Vehicle part manufacturing method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a vehicle part manufacturing method and device, electronic equipment and a storage medium. The method comprises the steps that machining parameters of a plurality of sample parts are obtained; the plurality of sample parts are made of composite materials according to different processing parameters; determining training set data and test set data according to the processing parameters of the plurality of sample parts; training a prediction model according to the training set data to train the prediction model to predict performance parameters of the sample parts based on the processing parameters of the sample parts; performing parameter optimization according to the test set data and the prediction model obtained by training to obtain a target processing parameter corresponding to the target performance parameter meeting a preset condition; and the vehicle parts are manufactured through the composite materials according to the target machining parameters. Therefore, the performance parameters of the manufactured vehicle parts are optimal, the performance of the vehicle is effectively improved, and the weight of the vehicle is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of parts manufacturing, and in particular to a method for manufacturing a vehicle part, a device for manufacturing a vehicle part, an electronic device, and a computer-readable storage medium. Background Art

[0002] With the development of automobiles, lightweighting has become an important means to improve fuel efficiency, reduce emissions, and enhance vehicle performance. Panoramic sunroofs are now a mainstream feature in automobiles, and their structural design and material selection are crucial for optimizing vehicle lightweighting and performance. However, traditional panoramic sunroof frames are typically made of plastic. While these panoramic sunroofs offer high strength and stability, they also have drawbacks such as being heavy and difficult to recycle, which hinders their ability to effectively improve vehicle performance and lightweighting. Summary of the Invention

[0003] In view of the above problems, embodiments of the present invention are proposed to provide a method for manufacturing a vehicle component, a device for manufacturing a vehicle component, an electronic device, and a computer-readable storage medium that overcome the above problems or at least partially solve the above problems.

[0004] In order to solve the above problems, an embodiment of the present invention discloses a method for manufacturing a vehicle component, the method comprising:

[0005] Obtaining processing parameters of a plurality of sample parts; the plurality of sample parts are made of composite materials according to different processing parameters;

[0006] Determining training set data and test set data based on the processing parameters of the plurality of sample parts;

[0007] Training a prediction model according to the training set data, so as to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts;

[0008] Optimizing parameters based on the test set data and the trained prediction model to obtain target processing parameters corresponding to target performance parameters that meet preset conditions;

[0009] The vehicle component is manufactured using the composite material according to the target processing parameters.

[0010] Optionally, the training of the prediction model according to the training set data to train the prediction model to predict the performance parameters of the sample component based on the processing parameters of the sample component includes:

[0011] Using the processing parameters as input neurons of a prediction model;

[0012] Using the performance parameters of the sample parts as output neurons of the prediction model;

[0013] Randomly initializing the neural network weights in the prediction model and performing forward propagation using the training set data to obtain performance data output by the prediction model;

[0014] The prediction model is iteratively trained according to the output performance data and the training set data, so as to train the prediction model to predict the performance parameters of the sample component based on the processing parameters of the sample component.

[0015] Optionally, the iterative training of the prediction model based on the output performance data and the training set data, so as to train the prediction model to predict the performance parameters of the sample component based on the processing parameters of the sample component, includes:

[0016] Determining the amount and data labels of the training set data, and determining a first loss value based on the amount of the training set data, the data labels, and the output performance data;

[0017] The weight echelon corresponding to the first loss value is determined by a back-propagation algorithm, and the neural network weight is updated according to the weight echelon. The prediction model is iteratively trained using the updated neural network weight and the training set data to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts.

[0018] Optionally, performing parameter optimization based on the test set data and the trained prediction model to obtain target processing parameters corresponding to target performance parameters that meet preset conditions includes:

[0019] Determining the characteristic values corresponding to the processing parameters in the test set data;

[0020] encoding the eigenvalues as planet elements in a Kepler optimization algorithm;

[0021] According to the planetary element, the prediction model and the test set data, the prediction model is optimized for parameters to obtain target processing parameters corresponding to target performance parameters that meet preset conditions.

[0022] Optionally, performing parameter optimization on the prediction model based on the planet element, the prediction model, and the test set data to obtain target processing parameters corresponding to target performance parameters that meet preset conditions includes:

[0023] Randomly initializing the position and velocity of the planetary element, and inputting the initialized parameters of the planetary element into the prediction model to obtain performance parameters output by the prediction model; the parameters of the planetary element correspond to the characteristic values;

[0024] Determining the amount and data labels of the test set data, and determining a second loss value based on the amount of the test set data, the data labels, and the performance parameter;

[0025] Determine the second loss value as the fitness of the planet element in the Kepler optimization algorithm;

[0026] Optimizing and updating the position and the velocity of the planetary element according to the fitness of the planetary element;

[0027] Outputting the optimized and updated position and velocity of the planetary element to the prediction model to obtain target performance parameters that meet preset conditions;

[0028] According to the target performance parameters that meet the preset conditions, corresponding target processing parameters are determined.

[0029] Optionally, the manufacturing of the vehicle component using the composite material according to the target processing parameters includes:

[0030] establishing a first three-dimensional model of the vehicle component according to the target processing parameters;

[0031] Establishing a first finite element model of the vehicle component based on the first three-dimensional model of the vehicle component, and performing morphology optimization and topology optimization on the first finite element model of the vehicle component to obtain an optimization result;

[0032] Establishing a second three-dimensional model of the vehicle component according to the optimization result, and establishing a second finite element model of the vehicle component according to the second three-dimensional model of the vehicle component;

[0033] When the second finite element model of the vehicle component passes verification, the second three-dimensional model of the vehicle component is determined as a target three-dimensional model, and the vehicle component is manufactured according to the composite material, the target three-dimensional model, and the target processing parameters.

[0034] Optionally, the processing parameters include: at least one of: ply angle, ply angle sequence, total ply thickness, number of plies, and total thickness of the composite material; and the performance parameters include: at least one of: mode, strength, and weight.

[0035] The embodiment of the present invention further discloses a device for manufacturing vehicle parts, the device comprising:

[0036] An acquisition module is used to acquire processing parameters of a plurality of sample parts; the plurality of sample parts are made of composite materials according to different processing parameters;

[0037] A determination module, configured to determine training set data and test set data based on the processing parameters of the plurality of sample parts;

[0038] A training module, configured to train a prediction model based on the training set data, so as to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts;

[0039] An optimization module, configured to perform parameter optimization based on the test set data and the trained prediction model to obtain target processing parameters corresponding to target performance parameters that meet preset conditions;

[0040] A manufacturing module is used to manufacture the vehicle component using the composite material according to the target processing parameters.

[0041] Optionally, the training module includes:

[0042] A first determination submodule is configured to use the processing parameters as input neurons of a prediction model;

[0043] A second determination submodule is configured to use the performance parameters of the sample components as output neurons of the prediction model;

[0044] An initialization submodule, configured to randomly initialize the neural network weights in the prediction model and perform forward propagation using the training set data to obtain performance data output by the prediction model;

[0045] A training submodule is used to iteratively train the prediction model based on the output performance data and the training set data, so as to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts.

[0046] Optionally, the training submodule includes:

[0047] a first determining unit, configured to determine the amount and data labels of the training set data, and determine a first loss value based on the amount of the training set data, the data labels, and the output performance data;

[0048] A training unit is used to determine the weight echelon corresponding to the first loss value through a back-propagation algorithm, and update the neural network weight according to the weight echelon, and iteratively train the prediction model using the updated neural network weight and the training set data to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts.

[0049] Optionally, the optimization module includes:

[0050] A third determining submodule is used to determine the characteristic values corresponding to the processing parameters in the test set data;

[0051] An encoding submodule, for encoding the eigenvalues into planet elements in a Kepler optimization algorithm;

[0052] The optimization submodule is used to optimize the parameters of the prediction model according to the planet element, the prediction model and the test set data, and obtain the target processing parameters corresponding to the target performance parameters that meet the preset conditions.

[0053] Optionally, the optimization submodule includes:

[0054] an output unit, configured to randomly initialize the position and velocity of the planetary element, and input the initialized parameters of the planetary element into the prediction model to obtain performance parameters output by the prediction model; the parameters of the planetary element correspond to the characteristic values;

[0055] a second determining unit, configured to determine the amount and data labels of the test set data, and determine a second loss value based on the amount of the test set data, the data labels, and the performance parameter;

[0056] a third determining unit, configured to determine the second loss value as the fitness of the planet element in the Kepler optimization algorithm;

[0057] an updating unit, configured to optimize and update the position and the velocity of the planet element according to the fitness of the planet element;

[0058] An input unit, configured to output the optimized updated position and velocity of the planetary element to the prediction model to obtain target performance parameters that meet preset conditions;

[0059] The fourth determining unit is configured to determine corresponding target processing parameters according to the target performance parameters that meet the preset conditions.

[0060] Optionally, the production module includes:

[0061] A first establishing submodule, configured to establish a first three-dimensional model of the vehicle component according to the target processing parameters;

[0062] a second establishing submodule, configured to establish a first finite element model of the vehicle component based on the first three-dimensional model of the vehicle component, and perform morphology optimization and topology optimization on the first finite element model of the vehicle component to obtain an optimization result;

[0063] A third establishing submodule is configured to establish a second three-dimensional model of the vehicle component according to the optimization result, and to establish a second finite element model of the vehicle component according to the second three-dimensional model of the vehicle component;

[0064] A production submodule is used to determine the second three-dimensional model of the vehicle component as a target three-dimensional model when the second finite element model of the vehicle component is verified, and to produce the vehicle component according to the composite material, the target three-dimensional model, and the target processing parameters.

[0065] The present invention also shows an electronic device, including a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned method for manufacturing vehicle parts when executing the computer program.

[0066] The present invention also shows a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method for manufacturing vehicle parts are implemented.

[0067] The embodiments of the present invention include the following advantages:

[0068] In an embodiment of the present invention, training set data and test set data can be determined from the processing parameters of multiple sample parts. A prediction model is then trained using the training set data to predict the performance parameters of the sample parts based on the sample parts' processing parameters. Parameter optimization is then performed based on the test set data and the trained prediction model to determine target processing parameters corresponding to target performance parameters that meet preset conditions. Finally, vehicle parts are manufactured using composite materials according to the target processing parameters. Consequently, the vehicle parts are manufactured using the target processing parameters determined through parameter optimization, further optimizing the performance parameters of the manufactured vehicle parts, effectively improving vehicle performance, and effectively reducing vehicle weight. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 This is a flowchart of a method for manufacturing a vehicle component provided by an embodiment of the present invention;

[0070] Figure 2 This is a structural block diagram of a prediction model provided by an embodiment of the present invention;

[0071] Figure 3 This is a structural block diagram of a vehicle parts manufacturing device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0072] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0073] In the related art, the material of the vehicle sunroof is generally based on plastic. Although such panoramic sunroofs have high strength and stability, they also have disadvantages such as being heavy and not easy to recycle, and cannot effectively improve the performance and lightweighting of the car. In order to solve the above technical problems, the present invention provides a method for manufacturing vehicle parts. The core concept is to determine training set data and test set data from the processing parameters of multiple sample parts, and then train a prediction model with the training set data to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts. Parameter optimization is then performed based on the test set data and the trained prediction model to determine the target processing parameters corresponding to the target performance parameters that meet the preset conditions. Finally, composite materials are used to manufacture vehicle parts according to the target processing parameters, so that the vehicle parts are manufactured with the target processing parameters determined by parameter optimization, further optimizing the performance parameters of the manufactured vehicle parts, effectively improving the performance of the vehicle, and effectively lightweighting the vehicle.

[0074] Reference Figure 1 , shows a flowchart of a method for manufacturing a vehicle component provided by an embodiment of the present invention, the method may specifically include the following steps:

[0075] Step 101 : obtaining processing parameters of a plurality of sample parts; the plurality of sample parts are made of composite materials according to different processing parameters.

[0076] In an embodiment of the present invention, when manufacturing a vehicle component, processing parameters of multiple sample components can be obtained, wherein the multiple sample components are manufactured using composite materials according to different processing parameters. The vehicle component in the present invention can be a vehicle sunroof.

[0077] Specifically, the present invention can utilize a composite material for manufacturing sample components by composite processing of a first material and a second material. The first material can be PLA (Polylactic Acid) material, and the second material can be a biofiber material, such as fast-growing bamboo fiber. When manufacturing the composite material for the sample component, the biofiber material can be selected to have a length of 10 cm to 20 cm and a diameter of 0.2 mm to 1.0 mm. When manufacturing the sample component, the second material can be first layered according to different processing parameters. After layering is completed, the first material can be hot-molded, i.e., the first material is injection molded and cured, to obtain sample components manufactured with different processing parameters.

[0078] In the present invention, the processing parameters include at least one of a layup angle, a layup angle sequence, a total layup thickness, a number of layup layers, and a total thickness of the composite material. The layup angle is the layup angle of the second material, the layup angle sequence is the layup angle sequence of the second material, the total layup thickness is the total thickness of the second material, the number of layup layers is the number of layup layers of the second material, and the total thickness of the composite material is the total thickness of the composite material after the first material and the second material are combined.

[0079] Step 102: Determine training set data and test set data based on the processing parameters of the plurality of sample parts.

[0080] In an embodiment of the present invention, after obtaining the processing parameters of a plurality of sample parts, training set data and test set data may be determined based on the processing parameters of the plurality of sample parts.

[0081] In the present invention, an optimal Latin square design sampling method can be used to sample processing parameters. Specifically, the ranges of values for the ply angle, ply angle sequence, total ply thickness, number of ply layers, and total thickness of the composite material can be determined based on actual requirements. For example, the ply angle can have four levels: 0°, 45°, 90°, and 135°; the ply sequence can have three levels: 0°-45°-90°-135°, 0°-90°-45°-135°, and 0°-135°-45°-90°; the total ply thickness can have four levels: 1 mm, 2 mm, 3 mm, and 4 mm; the number of ply layers can have four levels: 1 layer, 3 layers, 5 layers, and 7 layers; and the total thickness of the composite material can have three levels: 5 mm, 10 mm, and 15 mm. In the example of the present invention, in the case of mixed levels, each processing parameter has different numbers of levels, so a fractional factorial design or other optimization method is required.

[0082] The optimal Latin square matrix corresponding to the optimal Latin square refers to each row, each column, and the main and secondary diagonals containing all possible horizontal combinations. For the five processing parameters in the present invention, it is necessary to construct a 5*5 Latin square matrix. Since the number of levels of each processing parameter in the present invention is different, a mixed horizontal Latin square method can be adopted.

[0083] In the present invention, sampling is performed based on the processing parameters of multiple sample parts to obtain multiple sampling data sets, and then the multiple sampling data sets are divided into training set data and test set data according to actual needs or a certain proportion. For example, 60 groups of sampling data sets are sampled, 40 groups of sampling data sets can be determined as training set data, and 20 groups of sampling data sets can be determined as test set data.

[0084] Step 103 : training a prediction model according to the training set data, so as to train the prediction model to predict the performance parameters of the sample component based on the processing parameters of the sample component.

[0085] In the present invention, after determining the training set data and the prediction set data, the prediction model can be trained according to the training set data to obtain a trained prediction model, and the performance parameters corresponding to the sample parts and the processing parameters can be predicted based on the processing parameters of the sample parts and the trained prediction model.

[0086] In one embodiment, a prediction model is trained based on training set data to train the prediction model to predict performance parameters of sample parts based on the processing parameters of the sample parts, which may include: using the processing parameters as input neurons of the prediction model; using the performance parameters of the sample parts as output neurons of the prediction model; randomly initializing the weights of the neural network in the prediction model, and using the training set data for forward propagation to obtain performance data output by the prediction model; and iteratively training the prediction model based on the output performance data and the training set data to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts.

[0087] Specifically, the processing parameters in the present invention include: at least one of the ply angle, ply angle sequence, total ply thickness, number of ply layers and total thickness of the composite material; the performance parameters may include: at least one of the mode, strength and weight. The present invention can use the ply angle, ply angle sequence, total ply thickness, number of ply layers and total thickness of the composite material in the processing parameters as input neurons of the prediction model, and use the mode, strength and weight in the performance parameters of the sample parts as output neurons of the prediction model. That is, the number of neurons in the input layer of the prediction model in the present invention can be 5, the number of neurons in the output layer can be 3, and the number of hidden neurons can be set to 12. In the present invention, after many experiments, it was found that 12 hidden neurons can provide sufficient model complexity without causing the fitting and training time to be too long. The number of hidden neurons in the present invention can also be set according to actual needs, and the present invention does not make specific limitations here. For example Figure 2 , which shows a structural block diagram of a prediction model provided by an embodiment of the present invention.

[0088] After determining the input neurons and output neurons of the prediction model, the neural network weights of the prediction model can be randomly initialized, and the training set data can be used for forward propagation to obtain the performance data output by the prediction model. Then, based on the output performance data and the training set data, the prediction model is iteratively trained to obtain the trained prediction model. The performance parameters of the sample parts corresponding to the processing parameters can be predicted based on the processing parameters of the sample parts and the trained prediction model.

[0089] In one embodiment, the prediction model is iteratively trained based on the output performance data and training set data to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts, which may include: determining the number and data labels of the training set data, and determining a first loss value based on the number, data labels and output performance data of the training set data; determining a weight echelon corresponding to the first loss value through a back propagation algorithm, and updating the neural network weights according to the weight echelon, and iteratively training the prediction model using the updated neural network weights and training set data to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts.

[0090] Specifically, the number of training set data and data labels are first determined, and then a first loss value is determined based on the number of training set data, data labels, and performance data output by the prediction model. In the present invention, the first loss value can use the MSE root mean square error method to measure the mean square error between the performance parameters output by the prediction model and the actual performance parameters to improve the prediction accuracy of the model. The present invention can calculate the first loss value using the following formula (1):

[0091]

[0092] Among them, F1 (MSE) is the first loss value, n1 is the number of training set data, T1 i is the data label of the training set data, S1 i This is the performance parameter output by the prediction model. The weighted echelon corresponding to the first loss value is determined through the backpropagation algorithm. The neural network weights of the prediction model are updated based on the determined weighted echelon. The prediction model is iteratively trained with the goal of minimizing the first loss value.

[0093] In the present invention, the process of forward propagation, determining the first loss value, and backpropagation optimization can be repeated until a preset stopping condition is met, thereby obtaining a trained prediction model. Based on the processing parameters of the sample component and the trained prediction model, performance parameters corresponding to the processing parameters of the sample component can be predicted. The preset stopping condition in the present invention can be when the first loss value converges or when the number of iterations reaches a preset number.

[0094] Step 104 : Optimizing parameters based on the test set data and the trained prediction model to obtain target processing parameters corresponding to target performance parameters that meet preset conditions.

[0095] In an embodiment of the present invention, after the prediction model is trained based on the training set data to obtain a trained prediction model, parameter optimization can be performed based on the test set data and the trained prediction model to determine target processing parameters corresponding to target performance parameters that meet preset conditions. The prediction condition can be the optimal result obtained through the parameter optimization.

[0096] In one embodiment, performing parameter optimization based on the test set data and the trained prediction model to obtain target processing parameters corresponding to the target performance parameters that meet preset conditions may include: determining eigenvalues corresponding to the processing parameters in the test set data; encoding the eigenvalues as planetary elements in the Kepler optimization algorithm; and performing parameter optimization on the prediction model based on the planetary elements, the prediction model, and the test set data to obtain target processing parameters corresponding to the target performance parameters that meet the preset conditions.

[0097] In the present invention, the Kepler optimization algorithm can be used to optimize the parameters of a trained prediction model. Specifically, the five eigenvalues corresponding to the ply angle, ply angle sequence, total ply thickness, number of plies, and total thickness of the composite material in the test data set are first determined and encoded as planetary elements in the Kepler optimization algorithm. During the encoding of the eigenvalues, the eigenvalues of the five processing parameters can be normalized, the minimum and maximum values for each processing parameter can be calculated, and all eigenvalues can be mapped to the range [0, 1]. After obtaining the planetary elements, the prediction model can be optimized based on the planetary elements, the prediction model, and the test data set to obtain the target processing parameters corresponding to the target performance parameters that meet the preset conditions.

[0098] In one embodiment, based on the planetary element, the prediction model and the test set data, the parameters of the prediction model are optimized to obtain the target processing parameters corresponding to the target performance parameters that meet the preset conditions, which may include: randomly initializing the position and velocity of the planetary element, and inputting the parameters of the initialized planetary element into the prediction model to obtain the performance parameters output by the prediction model; the parameters of the planetary element correspond to the characteristic values; determining the number and data labels of the test set data, and determining the second loss value based on the number, data labels and performance parameters of the test set data; determining the second loss value as the fitness of the planetary element in the Kepler optimization algorithm; optimizing and updating the position and velocity of the planetary element based on the fitness of the planetary element; outputting the optimized and updated position and velocity of the planetary element to the prediction model to obtain the target performance parameters that meet the preset conditions; and determining the corresponding target processing parameters based on the target performance parameters that meet the preset conditions.

[0099] Specifically, the position and velocity of the planet element are randomly initialized first, and then the parameters of the planet element are input into the trained prediction model to obtain the performance data output by the prediction model. Then, the second loss value is calculated using the test set data. The second loss value can be calculated using the following formula (2):

[0100]

[0101] Among them, F2 (MSE) is the second loss value, n2 is the number of prediction set data, T2 i is the data label of the training set data, S2 i It is the performance parameter output by the prediction model.

[0102] The second loss value is then used as the fitness of the planetary element. According to the fitness of the planetary element, the position and velocity of the planetary element are optimized and updated. That is, the explorable domain is continuously searched through the Kepler optimization algorithm, the position and velocity of the planetary element are updated, the fitness and Kepler motion simulation process are repeatedly calculated, and the position and velocity of the planetary element are gradually optimized until the stopping condition is met. At this time, the Pareto solution set of the prediction model optimization can be obtained according to the output of the prediction model. The Pareto solution set contains the target processing parameters corresponding to the target performance parameters that meet the preset conditions, that is, the optimal laying angle, laying angle sequence, total laying thickness, number of laying layers and total thickness of the composite material are obtained.

[0103] It should be noted that the training set data in the present invention is used to train and optimize the prediction model, and the test set data is used to evaluate the generalization ability of the prediction model. The test set data does not participate in the training of the model to avoid overfitting.

[0104] Step 105 : Using the composite material to manufacture the vehicle component according to the target processing parameters.

[0105] In the embodiment of the present invention, after the target processing parameters are obtained, that is, after the optimal target processing parameters are obtained, the composite material can be used to manufacture vehicle parts according to the target processing parameters.

[0106] In one embodiment, manufacturing a vehicle part using a composite material according to target processing parameters may include: establishing a first three-dimensional model of the vehicle part according to the target processing parameters; establishing a first finite element model of the vehicle part according to the first three-dimensional model of the vehicle part, and performing morphology optimization and topology optimization on the first finite element model of the vehicle part to obtain an optimization result; establishing a second three-dimensional model of the vehicle part according to the optimization result, and establishing a second finite element model of the vehicle part according to the second three-dimensional model of the vehicle part; when the second finite element model of the vehicle part is verified, determining the second three-dimensional model of the vehicle part as the target three-dimensional model, and manufacturing the vehicle part according to the composite material, the target three-dimensional model, and the target processing parameters.

[0107] Specifically, a first three-dimensional model of the vehicle parts is first established according to the target processing parameters, and HyperWorks software is used to establish the optimal first finite element model of the vehicle parts through the first three-dimensional model of the vehicle parts. OptiStruct software is used to perform morphology optimization and topology optimization on the first finite element model of the vehicle parts to obtain optimization results, so as to achieve local weight reduction and performance improvement of the vehicle parts.

[0108] In the present invention, the constraints for morphology optimization can be: no weight increase, and a certain percentage increase in the first-order mode and strength of the vehicle component frame. The constraints for topology optimization can be: maintaining the same mode and strength, while achieving topological structure changes to reduce assembly weight.

[0109] After morphology optimization and topology optimization, the optimization results are obtained and based on the optimization results, the vehicle parts are re-modeled to obtain a second three-dimensional model of the vehicle parts, and the second three-dimensional model is imported into the HyperWorks software again to obtain a second finite element model of the vehicle parts. The second finite element model of the vehicle parts is then verified to see whether it meets the restriction conditions. When the second finite element model of the vehicle parts is verified, the second three-dimensional model of the vehicle parts is determined as the target three-dimensional model. Finally, the final vehicle parts can be manufactured according to the first material, the second material, the target three-dimensional model and the target processing parameters, so that the final vehicle parts are made of polylactic acid material, biofiber material, the verified target three-dimensional model and the target processing parameters obtained by parameter optimization, so that the manufactured vehicle parts have higher strength and stability, and at the same time, they are lighter in weight and easier to recycle than plastic-based vehicle parts, which effectively improves the performance of the vehicle and makes the vehicle more lightweight.

[0110] In an embodiment of the present invention, processing parameters for a plurality of sample parts are obtained; the plurality of sample parts are manufactured using composite materials according to different processing parameters; training set data and test set data are determined based on the processing parameters of the plurality of sample parts; a prediction model is trained based on the training set data, so that the trained prediction model predicts performance parameters of the sample parts based on the processing parameters of the sample parts; parameter optimization is performed based on the test set data and the trained prediction model to obtain target processing parameters corresponding to target performance parameters that meet preset conditions; and vehicle parts are manufactured using composite materials according to the target processing parameters. Thus, the vehicle parts are manufactured using the target processing parameters determined by the parameter optimization, further optimizing the performance parameters of the manufactured vehicle parts, effectively improving vehicle performance, and effectively reducing vehicle weight.

[0111] It should be noted that for the sake of simplicity, the method embodiments are described as a series of actions. However, those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0112] Reference Figure 3 , shows a structural block diagram of a vehicle parts manufacturing device provided by an embodiment of the present invention, which may specifically include the following modules:

[0113] An acquisition module 301 is used to acquire processing parameters of a plurality of sample parts; the plurality of sample parts are made of composite materials according to different processing parameters;

[0114] A determination module 302 is configured to determine training set data and test set data based on the processing parameters of the plurality of sample parts;

[0115] A training module 303 is configured to train a prediction model based on the training set data, so as to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts;

[0116] An optimization module 304 is configured to perform parameter optimization based on the test set data and the trained prediction model to obtain target processing parameters corresponding to target performance parameters that meet preset conditions;

[0117] The manufacturing module 305 is configured to manufacture the vehicle component using the composite material according to the target processing parameters.

[0118] In one embodiment, the training module 303 includes:

[0119] A first determination submodule is configured to use the processing parameters as input neurons of a prediction model;

[0120] A second determination submodule is configured to use the performance parameters of the sample components as output neurons of the prediction model;

[0121] An initialization submodule, configured to randomly initialize the neural network weights in the prediction model and perform forward propagation using the training set data to obtain performance data output by the prediction model;

[0122] A training submodule is used to iteratively train the prediction model based on the output performance data and the training set data, so as to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts.

[0123] In one embodiment, the training submodule includes:

[0124] a first determining unit, configured to determine the amount and data labels of the training set data, and determine a first loss value based on the amount of the training set data, the data labels, and the output performance data;

[0125] A training unit is used to determine the weight echelon corresponding to the first loss value through a back-propagation algorithm, and update the neural network weight according to the weight echelon, and iteratively train the prediction model using the updated neural network weight and the training set data to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts.

[0126] In one embodiment, the optimization module 304 includes:

[0127] A third determining submodule is used to determine the characteristic values corresponding to the processing parameters in the test set data;

[0128] An encoding submodule, for encoding the eigenvalues into planet elements in a Kepler optimization algorithm;

[0129] The optimization submodule is used to optimize the parameters of the prediction model according to the planet element, the prediction model and the test set data, and obtain the target processing parameters corresponding to the target performance parameters that meet the preset conditions.

[0130] In one embodiment, the optimal search submodule includes:

[0131] an output unit, configured to randomly initialize the position and velocity of the planetary element, and input the initialized parameters of the planetary element into the prediction model to obtain performance parameters output by the prediction model; the parameters of the planetary element correspond to the characteristic values;

[0132] a second determining unit, configured to determine the amount and data labels of the test set data, and determine a second loss value based on the amount of the test set data, the data labels, and the performance parameter;

[0133] a third determining unit, configured to determine the second loss value as the fitness of the planet element in the Kepler optimization algorithm;

[0134] an updating unit, configured to optimize and update the position and the velocity of the planet element according to the fitness of the planet element;

[0135] An input unit, configured to output the optimized updated position and velocity of the planetary element to the prediction model to obtain target performance parameters that meet preset conditions;

[0136] The fourth determining unit is configured to determine corresponding target processing parameters according to the target performance parameters that meet the preset conditions.

[0137] In one embodiment, the production module 305 includes:

[0138] A first establishing submodule, configured to establish a first three-dimensional model of the vehicle component according to the target processing parameters;

[0139] a second establishing submodule, configured to establish a first finite element model of the vehicle component based on the first three-dimensional model of the vehicle component, and perform morphology optimization and topology optimization on the first finite element model of the vehicle component to obtain an optimization result;

[0140] A third establishing submodule is configured to establish a second three-dimensional model of the vehicle component according to the optimization result, and to establish a second finite element model of the vehicle component according to the second three-dimensional model of the vehicle component;

[0141] A production submodule is used to determine the second three-dimensional model of the vehicle component as a target three-dimensional model when the second finite element model of the vehicle component is verified, and to produce the vehicle component according to the composite material, the target three-dimensional model, and the target processing parameters.

[0142] In an embodiment of the present invention, an acquisition module is used to obtain processing parameters of a plurality of sample parts; the plurality of sample parts are manufactured using composite materials according to different processing parameters; a determination module is used to determine training set data and test set data based on the processing parameters of the plurality of sample parts; a training module is used to train a prediction model based on the training set data, so that the trained prediction model predicts the performance parameters of the sample parts based on the processing parameters of the sample parts; an optimization module is used to perform parameter optimization based on the test set data and the trained prediction model to obtain target processing parameters corresponding to target performance parameters that meet preset conditions; and a manufacturing module is used to manufacture vehicle parts using composite materials according to the target processing parameters. Thus, the vehicle parts are manufactured using the target processing parameters determined by parameter optimization, further optimizing the performance parameters of the manufactured vehicle parts, effectively improving the performance of the vehicle, and effectively reducing the weight of the vehicle.

[0143] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0144] An embodiment of the present invention further provides an electronic device, including:

[0145] It includes a processor, a memory, and a computer program stored in the memory and capable of running on the processor. When the computer program is executed by the processor, the various processes of the above-mentioned vehicle parts manufacturing method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0146] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the various processes of the above-mentioned vehicle parts manufacturing method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0147] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0148] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, apparatus, or computer program products. Thus, embodiments of the present invention may take the form of a fully hardware embodiment, a fully software embodiment, or an embodiment combining software and hardware. Furthermore, embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0149] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the process in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0150] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0151] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0152] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they become aware of the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0153] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or terminal device that includes the element.

[0154] The above is a detailed introduction to a method for manufacturing a vehicle component, a device for manufacturing a vehicle component, an electronic device, and a computer-readable storage medium provided by the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the ideas of the present invention, there may be changes in the specific implementation methods and application scopes. In summary, the contents of this specification should not be understood as limiting the present invention.

Claims

1. A method for manufacturing a vehicle component, characterized in that: The method comprises: Obtaining processing parameters of a plurality of sample parts; the plurality of sample parts are made of composite materials according to different processing parameters; Determining training set data and test set data based on the processing parameters of the plurality of sample parts; Training a prediction model according to the training set data, so as to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts; Optimizing parameters based on the test set data and the trained prediction model to obtain target processing parameters corresponding to target performance parameters that meet preset conditions; The vehicle component is manufactured using the composite material according to the target processing parameters.

2. The method for manufacturing a vehicle component according to claim 1, wherein: The step of training the prediction model according to the training set data to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts includes: Using the processing parameters as input neurons of a prediction model; Using the performance parameters of the sample parts as output neurons of the prediction model; Randomly initializing the neural network weights in the prediction model and performing forward propagation using the training set data to obtain performance data output by the prediction model; The prediction model is iteratively trained according to the output performance data and the training set data, so as to train the prediction model to predict the performance parameters of the sample component based on the processing parameters of the sample component.

3. The method for manufacturing a vehicle component according to claim 2, wherein: The iterative training of the prediction model based on the output performance data and the training set data to train the prediction model to predict the performance parameters of the sample component based on the processing parameters of the sample component includes: Determining the amount and data labels of the training set data, and determining a first loss value based on the amount of the training set data, the data labels, and the output performance data; The weight echelon corresponding to the first loss value is determined by a back-propagation algorithm, and the neural network weight is updated according to the weight echelon. The prediction model is iteratively trained using the updated neural network weight and the training set data to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts.

4. The method for manufacturing a vehicle component according to claim 1, wherein: The parameter optimization is performed based on the test set data and the prediction model obtained through training to obtain target processing parameters corresponding to target performance parameters that meet preset conditions, including: Determining the characteristic values corresponding to the processing parameters in the test set data; encoding the eigenvalues as planet elements in a Kepler optimization algorithm; According to the planetary element, the prediction model and the test set data, the prediction model is optimized for parameters to obtain target processing parameters corresponding to target performance parameters that meet preset conditions.

5. The method for manufacturing a vehicle component according to claim 4, wherein: The step of optimizing parameters of the prediction model based on the planet element, the prediction model, and the test set data to obtain target processing parameters corresponding to target performance parameters that meet preset conditions includes: Randomly initializing the position and velocity of the planetary element, and inputting the initialized parameters of the planetary element into the prediction model to obtain performance parameters output by the prediction model; the parameters of the planetary element correspond to the characteristic values; Determining the amount and data labels of the test set data, and determining a second loss value based on the amount of the test set data, the data labels, and the performance parameter; Determine the second loss value as the fitness of the planet element in the Kepler optimization algorithm; Optimizing and updating the position and the velocity of the planetary element according to the fitness of the planetary element; Outputting the optimized and updated position and velocity of the planetary element to the prediction model to obtain target performance parameters that meet preset conditions; According to the target performance parameters that meet the preset conditions, corresponding target processing parameters are determined.

6. The method for manufacturing a vehicle component according to claim 1, wherein: The method of using the composite material to manufacture the vehicle component according to the target processing parameters includes: establishing a first three-dimensional model of the vehicle component according to the target processing parameters; Establishing a first finite element model of the vehicle component based on the first three-dimensional model of the vehicle component, and performing morphology optimization and topology optimization on the first finite element model of the vehicle component to obtain an optimization result; Establishing a second three-dimensional model of the vehicle component according to the optimization result, and establishing a second finite element model of the vehicle component according to the second three-dimensional model of the vehicle component; When the second finite element model of the vehicle component passes verification, the second three-dimensional model of the vehicle component is determined as a target three-dimensional model, and the vehicle component is manufactured according to the composite material, the target three-dimensional model, and the target processing parameters.

7. The method for manufacturing a vehicle component according to claim 1, wherein: The processing parameters include: at least one of: ply angle, ply angle sequence, total ply thickness, number of plies, and total thickness of the composite material; the performance parameters include: at least one of: mode, strength, and weight.

8. A vehicle parts manufacturing device, characterized in that: The device comprises: An acquisition module is used to acquire processing parameters of a plurality of sample parts; the plurality of sample parts are made of composite materials according to different processing parameters; A determination module, configured to determine training set data and test set data based on the processing parameters of the plurality of sample parts; A training module, configured to train a prediction model based on the training set data, so as to train the prediction model to predict the performance parameters of the sample parts based on the processing parameters of the sample parts; An optimization module, configured to perform parameter optimization based on the test set data and the trained prediction model to obtain target processing parameters corresponding to target performance parameters that meet preset conditions; A manufacturing module is used to manufacture the vehicle component using the composite material according to the target processing parameters.

9. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein when the computer program is executed by the processor, the steps of the method for manufacturing a vehicle component according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for manufacturing a vehicle component according to any one of claims 1 to 7 are implemented.

Images