Automobile sheet material selection method and device, electronic equipment and storage medium

Abstract

The application discloses a method and device for selecting automobile plate, an electronic device and a storage medium, wherein the method comprises: obtaining target type information and a target performance range of a target automobile plate to be selected; obtaining a plurality of candidate automobile plates conforming to the target type information, and obtaining mechanical index parameters corresponding to each candidate automobile plate; for each candidate automobile plate, predicting a predicted performance range of the candidate automobile plate by using a trained performance prediction model based on the target type information and the mechanical index parameters; and selecting the target automobile plate from the plurality of candidate automobile plates based on the target performance range and the predicted performance range corresponding to each candidate automobile plate. The accuracy of selecting the automobile plate can be improved.

Description

Technical Field

[0001] This application belongs to the field of artificial intelligence technology, and in particular relates to a method, apparatus, electronic device and storage medium for selecting automotive sheet metal. Background Technology

[0002] In the steel industry, when supplying sheet metal to automobile manufacturers, automotive sheet metal needs to possess sufficient strength to meet the safety requirements of the vehicle body structure. For example, high-strength automotive sheet metal can effectively resist deformation during a collision, protecting the passengers inside the vehicle. However, at the same time, automotive sheet metal also requires good formability during forming processes such as stamping. If the strength is too high, the elongation of the material may decrease, making it prone to cracking when stamping complex-shaped parts (such as car hoods, doors, etc.).

[0003] Currently, different automotive sheet materials are usually selected according to pre-defined rules to meet different needs, resulting in a low accuracy rate in sheet material selection. Summary of the Invention

[0004] The embodiments of this application provide a method, apparatus, electronic device, and computer storage medium for selecting automotive sheet metal, which can at least improve the accuracy of automotive sheet metal selection to a certain extent.

[0005] Other features and advantages of this application will become apparent from the following detailed description, or may be learned in part from practice of this application.

[0006] According to a first aspect of the embodiments of this application, a method for selecting automotive sheet metal is provided, comprising:

[0007] Obtain the target type information and target performance range of the target automotive sheet metal to be selected;

[0008] Obtain multiple candidate automotive sheet materials that meet the target type information, and obtain the mechanical index parameters corresponding to each candidate automotive sheet material;

[0009] For each candidate automotive sheet metal, based on target type information and mechanical index parameters, the predicted performance range of the candidate automotive sheet metal is predicted through the trained performance prediction model.

[0010] Based on the target performance range and the predicted performance range of each candidate automotive sheet, the target automotive sheet is selected from multiple candidate automotive sheet materials.

[0011] In some possible implementations, the performance prediction model is obtained in the following way:

[0012] Obtain the sample dataset; the sample dataset includes multiple sample automotive sheet metals, each sample automotive sheet metal includes corresponding sample type information, sample mechanical index parameters and sample reference performance range;

[0013] The initial prediction model is optimized at least once based on the sample dataset until a preset termination condition is met. The prediction model is then obtained based on the initial prediction model that meets the preset termination condition.

[0014] In some possible implementations, the sample dataset includes a training dataset, and the initial prediction model is optimized at least once based on the sample dataset, including:

[0015] The initial prediction model is trained at least once based on the training dataset. The preset termination condition includes that at least one training operation satisfies the training termination condition.

[0016] Training operations include:

[0017] Input the sample type information and mechanical index parameters of each sample of automobile sheet metal in the training dataset into the initial prediction model to obtain sample prediction information, which includes the first sample prediction performance range.

[0018] For each sample of automotive sheet metal, the first training loss of the sample automotive sheet metal is determined based on the difference between the predicted performance range of the first sample and the reference performance range of the sample.

[0019] The total training loss is determined based on the first training loss of each sample of automotive sheet metal.

[0020] The parameters of the initial prediction model are adjusted based on the total training loss, and the adjusted initial prediction model is used as the initial prediction model for the next training operation.

[0021] In some possible implementations, each sample automotive sheet metal also includes sample forming limit data, the sample prediction information also includes predicted forming limit data, and the training operation further includes:

[0022] For each sample of automotive sheet metal, a second training loss for the sample automotive sheet metal is determined based on the difference between the sample forming limit data and the predicted forming limit data.

[0023] Based on the first training loss of each sample of automotive sheet metal, the total training loss is determined, including:

[0024] The total training loss is determined based on the first training loss and the second training loss of each sample of car body panels.

[0025] In some possible implementations, the sample dataset also includes a validation dataset, and performing at least one optimization operation on the initial prediction model based on the sample dataset further includes:

[0026] Based on the validation dataset, perform at least one validation operation on the initial prediction model that meets the training termination condition. The preset termination condition also includes at least one validation operation that meets the validation termination condition.

[0027] The verification process includes:

[0028] The sample type information and mechanical index parameters of each sample of automobile sheet metal in the validation dataset are input into the first prediction model to obtain the second sample prediction performance range. The first prediction model is the initial prediction model that meets the training termination condition.

[0029] The validation results are determined based on the difference between the second sample predicted performance range and the sample reference performance range of each sample automotive sheet metal in the validation dataset.

[0030] The hyperparameters of the initial prediction model are adjusted based on the validation results, and the adjusted initial prediction model is used as the initial prediction model for the next optimization operation.

[0031] In some possible implementations, the initial prediction model includes an input layer, a hidden layer, and an output layer, and the hyperparameters of the initial prediction model include the number of nodes in the hidden layer;

[0032] The hyperparameters of the initial prediction model were adjusted based on the validation results, including:

[0033] Based on the verification results, the number of nodes in the hidden layer is adjusted within a preset range, where the preset range is determined based on the number of nodes in the input layer and the number of nodes in the output layer.

[0034] In some possible implementations, the number of nodes in the input layer is determined based on the number of parameters corresponding to the mechanical index parameters. The median value of the preset range is positively correlated with the number of nodes in the input layer, and the median value of the preset range is positively correlated with the number of nodes in the output layer.

[0035] According to a second aspect of the embodiments of this application, an apparatus for selecting automotive sheet metal is provided, comprising:

[0036] The information acquisition module is used to acquire information on the target type and target performance range of the target automotive sheet metal to be selected;

[0037] The sheet metal acquisition module is used to acquire multiple candidate automotive sheet metals that meet the target type information, and to acquire the mechanical index parameters corresponding to each candidate automotive sheet metal.

[0038] The prediction module is used to predict the predicted performance range of each candidate automotive sheet material based on target type information and mechanical index parameters, using a trained performance prediction model.

[0039] The selection module is used to select the target automotive sheet material from multiple candidate automotive sheet materials based on the target performance range and the predicted performance range corresponding to each candidate automotive sheet material.

[0040] In some possible implementations, the apparatus further includes a training module for:

[0041] Obtain the sample dataset; the sample dataset includes multiple sample automotive sheet metals, each sample automotive sheet metal includes corresponding sample type information, sample mechanical index parameters and sample reference performance range;

[0042] The initial prediction model is optimized at least once based on the sample dataset until a preset termination condition is met. The prediction model is then obtained based on the initial prediction model that meets the preset termination condition.

[0043] In some possible implementations, the sample dataset includes a training dataset, and the training module, when performing at least one optimization operation on the initial prediction model based on the sample dataset, is specifically used for:

[0044] The initial prediction model is trained at least once based on the training dataset. The preset termination condition includes that at least one training operation satisfies the training termination condition.

[0045] Training operations include:

[0046] Input the sample type information and mechanical index parameters of each sample of automobile sheet metal in the training dataset into the initial prediction model to obtain sample prediction information, which includes the first sample prediction performance range.

[0047] For each sample of automotive sheet metal, the first training loss of the sample automotive sheet metal is determined based on the difference between the predicted performance range of the first sample and the reference performance range of the sample.

[0048] The total training loss is determined based on the first training loss of each sample of automotive sheet metal.

[0049] The parameters of the initial prediction model are adjusted based on the total training loss, and the adjusted initial prediction model is used as the initial prediction model for the next training operation.

[0050] In some possible implementations, each sample automotive sheet metal also includes sample forming limit data, and the sample prediction information also includes predicted forming limit data. The training module, when performing training operations, is also used for:

[0051] For each sample of automotive sheet metal, a second training loss for the sample automotive sheet metal is determined based on the difference between the sample forming limit data and the predicted forming limit data.

[0052] When determining the total training loss based on the first training loss of each sample car body panel, the training module is specifically used for:

[0053] The total training loss is determined based on the first training loss and the second training loss of each sample of car body panels.

[0054] In some possible implementations, the sample dataset also includes a validation dataset, and the training module is further used for:

[0055] Based on the validation dataset, perform at least one validation operation on the initial prediction model that meets the training termination condition. The preset termination condition also includes at least one validation operation that meets the validation termination condition.

[0056] When performing the validation operation, the training module is specifically used for:

[0057] The sample type information and mechanical index parameters of each sample of automobile sheet metal in the validation dataset are input into the first prediction model to obtain the second sample prediction performance range. The first prediction model is the initial prediction model that meets the training termination condition.

[0058] The validation results are determined based on the difference between the second sample predicted performance range and the sample reference performance range of each sample automotive sheet metal in the validation dataset.

[0059] The hyperparameters of the initial prediction model are adjusted based on the validation results, and the adjusted initial prediction model is used as the initial prediction model for the next optimization operation.

[0060] In some possible implementations, the initial prediction model includes an input layer, a hidden layer, and an output layer, and the hyperparameters of the initial prediction model include the number of nodes in the hidden layer;

[0061] When adjusting the hyperparameters of the initial prediction model based on the validation results, the training module is specifically used for:

[0062] Based on the verification results, the number of nodes in the hidden layer is adjusted within a preset range, where the preset range is determined based on the number of nodes in the input layer and the number of nodes in the output layer.

[0063] In some possible implementations, the number of nodes in the input layer is determined based on the number of parameters corresponding to the mechanical index parameters. The median value of the preset range is positively correlated with the number of nodes in the input layer, and the median value of the preset range is positively correlated with the number of nodes in the output layer.

[0064] According to a third aspect of the present application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the methods described in the above embodiments.

[0065] According to a fourth aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the steps of the method described in the above embodiments.

[0066] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application.

[0067] The beneficial effects of the technical solutions provided in this application are:

[0068] If the target type information of the target automotive sheet material required by the user is obtained, multiple candidate automotive sheet materials that meet the target type information can be obtained first. Then, through the trained performance prediction model, the target type information and mechanical index parameters of each candidate automotive sheet material can be combined to predict the predicted performance range of each mechanical performance index of each candidate automotive sheet material. Then, based on the target performance range required by the user and the predicted performance range of each candidate automotive sheet material, the target automotive sheet material can be selected from multiple candidate automotive sheet materials. This can more accurately select the target automotive sheet material that meets both the target type information and the target performance range required by the user. Attached Figure Description

[0069] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings:

[0070] Figure 1 A flowchart illustrating a method for selecting automotive sheet metal according to an embodiment of this application;

[0071] Figure 2 A schematic diagram illustrating a data standardization scheme as an example provided in this application;

[0072] Figure 3 A schematic diagram illustrating a data standardization scheme as an example provided in this application;

[0073] Figure 4 A schematic diagram of a device for selecting automotive sheet metal according to an embodiment of this application;

[0074] Figure 5 This is a schematic diagram of the structure of an electronic device for selecting automotive sheet metal, provided as an embodiment of this application. Detailed Implementation

[0075] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0076] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this application. However, those skilled in the art will recognize that the technical solutions of this application can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this application.

[0077] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0078] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily need to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0079] It should also be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such uses of these terms can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described.

[0080] The technical solutions of this application and their effects are described below through several exemplary embodiments. It should be noted that the following embodiments can be referenced, borrowed from, or combined with each other. Identical terms, similar features, and similar implementation steps in different embodiments will not be repeated.

[0081] The method for selecting automotive sheet metal in this application can be performed by any electronic device, which may include a server or a terminal.

[0082] Those skilled in the art will understand that a server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server or server cluster that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. Terminals can be smartphones (such as Android phones, iOS phones, etc.), tablets, laptops, digital broadcast receivers, MIDs (Mobile Internet Devices), PDAs (Personal Digital Assistants), desktop computers, smart home appliances, in-vehicle terminals (such as in-vehicle navigation terminals, in-vehicle computers, etc.), smart speakers, smartwatches, etc. Terminals and servers can be directly or indirectly connected via wired or wireless communication, but are not limited to these methods. Embodiments of this invention can be applied to various scenarios, including but not limited to cloud technology, artificial intelligence, smart transportation, and assisted driving. Specific applications can be determined based on actual application scenario requirements and are not limited here. A terminal (also known as a user terminal or user device) can be a smartphone, tablet, laptop, desktop computer, smart voice interaction device (such as a smart speaker), wearable electronic device (such as a smartwatch), in-vehicle terminal, smart home appliance (such as a smart TV), AR / VR device, aircraft, etc., but is not limited to these.

[0083] like Figure 1 As shown, in some possible implementations, embodiments of this application provide a method for selecting automotive sheet metal. Taking a server as the executing entity, the method may include the following steps:

[0084] Step S101: Obtain the target type information and target performance range of the target automotive sheet metal to be selected.

[0085] Among them, type information can be used to characterize the brand, model, parts, grade, thickness, etc. of automotive sheet metal; performance range can include the applicable range corresponding to at least one mechanical performance index of automotive sheet metal.

[0086] Specifically, the target type information is the type information required by the automotive sheet material to be selected, and the target performance range can include the performance range required by the automotive sheet material to be selected.

[0087] Among them, mechanical performance indicators may include yield strength, tensile strength, elongation, R value (work hardening index), N value (plastic strain ratio), etc.

[0088] Specifically, yield strength is the stress at which a material begins to undergo significant plastic deformation; the yield strength of automotive sheet metal cannot be too high, otherwise a large force will be required in the initial stage of stamping, which can easily lead to increased wear of the mold.

[0089] Tensile strength reflects the maximum stress a material can withstand during stretching. Appropriate yield strength and tensile strength range are crucial for ensuring the smooth progress of the stamping process and the final performance of the stamped parts.

[0090] Elongation is the ratio of the amount of material elongated at tensile fracture to its original length. It directly reflects the material's ability to undergo plastic deformation. In the stamping process of automotive sheet metal, especially for parts with complex shapes that require large deformation (such as the curved parts of a car body), materials with high elongation can better adapt to the shape changes of the stamping die and avoid cracking during the stamping process.

[0091] The N value represents the work hardening characteristics of a material during plastic deformation. A higher N value means that the strength of the material increases relatively slowly as the degree of deformation increases during stamping, which is conducive to uniform deformation of the material and improves the shape accuracy of the stamped parts.

[0092] The R-value reflects the difference in the plastic strain capacity of a material in the plane of the sheet and in the direction perpendicular to the plane of the sheet. A larger R-value indicates that the material is more likely to deform in the plane of the sheet during the stamping process, which helps to prevent the stamped parts from thinning in the thickness direction and improve the quality of the stamped parts.

[0093] Step S102: Obtain multiple candidate automotive sheet materials that meet the target type information, and obtain the mechanical index parameters corresponding to each candidate automotive sheet material.

[0094] Specifically, from a pool of candidate automotive sheet materials corresponding to various types of information, multiple candidate automotive sheet materials that meet the target type information can be selected first. Each candidate automotive sheet material has corresponding mechanical index parameters.

[0095] Step S103: For each candidate automotive sheet metal, based on the target type information and mechanical index parameters, predict the predicted performance range of the candidate automotive sheet metal using the trained performance prediction model.

[0096] Specifically, a performance prediction model can be pre-trained, which can predict the performance range of automotive sheet metal based on the type information and mechanical parameters of the sheet metal.

[0097] In other words, for the performance prediction model, the input data includes the type information and mechanical parameters of the automotive sheet metal, and the output data includes the predicted performance range corresponding to at least one mechanical parameter of the automotive sheet metal.

[0098] For example, the input data for a performance prediction model may include: brand A, vehicle model B, front door inner panel (part), thickness 0.8, yield strength 153.53, tensile strength 39.45, elongation 46.35, N value 2.65, and R value 0.21.

[0099] The output data of the performance prediction model may include: yield strength ranging from 162.4 to 169.7, tensile strength ranging from 35.72 to 313.7, elongation ranging from 46.18 to 50.7, N value ranging from 2.59 to 2.78, and R value ranging from 0.24 to 0.25.

[0100] The specific training process for the performance prediction model will be described in more detail below.

[0101] Step S104: Select the target automotive sheet material from multiple candidate automotive sheet materials based on the target performance range and the predicted performance range corresponding to each candidate automotive sheet material.

[0102] Specifically, the predicted performance range of each candidate automotive sheet can be matched with the target performance range, and the candidate automotive sheet corresponding to the predicted performance range with the highest degree of matching with the target performance range can be used as the target automotive sheet.

[0103] The above-described method for selecting automotive sheet metal allows for the following steps: When the target type information of the desired automotive sheet metal is obtained, multiple candidate automotive sheet metals matching the target type information can be acquired first. Then, through a trained performance prediction model, the target type information and mechanical parameters of each candidate automotive sheet metal can be combined to predict the predicted performance range of each mechanical performance index of each candidate automotive sheet metal. Finally, based on the target performance range required by the user and the predicted performance range of each candidate automotive sheet metal, the target automotive sheet metal can be selected from the multiple candidate automotive sheet metals. This method can more accurately select target automotive sheet metals that meet both the target type information and the target performance range required by the user.

[0104] The specific training process of the performance prediction model will be further explained below with reference to the embodiments.

[0105] In some possible implementations, the performance prediction model is obtained in the following way:

[0106] (1) Obtain the sample dataset.

[0107] The sample dataset includes multiple sample automotive sheet metals, each of which includes corresponding sample type information, sample mechanical index parameters, and sample reference performance range.

[0108] In the specific implementation process, the sample reference performance range of the sample automotive sheet can be obtained by testing the sample automotive sheet.

[0109] Specifically, obtaining a sample dataset can include:

[0110] Obtain the initial sample dataset;

[0111] The initial sample dataset is standardized to obtain the sample dataset.

[0112] Specifically, in the initial sample dataset collected, the initial mechanical parameters of the automotive sheet metal samples may vary greatly. For example, the tensile strength value is around 200 to 350, while the R value is around 0.1 to 0.3.

[0113] To improve the training efficiency and stability of the initial prediction model, the data can be standardized.

[0114] Specifically, the standardized formula can be found in the following formula:

[0115]

[0116] in ... The mean, The standard deviation is denoted as .

[0117] like Figure 2 As shown, the initial yield strength of the sample automotive sheet metal ranges from 140 to 180, and the yield strength after standardization ranges from -0.15 to 0.1.

[0118] (2) Perform at least one optimization operation on the initial prediction model based on the sample dataset until the preset termination condition is met, and obtain the prediction model based on the initial prediction model that meets the preset termination condition.

[0119] Specifically, optimization operations can include training operations, which are used to obtain a prediction model after training; optimization operations can also include validation operations, which are used to perform validation operations after at least one training operation.

[0120] In some possible implementations, the sample dataset includes a training dataset, and the initial prediction model is optimized at least once based on the sample dataset, including:

[0121] The initial prediction model is trained at least once based on the training dataset. The preset termination condition includes that at least one training operation satisfies the training termination condition.

[0122] Training operations include:

[0123] ① Input the sample type information and mechanical index parameters of each sample of automobile sheet metal in the training dataset into the initial prediction model to obtain sample prediction information.

[0124] ② For each sample of automotive sheet metal, the first training loss of the sample automotive sheet metal is determined based on the difference between the predicted performance range of the first sample and the reference performance range of the sample.

[0125] ③ Determine the total training loss based on the first training loss of each sample of automotive sheet metal.

[0126] ④ Adjust the parameters of the initial prediction model based on the total training loss, and use the adjusted initial prediction model as the initial prediction model for the next training operation.

[0127] The sample prediction information includes the first sample prediction performance range.

[0128] The initial prediction model can include a BP (Back-Propagation Network) model, which is a multi-layer feedforward network trained using the error backpropagation algorithm.

[0129] Specifically, the initial prediction model includes an input layer, a hidden layer, and an output layer.

[0130] In practice, the backpropagation algorithm can be used to train the neural network. That is, based on the error of the output layer, the error is backpropagated to the hidden layer, and the weights and biases of each layer are adjusted to reduce the error.

[0131] The total training loss can be calculated using the following formula:

[0132] (2)

[0133] in, This is the actual value. Here, n represents the predicted value, and n is the number of nodes in the output layer. This represents the total training loss.

[0134] The parameters of the initial prediction model may include the weights and bias gradients of each layer of the initial prediction model.

[0135] In practice, the parameters of the initial prediction model can be updated using the following formula:

[0136] (3)

[0137] in, The weights are from layer i to layer j. Let E be the learning rate and E be the total training loss.

[0138] The choice of learning rate is crucial during training. If the learning rate is too large, the model may diverge during training and fail to converge; if the learning rate is too small, the training speed will be very slow. An initial small value of 0.01 is chosen, and then adjusted based on the model's performance on the validation set. A learning rate of 0.03 is generally considered ideal.

[0139] Specifically, the training termination condition can be that the number of training iterations reaches a preset number, the total training loss converges, or the total training loss is less than or equal to a first threshold.

[0140] In some possible implementations, each sample automotive sheet metal also includes sample forming limit data, the sample prediction information also includes predicted forming limit data, and the training operation further includes:

[0141] For each sample automotive sheet metal, a second training loss for the sample automotive sheet metal is determined based on the difference between the sample forming limit data and the predicted forming limit data.

[0142] In the specific implementation process, the process of determining the difference between the sample forming limit data and the predicted forming limit data can be referred to the above formula (2), which will not be elaborated here.

[0143] When obtaining sample forming limit data, one can also first obtain initial sample forming limit data, and then standardize the initial sample forming limit data, such as... Figure 3 As shown, the initial sample forming limit data of the sample automotive sheet metal ranges from -0.1 to 0.1, and the sample forming limit data after standardization ranges from -2 to 2.

[0144] Based on the first training loss of each sample of automotive sheet metal, the total training loss is determined, including:

[0145] The total training loss is determined based on the first training loss and the second training loss of each sample of car body panels.

[0146] Specifically, in addition to determining the total training loss based on the difference between the predicted performance range of the first sample and the reference performance range of the sample, the total training loss can also be calculated by combining the difference between the sample forming limit data and the predicted forming limit data.

[0147] Specifically, the sum of the first training loss and the second training loss can be used as the total training loss.

[0148] In the above embodiments, a first training loss for the sample automotive sheet is determined based on the difference between the predicted performance range of the first sample and the reference performance range of the sample. A second training loss for the sample automotive sheet is determined based on the difference between the forming limit data of the sample and the predicted forming limit data. Then, based on the first training loss of each sample automotive sheet and the second training loss of each sample automotive sheet, the total training loss is determined, so that the trained performance prediction model can take into account both the prediction ability of the performance range and the forming limit data.

[0149] In the above embodiments, the initial prediction model is trained using a training dataset until the training termination condition is met, thus obtaining a trained performance prediction model.

[0150] In other implementations, the initial prediction model can be trained using a training dataset until the training termination condition is met, and then the initial prediction model that meets the training termination condition can be validated using a validation dataset to obtain a performance prediction model.

[0151] In some possible implementations, the sample dataset also includes a validation dataset, and performing at least one optimization operation on the initial prediction model based on the sample dataset further includes:

[0152] Based on the validation dataset, perform at least one validation operation on the initial prediction model that meets the training termination condition. The preset termination condition also includes at least one validation operation that meets the validation termination condition.

[0153] In other words, the preset termination conditions include: at least one training operation meets the training termination condition, and the initial prediction model that meets the training termination condition passes the verification of the verification dataset, that is, at least one verification operation meets the verification termination condition.

[0154] The verification termination condition can be either the verification result being less than or equal to the second threshold, or the verification result converging.

[0155] The verification process includes:

[0156] (1) Input the sample type information and mechanical index parameters of each sample of automobile sheet metal in the validation dataset into the first prediction model to obtain the second sample prediction performance range.

[0157] The first prediction model is the initial prediction model that meets the training termination condition.

[0158] In other words, once the initial prediction model has undergone at least one training operation and meets the training termination condition, the validation operation begins.

[0159] (2) Based on the difference between the second sample predicted performance range and the sample reference performance range of each sample automotive sheet in the validation dataset, the validation results are determined.

[0160] Specifically, the process of determining the difference between the second sample predicted performance range and the sample reference performance range of each sample automotive sheet can also refer to the above formula (2).

[0161] (3) Adjust the hyperparameters of the initial prediction model based on the verification results, and use the adjusted initial prediction model as the initial prediction model for the next optimization operation.

[0162] The initial prediction model consists of an input layer, a hidden layer, and an output layer. The hyperparameters of the initial prediction model include the number of nodes in the hidden layer.

[0163] It is important to note that the hyperparameters of the initial prediction model are adjusted based on the validation results. In other words, after each adjustment of the hyperparameters of the initial prediction model, the steps of training the initial prediction model at least once and validating the initial prediction model that meets the training termination conditions are repeated.

[0164] Specifically, the hyperparameters of the initial prediction model are adjusted based on the validation results, including:

[0165] Based on the verification results, adjust the number of nodes in the hidden layer within a preset range.

[0166] The preset range is determined based on the number of nodes in the input layer and the number of nodes in the output layer.

[0167] Specifically, the number of nodes in the input layer is determined based on the number of parameters corresponding to the mechanical index parameters. The median value of the preset range is positively correlated with the number of nodes in the input layer, and the median value of the preset range is positively correlated with the number of nodes in the output layer.

[0168] Specifically, you can refer to the following formula to determine the midpoint of the preset range:

[0169] (4)

[0170] Where n is the number of hidden layer nodes, m is the number of input layer nodes, l is the number of output layer nodes, and α is a constant between 1 and 10.

[0171] In the specific implementation process, the number of nodes in the hidden layer of the initial prediction model can be determined based on formula (4) and a preset range can be determined. After training and verification operations, the number of nodes in the hidden layer can be adjusted within the preset range, and training and verification operations can be performed again.

[0172] The following will illustrate the above-mentioned method for selecting automotive sheet metal with specific examples.

[0173] In one example, taking the target type information including brand A, vehicle model B, component being the front door inner panel, grade C, and thickness 0.8 as an example of material performance indicators, this application can be described in detail as follows:

[0174] Sample data acquisition involves obtaining the mechanical property indicators of multiple sample automotive sheet metals from the database, including yield strength, tensile strength, elongation, N value, and R value.

[0175] The acquired data is standardized.

[0176] The standardized data was randomly divided into training and validation datasets in an 8:2 ratio.

[0177] The initial prediction model is trained based on the training dataset; the initial prediction model includes an input layer, a hidden layer and an output layer. The input layer of the neural network structure used in this patent includes 5 neurons, each corresponding to the yield strength, tensile strength, elongation, N value and R value respectively; the output layer contains 1 neuron, namely the forming limit.

[0178] Import the training set data into the initial prediction model and train the initial prediction model. When the training termination condition is met, use the validation dataset to validate the initial prediction model that meets the training termination condition.

[0179] In the specific training process, the hidden layer is initially set to have 9 neurons, and the learning rate is set to 0.2. Layers are connected by weights w and biases b, while neurons within the same layer are not connected. The error backpropagation algorithm is used to calculate the update amounts of w and b for each sample; the weights are updated using the following formula:

[0180] (5)

[0181] (6)

[0182] Wherein, element i is connected to element j, where For learning rate, For the error of unit j, For the output of unit i, For the new weights, This represents a new deviation.

[0183] If the verification result meets the verification termination condition, then the optimal model, i.e. the final performance prediction model, can be obtained. If the verification result does not meet the verification termination condition, then the steps of adjusting the hyperparameters of the initial prediction model (adjusting the number of hidden layer nodes), training the initial prediction model, and verifying the initial prediction model that meets the training termination condition are repeated until the training termination condition and the verification termination condition are met, and the performance prediction model is obtained.

[0184] The following describes an apparatus embodiment of this application, which can be used to perform the methods described in the above embodiments of this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the methods described in the above embodiments of this application.

[0185] In some possible implementations of this application, such as Figure 4 As shown, a selection device 40 for automotive sheet metal is provided, comprising:

[0186] Information acquisition module 401 is used to acquire the target type information and target performance range of the target automotive sheet metal to be selected;

[0187] The sheet metal acquisition module 402 is used to acquire multiple candidate automotive sheet metals that meet the target type information, and to acquire the mechanical index parameters corresponding to each candidate automotive sheet metal.

[0188] The prediction module 403 is used to predict the predicted performance range of each candidate automotive sheet material based on target type information and mechanical index parameters, using a trained performance prediction model.

[0189] The selection module 404 is used to select the target automotive sheet material from multiple candidate automotive sheet materials based on the target performance range and the predicted performance range corresponding to each candidate automotive sheet material.

[0190] In some possible implementations, the apparatus further includes a training module for:

[0191] Obtain the sample dataset; the sample dataset includes multiple sample automotive sheet metals, each sample automotive sheet metal includes corresponding sample type information, sample mechanical index parameters and sample reference performance range;

[0192] The initial prediction model is optimized at least once based on the sample dataset until a preset termination condition is met. The prediction model is then obtained based on the initial prediction model that meets the preset termination condition.

[0193] In some possible implementations, the sample dataset includes a training dataset, and the training module, when performing at least one optimization operation on the initial prediction model based on the sample dataset, is specifically used for:

[0194] The initial prediction model is trained at least once based on the training dataset. The preset termination condition includes that at least one training operation satisfies the training termination condition.

[0195] Training operations include:

[0196] Input the sample type information and mechanical index parameters of each sample of automobile sheet metal in the training dataset into the initial prediction model to obtain sample prediction information, which includes the first sample prediction performance range.

[0197] For each sample of automotive sheet metal, the first training loss of the sample automotive sheet metal is determined based on the difference between the predicted performance range of the first sample and the reference performance range of the sample.

[0198] The total training loss is determined based on the first training loss of each sample of automotive sheet metal.

[0199] The parameters of the initial prediction model are adjusted based on the total training loss, and the adjusted initial prediction model is used as the initial prediction model for the next training operation.

[0200] In some possible implementations, each sample automotive sheet metal also includes sample forming limit data, and the sample prediction information also includes predicted forming limit data. The training module, when performing training operations, is also used for:

[0201] For each sample of automotive sheet metal, a second training loss for the sample automotive sheet metal is determined based on the difference between the sample forming limit data and the predicted forming limit data.

[0202] When determining the total training loss based on the first training loss of each sample car body panel, the training module is specifically used for:

[0203] The total training loss is determined based on the first training loss and the second training loss of each sample of car body panels.

[0204] In some possible implementations, the sample dataset also includes a validation dataset, and the training module is further used for:

[0205] Based on the validation dataset, perform at least one validation operation on the initial prediction model that meets the training termination condition. The preset termination condition also includes at least one validation operation that meets the validation termination condition.

[0206] When performing the validation operation, the training module is specifically used for:

[0207] The sample type information and mechanical index parameters of each sample of automobile sheet metal in the validation dataset are input into the first prediction model to obtain the second sample prediction performance range. The first prediction model is the initial prediction model that meets the training termination condition.

[0208] The validation results are determined based on the difference between the second sample predicted performance range and the sample reference performance range of each sample automotive sheet metal in the validation dataset.

[0209] The hyperparameters of the initial prediction model are adjusted based on the validation results, and the adjusted initial prediction model is used as the initial prediction model for the next optimization operation.

[0210] In some possible implementations, the initial prediction model includes an input layer, a hidden layer, and an output layer, and the hyperparameters of the initial prediction model include the number of nodes in the hidden layer;

[0211] When adjusting the hyperparameters of the initial prediction model based on the validation results, the training module is specifically used for:

[0212] Based on the verification results, the number of nodes in the hidden layer is adjusted within a preset range, where the preset range is determined based on the number of nodes in the input layer and the number of nodes in the output layer.

[0213] In some possible implementations, the number of nodes in the input layer is determined based on the number of parameters corresponding to the mechanical index parameters. The median value of the preset range is positively correlated with the number of nodes in the input layer, and the median value of the preset range is positively correlated with the number of nodes in the output layer.

[0214] The aforementioned automotive sheet material selection device, if it obtains the target type information of the target automotive sheet material required by the user, can first obtain multiple candidate automotive sheet materials that meet the target type information. Then, through the trained performance prediction model, it can combine the target type information and mechanical index parameters of each candidate automotive sheet material to predict the predicted performance range corresponding to each mechanical performance index of each candidate automotive sheet material. Then, based on the target performance range required by the user and the predicted performance range of each candidate automotive sheet material, it can select the target automotive sheet material from multiple candidate automotive sheet materials. This allows for more accurate selection of the target automotive sheet material that meets both the target type information and the target performance range required by the user.

[0215] In one alternative embodiment, an electronic device is provided, such as Figure 5 As shown, Figure 5 The illustrated electronic device 4000 includes a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, for example, via a bus 4002. Optionally, the electronic device 4000 may further include a transceiver 4004, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 4004 is not limited to one type, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of this application.

[0216] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 4001 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0217] Bus 4002 may include a pathway for transmitting information between the aforementioned components. Bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 4002 can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the figure, but this does not indicate that there is only one bus or one type of bus.

[0218] The memory 4003 may be ROM (Read Only Memory) or other types of static storage devices capable of storing static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices capable of storing information and instructions, or EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media, other magnetic storage devices, or any other medium capable of carrying or storing computer programs and capable of being read by a computer, without limitation herein.

[0219] The memory 4003 stores computer programs that execute embodiments of this application, and its execution is controlled by the processor 4001. The processor 4001 executes the computer programs stored in the memory 4003 to implement the steps shown in the foregoing method embodiments.

[0220] This application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it can implement the steps and corresponding content of the aforementioned method embodiments.

[0221] This application also provides a computer program product, including a computer program that, when executed by a processor, can implement the steps and corresponding content of the aforementioned method embodiments.

[0222] It should be understood that although arrows indicate various operation steps in the flowcharts of this application's embodiments, the order in which these steps are implemented is not limited to the order indicated by the arrows. Unless explicitly stated herein, in some implementation scenarios of this application's embodiments, the implementation steps in each flowchart can be executed in other orders as required. Furthermore, some or all steps in each flowchart, based on the actual implementation scenario, may include multiple sub-steps or multiple stages. Some or all of these sub-steps or stages can be executed at the same time, and each sub-step or stage can also be executed at different times. In scenarios where execution times differ, the execution order of these sub-steps or stages can be flexibly configured according to requirements, and this application's embodiments do not limit this.

[0223] The above are only optional implementation methods for some implementation scenarios of this application. It should be noted that for those skilled in the art, other similar implementation methods based on the technical concept of this application, without departing from the technical concept of this application, also fall within the protection scope of the embodiments of this application.

Claims

1. A method of selecting an automotive panel, characterized by, include: Obtain the target type information and target performance range of the target automotive sheet metal to be selected; Obtain multiple candidate automotive sheet materials that meet the target type information, and obtain the mechanical index parameters corresponding to each candidate automotive sheet material; For each candidate automotive sheet metal, based on the target type information and the mechanical index parameters, the predicted performance range of the candidate automotive sheet metal is predicted by the trained performance prediction model. Based on the target performance range and the predicted performance range corresponding to each candidate automotive sheet, a target automotive sheet is selected from multiple candidate automotive sheet materials. The performance prediction model is trained in the following manner: Obtain a sample dataset; the sample dataset includes multiple sample automotive sheet metals, each sample automotive sheet metal including corresponding sample type information, sample mechanical index parameters and sample reference performance range; The initial prediction model is optimized at least once based on the sample dataset until a preset termination condition is met, and the prediction model is obtained based on the initial prediction model that meets the preset termination condition. The sample dataset further includes a training dataset, and the step of performing at least one optimization operation on the initial prediction model based on the sample dataset includes: The initial prediction model is trained at least once based on the training dataset, wherein the preset termination condition includes the training at least once satisfying the training termination condition. The training operations include: The sample type information and mechanical index parameters of each sample of automobile sheet metal in the training dataset are input into the initial prediction model to obtain sample prediction information, wherein the sample prediction information includes the first sample prediction performance range. For each sample of automotive sheet metal, a first training loss for the sample automotive sheet metal is determined based on the difference between the predicted performance range of the first sample and the reference performance range of the sample. The total training loss is determined based on the first training loss of each sample of automotive sheet metal. The parameters of the initial prediction model are adjusted based on the total training loss, and the adjusted initial prediction model is used as the initial prediction model for the next training operation.

2. The method of claim 1, wherein, Each sample automotive sheet metal also includes sample forming limit data, the sample prediction information also includes predicted forming limit data, and the training operation also includes: For each sample automotive sheet metal, a second training loss for the sample automotive sheet metal is determined based on the difference between the sample forming limit data and the predicted forming limit data. The first training loss based on each sample of automotive sheet metal is used to determine the total training loss, including: The total training loss is determined based on the first training loss and the second training loss of each sample of car body panels.

3. The method of claim 1, wherein, The sample dataset also includes a validation dataset, and the step of performing at least one optimization operation on the initial prediction model based on the sample dataset further includes: Based on the validation dataset, at least one validation operation is performed on the initial prediction model that meets the training termination condition, and the preset termination condition further includes that the at least one validation operation meets the validation termination condition. The verification operation includes: The sample type information and mechanical index parameters of each sample of automobile sheet metal in the validation dataset are input into the first prediction model to obtain the second sample prediction performance range. The first prediction model is the initial prediction model that meets the training termination condition. The validation results are determined based on the difference between the second sample predicted performance range and the sample reference performance range of each sample automotive sheet metal in the validation dataset. Based on the verification results, the hyperparameters of the initial prediction model are adjusted, and the adjusted initial prediction model is used as the initial prediction model for the next optimization operation.

4. The method of claim 3, wherein, The initial prediction model includes an input layer, a hidden layer, and an output layer, and the hyperparameters of the initial prediction model include the number of nodes in the hidden layer; The adjustment of the hyperparameters of the initial prediction model based on the verification results includes: Based on the verification results, the number of nodes in the hidden layer is adjusted within a preset range, wherein the preset range is determined based on the number of nodes in the input layer and the number of nodes in the output layer.

5. The method of claim 4, wherein, The number of nodes in the input layer is determined based on the number of parameters corresponding to the mechanical index parameters. The median value of the preset range is positively correlated with the number of nodes in the input layer, and the median value of the preset range is positively correlated with the number of nodes in the output layer.

6. A device for selecting automotive sheet metal, characterized in that, include: The information acquisition module is used to acquire information on the target type and target performance range of the target automotive sheet metal to be selected; The sheet metal acquisition module is used to acquire multiple candidate automotive sheet metals that meet the target type information, and to acquire the mechanical index parameters corresponding to each candidate automotive sheet metal. The training module is used to acquire a sample dataset; the sample dataset includes multiple sample automotive sheet metals, each sample automotive sheet metal including corresponding sample type information, sample mechanical index parameters and sample reference performance range; the initial prediction model is optimized at least once based on the sample dataset until a preset termination condition is met, and a prediction model is acquired based on the initial prediction model that meets the preset termination condition. The prediction module is used to predict the predicted performance range of each candidate automotive sheet material based on the target type information and the mechanical index parameters, using a trained performance prediction model. The selection module is used to select a target automotive material from multiple candidate automotive materials based on the target performance range and the predicted performance range corresponding to each candidate automotive material. The sample dataset further includes a training dataset, and the training module, when performing at least one optimization operation on the initial prediction model based on the sample dataset, is used to: The initial prediction model is trained at least once based on the training dataset, wherein the preset termination condition includes the training at least once satisfying the training termination condition. The training operations include: The sample type information and mechanical index parameters of each sample of automotive sheet metal in the training dataset are input into the initial prediction model to obtain sample prediction information, wherein the sample prediction information includes the first sample prediction performance range. For each sample of automotive sheet metal, a first training loss for the sample automotive sheet metal is determined based on the difference between the predicted performance range of the first sample and the reference performance range of the sample. The total training loss is determined based on the first training loss of each sample of automotive sheet metal. The parameters of the initial prediction model are adjusted based on the total training loss, and the adjusted initial prediction model is used as the initial prediction model for the next training operation.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1-5.

8. A computer readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-5.

Images