Tire wear life prediction method and device

Abstract

The invention discloses a tire wear life prediction method and device. The method comprises the following steps: acquiring first driving working condition data and first tire parameters of a target tire under a target driving working condition; a pre-trained tire wear prediction model is called to analyze the first driving condition data and the first tire parameters, a first wear data sequence of the target tire in a future preset time period is obtained, the tire wear prediction model is a long short-term memory (LSTM) network model, and a second wear data sequence of the target tire in the future preset time period is obtained; the tire wear prediction model is obtained by training driving condition data, tire parameters and a time-based wear data sequence of different tires under different driving conditions; and determining the wear life of the target tire under the target driving condition according to the first wear data sequence. The technical problem that it is difficult to efficiently and accurately predict the wear life of the tire in related schemes is solved.

Description

Technical Field

[0001] The present application relates to the field of tire detection technology, and in particular to a tire wear life prediction method and device. Background Art

[0002] Tires often experience irregular wear during use, such as bilateral shoulder wear caused by insufficient inflation pressure, or center wear of the tread caused by excessive inflation pressure. The main causes of wear are improper driving behavior (such as sudden braking and sharp turns), overloading, and tire design and material problems. In addition, tire wear is also related to factors such as drive braking and continuous slipping, which can increase shear stress, cause fatigue softening and even fracture of the rubber material, and affect the driving performance of the vehicle. Therefore, studying the wear state of tires is of great significance for improving tire performance and enhancing driving experience.

[0003] Traditional tire wear assessment relies on manual visual inspection and periodic measurement, which is inefficient and inaccurate. At present, tire wear life assessment mostly relies on real vehicle testing, estimating the tire service life based on the experience of drivers and tire design engineers, which is not only time-consuming and labor-intensive, but also the results are greatly affected by subjective factors. Although finite element simulation analysis can predict tire wear, the complexity of the operation and the low correlation with actual road tests have led to insufficient utilization of tire development. Although some studies have begun to try to use machine learning technology to predict tire life, it only focuses on tire tread wear and requires a large amount of image sample data, and the learning process is long, so it still needs further improvement.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present application provide a tire wear life prediction method and device to at least solve the technical problem that related solutions are difficult to efficiently and accurately predict the tire wear life.

[0006] According to one aspect of an embodiment of the present application, a tire wear life prediction method is provided, comprising: obtaining first driving condition data and first tire parameters of a target tire under a target driving condition; calling a pre-trained tire wear prediction model to analyze the first driving condition data and the first tire parameter to obtain a first wear data sequence of the target tire within a preset time period in the future, wherein the tire wear prediction model is a long short-term memory (LSTM) network model, and the tire wear prediction model is trained using driving condition data, tire parameters and time-based wear data sequences of different tires under different driving conditions; and determining the wear life of the target tire under the target driving condition based on the first wear data sequence.

[0007] Optionally, the characteristic dimensions of the driving condition data include at least one of the following: road condition, ambient temperature, vehicle speed, tire pressure, tire load, tire camber angle, and tire toe angle; the characteristic dimensions of tire parameters include at least one of the following: tire size and tire material; the wear data sequence includes: data on the depth of each node on the tire changing over time.

[0008] Optionally, the training process of the tire wear prediction model includes: constructing an initial prediction model, wherein the initial prediction model includes in sequence: an input layer, an LSTM layer, a linear rectified unit (ReLU) activation function layer, a random dropout layer, a fully connected layer, and a linear activation function layer; obtaining multiple groups of sample data of multiple tires under multiple driving conditions from historical road test data and historical finite element simulation data, wherein each group of sample data includes: second driving condition data of a tire under one driving condition, second tire parameters, and a time-based second wear data sequence, the second driving condition data and the second tire parameters are used as input samples, and the second wear data sequence is used as a sample label; dividing the multiple groups of sample data into a training set, a validation set, and a test set; using the training set to iteratively train the initial prediction model, using the validation set to verify the model after each training cycle, using the test set to test the model after training, and using the model after the test as the tire wear prediction model.

[0009] Optionally, before dividing the multiple groups of sample data into training sets, validation sets and test sets, the method also includes: performing data cleaning on the multiple groups of sample data, wherein data cleaning includes: duplicate value processing, missing value processing, and outlier processing; and normalizing the feature data of the same dimension in the multiple groups of sample data after data cleaning.

[0010] Optionally, the initial prediction model is iteratively trained using the training set, the model is verified using the validation set after each training cycle, and the model is tested using the test set after the training, including: determining a preset number of training cycles, a preset training batch size, and a preset learning rate; during the iterative training process, the training set is divided according to the preset training batch size, and in each training batch, the input samples corresponding to the training batch are input into the model, a mean square error loss function is constructed based on the model output and the sample labels corresponding to the training batch, and the model parameters are adjusted based on the mean square error loss function; after each training cycle is completed, the input samples corresponding to the validation set are input into the model, the validation loss is determined based on the model output and the sample labels corresponding to the validation set, and if the decrease in the validation loss for a preset number of consecutive adjacent training cycles is less than a preset threshold, the preset learning rate is attenuated based on a preset step size; after the number of training cycles reaches the preset number of training cycles, the input samples corresponding to the test set are input into the model, and the model performance evaluation indicators are determined based on the model output and the sample labels corresponding to the test set, wherein the model performance evaluation indicators include: mean absolute error, correlation coefficient, and root mean square error.

[0011] Optionally, after the tire wear prediction model is trained, the tire wear prediction model is stored as a file in a target format for calling, wherein the target format includes: HDF5 format, SavedModel format, ONNX format, PMML format.

[0012] Optionally, the target driving condition includes multiple preset driving conditions. After determining the wear life of the target tire under the target driving condition based on the first wear data sequence, the method also includes: sorting the multiple preset driving conditions according to the size of the corresponding wear life; determining the preset driving condition with the largest corresponding wear life as the optimal driving condition for the target tire.

[0013] According to another aspect of the embodiment of the present application, there is also provided a tire wear life prediction device, comprising: an acquisition module, for acquiring first driving condition data and first tire parameters of a target tire under a target driving condition;

[0014] A prediction module is used to call a pre-trained tire wear prediction model to analyze the first driving condition data and the first tire parameter to obtain a first wear data sequence of the target tire in a preset time period in the future, wherein the tire wear prediction model is an LSTM network model, and the tire wear prediction model is trained using driving condition data of different tires under different driving conditions, tire parameters and time-based wear data sequences; a determination module is used to determine the wear life of the target tire under the target driving condition based on the first wear data sequence.

[0015] According to another aspect of an embodiment of the present application, a computer program product is further provided, the computer program product comprising: a computer program, wherein when the computer program is executed by a processor, the above-mentioned tire wear life prediction method is implemented.

[0016] According to another aspect of an embodiment of the present application, an electronic device is also provided, which includes: a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to execute the above-mentioned tire wear life prediction method through the computer program.

[0017] In an embodiment of the present application, the first driving condition data and the first tire parameter of the target tire under the target driving condition are obtained; the pre-trained tire wear prediction model is called to analyze the first driving condition data and the first tire parameter to obtain the first wear data sequence of the target tire in the future preset time period, wherein the tire wear prediction model is a long short-term memory LSTM network model, and the tire wear prediction model is trained using the driving condition data, tire parameters and time-based wear data sequence of different tires under different driving conditions; the wear life of the target tire under the target driving condition is determined according to the first wear data sequence. In this way, the first driving condition data and tire parameters of the target tire are deeply analyzed using the tire wear prediction model obtained by training the driving condition data, tire parameters and time-based wear data sequence of different tires under different driving conditions, so as to accurately predict the wear data sequence of the target tire in the future time period and the wear life under the target driving condition, thereby solving the technical problem that the related schemes are difficult to efficiently and accurately predict the wear life of the tire. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0019] Figure 1 is a flow chart of an optional tire wear life prediction method according to an embodiment of the present application;

[0020] Figure 2 is a schematic structural diagram of an optional tire wear life prediction device according to an embodiment of the present application;

[0021] Figure 3 It is a schematic diagram of the structure of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0023] It should be noted that the terms "first", "second", etc. in the specification, claims and drawings of the present application are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0024] Example 1

[0025] According to an embodiment of the present application, a tire wear life prediction method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0026] Figure 1 is a flow chart of a tire wear life prediction method provided according to an embodiment of the present application, such as Figure 1 As shown, the method comprises the following steps:

[0027] Step S102, obtaining first driving condition data and first tire parameters of a target tire under a target driving condition;

[0028] Step S104, calling a pre-trained tire wear prediction model to analyze the first driving condition data and the first tire parameter to obtain a first wear data sequence of the target tire in a future preset time period, wherein the tire wear prediction model is a long short-term memory (LSTM) network model, and the tire wear prediction model is trained using driving condition data of different tires under different driving conditions, tire parameters, and a time-based wear data sequence;

[0029] Step S106, determining the wear life of the target tire under the target driving condition according to the first wear data sequence.

[0030] The following describes the various steps of the tire wear life prediction method in conjunction with a specific implementation process.

[0031] In order to improve the efficiency and accuracy of tire wear assessment and more accurately predict the wear life of tires, this application proposes a tire wear prediction model based on a long short-term memory (LSTM) network. The model can effectively predict the service life of tires by analyzing driving condition data and tire parameters.

[0032] Optionally, before analyzing the first driving condition data and the first tire parameter using the tire wear prediction model, the driving condition data of different tires under different driving conditions, tire parameters and a time-based wear data sequence are used to train the model. The characteristic dimensions of the driving condition data include at least one of the following: road surface conditions, ambient temperature, vehicle driving speed, tire pressure, tire load, tire camber angle, and tire toe angle; the characteristic dimensions of the tire parameters include at least one of the following: tire size and tire material; and the wear data sequence includes: data on the depth of each node on the tire changing over time.

[0033] Specifically, the training process of the tire wear prediction model can be carried out in the following manner:

[0034] First, an initial prediction model is constructed, which includes: input layer, LSTM layer, linear rectified unit ReLU activation function layer, random dropout layer, fully connected layer and linear activation function layer.

[0035] Among them, the input layer is responsible for receiving input data. The dimension of the input data is set according to the actual situation, usually including tire driving condition data and tire parameters; the LSTM layer is used to process time-based wear data sequences and capture long-term dependencies in the data. The appropriate number of hidden units (for example, 128 hidden units) can be selected to ensure that the model has sufficient capacity to learn complex patterns; the ReLU (rectified linear unit) activation function layer is used to increase the nonlinear characteristics of the model, and the activation function helps the model better fit the nonlinear relationship; the Dropout layer prevents overfitting by randomly closing a certain proportion of neurons (for example, the proportion is set to 0.5) to reduce the model's excessive dependence on specific input features; the fully connected layer usually contains multiple neurons for extracting higher-level features; the linear activation function layer is used for the final output, and usually no activation function is used so that the model can output a series of continuous values, such as tire wear data sequences.

[0036] As an optional implementation, in order to ensure that the training data set covers a variety of driving conditions and tire parameters, multiple groups of sample data of multiple tires under a variety of driving conditions can be obtained from historical road test data and historical finite element simulation data, wherein each group of sample data includes: second driving condition data, second tire parameters and a time-based second wear data sequence of a tire under a driving condition, the second driving condition data and the second tire parameters are used as input samples, and the second wear data sequence is used as a sample label.

[0037] Optionally, after obtaining the above sample data, firstly perform data cleaning on multiple groups of sample data, including duplicate value processing, missing value processing, and outlier processing. In order to eliminate the influence of data dimension and ensure that different features are compared and calculated on the same scale, the feature data of the same dimension in the multiple groups of sample data after data cleaning are normalized; then, the multiple groups of sample data are divided into training set, validation set and test set.

[0038] Before the model training begins, the preset number of training cycles, the preset training batch size, and the preset learning rate are determined. Then, the initial prediction model is iteratively trained using the training set. The specific steps are as follows: the training set is divided according to the preset training batch size, and in each training batch, the input samples corresponding to the training batch are input into the model. The mean square error loss function is constructed based on the model output and the sample labels corresponding to the training batch, and the model parameters are adjusted based on the mean square error loss function.

[0039] Optionally, after each training cycle is completed, the validation set is used to verify the model after each training cycle is completed. The specific steps are as follows: the input samples corresponding to the validation set are input into the model, and the validation loss is determined based on the model output and the sample labels corresponding to the validation set. If the decrease in the validation loss for a preset number of consecutive adjacent training cycles is less than a preset threshold, the preset learning rate is decayed according to the preset step size.

[0040] For example, create a learning rate adjuster ReduceLROnPlateau, which automatically reduces the learning rate when the validation loss does not improve within a training round (i.e. reaches a plateau). The maximum round set here is 5, which means that if the validation loss does not decrease for 5 consecutive rounds, i.e. 5 consecutive training cycles, the learning rate reduction is triggered.

[0041] As an optional implementation, after the number of training cycles reaches a preset number of training cycles, the input samples corresponding to the test set are input into the model, and the model performance evaluation indicators are determined based on the model output and the sample labels corresponding to the test set, where the model performance evaluation indicators include: mean absolute error, correlation coefficient and root mean square error.

[0042] When the above-mentioned model performance evaluation index reaches the qualified standard, the model after the test is used as the tire wear prediction model, and the tire wear prediction model can be stored as a file in the target format for calling. Among them, the target format can be: HDF5 format based on Python deep learning framework, SavedModel format based on TensorFlow deep learning framework, ONNX (Open Neural Network Exchange) format that supports multiple deep learning frameworks, PMML (Predictive Model Markup Language) format based on XML (Extensible Markup Language), etc. The specific model file format can be selected according to the actual application environment of the subsequent model, including but not limited to the above.

[0043] When using the tire wear prediction model for analysis, first obtain the first driving condition data and the first tire parameter of the target tire under the target driving condition; then call the pre-trained tire wear prediction model to analyze the first driving condition data and the first tire parameter to obtain a first wear data sequence of the target tire in a future preset time period; finally, determine the wear life of the target tire under the target driving condition based on the first wear data sequence.

[0044] Optionally, the target driving condition includes multiple preset driving conditions. After determining the wear life of the target tire under the target driving condition based on the first wear data sequence, the multiple preset driving conditions can also be sorted according to the size of the corresponding wear life; and the preset driving condition with the largest corresponding wear life is determined as the optimal driving condition for the target tire.

[0045] In an embodiment of the present application, the first driving condition data and the first tire parameter of the target tire under the target driving condition are obtained; the pre-trained tire wear prediction model is called to analyze the first driving condition data and the first tire parameter to obtain the first wear data sequence of the target tire in the future preset time period, wherein the tire wear prediction model is a long short-term memory LSTM network model, and the tire wear prediction model is trained using the driving condition data, tire parameters and time-based wear data sequence of different tires under different driving conditions; the wear life of the target tire under the target driving condition is determined according to the first wear data sequence. In this way, the first driving condition data and tire parameters of the target tire are deeply analyzed using the tire wear prediction model obtained by training the driving condition data, tire parameters and time-based wear data sequence of different tires under different driving conditions, so as to accurately predict the wear data sequence of the target tire in the future time period and the wear life under the target driving condition, thereby solving the technical problem that the related schemes are difficult to efficiently and accurately predict the wear life of the tire.

[0046] Example 2

[0047] According to an embodiment of the present application, a tire wear life prediction device for implementing the tire wear life prediction method in embodiment 1 is also provided. Figure 2 As shown, the tire wear life prediction device at least includes: an acquisition module 21, a prediction module 22 and a determination module 23, wherein:

[0048] An acquisition module 21, used for acquiring first driving condition data and first tire parameters of a target tire under a target driving condition;

[0049] The prediction module 22 is used to call a pre-trained tire wear prediction model to analyze the first driving condition data and the first tire parameter to obtain a first wear data sequence of the target tire in a preset time period in the future, wherein the tire wear prediction model is an LSTM network model, and the tire wear prediction model is trained using driving condition data of different tires under different driving conditions, tire parameters and a time-based wear data sequence;

[0050] The determination module 23 is used to determine the wear life of the target tire under the target driving condition according to the first wear data sequence.

[0051] The functions of each module of the tire wear life prediction device are explained below in conjunction with a specific implementation process.

[0052] In order to improve the efficiency and accuracy of tire wear assessment and more accurately predict the wear life of tires, this application proposes a tire wear prediction model based on a long short-term memory (LSTM) network. The model can effectively predict the service life of tires by analyzing driving condition data and tire parameters.

[0053] Optionally, before analyzing the first driving condition data and the first tire parameter using the tire wear prediction model, the driving condition data of different tires under different driving conditions, tire parameters and a time-based wear data sequence are used to train the model. The characteristic dimensions of the driving condition data include at least one of the following: road surface conditions, ambient temperature, vehicle driving speed, tire pressure, tire load, tire camber angle, and tire toe angle; the characteristic dimensions of the tire parameters include at least one of the following: tire size and tire material; and the wear data sequence includes: data on the depth of each node on the tire changing over time.

[0054] Specifically, the training process of the tire wear prediction model can be carried out in the following manner:

[0055] First, an initial prediction model is constructed, which includes: input layer, LSTM layer, linear rectified unit ReLU activation function layer, random dropout layer, fully connected layer and linear activation function layer.

[0056] Among them, the input layer is responsible for receiving input data. The dimension of the input data is set according to the actual situation, usually including tire driving condition data and tire parameters; the LSTM layer is used to process time-based wear data sequences and capture long-term dependencies in the data. The appropriate number of hidden units (for example, 128 hidden units) can be selected to ensure that the model has sufficient capacity to learn complex patterns; the ReLU (rectified linear unit) activation function layer is used to increase the nonlinear characteristics of the model, and the activation function helps the model better fit the nonlinear relationship; the Dropout layer prevents overfitting by randomly closing a certain proportion of neurons (for example, the proportion is set to 0.5) to reduce the model's excessive dependence on specific input features; the fully connected layer usually contains multiple neurons for extracting higher-level features; the linear activation function layer is used for the final output, and usually no activation function is used so that the model can output a series of continuous values, such as tire wear data sequences.

[0057] As an optional implementation, in order to ensure that the training data set covers a variety of driving conditions and tire parameters, multiple groups of sample data of multiple tires under a variety of driving conditions can be obtained from historical road test data and historical finite element simulation data, wherein each group of sample data includes: second driving condition data, second tire parameters and a time-based second wear data sequence of a tire under a driving condition, the second driving condition data and the second tire parameters are used as input samples, and the second wear data sequence is used as a sample label.

[0058] Optionally, after obtaining the above sample data, firstly perform data cleaning on multiple groups of sample data, including duplicate value processing, missing value processing, and outlier processing. In order to eliminate the influence of data dimension and ensure that different features are compared and calculated on the same scale, the feature data of the same dimension in the multiple groups of sample data after data cleaning are normalized; then, the multiple groups of sample data are divided into training set, validation set and test set.

[0059] Before the model training begins, the preset number of training cycles, the preset training batch size, and the preset learning rate are determined. Then, the initial prediction model is iteratively trained using the training set. The specific steps are as follows: the training set is divided according to the preset training batch size, and in each training batch, the input samples corresponding to the training batch are input into the model. The mean square error loss function is constructed based on the model output and the sample labels corresponding to the training batch, and the model parameters are adjusted based on the mean square error loss function.

[0060] Optionally, after each training cycle is completed, the validation set is used to verify the model after each training cycle is completed. The specific steps are as follows: the input samples corresponding to the validation set are input into the model, and the validation loss is determined based on the model output and the sample labels corresponding to the validation set. If the decrease in the validation loss for a preset number of consecutive adjacent training cycles is less than a preset threshold, the preset learning rate is decayed according to the preset step size.

[0061] As an optional implementation, after the number of training cycles reaches a preset number of training cycles, the input samples corresponding to the test set are input into the model, and the model performance evaluation indicators are determined based on the model output and the sample labels corresponding to the test set, where the model performance evaluation indicators include: mean absolute error, correlation coefficient and root mean square error.

[0062] When the above-mentioned model performance evaluation indicators meet the qualified standards, the model after the test is used as a tire wear prediction model, and the tire wear prediction model can be stored as a file in a target format for calling, wherein the target format can be: HDF5 format based on Python deep learning framework, SavedModel format based on TensorFlow deep learning framework, ONNX format supporting multiple deep learning frameworks, PMML format based on XML, etc. The specific model file format can be selected according to the actual application environment of the subsequent model, including but not limited to the above.

[0063] When using the tire wear prediction model for analysis, the acquisition module first obtains the first driving condition data and the first tire parameter of the target tire under the target driving condition; the prediction module then calls the pre-trained tire wear prediction model to analyze the first driving condition data and the first tire parameter to obtain a first wear data sequence of the target tire in a preset time period in the future; finally, the determination module determines the wear life of the target tire under the target driving condition based on the first wear data sequence.

[0064] Optionally, the target driving condition includes multiple preset driving conditions. After determining the wear life of the target tire under the target driving condition based on the first wear data sequence, the multiple preset driving conditions can also be sorted according to the size of the corresponding wear life; and the preset driving condition with the largest corresponding wear life is determined as the optimal driving condition for the target tire.

[0065] It should be noted that each module in the tire wear life prediction device in the embodiment of the present application corresponds one by one to each implementation step of the tire wear life prediction method in Example 1. Since a detailed description has been given in Example 1, some details not reflected in this embodiment can be referred to Example 1 and will not be repeated here.

[0066] Example 3

[0067] According to an embodiment of the present application, a computer program product is also provided. The computer program product includes a computer program, wherein when the computer program is executed by a processor, the tire wear life prediction method in Example 1 is implemented.

[0068] According to an embodiment of the present application, a non-volatile storage medium is also provided, which includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the tire wear life prediction method in Example 1 by running the computer program.

[0069] According to an embodiment of the present application, a processor is also provided, which is used to run a computer program, wherein the tire wear life prediction method in Example 1 is executed when the computer program is running.

[0070] According to an embodiment of the present application, an electronic device is also provided, which includes: a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to execute the tire wear life prediction method in Example 1 through the computer program.

[0071] Specifically, the following steps are executed when the computer program is running: obtaining first driving condition data and first tire parameters of a target tire under a target driving condition; calling a pre-trained tire wear prediction model to analyze the first driving condition data and the first tire parameters to obtain a first wear data sequence of the target tire within a preset time period in the future, wherein the tire wear prediction model is a long short-term memory (LSTM) network model, and the tire wear prediction model is trained using driving condition data, tire parameters, and time-based wear data sequences of different tires under different driving conditions; and determining the wear life of the target tire under the target driving condition based on the first wear data sequence.

[0072] As an optional implementation, the electronic device may be in the form of a mobile terminal, a computer terminal or a similar computing device. Figure 3 FIG. 1 shows a hardware structure block diagram of an electronic device for implementing a tire wear life prediction method. Figure 3 As shown, the electronic device 30 may include one or more (302a, 302b, ..., 302n are used to illustrate) processors 302 (the processor 302 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 304 for storing data, and a transmission device 306 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 3 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 3 More or fewer components as shown, or with Figure 3 Different configurations are shown.

[0073] It should be noted that the one or more processors 302 and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the electronic device 30. As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0074] The memory 304 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the tire wear life prediction method in the embodiment of the present application. The processor 302 executes various functional applications and data processing by running the software programs and modules stored in the memory 304, that is, to implement the vulnerability detection method of the above-mentioned application. The memory 304 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 304 may further include a memory remotely arranged relative to the processor 302, and these remote memories may be connected to the electronic device 30 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0075] The transmission device 306 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the electronic device 30. In one example, the transmission device 306 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 306 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0076] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the electronic device 30 .

[0077] The serial numbers of the above embodiments are only for description and do not represent the advantages or disadvantages of the embodiments.

[0078] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0079] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0080] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed over multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0081] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0082] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or all or part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of each embodiment method of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), mobile hard disk, disk or optical disk, etc., which can store program code.

[0083] The above are only preferred implementations of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A tire wear life prediction method, characterized in that: include: Acquiring first driving condition data and first tire parameters of a target tire under a target driving condition; Calling a pre-trained tire wear prediction model to analyze the first driving condition data and the first tire parameter to obtain a first wear data sequence of the target tire in a future preset time period, wherein the tire wear prediction model is a long short-term memory (LSTM) network model, and the tire wear prediction model is trained using driving condition data of different tires under different driving conditions, tire parameters, and a time-based wear data sequence; The wear life of the target tire under the target driving condition is determined according to the first wear data sequence.

2. The method according to claim 1, characterized in that The characteristic dimensions of the driving condition data include at least one of the following: road surface conditions, ambient temperature, vehicle speed, tire pressure, tire load, tire camber angle, and tire toe angle; The characteristic dimensions of the tire parameters include at least one of the following: tire size, tire material; The wear data sequence includes data on the depth of each node on the tire changing with time.

3. The method according to claim 1, characterized in that The training process of the tire wear prediction model includes: Constructing an initial prediction model, wherein the initial prediction model includes in sequence: an input layer, an LSTM layer, a linear rectified unit ReLU activation function layer, a random dropout layer, a fully connected layer, and a linear activation function layer; Acquire multiple groups of sample data of multiple tires under multiple driving conditions from historical road test data and historical finite element simulation data, wherein each group of sample data includes: second driving condition data of a tire under one driving condition, second tire parameters, and a second wear data sequence based on time, the second driving condition data and the second tire parameters are used as input samples, and the second wear data sequence is used as a sample label; Dividing the multiple groups of sample data into a training set, a validation set and a test set; The initial prediction model is iteratively trained using the training set, the model after each training cycle is verified using the verification set, the model after training is tested using the test set, and the model after the test is used as the tire wear prediction model.

4. The method according to claim 3, characterized in that: Before dividing the plurality of groups of sample data into a training set, a validation set and a test set, the method further comprises: Performing data cleaning on the multiple groups of sample data, wherein the data cleaning includes: duplicate value processing, missing value processing, and outlier value processing; The feature data of the same dimension in multiple groups of sample data after data cleaning are normalized respectively.

5. The method according to claim 3, characterized in that: Iteratively training the initial prediction model using the training set, verifying the model after each training cycle using the verification set, and testing the model after the training using the test set, including: Determine the preset number of training epochs, the preset training batch size, and the preset learning rate; During the iterative training process, the training set is divided according to the preset training batch size, in each training batch, the input samples corresponding to the training batch are input into the model, a mean square error loss function is constructed according to the model output and the sample labels corresponding to the training batch, and the model parameters are adjusted according to the mean square error loss function; After each training cycle is completed, the input sample corresponding to the verification set is input into the model, and the verification loss is determined according to the model output and the sample label corresponding to the verification set. If the decrease value of the verification loss for a preset number of consecutive adjacent training cycles is less than a preset threshold, the preset learning rate is decayed according to a preset step size; After the number of training cycles reaches the preset number of training cycles, the input samples corresponding to the test set are input into the model, and the model performance evaluation indicators are determined based on the model output and the sample labels corresponding to the test set, wherein the model performance evaluation indicators include: mean absolute error, correlation coefficient and root mean square error.

6. The method according to claim 3, characterized in that The method further comprises: After the tire wear prediction model is obtained through training, the tire wear prediction model is stored as a file in a target format for calling, wherein the target format includes: HDF5 format, SavedModel format, open neural network exchange ONNX format, and prediction model markup language PMML format.

7. The method according to claim 1, characterized in that The target driving condition includes a plurality of preset driving conditions. After determining the wear life of the target tire under the target driving condition according to the first wear data sequence, the method further includes: sorting the plurality of preset driving conditions according to the size of the corresponding wear life; The preset driving condition corresponding to the maximum wear life is determined as the optimal driving condition for the target tire.

8. A tire wear life prediction device, characterized in that: include: An acquisition module, used for acquiring first driving condition data and first tire parameters of a target tire under a target driving condition; A prediction module, configured to call a pre-trained tire wear prediction model to analyze the first driving condition data and the first tire parameter to obtain a first wear data sequence of the target tire in a future preset time period, wherein the tire wear prediction model is an LSTM network model, and the tire wear prediction model is trained using driving condition data of different tires under different driving conditions, tire parameters, and a time-based wear data sequence; A determination module is used to determine the wear life of the target tire under the target driving condition according to the first wear data sequence.

9. A computer program product, characterized in that include: A computer program, wherein when the computer program is executed by a processor, the tire wear life prediction method according to any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: include: A memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the tire wear life prediction method according to any one of claims 1 to 7 through the computer program.

Images