is greater than a preset difference threshold, the training sample data of the first standby performance parameter is used as abnormal training data.

[0117] The implementation of Z-score (standard score) can be: calculate the average value and standard deviation of the first standby performance parameter, obtain the difference between each first standby performance parameter and the average value of the first standby performance parameter, and if the ratio between the difference and the standard deviation of the first standby performance parameter is greater than a preset ratio threshold, the training sample data corresponding to the first standby performance parameter is used as abnormal training data. The formula can be:

[0118]

[0119] Wherein, X may be the first standby performance parameter, It can be the average value of the first standby performance parameter, and S can be the standard deviation of the first standby performance parameter.

[0120] Implementation method 2 is to determine sensitive feature design parameters in one or more vehicle design parameters; obtain a supplementary training data set corresponding to the sensitive feature design parameters; and use the sensitive feature design parameters and the supplementary training data set in a preset initial training data set as target training data.

[0121] It should be noted that the sensitive characteristic design parameters in one or more vehicle design parameters may be determined by using sensitivity analysis methods such as Sobol Sensitivity Analysis, Local Sensitivity Analysis or perturbation analysis.

[0122] Implementation method three, obtaining preset constraint data; inputting the preset initial model and preset constraints into the preset multi-objective optimization algorithm to obtain the optimal solution set of vehicle design parameters output by the preset multi-objective optimization algorithm, the optimal solution set including multiple groups of optimal solutions; inputting each group of optimal solutions into the preset simulation model to obtain the second standby performance parameter output by the preset simulation model; obtaining the second predicted performance parameter corresponding to the optimal solution; if the difference between the second standby performance parameter and the second predicted performance parameter is greater than the preset deviation threshold, using the optimal solution as the pending training data; using the preset initial training data set and the pending training data as the target training data.

[0123] Among them, the preset constraint data may include the value range of each vehicle design parameter, and may also include that the prediction accuracy of the preset initial model is greater than the preset accuracy threshold; the preset multi-objective optimization algorithm may be a multi-objective genetic algorithm (i.e., a multi-objective evolutionary algorithm), a multi-objective particle swarm algorithm, a multi-objective differential evolution algorithm or a multi-objective simulated annealing algorithm, etc.

[0124] It should be noted that by inputting the preset initial model and the preset constraints into the preset multi-objective optimization algorithm, the optimal solution set of the vehicle design parameters can be inferred. In the process of inferring the optimal solution set, the second predicted performance parameter corresponding to the optimal solution set can be obtained through the preset initial model, and the optimal solution set is input into the preset simulation model, the second standby performance parameter is calculated, and the difference between the second standby performance parameter and the second predicted performance parameter is obtained. The optimal solution corresponding to the difference between the second standby performance parameter and the second predicted performance parameter is greater than the preset deviation threshold as the pending training data, and the pending training data is added to the preset initial training data set to obtain the target training data. It should be pointed out that the above process of using the multi-objective optimization algorithm to infer the optimal solution set of vehicle design parameters belongs to a mature technology in the prior art, and the present disclosure does not limit this.

[0125] In addition, it should be noted that the above three implementation methods of determining the target training data according to the preset initial training data set can be used in combination with each other, or one or two of them can be used. In addition, the implementation method of determining the target training data according to the preset initial training data set can also include: Latin hypercube sampling, Monte Carlo sampling and other implementation methods, which are prior art and are not limited in this disclosure.

[0126] In step 152, the preset initial model is trained with target training data to obtain an updated pending performance prediction model.

[0127] In step 153, a target vehicle performance prediction model is determined based on the pending performance prediction model.

[0128] In this step, when it is determined that the prediction accuracy is less than a preset accuracy threshold, the preset initial model can be used as the target vehicle performance prediction model.

[0129] The above scheme obtains a test data set, which includes multiple groups of test sample data. Each group of test sample data is input into a preset initial model in turn to obtain the first prediction performance parameters corresponding to each group of test sample data output by the preset initial model, and obtains the simulation performance parameters corresponding to each group of test sample data. The prediction accuracy of the preset initial model is determined according to the first prediction performance parameters and simulation performance parameters corresponding to each group of test sample data in the multiple groups of test sample data. The target vehicle performance prediction model is determined according to the prediction accuracy and the preset initial model. A target vehicle performance prediction model with high vehicle performance prediction accuracy can be effectively obtained. The performance of the vehicle corresponding to the vehicle design parameters is predicted by the target vehicle performance prediction model, which can effectively improve the efficiency of determining the vehicle design parameters, and thus can effectively shorten the time cost required for the vehicle from design to processing, which is conducive to reducing the design cost of high-performance vehicles.

[0130] Another exemplary embodiment of the present disclosure provides a vehicle performance prediction method, which may include:

[0131] Obtain target design parameters;

[0132] The target design parameters are input through the above Figure 1 to Figure 2 any of the target vehicle performance prediction models described in , to obtain target performance parameters corresponding to the target design parameters output by the target vehicle performance prediction model.

[0133] Figure 3 is a block diagram of a vehicle performance prediction model training device according to an exemplary embodiment. Figure 3 As shown, the device 300 includes:

[0134] A first acquisition module 310 is configured to acquire a test data set, the test data set includes multiple groups of test sample data, each group of test sample data includes one or more vehicle design parameters;

[0135] A second acquisition module 320 is configured to sequentially input each set of test sample data into a preset initial model to obtain a first predicted performance parameter corresponding to each set of test sample data output by the preset initial model, wherein the preset initial model is a machine learning model for predicting a target performance parameter of a vehicle;

[0136] The third acquisition module 330 is configured to obtain the simulation performance parameters corresponding to each set of test sample data;

[0137] A first determination module 340 is configured to determine the prediction accuracy of a preset initial model according to the first prediction performance parameter and the simulation performance parameter corresponding to each group of test sample data in the plurality of groups of test sample data;

[0138] The second determination module 350 is configured to determine the target vehicle performance prediction model according to the prediction accuracy and the preset initial model.

[0139] Optionally, the preset initial model is a machine learning model obtained by training a preset initial training data set, and the second determination module 350 is configured to:

[0140] When it is determined that the prediction accuracy is greater than or equal to a preset accuracy threshold, determining target training data according to a preset initial training data set;

[0141] The preset initial model is trained by target training data to obtain an updated pending performance prediction model;

[0142] The target vehicle performance prediction model is determined according to the pending performance prediction model.

[0143] Optionally, the preset initial training data set includes multiple sets of training sample data, and the second determining module is configured to:

[0144] Inputting each set of training sample data in the plurality of sets of training sample data into a preset initial model in turn to obtain a first standby performance parameter corresponding to each set of training sample data output by the preset initial model;

[0145] In the case where it is determined that the first standby performance parameter does not belong to the preset performance parameter range, the training sample data corresponding to the first standby performance parameter is used as abnormal training data;

[0146] Abnormal training data is deleted from the preset initial training data set to obtain target training data.

[0147] Optionally, the second determining module 350 is configured to:

[0148] determining a sensitive characteristic design parameter of one or more vehicle design parameters;

[0149] Obtaining a supplementary training data set corresponding to sensitive feature design parameters;

[0150] The sensitive feature design parameters in the preset initial training data set and the supplementary training data set are used as the target training data.

[0151] Optionally, the second determining module 350 is configured to:

[0152] Acquiring preset constraint condition data, the preset constraint condition data including a value range of each vehicle design parameter;

[0153] Inputting a preset initial model and preset constraints into a preset multi-objective optimization algorithm to obtain an optimal solution set of vehicle design parameters output by the preset multi-objective optimization algorithm, wherein the optimal solution set includes multiple groups of optimal solutions;

[0154] Inputting each set of optimal solutions into a preset simulation model to obtain a second standby performance parameter output by the preset simulation model;

[0155] Obtaining a second prediction performance parameter corresponding to the optimal solution;

[0156] If the difference between the second standby performance parameter and the second predicted performance parameter is greater than a preset deviation threshold, the optimal solution is used as the pending training data;

[0157] The preset initial training data set and the pending training data are used as the target training data.

[0158] Optionally, the first determining module 340 is configured to:

[0159] Determine the difference between the first predicted performance parameter and the simulated performance parameter for each set of test sample data;

[0160] Based on the difference, the prediction accuracy of the preset initial model is determined.

[0161] Optionally, the second determining module 350 is configured to:

[0162] When it is determined that the prediction accuracy is less than the preset accuracy threshold, the preset initial model is used as the target vehicle performance prediction model.

[0163] Optionally, the third acquisition module 330 is configured to:

[0164] Each set of test sample data is input into the preset simulation model in turn to obtain simulation performance parameters corresponding to the test sample data output by the preset simulation model.

[0165] The above scheme obtains a test data set, which includes multiple groups of test sample data. Each group of test sample data is input into a preset initial model in turn to obtain the first prediction performance parameters corresponding to each group of test sample data output by the preset initial model, and obtains the simulation performance parameters corresponding to each group of test sample data. The prediction accuracy of the preset initial model is determined according to the first prediction performance parameters and simulation performance parameters corresponding to each group of test sample data in the multiple groups of test sample data. The target vehicle performance prediction model is determined according to the prediction accuracy and the preset initial model. A target vehicle performance prediction model with high vehicle performance prediction accuracy can be effectively obtained. The performance of the vehicle corresponding to the vehicle design parameters is predicted by the target vehicle performance prediction model, which can effectively improve the efficiency of determining the vehicle design parameters, and thus can effectively shorten the time cost required for the vehicle from design to processing, which is conducive to reducing the design cost of high-performance vehicles.

[0166] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0167] The present disclosure also provides a computer-readable storage medium having computer program instructions stored thereon. When the program instructions are executed by a processor, the steps of the vehicle performance prediction model training method provided by the present disclosure are implemented, or the steps of the vehicle performance prediction method provided by the present disclosure are implemented.

[0168] Figure 4 4 is a block diagram of a device for training a vehicle performance prediction model according to an exemplary embodiment. For example, the device 400 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0169] Reference Figure 4 , the device 400 may include one or more of the following components: a processing component 402 , a memory 404 , a power component 406 , a multimedia component 408 , an audio component 410 , an input / output interface 412 , a sensor component 414 , and a communication component 416 .

[0170] The processing component 402 generally controls the overall operation of the device 400, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 402 may include one or more processors 420 to execute instructions to complete all or part of the steps of the above-mentioned method. In addition, the processing component 402 may include one or more modules to facilitate the interaction between the processing component 402 and other components. For example, the processing component 402 may include a multimedia module to facilitate the interaction between the multimedia component 408 and the processing component 402.

[0171] The memory 404 is configured to store various types of data to support operations on the device 400. Examples of such data include instructions for any application or method operating on the device 400, contact data, phone book data, messages, pictures, videos, etc. The memory 404 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0172] The power supply component 406 provides power to the various components of the device 400. The power supply component 406 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device 400.

[0173] The multimedia component 408 includes a screen that provides an output interface between the device 400 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 408 includes a front camera and / or a rear camera. When the device 400 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0174] The audio component 410 is configured to output and / or input audio signals. For example, the audio component 410 includes a microphone (MIC), and when the device 400 is in an operating mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 404 or sent via the communication component 416. In some embodiments, the audio component 410 also includes a speaker for outputting audio signals.

[0175] The input / output interface 412 provides an interface between the processing component 402 and the peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: a home button, a volume button, a start button, and a lock button.

[0176] The sensor assembly 414 includes one or more sensors for providing various aspects of the status assessment of the device 400. For example, the sensor assembly 414 can detect the open / closed state of the device 400, the relative positioning of components, such as the display and keypad of the device 400, and the sensor assembly 414 can also detect the position change of the device 400 or a component of the device 400, the presence or absence of user contact with the device 400, the orientation or acceleration / deceleration of the device 400, and the temperature change of the device 400. The sensor assembly 414 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 414 may also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 414 may also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0177] The communication component 416 is configured to facilitate wired or wireless communication between the device 400 and other devices. The device 400 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 416 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 416 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0178] In an exemplary embodiment, the device 400 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform a vehicle performance prediction model training method.

[0179] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 404 including instructions, and the instructions can be executed by a processor 420 of the device 400 to complete the vehicle performance prediction model training method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0180] In addition to being an independent electronic device, the above-mentioned device can also be a part of an independent electronic device. For example, in one embodiment, the device can be an integrated circuit (IC) or a chip, wherein the integrated circuit can be an IC or a collection of multiple ICs; the chip can include but is not limited to the following types: GPU (Graphics Processing Unit), CPU (Central Processing Unit), FPGA (Field Programmable Gate Array), DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), SOC (System on Chip, SoC), etc. The above-mentioned integrated circuit or chip can be used to execute executable instructions (or codes) to implement the above-mentioned vehicle performance prediction model training method. The executable instructions can be stored in the integrated circuit or chip, or can be obtained from other devices or equipment, such as the integrated circuit or chip includes a processor, a memory, and an interface for communicating with other devices. The executable instruction can be stored in the vehicle performance prediction model training, and when the executable instruction is executed by the processor, the above-mentioned vehicle performance prediction model training method is implemented; alternatively, the integrated circuit or chip can receive the executable instruction through the interface and transmit it to the processor for execution, so as to implement the above-mentioned vehicle performance prediction model training method.

[0181] In another exemplary embodiment, a computer program product is also provided, which includes a computer program that can be executed by a programmable device, and the computer program has a code portion for executing the above-mentioned vehicle performance prediction model training method or vehicle performance prediction method when executed by the programmable device.

[0182] Figure 5 5 is a block diagram of a device for training a vehicle performance prediction model according to an exemplary embodiment. For example, the device 500 may be provided as a server. Figure 5 The apparatus 500 includes a processing component 522, which further includes one or more processors, and a memory resource represented by a memory 532 for storing instructions executable by the processing component 522, such as an application. The application stored in the memory 532 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 522 is configured to execute the instructions to perform the vehicle performance prediction model training method.

[0183] The device 500 may also include a power supply component 526 configured to perform power management of the device 500, a wired or wireless network interface 550 configured to connect the device 500 to a network, and an input / output interface 558. The device 500 may operate based on an operating system stored in the memory 532.

[0184] In another exemplary embodiment, a computer program product is also provided, which includes a computer program that can be executed by a programmable device, and the computer program has a code portion for executing the above-mentioned vehicle performance prediction model training method or vehicle performance prediction method when executed by the programmable device.

[0185] Figure 6 600 is a block diagram of a vehicle according to an exemplary embodiment. For example, vehicle 600 may be a hybrid vehicle, a non-hybrid vehicle, an electric vehicle, a fuel cell vehicle, or other types of vehicles. Vehicle 600 may be an autonomous vehicle, a semi-autonomous vehicle, or a non-autonomous vehicle.

[0186] Reference Figure 6 , the vehicle 600 may include various subsystems, for example, an infotainment system 610, a perception system 620, a decision control system 630, a drive system 640, and a computing platform 650. The vehicle 600 may also include more or fewer subsystems, and each subsystem may include multiple components. In addition, each subsystem and each component of the vehicle 600 may be interconnected by wire or wireless means.

[0187] In some embodiments, the infotainment system 610 may include a communication system, an entertainment system, and a navigation system, among others.

[0188] The perception system 620 may include several sensors for sensing information about the environment around the vehicle 600. For example, the perception system 620 may include a global positioning system (the global positioning system may be a GPS system, or a Beidou system or other positioning systems), an inertial measurement unit (IMU), a laser radar, a millimeter wave radar, an ultrasonic radar, and a camera.

[0189] The decision control system 630 may include a computing system, a vehicle controller, a steering system, a throttle, and a braking system.

[0190] The drive system 640 may include components that provide powered motion for the vehicle 600. In one embodiment, the drive system 640 may include an engine, an energy source, a transmission system, and wheels. The engine may be one or a combination of an internal combustion engine, an electric motor, and an air compression engine. The engine is capable of converting energy provided by the energy source into mechanical energy.

[0191] Some or all functions of the vehicle 600 are controlled by a computing platform 650. The computing platform 650 may include at least one processor 651 and a memory 652, and the processor 651 may execute instructions 653 stored in the memory 652.

[0192] The processor 651 may be any conventional processor, such as a commercially available CPU. The processor may also include a graphics processor (Graphic Process Unit, GPU), a field programmable gate array (Field Programmable Gate Array, FPGA), a system on chip (System on Chip, SOC), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC) or a combination thereof.

[0193] The memory 652 may be implemented by any type of volatile or nonvolatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0194] In addition to the instructions 653 , the memory 652 may also store data, such as road maps, route information, and data such as the location, direction, and speed of the vehicle. The data stored in the memory 652 may be used by the computing platform 650 .

[0195] In the disclosed embodiment, the processor 651 may execute instruction 653 to complete the above-mentioned vehicle performance prediction model training method, or all or part of the steps of the vehicle performance prediction method.

[0196] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the present disclosure. This application is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The specification and examples are to be considered as exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0197] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A vehicle performance prediction model training method, characterized in that: The method comprises: Acquire a test data set, wherein the test data set includes multiple groups of test sample data, each group of test sample data includes one or more vehicle design parameters; Inputting each group of the test sample data into a preset initial model in sequence to obtain a first predicted performance parameter corresponding to each group of the test sample data output by the preset initial model, wherein the preset initial model is a machine learning model for predicting a target performance parameter of a vehicle; Obtaining simulation performance parameters corresponding to each group of test sample data; Determining the prediction accuracy of the preset initial model according to the first prediction performance parameter and the simulation performance parameter corresponding to each group of the test sample data in the multiple groups of test sample data; A target vehicle performance prediction model is determined according to the prediction accuracy and the preset initial model.

2. The method according to claim 1, characterized in that: The preset initial model is a machine learning model obtained by training a preset initial training data set, and determining the target vehicle performance prediction model according to the prediction accuracy and the preset initial model includes: When it is determined that the prediction accuracy is greater than or equal to a preset accuracy threshold, determining target training data according to the preset initial training data set; Training the preset initial model with the target training data to obtain an updated pending performance prediction model; The target vehicle performance prediction model is determined according to the pending performance prediction model.

3. The method according to claim 2, characterized in that The preset initial training data set includes multiple groups of training sample data, and determining target training data according to the preset initial training data set includes: Inputting each set of training sample data in the plurality of sets of training sample data into the preset initial model in sequence to obtain a first standby performance parameter corresponding to each set of training sample data output by the preset initial model; In the case where it is determined that the first standby performance parameter does not belong to the preset performance parameter range, taking the training sample data corresponding to the first standby performance parameter as abnormal training data; The abnormal training data is deleted from the preset initial training data set to obtain the target training data.

4. The method according to claim 2, characterized in that: The determining target training data according to the preset initial training data set includes: determining a sensitive characteristic design parameter of the one or more vehicle design parameters; Obtaining a supplementary training data set corresponding to the sensitive feature design parameters; The sensitive feature design parameters and the supplementary training data set in the preset initial training data set are used as the target training data.

5. The method according to claim 2, characterized in that: The determining target training data according to the preset initial training data set includes: Acquiring preset constraint condition data, wherein the preset constraint condition data includes a value range of each vehicle design parameter; Inputting the preset initial model and the preset constraint conditions into a preset multi-objective optimization algorithm to obtain an optimal solution set of vehicle design parameters output by the preset multi-objective optimization algorithm, wherein the optimal solution set includes multiple groups of optimal solutions; Inputting each set of optimal solutions into a preset simulation model to obtain a second standby performance parameter output by the preset simulation model; Obtaining a second prediction performance parameter corresponding to the optimal solution; If the difference between the second standby performance parameter and the second predicted performance parameter is greater than a preset deviation threshold, using the optimal solution as pending training data; The preset initial training data set and the pending training data are used as the target training data.

6. The method according to claim 1, characterized in that The determining the prediction accuracy of the preset initial model according to the first prediction performance parameter and the simulation performance parameter corresponding to each group of the test sample data in the plurality of groups of test sample data includes: Determine, for each set of test sample data, a difference between the first predicted performance parameter and the simulated performance parameter; The prediction accuracy of the preset initial model is determined according to the difference.

7. The method according to claim 1, characterized in that The step of determining the target vehicle performance prediction model according to the prediction accuracy and the preset initial model includes: When it is determined that the prediction accuracy is less than a preset accuracy threshold, the preset initial model is used as the target vehicle performance prediction model.

8. The method according to claim 1, characterized in that: The obtaining of simulation performance parameters corresponding to each group of test sample data includes: Each group of the test sample data is input into a preset simulation model in turn to obtain the simulation performance parameters corresponding to the test sample data output by the preset simulation model.

9. A vehicle performance prediction method, characterized in that: The method comprises: Obtain target design parameters; The target design parameters are input into the target vehicle performance prediction model described in any one of claims 1 to 8 to obtain target performance parameters corresponding to the target design parameters output by the target vehicle performance prediction model.

10. A vehicle performance prediction model training device, characterized in that: The device comprises: A first acquisition module is configured to acquire a test data set, wherein the test data set includes multiple groups of test sample data, each group of test sample data includes one or more vehicle design parameters; A second acquisition module is configured to input each group of the test sample data into a preset initial model in sequence to obtain a first predicted performance parameter corresponding to each group of the test sample data output by the preset initial model, wherein the preset initial model is a machine learning model for predicting a target performance parameter of a vehicle; A third acquisition module is configured to acquire simulation performance parameters corresponding to each group of test sample data; A first determination module is configured to determine the prediction accuracy of the preset initial model according to the first prediction performance parameter and the simulation performance parameter corresponding to each group of the test sample data in the multiple groups of test sample data; The second determination module is configured to determine a target vehicle performance prediction model according to the prediction accuracy and the preset initial model.

11. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to execute the steps of the method described in any one of claims 1 to 8, or to execute the steps of the method described in claim 9.

12. A vehicle, characterized in that: include: The electronic device as claimed in claim 11.

13. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the program instructions are executed by a processor, the steps of the method described in any one of claims 1 to 8 are implemented, or the steps of the method described in claim 9 are executed.

Images