Vehicle parameter determination method and device, electronic equipment and storage medium

Abstract

The invention relates to a vehicle parameter determination method and device, electronic equipment and a storage medium, and relates to the technical field of data processing. Comprising the steps of obtaining a multi-objective optimization function according to at least one performance prediction model and constraint conditions; the performance prediction model is used for obtaining a vehicle performance value according to vehicle parameters, and the constraint condition is used for constraining the vehicle parameters and / or the vehicle performance value; and at least one group of vehicle parameters meeting the constraint condition is obtained by taking minimization or maximization of the output value of the multi-objective optimization function as a target. By using the vehicle parameter determination method provided by the invention, the vehicle parameters meeting the vehicle performance value can be quickly obtained.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, and in particular to a vehicle parameter determination method, device, electronic device and storage medium. Background Art

[0002] In the related art, the user will input vehicle parameters into the simulation software to obtain vehicle performance values ​​related to the vehicle. When the user wants the vehicle parameters corresponding to the maximum vehicle performance value, the user will adjust the vehicle parameters input into the simulation software to make the simulation software output the maximum vehicle performance value in multiple simulations. The vehicle parameters corresponding to the maximum vehicle performance value are then applied to the actual vehicle, so that the actual vehicle can produce a larger vehicle performance value.

[0003] Normally, simulation software takes a long time to perform a simulation. Every time a user wants to achieve a certain vehicle performance value, he needs to constantly adjust the vehicle parameters. The simulation software will go through multiple simulations to obtain the vehicle performance value that meets the user's needs. Each simulation takes a long time. Therefore, after the user adjusts the required vehicle performance value, the process of obtaining the vehicle parameters is also long. Summary of the invention

[0004] In order to overcome the problems existing in the related art, the present disclosure provides a vehicle parameter determination method, device, electronic device and storage medium.

[0005] According to a first aspect of an embodiment of the present disclosure, a method for determining vehicle parameters is provided, comprising:

[0006] A multi-objective optimization function is obtained according to at least one performance prediction model and constraint conditions; the performance prediction model is used to obtain a vehicle performance value according to vehicle parameters, and the constraint conditions are used to constrain the vehicle parameters and / or the vehicle performance value;

[0007] With the goal of minimizing or maximizing the output value of the multi-objective optimization function, at least one set of vehicle parameters that meets the constraint conditions is obtained.

[0008] Optionally, a multi-objective optimization function is obtained according to at least one performance prediction model and constraint conditions, including:

[0009] According to the performance prediction model, a performance prediction function is obtained, wherein the performance prediction function is used to obtain the vehicle performance value according to the vehicle parameters;

[0010] The multi-objective optimization function is obtained according to the performance prediction function and the constraint conditions.

[0011] Optionally, at least one of the performance prediction models includes a stiffness prediction model and a modal prediction model; a multi-objective optimization function is obtained according to at least one performance prediction model and constraint conditions, including:

[0012] The multi-objective optimization function is obtained according to the stiffness prediction model, the modal prediction model and the constraint conditions; wherein the stiffness prediction model is used to output a stiffness performance value according to the vehicle parameters, the modal prediction model is used to output a modal performance value according to the vehicle parameters, and the constraint conditions are used to constrain at least one of the vehicle parameters, the stiffness performance value and the modal performance value.

[0013] Optionally, a multi-objective optimization function is obtained according to at least one performance prediction model and constraint conditions, including:

[0014] In response to the modification operation on the constraint condition, a modified target constraint condition is obtained;

[0015] The multi-objective optimization function is obtained according to at least one of the performance prediction models and the objective constraints.

[0016] Optionally, the method further comprises:

[0017] Using vehicle parameter samples as training samples of the initial model, using vehicle performance value samples corresponding to the vehicle parameter samples as labels of the initial model, and training the initial model;

[0018] When a first error between the vehicle performance value predicted by the initial model and the vehicle performance value sample satisfies a first convergence condition, the initial model when the first error satisfies the first convergence condition is used as the performance prediction model.

[0019] Optionally, the method further comprises:

[0020] In the case where the vehicle parameters are changed, the changed target vehicle parameters are used as training samples of the performance prediction model, and the target vehicle performance values ​​corresponding to the target vehicle parameters are used as labels of the performance prediction model to train the performance prediction model;

[0021] The network parameters of the performance prediction model are adjusted until a second error between the vehicle performance value output by the performance prediction model and the target vehicle performance value satisfies a second convergence condition.

[0022] Optionally, the method further comprises:

[0023] The vehicle parameter samples are used as input parameters of simulation software to obtain vehicle performance value samples corresponding to the vehicle parameter samples.

[0024] According to a second aspect of an embodiment of the present disclosure, there is provided a vehicle parameter determination device, comprising:

[0025] A determination module is configured to obtain a multi-objective optimization function according to at least one performance prediction model and a constraint condition; the performance prediction model is used to obtain a vehicle performance value according to a vehicle parameter, and the constraint condition is used to constrain the vehicle parameter and / or the vehicle performance value;

[0026] The optimization module is configured to obtain at least one set of vehicle parameters that meet the constraint conditions by minimizing or maximizing the output value of the multi-objective optimization function.

[0027] According to a third aspect of an embodiment of the present disclosure, there is provided an electronic device, including:

[0028] processor;

[0029] a memory for storing processor-executable instructions;

[0030] Wherein, the processor is configured to:

[0031] Execute the steps of the vehicle parameter determination method provided in the first aspect of the embodiment of the present disclosure.

[0032] According to a fourth 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 parameter determination method provided in the first aspect of the present disclosure are implemented.

[0033] According to a fifth aspect of an embodiment of the present disclosure, a chip is provided, including a processor and an interface; the processor is used to read instructions to execute the steps of the vehicle parameter determination method provided in the first aspect of the present disclosure.

[0034] According to a sixth aspect of an embodiment of the present disclosure, a vehicle is provided, on which the electronic device provided by the third aspect of the embodiment of the present disclosure is configured, or connected to the electronic device provided by the third aspect of the embodiment of the present disclosure.

[0035] The technical solution provided by the embodiments of the present disclosure may have the following beneficial effects:

[0036] In the present disclosure, the goal is to maximize or minimize the output value of the multi-objective optimization function, which can be regarded as maximizing the vehicle performance value output by the performance prediction model. In this way, on the basis of maximizing the vehicle performance value, at least one set of vehicle parameters is found through the multi-objective optimization function to make the vehicle performance value better. There is no need to perform multiple simulations through the simulation software to solve the vehicle parameters that make the vehicle performance value better, thereby saving the time brought by multiple simulations and improving the efficiency of obtaining vehicle parameters.

[0037] And when the user's needs change, the constraints can be modified directly, and the multi-objective optimization function can also solve at least one set of vehicle parameters according to the modified constraints to adapt to the user's current modified constraints without repeated simulations.

[0038] 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

[0039] 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.

[0040] Figure 1 The figure is a flow chart of a method for determining vehicle parameters according to an exemplary embodiment.

[0041] Figure 2 It is a schematic diagram showing a multi-objective optimization function combined with a performance prediction model according to an exemplary embodiment.

[0042] Figure 3 The diagram is a schematic diagram showing a method of training a performance prediction model according to an exemplary embodiment.

[0043] Figure 4 The figure is a block diagram of a vehicle parameter determination device according to an exemplary embodiment.

[0044] Figure 5 It is a functional block diagram of an electronic device according to an exemplary embodiment.

[0045] Figure 6 is a functional block diagram of a vehicle according to an exemplary embodiment. DETAILED DESCRIPTION

[0046] 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.

[0047] 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.

[0048] In the related technology, the function of the simulation software is to obtain the vehicle performance value based on the input vehicle parameters. Every time the user wants a certain vehicle performance value to be maximum, the vehicle parameters will be continuously adjusted. The simulation software will go through multiple simulations to obtain the maximum vehicle performance value that meets the user's needs. The user will then apply the vehicle parameters when the maximum vehicle performance value is reached to the actual vehicle, so that the actual vehicle can produce the maximum vehicle performance value required by the user.

[0049] For example, vehicle performance values ​​include the vehicle's stiffness performance value and modal performance value. When the user wants the maximum modal performance value, the vehicle parameters will be continuously adjusted to obtain the maximum modal performance value through multiple simulations of the simulation software. The user then applies the vehicle parameters corresponding to the maximum modal performance value to the actual vehicle, so that the actual vehicle can use the vehicle parameters to produce greater modal performance.

[0050] It can be seen that obtaining the vehicle parameters corresponding to the maximum vehicle performance value that meets user needs requires multiple simulations, and the simulation software takes a long time to perform a simulation, so it takes a long time to obtain the vehicle parameters corresponding to the maximum vehicle performance value that meets user needs.

[0051] In addition, the vehicle performance values ​​required by users will change frequently. After each change in the vehicle performance value, if you want to obtain the vehicle parameters corresponding to the changed vehicle performance value, you need to simulate multiple times, which means that each time the user changes the demand, you need to wait for a long time to obtain the required vehicle parameters. For example, when the user needs the maximum stiffness performance value for the first time, the vehicle parameters input into the simulation software will be adjusted, and the vehicle parameters will be adjusted continuously to obtain the vehicle parameters corresponding to the maximum stiffness performance value through multiple simulations; when the user needs the maximum modal performance value for the second time, the vehicle parameters input into the simulation software will also be adjusted, and the vehicle parameters will be adjusted continuously to obtain the vehicle parameters corresponding to the maximum modal performance value through multiple simulations.

[0052] It can be seen that the use of simulation software in the related art to obtain vehicle parameters corresponding to vehicle performance values ​​that meet user needs requires users to wait for a long time and its speed is slow.

[0053] Based on this, see Figure 1 , Figure 1 A vehicle parameter determination method is shown, and the vehicle parameter determination method is applied to an electronic device, which may be a terminal such as a computer or a mobile phone. The vehicle parameter determination method includes the following steps:

[0054] In step S11, a multi-objective optimization function is obtained according to at least one performance prediction model and constraint conditions.

[0055] The performance prediction model is used to obtain vehicle performance values ​​based on vehicle parameters.

[0056] Vehicle parameters refer to information such as the size, weight, and material of sheet metal parts and other components on the vehicle. The size includes the length, thickness, and height of the components, and the materials include plastic, metal, etc.

[0057] The vehicle performance value is a parameter related to vehicle performance, and the vehicle performance value includes stiffness performance value, modal performance value, collision safety performance value, weight, etc.

[0058] The stiffness performance value reflects the rigidity or ability to resist deformation of a component under stress. The higher the stiffness, the smaller the deformation of the component under stress, and the higher the stability of the vehicle. The stiffness performance value includes torsional stiffness and bending stiffness. Torsional stiffness is the ability of a component to resist torsional force, which can also be understood as the degree of deformation of a component under torsional load. The higher the torsional stiffness of a component, the smaller the deformation. Bending stiffness refers to the ability of a component to resist bending force, which can also be understood as the degree of deformation of a component under bending load. The higher the bending stiffness of a component, the smaller the deformation.

[0059] The modal performance value reflects the characteristics and behaviors of different vibration modes generated by the components when they are stimulated by external factors. The quality of modal performance is affected by factors such as natural frequency, modal shape, modal damping, and modal coupling. The modal performance values ​​include torsional mode and bending mode. The torsional mode is the vibration mode generated by the component when it is subjected to torsional force. In torsional vibration, different parts of the component rotate around its longitudinal axis or twist relatively. The natural frequency and vibration mode of the torsional mode depend on factors such as the geometry, material properties, and boundary conditions of the component. The bending mode is the vibration mode generated by the component when it is subjected to bending force. In bending vibration, the component bends or bends along its transverse axis. Similar to the torsional mode, the natural frequency and vibration mode of the bending mode also depend on factors such as the geometry, material properties, and boundary conditions of the component.

[0060] The collision safety performance value refers to the vehicle's ability to protect the user from injury in the event of a collision or impact. The higher the collision safety performance value, the greater the protection for the user.

[0061] Weight refers to the weight of the vehicle. In order to ensure the lightweight of the vehicle, generally speaking, the lighter the vehicle weight, the higher the performance of the vehicle.

[0062] Constraints are used to constrain vehicle parameters and / or vehicle performance values. Constraints can also be understood as being used to constrain input parameters input into a performance prediction model and output parameters output from a performance prediction model. For example, when constraints are used to constrain vehicle parameters, taking the vehicle parameter as the thickness of a component as an example, the constraints can be constraining the width of the component, limiting the thickness of the component within the range of [0, 10], and [0, 10] is a constraint for constraining the thickness of the component. For another example, when constraints are used to constrain vehicle performance values, taking the vehicle performance value as the stiffness performance value as an example, the constraints can be constraining the stiffness performance value, limiting the stiffness performance value within the range of [50, +∞], and [50, +∞] is a constraint for constraining the stiffness performance value.

[0063] The constraint condition is a constraint condition that constrains the vehicle performance value, and the constraint condition can also be a constraint condition set by the vehicle in order to balance the performance values ​​of other vehicles.

[0064] For example, if the multi-objective optimization function is composed of the performance prediction function corresponding to the modal performance prediction model and the performance prediction function corresponding to the weight prediction function, the greater the modal performance value output by the modal performance prediction model, the greater the vehicle weight output by the corresponding weight prediction function. If the optimization goal set by the user is to maximize the modal performance value, in the process of ensuring the maximum modal performance value as much as possible, the vehicle weight will not be increased indiscriminately to ensure the improvement of the modal performance value. The vehicle will preset the constraint condition that the vehicle weight is less than the preset weight, so that the modal performance value is guaranteed to reach the maximum value under the constraint condition that the vehicle weight is less than the preset weight.

[0065] Constraints can also be optimization goals set by users.

[0066] For example, the terminal preset constraints are that the stiffness performance value is greater than 30 and the vehicle weight is equal to 30. At this time, the user can also set an additional constraint that the modal performance value is greater than 20. In this way, the multi-objective optimization function will use the stiffness performance value greater than 30, the vehicle weight equal to 30 and the modal performance value greater than 20 as optimization goals to obtain at least one set of vehicle parameters. These at least one set of vehicle parameters can meet the constraints preset by the terminal and set by the user.

[0067] When the constraint conditions for the vehicle performance value change, it indicates that the vehicle performance value required by the user has changed, that is, the optimization target has changed. For example, the user changes the constraint conditions from maximum stiffness performance value to maximum modal performance value, or to minimum mass performance value.

[0068] A performance prediction function can be obtained based on the performance prediction model, and the performance prediction function is used to obtain vehicle performance values ​​based on vehicle parameters; and then the multi-objective optimization function is obtained based on the performance prediction function and the constraint conditions.

[0069] The multi-objective optimization function is obtained by the following formula:

[0070] A=f1(x),f2(x),...,f m (x)

[0071] Where A is the output value of the multi-objective optimization function, f m (x) is a performance prediction function, which can be a prediction function obtained through a performance prediction model or a weight prediction function. For example, f1(x), f2(x), and f3(x) are the performance prediction functions corresponding to the stiffness prediction model, the modal prediction model, and the collision performance prediction model, respectively, and f4(x) is a weight prediction function. x is a vehicle parameter.

[0072] It can be understood that in this formula, the type of vehicle parameters input into each performance prediction function is consistent. For example, if the input into f1(x) is the thickness of sheet metal A, the thickness of sheet metal B, and the thickness of sheet metal C, then the vehicle parameters input into the remaining performance prediction functions are also the thickness of sheet metal A, the thickness of sheet metal B, and the thickness of sheet metal C.

[0073] The multi-objective optimization function is a set of performance prediction functions. Each performance prediction function in the multi-objective optimization function represents an optimization target (the optimization target is the vehicle performance value). There are correlations or conflicts between these optimization targets. The purpose of the multi-objective optimization function is to find at least one set of solutions (the solutions are vehicle parameters) so as to solve the multi-objective optimization problem. For example, if the stiffness performance value f1(x) is required to be greater than the first preset value, the modal performance value f2(x) is required to be greater than the second preset value, the collision safety performance value f3(x) is required to be greater than the third preset value, and the mass f4(x) of the vehicle is equal to 0, the multi-objective optimization function can find at least one set of vehicle parameters based on these performance prediction functions, which can simultaneously meet the conditions of these performance prediction functions.

[0074] The performance prediction model includes a stiffness prediction model and a modal prediction model. It can be seen from the above formula that the performance prediction function in the multi-objective optimization function can be obtained according to the stiffness prediction model, the modal prediction model and the collision performance prediction model.

[0075] In step S12, at least one set of vehicle parameters satisfying the constraint conditions is obtained with the goal of minimizing or maximizing the output value of the multi-objective optimization function.

[0076] In the field of vehicle technology, it is usually necessary to maximize the modal performance value, stiffness performance value and collision safety performance value of the vehicle and minimize the weight of the vehicle, so that the quality of the vehicle will be better. In order to unify the various performance prediction functions in the multi-objective optimization function, the above multi-objective optimization function has the following two expressions.

[0077] Method 1: The multi-objective optimization function is:

[0078] min A=f1(x), f2(x),...,f m (x)

[0079] The constraints are:

[0080] g j (x)≤0,j=1,2,...,p

[0081] h k (x) = 0, k = 1, 2, ..., p

[0082] In the formula, g j (x) is an inequality constraint condition, which may represent that the negative value of the constraint stiffness performance value, the negative value of the modal performance value, or the negative value of the collision safety performance value is less than or equal to 0, that is, the constraint stiffness performance value is greater than 0, or the constraint modal performance value is greater than 0, or the collision safety performance value is greater than 0; h k (x) is an equality constraint, which means that the mass of the vehicle is constrained to be equal to 0.

[0083] It can be understood that the constraints here can also be understood as the preset stiffness performance value, modal performance value or collision safety performance value being greater than 0, and the mass of the vehicle tends to 0. Therefore, the desired conditions are used as constraints to constrain the multi-objective optimization function, so that the obtained vehicle parameters can meet the constraints required by the user.

[0084] In method 1, if the function of the stiffness performance prediction model is Y1(x), and f1(x) is used to express the performance prediction function corresponding to the stiffness performance prediction model, f1(x) = -Y1(x); if the function of the modal performance prediction model is Y2(x), and f2(x) is used to express the performance prediction function corresponding to the modal performance prediction model, f2(x) = -Y2(x); if the function of the collision safety performance prediction model is Y3(x), and f3(x) is used to express the performance prediction function corresponding to the collision safety performance prediction model, f3(x) = -Y3(x); if the weight prediction function is Y4(x), and f4(x) is used to express the performance prediction function of the weight, f4(x) = Y4(x).

[0085] On the basis of the function of the stiffness performance prediction model, the function of the modal performance prediction model and the function of the collision safety performance prediction model, negative values ​​are added to transform them into performance prediction functions; and the reason why negative values ​​are not added to the weight prediction function is that when minimizing the output value of the multi-objective optimization function is the goal, it can be regarded as minimizing the performance prediction function, and the performance prediction function is the negative value of the function of the stiffness performance prediction model, the function of the modal performance prediction model and the function of the collision safety performance prediction model, so minimizing the performance prediction function can be regarded as maximizing the stiffness performance value output by the stiffness performance prediction model, the modal performance value output by the modal performance prediction model and the collision safety performance value output by the collision safety performance prediction model; and the performance prediction function is also a weight prediction function, so minimizing the performance prediction function can also be regarded as minimizing the vehicle weight output by the weight prediction function.

[0086] Through method 1, negative values ​​are added to the function of the performance prediction model to transform it into a performance prediction function, and the weight prediction function is not deformed. The weight prediction function is directly used as the performance prediction function. When the goal is to minimize the multi-objective function, it can be ensured that at least one set of vehicle parameters is obtained by maximizing the vehicle performance value output by the performance prediction model and minimizing the vehicle weight.

[0087] In method 1, the user can also modify the constraint conditions. The electronic device responds to the user's modification operation on the constraint conditions, obtains the modified target constraint conditions, and obtains a multi-objective optimization function based on at least one performance prediction model and the target constraint conditions.

[0088] For example, if the user wants the stiffness performance value to be greater than 50, the modal performance value to be greater than 30, and the vehicle weight to be less than 100, the terminal will automatically generate the constraint conditions: g j 1(x)≤-50, g j 1(x) is the negative value of the stiffness performance value; g j 2(x)≤-30, g j 2(x) is the negative value of the modal performance value; h k (x)<100, h k (x) is the vehicle weight. In this way, the terminal will obtain at least one set of vehicle parameters based on the constraints and with the goal of minimizing the output value of the multi-objective optimization function.

[0089] Method 2, the multi-objective optimization function is:

[0090] max A=f1(x), f2(x),...,f m (x)

[0091] The constraints are:

[0092] g j(x)>0,j=1,2,...,p

[0093] h k (x) = 0, k = 1, 2, ..., p

[0094] In the formula, g j (x) is an inequality constraint condition, which may represent a constraint stiffness performance value, a modal performance value, or a collision safety performance value greater than 0; h k (x) is an equality constraint, which means that the mass of the vehicle is constrained to be equal to 0.

[0095] It can be understood that the constraints here can also be understood as the stiffness performance value, modal performance value or collision safety performance value that the user wants to achieve is greater than 0, and the mass of the vehicle tends to 0. Therefore, the desired conditions are used as constraints to constrain the multi-objective optimization function, so that the obtained vehicle parameters can meet the constraints required by the user.

[0096] In method 2, if the function of the stiffness performance prediction model is Y1(x), and f1(x) is used to express the performance prediction function corresponding to the stiffness performance prediction model, f1(x) = Y1(x); if the function of the modal performance prediction model is Y2(x), and f2(x) is used to express the performance prediction function corresponding to the modal performance prediction model, f2(x) = Y2(x); if the function of the collision safety performance prediction model is Y3(x), and f3(x) is used to express the performance prediction function corresponding to the collision safety performance prediction model, f3(x) = Y3(x); if the weight prediction function is Y4(x), and f4(x) is used to express the performance prediction function of the weight, f4(x) = -Y4(x).

[0097] The functions of the stiffness performance prediction model, the modal performance prediction model and the collision safety performance prediction model are directly used as the performance prediction function; and the reason for adding a negative value on the basis of the weight prediction function is that when the goal is to maximize the output value of the multi-objective optimization function, it can be regarded as maximizing the performance prediction function, and the performance prediction function is a function of the stiffness performance prediction model, a function of the modal performance prediction model and a function of the collision safety performance prediction model, so the maximization performance prediction function can be regarded as maximizing the stiffness performance value output by the stiffness performance prediction model, the modal performance value output by the modal performance prediction model and the collision safety performance value output by the collision safety performance prediction model; and the performance prediction function is deformed by adding a negative value to the weight prediction function, so the maximization performance prediction function can also be regarded as minimizing the vehicle weight output by the weight prediction function.

[0098] In method 2, the user can also modify the constraint conditions. The electronic device responds to the user's modification operation on the constraint conditions, obtains the modified target constraint conditions, and obtains a multi-objective optimization function based on at least one performance prediction model and the target constraint conditions.

[0099] For example, if the user wants the stiffness performance value to be greater than 50, the modal performance value to be greater than 30, and the vehicle weight to be less than 100, the terminal will automatically generate the constraint conditions: g j 1(x)>50,g j 1(x) is the stiffness performance value; g j 2(x)>30,g j 2(x) is the modal performance value; h k (x)>-100;h k (x) is the negative value of the vehicle weight. In this way, the terminal will obtain at least one set of vehicle parameters based on the constraints and with the goal of maximizing the output value of the multi-objective optimization function.

[0100] Common multi-objective optimization functions include genetic functions, particle functions, simulated annealing functions, differential evolution functions, etc., and these multi-objective optimization functions can only obtain a set of solutions, that is, a set of vehicle parameters. In the process of multi-objective optimization, there may be multiple objectives that need to be optimized. For example, the stiffness performance value needs to be greater than the first preset value, the modal performance value needs to be greater than the second preset value, the collision safety performance value needs to be greater than the third preset value, and the vehicle weight needs to be less than the fourth preset value. There may be conflicts between these optimization objectives. The performance improvement of some optimization objectives will be accompanied by the deterioration of the performance of other optimization objectives. For example, after the stiffness performance value is improved, the modal performance value will decrease and the vehicle weight will increase. At this time, if the above-mentioned multi-objective optimization function is used, it is impossible to meet the needs of optimizing multiple optimization objectives at the same time.

[0101] Based on this, the present disclosure can adopt multi-objective optimization functions such as target fitness function, non-dominated sorting genetic function (NSGA), Pareto frontier (Pareto) to solve these problems.

[0102] The following will use a scenario to explain the process of the NSGA function obtaining at least one set of vehicle parameters in order to satisfy multiple vehicle performance values.

[0103] Assuming the constraints set by the user, that is, the optimization objectives set by the user include: Optimization objective 1: minimizing the vehicle weight; Optimization objective 2: maximizing the modal performance value.

[0104] When applying the NSGA function to meet multiple optimization objectives, you can follow the following sub-steps:

[0105] (1) Initialize the population: Randomly generate a set of individuals with vehicle parameters as the initial population, for example:

[0106] Thickness of sheet metal part A: 1mm;

[0107] Thickness of sheet metal part B: 2mm;

[0108] Thickness of sheet metal part C: 1.5 mm.

[0109] (2) Fitness evaluation: The fitness value of each individual is evaluated according to different performance test functions. The vehicle weight and modal performance value of each individual can be obtained through simulation and testing.

[0110] For example, the thickness of the sheet metal A is 1mm, the thickness of the sheet metal B is 2mm, and the thickness of the sheet metal C is 1.5mm. The three individuals are respectively input into the performance prediction function corresponding to the weight prediction function to obtain the vehicle weight corresponding to each individual; and then the three individuals are respectively input into the performance prediction function corresponding to the modal performance prediction model to obtain the modal performance value corresponding to each individual.

[0111] (3) Fast non-dominated sorting: According to the fitness value of the individual, the individuals in the initial population are non-dominated and sorted. The individuals are divided into different levels, where the first level is the non-dominated solution. For example, after sorting, the following individual sequence is obtained:

[0112] Individual 1: modal performance value = 10, vehicle weight = 200;

[0113] Individual 2: modal performance value = 5, vehicle weight = 100;

[0114] Instance 3: Modal performance value = 30, vehicle weight = 170.

[0115] In this example, both individuals 2 and 3 are first-level non-dominated solutions because individual 2 has the smallest vehicle weight and individual 3 has the largest modal performance value, and no other individual performs better in both modal performance value and vehicle weight.

[0116] (4) Calculate the crowding distance: Calculate the crowding distance of individuals in the first-level non-dominated solution. The crowding distance measures the distribution density of individuals in the solution space. A higher crowding distance means that the individual is farther away from the surrounding individuals and has better diversity.

[0117] Among the three individuals above, the crowding distance between individual 2 and individual 3 can be calculated.

[0118] (5) Selection: Select the next generation of individuals based on the individual ranking and crowding distance in (3).

[0119] For selecting the next generation of individuals: first, you can choose a non-dominated solution with a lower level, because a non-dominated solution with a lower level represents a better solution; secondly, if two individuals are evenly divided in the dominance relationship and the crowding distance of one of the individuals is larger, then this individual will be given priority as the next generation individual.

[0120] For example, among the three individuals above, individual 2 can be selected as the next generation individual because individual 2 is ranked at a lower level in the non-dominated sorting and has a larger crowding distance.

[0121] For retaining elite individuals: the elite individual is the best individual in the previous generation, that is, the individual with the best fitness value. For example, among the three individuals above, individual 1 can be retained as the elite individual.

[0122] (6) Crossover and mutation: The next generation of individuals after selection are subjected to crossover and mutation operator operations to generate new individuals. New solutions and genetic diversity can be introduced through crossover and mutation.

[0123] (7) Update the population: merge the newly generated individuals with the original population to update the population.

[0124] (8) Termination condition: If the termination condition is met (for example, the maximum number of iterations is reached or a sufficiently good solution is found), the algorithm stops; otherwise, return to substep (3).

[0125] By continuously iterating the above steps, the NSGA function will generate at least one set of vehicle parameters with high diversity and good non-dominance. These vehicle parameters can present the trade-off between modal performance values ​​and vehicle weight. Users can select vehicle parameters that meet actual needs from at least one set of vehicle parameters.

[0126] See also Figure 2 As shown, the present disclosure combines a multi-objective optimization algorithm with various types of performance prediction models, utilizing the characteristics of the performance prediction model that can quickly output vehicle performance values ​​based on vehicle input parameters, and automatically solves for at least one set of vehicle parameters that meet the requirements according to the optimization goals (constraints) set by the user.

[0127] After the user obtains at least one set of vehicle parameters, the user can still input the vehicle parameters into the simulation software to determine whether the vehicle performance values ​​output by the simulation software are consistent with the vehicle performance values ​​predicted by the performance prediction model, thereby verifying the accuracy of the performance prediction model's prediction.

[0128] In addition, if the multiple sets of vehicle parameters solved based on the multi-objective algorithm still do not meet user needs, the values ​​of the obtained vehicle parameters can be modified. For example, the sheet metal thickness can be changed from 4 to 5, and then check whether the vehicle performance value is improved. This allows users to dynamically adjust the values ​​of the solved vehicle parameters so that the obtained vehicle performance value meets user needs.

[0129] Through the above technical solution, maximizing or minimizing the output value of the multi-objective optimization function is the goal, which can be regarded as maximizing the vehicle performance value output by the performance prediction model and minimizing the vehicle weight. In this way, on the basis of maximizing the vehicle performance value and minimizing the vehicle weight, at least one set of vehicle parameters is found through the multi-objective optimization function to maximize the vehicle performance value and minimize the vehicle weight. There is no need to perform multiple simulations through simulation software to solve the vehicle parameters, thereby saving the time consumed by multiple simulations and improving the efficiency of obtaining vehicle parameters.

[0130] Furthermore, when user requirements change, the constraints can be modified directly, and the multi-objective optimization function can also solve at least one set of vehicle parameters based on the modified constraints to adapt to the user's current modified constraints without the need for repeated simulations.

[0131] The specific embodiments involved in the above steps S11 to S12 will be introduced below.

[0132] Figure 3 A schematic diagram of training the initial model to obtain a performance prediction model is shown. The training to obtain the performance prediction model includes the following steps:

[0133] In step S21, the vehicle parameter samples are used as training samples of the initial model, and the vehicle performance value samples corresponding to the vehicle parameter samples are used as labels of the initial model to train the initial model.

[0134] The performance prediction model may include a stiffness performance prediction model, a modal performance prediction model and a collision safety performance prediction model.

[0135] Vehicle parameter samples can be collected from the sample pool through Latin hypercube sampling, thereby generating multiple groups of non-repetitive vehicle parameter samples.

[0136] For example, when training the performance prediction model, the vehicle parameters include sheet metal A, sheet metal B, and sheet metal C. The parameter range corresponding to each sheet metal part is as follows:

[0137] Vehicle parameter 1 (length of sheet metal part A): value range is [0,10];

[0138] Vehicle parameter 2 (length of sheet metal part B): value range is [100,200];

[0139] Vehicle parameter 3 (length of sheet metal part C): the value range is [0,1].

[0140] Through Latin hypercube sampling, a sample set (sample 1 to sample 10) can be generated, in which each vehicle parameter has a unique value. The following is an example of Latin hypercube sampling:

[0141] Sample 1: vehicle parameter 1 = 2, vehicle parameter 2 = 140, vehicle parameter 3 = 0.5;

[0142] Sample 2: vehicle parameter 1 = 7, vehicle parameter 2 = 185, vehicle parameter 3 = -0.7;

[0143] Sample 3: vehicle parameter 1 = 4, vehicle parameter 2 = 115, vehicle parameter 3 = -0.2;

[0144] Sample 4: vehicle parameter 1 = 9, vehicle parameter 2 = 160, vehicle parameter 3 = 0.1;

[0145] Sample 5: vehicle parameter 1 = 1, vehicle parameter 2 = 175, vehicle parameter 3 = -0.9;

[0146] Sample 6: vehicle parameter 1 = 6, vehicle parameter 2 = 125, vehicle parameter 3 = 0.8;

[0147] Sample 7: vehicle parameter 1 = 3, vehicle parameter 2 = 155, vehicle parameter 3 = 0.3;

[0148] Sample 8: vehicle parameter 1 = 8, vehicle parameter 2 = 195, vehicle parameter 3 = -0.5;

[0149] Sample 9: vehicle parameter 1 = 0, vehicle parameter 2 = 130, vehicle parameter 3 = 0.4;

[0150] Sample 10: Vehicle parameter 1 = 5, vehicle parameter 2 = 170, vehicle parameter 3 = 0.1.

[0151] The above samples 1 to 10 may be input into the initial model to train the initial model to obtain a performance prediction model.

[0152] When training the initial model, the vehicle parameter samples input into the initial model can be first input into the simulation software, and the vehicle performance value samples can be obtained through the simulation software. The vehicle performance value samples are real and accurate vehicle performance values; then the vehicle parameter samples are used as input parameters of the initial model, and the vehicle performance value samples are used as labels of the initial model to train the initial model.

[0153] For example, when training the stiffness performance prediction model, the above 10 samples can be respectively input into the simulation software, and the simulation software obtains 10 corresponding stiffness performance values; then the above 10 samples and the stiffness performance values ​​corresponding to the 10 samples are respectively input into the initial model to train the initial model.

[0154] For the samples in the sample pool that do not conform to the normal distribution, we can focus on collecting these samples, that is, collect more samples that do not conform to the normal distribution. Because these samples will have a greater impact on the training results, we can focus on collecting these samples to train the initial model so that the initial model can recognize different types of data.

[0155] In step S22, when a first error between the vehicle performance value predicted by the initial model and the vehicle performance value sample satisfies a first convergence condition, the initial model when the first error satisfies the first convergence condition is used as the performance prediction model.

[0156] The first error between the vehicle performance value predicted by the initial model and the actual vehicle performance value sample output by the simulation software satisfies the first convergence condition. For example, the first error is less than the preset value 0.1, or the first error cannot continue to decrease, etc., which can be regarded as the first error satisfying the first convergence condition. The initial model when the first error satisfies the first convergence condition is used as the performance prediction model.

[0157] The performance prediction model when the first error satisfies the first convergence condition means that the difference between the predicted value output by the performance prediction model and the true actual value is small, so that the vehicle performance value output by the performance prediction model can be close to the true value.

[0158] Through the above technical solution, the performance prediction model obtained through training can be used to replace the simulation software to predict the vehicle performance value, without the need to input the vehicle parameters into the simulation software for simulation to obtain the vehicle performance value, thereby improving the speed of obtaining the vehicle performance value.

[0159] In the related art, when the vehicle design model changes, for example, when a sheet metal part on the vehicle is replaced from sheet metal part A to sheet metal part B, the trained performance prediction model can well predict the vehicle performance value corresponding to sheet metal part A, but cannot well predict the vehicle performance value corresponding to the changed sheet metal part B. The previously trained performance prediction model cannot use sheet metal part B.

[0160] The conventional solution is to reuse the vehicle parameters after the vehicle design model is changed, such as the vehicle parameters of sheet metal part B, to retrain the initial model to obtain a new performance prediction model. However, retraining a new performance prediction model requires the simulation software to simulate the vehicle performance values ​​corresponding to more vehicle parameters, which will also result in a slower speed in obtaining the new performance prediction model, and thus a slower speed in obtaining the vehicle performance values ​​after the vehicle design model is changed.

[0161] Based on this, the present disclosure also proposes: when the vehicle parameters change, the changed target vehicle parameters are used as training samples of the performance prediction model, and the target vehicle performance values ​​corresponding to the target vehicle parameters are used as labels of the performance prediction model to train the performance prediction model; the network parameters of the performance prediction model are adjusted until the second error between the vehicle performance value output by the performance prediction model and the target vehicle performance value meets the second convergence condition.

[0162] This method utilizes the idea of ​​transfer learning and uses the performance prediction model trained from step S21 to step S22 as the initial model. The simulation software only needs to simulate a small number of target vehicle performance values ​​corresponding to target vehicle parameters to train the performance prediction model. Based on the performance prediction model trained from step S21 to step S22, the network parameters of the performance prediction model are fine-tuned to obtain a new performance prediction model.

[0163] The network parameters include: weights and biases of the performance prediction model. Weights are parameters connecting the neurons of the performance prediction model, which are used to calculate the linear combination between input features and neurons; biases are constant terms on each neuron, which are used to introduce offsets in the calculation.

[0164] When fine-tuning the network parameters of the performance prediction model, a part of the network parameters of the performance prediction model can be fixed, and the other part of the network parameters can be trained. For example, the weights of a part of the network layers of the performance prediction model can be fixed, and the weights of another part of the network layers of the performance prediction model can be adjusted, so that the second error between the vehicle performance value output by the performance prediction model and the target vehicle performance value satisfies the second convergence condition. The performance prediction model when the second error satisfies the second convergence condition can be better applied to the changed vehicle design model.

[0165] The second error satisfies the second convergence condition including: the second error cannot continue to decrease or the second error is less than a preset value, etc.

[0166] Through the above method, when the vehicle design model changes, a smaller amount of data, that is, a smaller number of changed target vehicle parameters and target vehicle performance values ​​corresponding to the target vehicle parameters, can be used to train the previously trained performance prediction model, and the network parameters of the performance prediction model can be fine-tuned to quickly obtain a new performance prediction model to adapt to the changed vehicle design model, thereby quickly obtaining at least one set of target vehicle parameters that can meet the vehicle performance values ​​required by the user.

[0167] Figure 4 is a block diagram of a vehicle parameter determination device according to an exemplary embodiment. Figure 4 The vehicle parameter determination device 400 includes a determination module 410 and an optimization module 420 .

[0168] The determination module 410 is configured to obtain a multi-objective optimization function according to at least one performance prediction model and a constraint condition; the performance prediction model is used to obtain a vehicle performance value according to a vehicle parameter, and the constraint condition is used to constrain the vehicle parameter and / or the vehicle performance value;

[0169] The optimization module 420 is configured to obtain at least one set of vehicle parameters that meet the constraint conditions by minimizing or maximizing the output value of the multi-objective optimization function.

[0170] Optionally, the determination module 410 includes:

[0171] A first determination submodule is configured to obtain a performance prediction function according to the performance prediction model, wherein the performance prediction function is used to obtain the vehicle performance value according to the vehicle parameter;

[0172] The second determination submodule is configured to obtain the multi-objective optimization function according to the performance prediction function and the constraint conditions.

[0173] Optionally, at least one of the performance prediction models includes a stiffness prediction model and a modal prediction model;

[0174] Optionally, the determination module 410 includes:

[0175] The third determination submodule is configured to obtain the multi-objective optimization function according to the stiffness prediction model, the modal prediction model and the constraint conditions; wherein the stiffness prediction model is used to output the stiffness performance value according to the vehicle parameters, the modal prediction model is used to output the modal performance value according to the vehicle parameters, and the constraint conditions are used to constrain at least one of the vehicle parameters, the stiffness performance value and the modal performance value.

[0176] Optionally, the determination module 410 includes:

[0177] A modification submodule, configured to obtain a modified target constraint condition in response to a modification operation on the constraint condition;

[0178] The fourth determination submodule is configured to obtain the multi-objective optimization function according to at least one of the performance prediction models and the objective constraints.

[0179] Optionally, the vehicle parameter determination device 400 includes:

[0180] A first training module is configured to use vehicle parameter samples as training samples of an initial model, use vehicle performance value samples corresponding to the vehicle parameter samples as labels of the initial model, and train the initial model;

[0181] The first convergence module is configured to use the initial model when the first error between the vehicle performance value predicted by the initial model and the vehicle performance value sample satisfies a first convergence condition as the performance prediction model.

[0182] Optionally, the vehicle parameter determination device 400 includes:

[0183] A second training module is configured to, when a vehicle parameter changes, use the changed target vehicle parameter as a training sample of the performance prediction model, use the target vehicle performance value corresponding to the target vehicle parameter as a label of the performance prediction model, and train the performance prediction model;

[0184] The second convergence module is configured to adjust the network parameters of the performance prediction model until a second error between the vehicle performance value output by the performance prediction model and the target vehicle performance value satisfies a second convergence condition.

[0185] Optionally, the vehicle parameter determination device 400 includes:

[0186] The simulation module is configured to use the vehicle parameter sample as an input parameter of the simulation software to obtain a vehicle performance value sample corresponding to the vehicle parameter sample.

[0187] 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.

[0188] 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 parameter determination method provided by the present disclosure are implemented.

[0189] Figure 5is a block diagram of a vehicle parameter determination device 500 according to an exemplary embodiment. For example, the vehicle parameter determination device 500 may be an electronic device such as a mobile phone, a computer, a digital broadcast terminal, a message transceiver device, a tablet device, a medical device, a test device, a personal digital assistant, etc.

[0190] Reference Figure 5 , the device 500 may include one or more of the following components: a processing component 502 , a memory 504 , a power component 506 , a multimedia component 508 , an audio component 510 , an input / output interface 512 , a sensor component 514 , and a communication component 516 .

[0191] The processing component 502 generally controls the overall operation of the device 500, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 502 may include one or more processors 520 to execute instructions to complete all or part of the steps of the above-mentioned vehicle parameter determination method. In addition, the processing component 502 may include one or more modules to facilitate the interaction between the processing component 502 and other components. For example, the processing component 502 may include a multimedia module to facilitate the interaction between the multimedia component 508 and the processing component 502.

[0192] The memory 504 is configured to store various types of data to support operations on the device 500. Examples of such data include instructions for any application or method operating on the device 500, contact data, phone book data, messages, pictures, videos, etc. The memory 504 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.

[0193] The power supply component 506 provides power to the various components of the device 500. The power supply component 506 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 500.

[0194] The multimedia component 508 includes a screen that provides an output interface between the device 500 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 508 includes a front camera and / or a rear camera. When the device 500 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.

[0195] The audio component 510 is configured to output and / or input audio signals. For example, the audio component 510 includes a microphone (MIC), and when the device 500 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 504 or sent via the communication component 516. In some embodiments, the audio component 510 also includes a speaker for outputting audio signals.

[0196] The input / output interface 512 provides an interface between the processing component 502 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.

[0197] The sensor assembly 514 includes one or more sensors for providing various aspects of status assessment for the device 500. For example, the sensor assembly 514 can detect the open / closed state of the device 500, the relative positioning of components, such as the display and keypad of the device 500, the sensor assembly 514 can also detect the position change of the device 500 or a component of the device 500, the presence or absence of user contact with the device 500, the orientation or acceleration / deceleration of the device 500, and the temperature change of the device 500. The sensor assembly 514 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 514 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 514 may also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0198] The communication component 516 is configured to facilitate wired or wireless communication between the device 500 and other devices. The device 500 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 516 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 516 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.

[0199] In an exemplary embodiment, the device 500 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 execute the above-mentioned vehicle parameter determination method.

[0200] Figure 6 is a block diagram of a vehicle 600 according to an exemplary embodiment. The vehicle 600 may be configured with Figure 5 The electronic equipment shown in the vehicle 600 may also be used with Figure 5 The electronic device shown is connected to the vehicle 600, and the electronic device obtains various vehicle parameters on the vehicle, and obtains at least one set of vehicle parameters that meet the constraint conditions from the multiple vehicle parameters with the goal of maximizing or minimizing the output value of the multi-objective function. The electronic device can also return the obtained at least one set of vehicle parameters that meet the constraint conditions to the vehicle 600, so that the vehicle 600 performs a vehicle test according to the returned at least one set of vehicle parameters.

[0201] For example, the vehicle 600 may be a hybrid vehicle, a non-hybrid vehicle, an electric vehicle, a fuel cell vehicle, or other types of vehicles. The vehicle 600 may be an autonomous vehicle, a semi-autonomous vehicle, or a non-autonomous vehicle.

[0202] 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.

[0203] In some embodiments, the infotainment system 610 may include a communication system, an entertainment system, and a navigation system, among others.

[0204] 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.

[0205] The decision control system 630 may include a computing system, a vehicle controller, a steering system, a throttle, and a braking system.

[0206] 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.

[0207] 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.

[0208] 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.

[0209] 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.

[0210] 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 .

[0211] In the disclosed embodiment, the processor 651 may execute instruction 653 to complete all or part of the steps of the above-mentioned vehicle parameter determination method.

[0212] 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.

[0213] 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 method for determining vehicle parameters, characterized in that: include: A multi-objective optimization function is obtained according to at least one performance prediction model and constraint conditions; The performance prediction model is used to obtain a vehicle performance value according to vehicle parameters, and the constraint condition is used to constrain the vehicle parameters and / or the vehicle performance value; With the goal of minimizing or maximizing the output value of the multi-objective optimization function, at least one set of vehicle parameters satisfying the constraint conditions is obtained.

2. The method according to claim 1, characterized in that The multi-objective optimization function is obtained according to at least one performance prediction model and constraint conditions, including: According to the performance prediction model, a performance prediction function is obtained, wherein the performance prediction function is used to obtain the vehicle performance value according to the vehicle parameters; The multi-objective optimization function is obtained according to the performance prediction function and the constraint conditions.

3. The method according to claim 1, characterized in that At least one of the performance prediction models includes a stiffness prediction model and a modal prediction model; the multi-objective optimization function is obtained according to the at least one performance prediction model and the constraint conditions, including: The multi-objective optimization function is obtained according to the stiffness prediction model, the modal prediction model and the constraint conditions; wherein the stiffness prediction model is used to output a stiffness performance value according to the vehicle parameters, the modal prediction model is used to output a modal performance value according to the vehicle parameters, and the constraint conditions are used to constrain at least one of the vehicle parameters, the stiffness performance value and the modal performance value.

4. The method according to claim 1, characterized in that: The multi-objective optimization function is obtained according to at least one performance prediction model and constraint conditions, including: In response to the modification operation on the constraint condition, a modified target constraint condition is obtained; The multi-objective optimization function is obtained according to at least one of the performance prediction models and the objective constraints.

5. The method according to claim 1, characterized in that The method further comprises: Using vehicle parameter samples as training samples of the initial model, using vehicle performance value samples corresponding to the vehicle parameter samples as labels of the initial model, and training the initial model; When a first error between the vehicle performance value predicted by the initial model and the vehicle performance value sample satisfies a first convergence condition, the initial model when the first error satisfies the first convergence condition is used as the performance prediction model.

6. The method according to claim 5, characterized in that The method further comprises: In the case where the vehicle parameters are changed, the changed target vehicle parameters are used as training samples of the performance prediction model, and the target vehicle performance values ​​corresponding to the target vehicle parameters are used as labels of the performance prediction model to train the performance prediction model; The network parameters of the performance prediction model are adjusted until a second error between the vehicle performance value output by the performance prediction model and the target vehicle performance value satisfies a second convergence condition.

7. The method according to claim 5, characterized in that The method further comprises: The vehicle parameter samples are used as input parameters of simulation software to obtain vehicle performance value samples corresponding to the vehicle parameter samples.

8. A vehicle parameter determination device, characterized in that: include: A determination module configured to obtain a multi-objective optimization function according to at least one performance prediction model and constraint conditions; The performance prediction model is used to obtain a vehicle performance value according to vehicle parameters, and the constraint condition is used to constrain the vehicle parameters and / or the vehicle performance value; The optimization module is configured to obtain at least one set of vehicle parameters that meet the constraint conditions by minimizing or maximizing the output value of the multi-objective optimization function.

9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to: Execute the steps of any one of the methods of claims 1 to 7.

10. 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 7 are implemented.

11. A chip, characterized in that: The method comprises a processor and an interface; the processor is used to read instructions to execute the method according to any one of claims 1 to 7.

12. A vehicle, characterized in that: The electronic device according to claim 8 is arranged on the vehicle, or the vehicle is connected to the electronic device according to claim 8.

Images