Method, device and electronic equipment for determining tire inflation size

Abstract

This application discloses a method, apparatus, and electronic device for determining tire inflation dimensions. The method includes: acquiring geometric design parameters of the tire to be predicted, wherein the geometric design parameters include initial dimensional information of the tire in its uninflated state; and analyzing the geometric design parameters using a target model to obtain target dimensional information of the tire to be predicted, wherein the target model includes a learning model for learning the changes in the initial dimensional information of the tire under inflated conditions. This application solves the technical problem of related technologies where the prediction of tire inflation section width relies on the designer's experience and repeated finite element simulation verification, resulting in a long prediction cycle and low accuracy.

Description

Technical Field

[0001] This application relates to the field of tire engineering technology, and more specifically, to a method, apparatus, and electronic device for determining tire inflation size. Background Technology

[0002] As a core component of automobiles, the cross-sectional width and outer diameter of tires after inflation are key geometric parameters that determine tire fitability, driving safety, load-bearing capacity, and NVH performance. Related technologies, through the experience of designers and finite element simulation analysis, pre-consider deformation to ensure that indicators such as the cross-sectional width after inflation meet standards during product development. After product manufacturing, experimental verification is then conducted. However, this method requires product personnel to have extensive development experience, and the simulation and experimental verification methods are time-consuming and costly. If the development fails, the cost of iterating the product becomes prohibitively high.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This application provides a method, apparatus, and electronic device for determining tire inflation dimensions, which at least solves the technical problem that the prediction of tire inflation cross-sectional width in related technologies relies on the experience judgment of designers and repeated finite element simulation verification, resulting in a long prediction cycle and low accuracy.

[0005] According to one aspect of the embodiments of this application, a method for determining the inflation size of a tire is provided, comprising: obtaining geometric design parameters of a tire to be predicted, wherein the geometric design parameters include initial size information of the tire to be predicted in an uninflated state; and analyzing the geometric design parameters using a target model to obtain target size information of the tire to be predicted, wherein the target model includes a learning model for learning the changes in the initial size information of the tire to be predicted in an inflated state.

[0006] In some embodiments of this application, the initial size information includes the mold cross-sectional width of the tire to be predicted; the target size information of the tire to be predicted is obtained by analyzing the geometric design parameters using a target model, including: predicting the geometric design parameters using a first target model to obtain the cross-sectional width increment of the tire to be predicted, wherein the first target model is trained based on historical geometric design parameters and corresponding historical cross-sectional widths, and the cross-sectional width increment is used to characterize the degree of deformation of the tire structure to be predicted expanding to the left and right sides under internal pressure; the inflation cross-sectional width of the tire to be predicted is determined based on the mold cross-sectional width and the cross-sectional width increment, and the inflation cross-sectional width is used as the first target size information.

[0007] In some embodiments of this application, the initial size information includes the initial outer diameter of the tire to be predicted; the target size information of the tire to be predicted is obtained by analyzing the geometric design parameters using a target model, including: using a second target model to predict the geometric design parameters to obtain the inflated outer diameter of the tire to be predicted, wherein the second target model is trained based on historical geometric design parameters and corresponding historical outer diameters, and the inflated outer diameter is used to characterize the deformation behavior of the structure of the tire to be predicted expanding to the upper and lower sides under the action of internal pressure; the inflated outer diameter is used as the target size information.

[0008] In some embodiments of this application, the target model is trained by: acquiring historical geometric design parameters of multiple tires; determining a geometric model corresponding to the historical geometric design parameters, wherein the geometric model is used to characterize the parameterized geometric profile of the tire cross section; converting the geometric model into a simulation model, and performing simulation analysis on the simulation model to obtain simulation dimension information corresponding to the historical geometric design parameters, wherein the simulation dimension information is used to quantitatively represent the deformation behavior of the tire in the inflated state; and training the initial model using the historical geometric design parameters and the simulation dimension information to obtain the target model.

[0009] In some embodiments of this application, the simulation dimensional information includes the simulation inflatable cross-sectional width and the simulation inflatable outer diameter; training the initial model using historical geometric design parameters and simulation dimensional information to obtain a target model includes: training the initial model based on the simulation inflatable cross-sectional width to obtain a first target model for predicting the inflatable cross-sectional width; and training the initial model based on the simulation inflatable outer diameter to obtain a second target model for predicting the inflatable outer diameter.

[0010] In some embodiments of this application, the method further includes: obtaining the predicted demand of the tire to be predicted; and selecting at least one of the first target model and the second target model for prediction based on the predicted demand.

[0011] In some embodiments of this application, the simulation model includes a two-dimensional tire model; the simulation model is analyzed to obtain simulation size information corresponding to historical geometric design parameters, including: determining the inner surface and outer surface of the tire corresponding to the two-dimensional tire model; applying radial surface pressure to the inner surface of the tire to simulate the tire inflation process, and extracting the simulation size information of the two-dimensional tire model after inflation based on the outer surface of the tire.

[0012] According to another aspect of the embodiments of this application, a device for determining tire inflation size is also provided, comprising: an acquisition module for acquiring geometric design parameters of a tire to be predicted, wherein the geometric design parameters include initial size information of the tire to be predicted in an uninflated state; and an analysis module for analyzing the geometric design parameters using a target model to obtain target size information of the tire to be predicted, wherein the target model includes a learning model for learning the changes in the initial size information of the tire to be predicted in an inflated state.

[0013] According to another aspect of the embodiments of this application, an electronic device is also provided, including: a memory and a processor, wherein the memory is used to store program instructions; the processor is connected to the memory and is used to execute the above-described method for determining tire inflation size.

[0014] According to another aspect of the embodiments of this application, a non-volatile storage medium is also provided, the non-volatile storage medium including a stored computer program, wherein the device containing the non-volatile storage medium executes the above-described method for determining tire inflation size by running the computer program.

[0015] According to another aspect of the embodiments of this application, a computer program product is also provided, including computer instructions that, when executed by a processor, implement the above-described method for determining tire inflation size.

[0016] In this embodiment, by extracting the geometric design parameters of the tire in its uninflated state and inputting them into a target model with the ability to learn the inflation deformation law, the target size information after inflation is automatically output. This achieves the goal of quickly obtaining accurate inflation size without human experience intervention and physical simulation experiments, thereby realizing the technical effect of efficient and automated prediction of tire inflation section width. This solves the technical problem that the prediction of tire inflation section width in related technologies relies on the experience judgment of designers and repeated finite element simulation verification, resulting in a long prediction cycle and low accuracy. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0018] Figure 1 This is a hardware structure block diagram of a computer terminal for a method of determining tire inflation size according to an embodiment of this application.

[0019] Figure 2 This is a flowchart of a method for determining tire inflation size according to an embodiment of this application;

[0020] Figure 3This is a schematic diagram of the overall process of a method for determining tire inflation size according to an embodiment of this application;

[0021] Figure 4 This is a finite element simulation flowchart of a method for determining tire inflation size according to an embodiment of this application;

[0022] Figure 5 This is a schematic diagram illustrating the prediction accuracy of a method for determining tire inflation size according to an embodiment of this application;

[0023] Figure 6 This is a schematic diagram of a tire inflation size determination device according to an embodiment of this application. Detailed Implementation

[0024] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0025] It should 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 data 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 herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0026] To better understand the embodiments of this application, the technical terms involved in the embodiments of this application are explained below:

[0027] Geometric design parameters refer to the set of quantitative variables that describe the size, position, and shape of each structural component of a tire in its uninflated state. These include structural and operational parameters such as tread rubber thickness, sidewall rubber thickness, belt layer angle, tire body positioning point coordinates, triangular rubber thickness, mold cross-section width, rim width, and air pressure.

[0028] Target Dimensional Information refers to the key physical dimensions of a tire when it is inflated, including outputs characterizing the tire's spatial shape such as the inflatable cross-sectional width and inflatable outer diameter. In this embodiment, the target dimensional information is the output of the model, used to directly reflect the tire's deformation response under actual usage conditions.

[0029] A Learning Model (LM) is a data-driven statistical machine learning algorithm that automatically constructs nonlinear mapping relationships from historical input-output samples, predicting system responses without explicit physical equations. In this embodiment, the learning model obtains the complex correlation between initial size information and target size information after inflation through training, achieving rapid end-to-end extrapolation from parameter input to size output.

[0030] Target Model (TM): An integrated prediction system consisting of one or more learning models, dedicated to outputting the target size information of the inflated tire based on the input geometric design parameters.

[0031] Tires are a core component of automobiles, playing a role in bearing loads, braking, driving, and cushioning during vehicle operation. The inflated section width and outer diameter of a tire are crucial aspects of its performance. Appropriate inflated section width and outer diameter can effectively ensure vehicle safety. Excessively large or small inflated section width / outer diameter not only affect tire performance but also significantly impact vehicle safety. Therefore, many countries and regions have mandatory standards for the inflated section width of tires, requiring it to remain within the standard range.

[0032] Currently, the prediction of tire inflation section width mainly relies on the experience and judgment of designers and repeated finite element simulation verification. This requires manual modeling, manual parameter adjustment, successive simulation calculations, and verification by physical tests. The process is lengthy (a single simulation takes several hours to several days), costly, and has low iteration efficiency. Furthermore, due to the complexity of the tire structure (such as multiple design factors including rubber, cords, carcass, and belt layers) and operating condition variables (such as rim specifications and air pressure), traditional methods cannot systematically quantify the nonlinear influence of each parameter on the inflation section width, making it difficult to quickly locate key control factors when the design deviates from the standard range.

[0033] To address the aforementioned technical problems, this application provides corresponding solutions, which are detailed below.

[0034] The method for determining tire inflation size provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1A hardware block diagram of a computer terminal for implementing a method for determining tire inflation size is shown. Figure 1 As shown, the computer terminal 10 may include one or more processors (shown as 102a, 102b, ..., 102n in the figure) (the processor may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission module 106 for communication functions connected via wired and / or wireless networks. In addition, it may also include: a display, a keyboard, a cursor control device, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, and a BUS bus. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0035] It should be noted that the aforementioned one or more processors and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10. As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0036] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the tire inflation size determination method in this embodiment. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned tire inflation size determination method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0037] The transmission module 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission module 106 includes a network interface controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission module 106 may be a radio frequency (RF) module, used for wireless communication with the Internet.

[0038] The display can be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10.

[0039] It should be noted here that, in some optional embodiments, the above... Figure 1 The computer terminal shown may include hardware elements (including circuitry), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that... Figure 1 This is only one instance of a specific particular instance, and is intended to illustrate the types of components that may exist in the aforementioned computer terminal.

[0040] In the above operating environment, this application provides an embodiment of a method for determining tire inflation size. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Also, although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than that shown here.

[0041] Figure 2 This is a flowchart of a method for determining tire inflation size according to an embodiment of this application, such as... Figure 2 As shown, the method includes the following steps:

[0042] Step S202: Obtain the geometric design parameters of the tire to be predicted, wherein the geometric design parameters include the initial size information of the tire to be predicted in the uninflated state.

[0043] In step S202 above, geometric design parameters refer to a set of quantifiable parameters that describe the geometric dimensions, spatial distribution, and material configuration of each structural component of the tire in an uninflated state. Initial size information refers to the structural size data of the tire in an uninflated, unloaded, and unassembled rim state. It is a core subset of geometric design parameters, including but not limited to geometric quantities that can be directly obtained from design drawings, such as mold cross-section width, tread running width, and tire end position.

[0044] It should be noted that the initial dimensional information is not directly used for physical simulation, but rather serves as the basic input feature for training and inference of the learning model. Its numerical changes are implicitly correlated with the deformation response after inflation. For example, although a 1mm increase in the mold cross-sectional width is not directly equivalent to a 1mm increase in the inflation cross-sectional width (due to the nonlinearity of the elastic deformation of the cord and the rubber), this change will have a high correlation with the final cross-sectional width value.

[0045] In some embodiments of this application, the parametric design module integrated into the platform allows engineers to complete tire structure design in a graphical interface. The system then automatically extracts all preset geometric design parameters and writes them to the local database in real time in a structured data format (such as JSON or HDF5). This method relies on the API interface between the design software and the prediction platform. The parameter extraction process is fully automated and requires no manual intervention, making it suitable for platform-based development processes.

[0046] In a non-integrated environment, designers can also manually export key dimension parameter tables (such as in Excel format) from design drawings. These tables are then read by a data acquisition program and standardized (e.g., units are standardized to mm, and text-based material types are converted to unique thermal codes). After feature alignment, a vector consistent with the training data of the model is formed.

[0047] Because the parameter set used contains hundreds of variables, some of which are textual (e.g., cord material type), some are continuous (e.g., thickness), and some are discrete (e.g., number of belt layers), inconsistent dimensions of the input parameters or mismatched data formats can easily lead to the failure of the target model output or a sharp increase in prediction bias. To solve this problem, a parameter consistency verification and standardization preprocessing module can be introduced. This module performs several automatic processing steps before the parameters are input into the target model, including but not limited to:

[0048] ① Parameter integrity verification: compare the input parameter set with the union of the 220-dimensional parameters used in the training model, automatically fill missing items with default values ​​(such as the median) and mark warnings; ② Dimension and unit standardization: force all length units to be converted to millimeters, angles to degrees, and air pressure to MPa; ③ Text parameter one-hot encoding conversion: for example, "cord material = Nylon 66" is automatically mapped by the system to a binary vector of length 8 (corresponding to 8 common cord types).

[0049] Step S204: Analyze the geometric design parameters using the target model to obtain the target size information of the tire to be predicted. The target model includes a learning model for learning the changes in the initial size information of the tire to be predicted under the inflation state.

[0050] In step S204 above, the target model refers to an intelligent prediction system that encapsulates statistical learning capabilities. Essentially, it constructs a high-dimensional nonlinear mapping function between input and output by training on historical design parameters and corresponding inflation size response data. It should be noted that the target model does not rely on any physical equations or mechanical solvers, but is entirely based on a data-driven reasoning mechanism. Its input is the tire's geometric design parameters, and its output is the target size information after inflation, such as the inflation cross-sectional width and inflation outer diameter.

[0051] A learning model refers to a class of function approximators built based on machine learning algorithms, used to automatically identify implicit correlations between input and output variables from sample data. Typical examples include Random Forest, Gradient Boosting Machine (GBM), and Extreme Gradient Boosting (XGBoost). In some embodiments of this application, the learning model is the core component of the target model. Its function is to learn the changes in the initial size information of the tire to be predicted under the inflated state. That is, from a large number of paired samples of historical geometric design parameters and historical target size information, it extracts the statistical laws of how structural parameters, through complex coupling effects such as material elastic deformation, cord expansion, and rubber layer compression, ultimately lead to the nonlinear growth of cross-sectional width and outer diameter.

[0052] In some embodiments of this application, once the new tire design is completed and the complete geometric design parameters are extracted, the system automatically inputs the parameter vector into the trained target model. The target model internally calls its embedded learning model (such as XGBoost) to perform a forward propagation calculation: first, the input parameters are standardized (e.g., Z-score normalization) and text variables are one-hot encoded (e.g., material type "steel wire / nylon" is converted to a binary vector); then, feature importance weighting and nonlinear combination are performed through a multi-layer tree structure ensemble model, finally outputting continuous values ​​of the target size information. This process does not require calling any simulation solver, and a single prediction takes approximately 0.03 seconds, making it suitable for rapid scheme selection and iterative optimization during the design phase.

[0053] Before deploying the target model, a learning model can be trained and optimized using historical data. Specifically, over 10,000 sets of historical geometric design parameters and historical target size information samples can be extracted from the platform and historical simulation database, randomly dividing 80% into a training set and 20% into a validation set. Subsequently, the system trains multiple learning models (such as random forests, support vector machines, and neural networks) in parallel, using mean absolute error (MAE) and coefficient of determination (R²) as evaluation metrics for comprehensive comparison and selection. Ultimately, the XGBoost model was selected as the core architecture of the learning model due to its MAE=0.011mm and R²=0.994 on the validation set, and was packaged into a stable target model service.

[0054] In traditional methods, engineers need to set mold dimensions based on experience and then simulate inflation deformation through finite element simulation. If the results deviate from the target value, multiple parameters need to be manually adjusted and the simulation rerun. A single iteration cycle often takes several hours, and due to the complex parameter coupling, multiple rounds are often required to converge, resulting in a 2-4 week extension of the new product development cycle. This application, by introducing a target model, reconstructs the serial process of simulation → analysis → adjustment into a parallel decision-making process of input parameters → one-click prediction → real-time feedback. This achieves a prediction efficiency improvement of more than a thousand times, and the prediction accuracy is stable within ±0.012mm, completely changing the paradigm of tire inflation size prediction.

[0055] In some embodiments of this application, the initial size information includes the mold section width of the tire to be predicted. Based on this, the target size information of the tire to be predicted can be obtained in the following ways: the geometric design parameters are predicted using a first target model to obtain the section width increment of the tire to be predicted. The first target model is trained based on historical geometric design parameters and corresponding historical section widths. The section width increment is used to characterize the degree of deformation of the tire structure to be predicted as it expands to the left and right sides under internal pressure. The inflation section width of the tire to be predicted is determined based on the mold section width and the section width increment, and the inflation section width is used as the first target size information.

[0056] It should be noted that the mold cross-sectional width refers to the lateral width of the tire mold cavity at the center section of the tread when the mold is closed. It is one of the most basic and directly adjustable geometric input parameters in the tire design stage, and directly affects the profile of the tire when it is not inflated.

[0057] Section width increment refers to the lateral dimension increase of a tire as it expands from the mold section width to the actual inflated section width under the action of internal inflation pressure. It is used to quantitatively characterize the degree of nonlinear deformation response of the tire structure (including sidewall rubber elasticity, cord expansion, belt layer constraint, etc.) to both sides under internal pressure. Section width increment is not a fixed value, but is highly dependent on the coupling effect of various design parameters such as tire cord angle, belt layer modulus, triangular rubber stiffness, and air pressure.

[0058] The first target model refers to the dedicated machine learning model in this application specifically used to predict the cross-sectional width increment. Its training data comes from a large number of historical tire design parameters and the cross-sectional width increment values ​​(i.e., inflation cross-sectional width - mold cross-sectional width) obtained from corresponding finite element simulations. In some embodiments of this application, the model does not directly predict the final inflation cross-sectional width, but focuses on learning the statistical laws of how structural variables affect the physical mechanism of deformation increment. It decouples the high-dimensional, strongly coupled full-parameter prediction problem into a two-stage prediction of baseline size + incremental response, which significantly reduces model complexity, improves training convergence and engineering interpretability. For example, in the training data, when the "angle of the second belt layer" increases from 22° to 25°, the cross-sectional width increment increases by an average of 0.12 mm. This law is accurately captured by the first target model and used for incremental prediction of new designs.

[0059] In other embodiments of this application, the first target model can also be designed to directly predict the final inflatable cross-sectional width. Its prediction no longer depends on the addition operation of the mold cross-sectional width, but is directly generated by the model by learning the complex nonlinear relationship between the design parameter combination and the final deformation result in historical data.

[0060] In some embodiments of this application, the initial size information includes the initial outer diameter of the tire to be predicted. Based on this, the target size information of the tire to be predicted can be obtained in the following way: the geometric design parameters are predicted using a second target model to obtain the inflated outer diameter of the tire to be predicted. The second target model is trained based on historical geometric design parameters and corresponding historical outer diameters. The inflated outer diameter is used to characterize the deformation behavior of the tire to be predicted as it expands to the upper and lower sides under the action of internal pressure. The inflated outer diameter is used as the target size information.

[0061] It should be noted that the initial outer diameter refers to the maximum vertical diameter of the tire measured along the center section of the tread when the tire is uninflated, unloaded, and without a rim. It is one of the initial geometric references for the tire structure and is usually determined by the vertical height of the mold cavity. Although the initial outer diameter is recorded as part of the geometric design parameters, it is not directly used to calculate the final target size. Instead, it serves as an auxiliary reference feature for training the second target model. Its role is to provide the model with the vertical spatial profile of the tire in its undeformed state, assisting in learning the synergistic influence mechanisms of cord layer axial compression, sidewall rubber vertical buckling, and bead rigidity on the outer diameter change after inflation. For example, when the carcass cord angle is adjusted from 55° to 65°, even if the mold height remains unchanged, the outer diameter may actually decrease after inflation due to the increased axial component of cord tension. This non-intuitive response relies on the combined input of the initial outer diameter and other parameters to be captured by the learning model.

[0062] The inflatable outer diameter refers to the maximum diameter of the tire's outer contour measured along the vertical axis after it is inflated to standard pressure and installed on a specified rim. It is a core parameter for tire space adaptability, vehicle ground clearance, transmission ratio matching, and NVH performance. In some embodiments of this application, the inflatable outer diameter is directly predicted by the second target model, and its value is jointly controlled by multiple parameters such as tire carcass stiffness, belt layer constraint, rubber modulus, air pressure, and rim width. Unlike the section width, the deformation of the outer diameter is mainly manifested as the overall "stretching" or "compression" of the tire in the axial direction. Its changing trend is partially decoupled from the section width. Therefore, embodiments of this application use an independent modeling strategy for prediction. For example, when the tire carcass cord modulus increases by 20% and the air pressure increases from 0.8MPa to 1.0MPa, the inflatable outer diameter may decrease by 0.5mm due to the increased axial stiffness of the cord. This complex coupling relationship is learned by the second target model as a nonlinear mapping function through historical data.

[0063] The second target model refers to the independent machine learning model in this application specifically used to predict the inflation outer diameter of tires. Its training data comes from historical tire design parameters and the inflation outer diameter values ​​obtained from corresponding finite element simulations. This model does not rely on physical equations, but implicitly models the statistical laws of "structural design - axial deformation response" through a data-driven approach. By decoupling the outer diameter prediction from the overall deformation modeling, it avoids parameter interference or gradient conflicts with the cross-section width prediction model, thereby improving the model's convergence speed, prediction accuracy, and engineering robustness.

[0064] Traditional methods treat the outer diameter and cross-sectional width as coupled variables for unified prediction, resulting in high model complexity, unstable training, and weak generalization ability to extreme parameter combinations. In contrast, this application achieves decoupled modeling of the two major physical responses, lateral deformation and longitudinal deformation, by making the outer diameter prediction an independent second objective model and running it in parallel with the cross-sectional width prediction model. This not only improves the model's convergence efficiency and prediction accuracy but also enhances its robustness to high-dimensional parameter spaces.

[0065] The target model is trained as follows: historical geometric design parameters of multiple tires are obtained; a geometric model corresponding to the historical geometric design parameters is determined, wherein the geometric model is used to characterize the parametric geometric profile of the tire cross section; the geometric model is converted into a simulation model, and simulation analysis is performed on the simulation model to obtain simulation dimensional information corresponding to the historical geometric design parameters, wherein the simulation dimensional information is used to quantitatively represent the deformation behavior of the tire under inflation; the initial model is trained using the historical geometric design parameters and the simulation dimensional information to obtain the target model.

[0066] It should be noted that the geometric model refers to a digital structural model constructed based on parametric modeling technology to accurately represent the geometric contour of the tire's cross-section. Essentially, it is a two-dimensional cross-sectional wireframe diagram generated in a CAD platform through parameter-driven processing. It includes the relative positions, boundary shapes, and connection relationships of components such as the tread, sidewall, carcass, belt layers, triangular rubber, and bead. This model does not contain material constitutive or mechanical properties; it is merely a geometric and topological representation. The geometric model transforms abstract design variables into geometric inputs that can be recognized by finite element analysis software. For example, when the parameter "carcass positioning point 1" changes, the endpoint position of the carcass cord layer in the geometric model is automatically repositioned, causing the sidewall rubber contour to deform synchronously, thus realistically reflecting the impact of structural changes on the tire's cross-sectional shape.

[0067] A simulation model is a finite element model for mechanical calculations, formed by meshing a geometric model, assigning material properties, and setting boundary conditions. It includes the material constitutive models of each component (such as hyperelastic rubber models and orthotropic cord models), contact relationships, constraint conditions, and inflation pressure loads. The simulation model is a digital representation of physical behavior, used to simulate the structural deformation process of a tire under standard tire pressure in a virtual environment. It outputs quantitative response data such as inflation cross-sectional width and outer diameter, providing "ground truth" for machine learning models and ensuring the physical reliability of the training data. Specifically, a preset tire cross-section parameterized template can be called, and the corresponding two-dimensional geometric model can be generated in real time by driving the geometric parameter updates through scripts.

[0068] Simulation dimensional information refers to the quantitative results obtained by finite element simulation model, which characterize the structural deformation behavior of the tire in the inflated state. It mainly includes key dimensions such as the inflated cross-sectional width and the inflated outer diameter. This information is a direct reflection of tire performance and also serves as the supervised output label for training the target model.

[0069] Specifically, the generated geometric model can be automatically imported into a finite element analysis (FEA) simulation platform and perform four steps: ① Automatically generate axisymmetric two-dimensional meshes to ensure that critical areas (such as tire bead and belt layer ends) are finely meshed; ② Assign corresponding hyperelastic models and orthotropic constitutive models to each component (rubber compound, cord, and steel wire coil) according to a preset material library; ③ Apply rim constraints and internal pressure loads (standard air pressure 0.9MPa); ④ Start static solution and extract the inflation section width and outer diameter.

[0070] After obtaining the simulation size information, the dataset can be divided into a training set (80%) and a validation set (20%). Multiple machine learning algorithms (random forest, XGBoost, neural networks, and support vector regression) are then loaded as initial models. By calculating the mean absolute error (MAE) and coefficient of determination (R²) of each model on the validation set, the best-performing model is selected as the target model.

[0071] In traditional methods, designers need to manually modify CAD models, submit simulation experiments one by one, and manually analyze the results. A single optimization takes 3–7 days, and due to the limited sample size (usually only a few dozen sets), it is difficult to establish reliable predictive patterns. This application, however, transforms the original manual, single-point trial-and-error process into a batch data-driven process by automatically generating a "design-simulation" dataset. It builds over 10,000 high-quality samples within a few weeks, enabling machine learning models to capture complex nonlinear relationships and achieve a prediction accuracy of ±0.012 mm and a single response time of <0.03 seconds.

[0072] However, if the physical accuracy of the simulation model is insufficient or the mesh / material settings are inconsistent, it can easily lead to systematic deviations in the simulation dimensional information, contaminating the training data. To address this, simulation consistency verification and multi-source error control mechanisms can be employed, which may include at least one of the following measures:

[0073] ① The simulation standard template is forcibly bound. All simulation models must be executed based on the unified "Tire Finite Element Modeling Specification", including mesh density requirements (e.g., the minimum element in the bead area is ≤1mm), material model type (e.g., the rubber material uniformly adopts Mooney-Rivlin, and the cord uniformly adopts orthogonal anisotropy), and load application method (internal pressure is a uniform surface load). Any deviation will trigger a system alarm and refuse submission.

[0074] ②Simulation result consistency check: The system randomly selects 5% of the simulation samples, and CAE experts review their mesh quality, material assignment and convergence status. Abnormal samples are removed or recalculated.

[0075] ③ Simulation error compensation training: During the model training phase, simulation uncertainty weights are introduced. Samples with coarse meshes or high material parameter uncertainties are assigned lower training weights to reduce their impact on model parameter updates. Specifically, after each set of geometric models is converted into simulation models and finite element analysis is performed, the system automatically extracts key quality indicators from the simulation process, including: ① Whether the minimum element size of the tire carcass cord layer and belt layer region is lower than a preset threshold (e.g., 1.0 mm); ② Whether the contact penetration of the contact surface (bead-rim) exceeds the material thickness (e.g., 5%); ③ Whether the stress-strain convergence residual of the rubber constitutive model is greater than a preset threshold (e.g., 10 mm). -4 ④ Whether the material parameters are input from actual measured data or only use typical values ​​from literature (e.g., the Mooney viscosity of the tread rubber is not marked with a test standard).

[0076] Furthermore, the above indicators are weighted to form a comprehensive simulation confidence score for a single set of simulation samples, with a value range of 0.0–1.0. The higher the score, the more reliable the simulation results of the sample. The system dynamically assigns training weight coefficients (ω) to each set of training samples based on the comprehensive simulation confidence score. For example, the rules can be set as follows: when the comprehensive simulation confidence score is ≥0.95, ω=1.0 (full weight); when 0.85≤comprehensive simulation confidence score<0.95, ω=0.7 (medium weight); when the comprehensive simulation confidence score<0.85, ω=0.3 (low weight), and these samples are marked as samples requiring review.

[0077] The aforementioned weighting coefficients are directly embedded in the loss function of the machine learning model. During training, the model imposes a stronger penalty on the prediction error of high-weight samples, while the error impact of low-weight samples is significantly suppressed. For example, a set of samples whose cross-sectional width is predicted to be 0.05 mm smaller due to mesh coarseness (overall simulation confidence score = 0.78, ω = 0.3) contributes only 30% of the standard samples to the splitting of each tree node in the XGBoost model. This avoids the model misjudging the systematic bias as a real physical law. Finally, the model completes training under the weighted loss function, ensuring that it learns the real structural response law revealed by high-confidence samples, rather than the noise trend caused by simulation error.

[0078] To address the problem that three-dimensional full-size simulation in tire development is time-consuming and expensive, making it difficult to support high-frequency iterations of parametric design, in some embodiments of this application, the simulation model includes a two-dimensional tire model. Based on this, simulation analysis can be performed in the following ways: determining the inner and outer surfaces of the tire corresponding to the two-dimensional tire model; applying radial surface pressure to the inner surface of the tire to simulate the tire inflation process; and extracting the simulation size information of the two-dimensional tire model after inflation based on the outer surface of the tire.

[0079] It should be noted that in a two-dimensional tire model, the inner surface of the tire refers to the inner contour line that contacts the rim and bears the inflation pressure, usually the innermost edge of the tire carcass cord layer; the outer surface of the tire refers to the outermost contour line of the tread that contacts the road surface, representing the external deformation boundary of the tire. These two surfaces define the physical boundaries of the tire's deformation response, and in the simulation, they respectively perform the functions of load application and result extraction. The inner surface is used to apply radial surface pressure to simulate the effect of internal tire pressure, while the outer surface is used to extract the final geometry of the inflated tire.

[0080] Specifically, a standardized two-dimensional tire cross-section template can be pre-constructed. Its geometry is driven by over 20 key parameters, including tread rubber thickness, sidewall rubber thickness, tire carcass cord positioning point coordinates, belt layer angles and widths, and the position of the triangular rubber vertices. When the system receives a set of historical geometric design parameters, it automatically calls the template via script, dynamically adjusting the geometric dimensions and relative positions of each component to generate a complete two-dimensional cross-sectional geometry that conforms to the design intent. This process ensures that all simulation models are completely consistent in topology, with only geometric parameters changing, thus guaranteeing the comparability of simulation results.

[0081] Furthermore, after generating the two-dimensional tire model, the software's geometric analysis function is invoked to automatically identify and mark the inner and outer surfaces based on preset component topological relationships. The inner surface is defined as the innermost boundary of the tire carcass cord layer in contact with the air, and the outer surface is defined as the outermost boundary of the tread rubber in contact with the environment. The system assigns these two geometric contours to independent node sets or edge sets, serving as dedicated areas for subsequent load application and result extraction. For example, a Python script traverses the unit nodes to identify nodes connected to the "carcass" component but without adjacent rubber coverage, forming the inner surface node set; similarly, the outermost nodes of the tread rubber are identified to form the outer surface node set.

[0082] Furthermore, a uniformly distributed radial surface pressure is applied to the inner surface node set. The pressure values ​​are set according to historical working conditions (e.g., 0.8MPa, 1.0MPa, 1.2MPa), and the direction is perpendicular to the inner surface tangent, simulating the tire expansion process under standard air pressure. After the solution is completed, the system automatically calls the post-processing module to read the displacement field of all nodes on the outer surface, extracts the maximum coordinate difference in the vertical direction as the inflation outer diameter, and extracts the maximum span in the horizontal direction as the inflation cross-sectional width. It should be noted that this extraction process is based on a high-precision contour fitting algorithm: the system performs spline interpolation on the original node coordinates to reconstruct a smooth contour curve of the outer surface, and then obtains accurate size values ​​through an extreme value search algorithm, ensuring an accuracy better than ±0.005mm.

[0083] In traditional methods, if a three-dimensional full model is used for inflation analysis, a single simulation takes several hours and the development cycle is measured in weeks. Relying solely on empirical formulas (such as cross-sectional width = mold width × 1.08) cannot cope with the nonlinear deformation behavior brought about by new structures (such as ultra-low aspect ratio and multi-layer belt layers). The embodiments of this application achieve the optimal balance between efficiency and accuracy through a two-dimensional axisymmetric model. While retaining the key mechanical response, the simulation time is reduced from several hours to less than 15 minutes, making the data collection of tens of thousands of samples a reality, thereby supporting the construction of high-precision machine learning models.

[0084] It should be noted that if the axisymmetry assumption of the two-dimensional model fails under asymmetric structures or extreme conditions, it can easily lead to systematic deviations in the simulation dimensional information, contaminating the training data. To address this, an adaptive selection mechanism for the simulation model based on structural symmetry evaluation can be employed. This mechanism comprises three levels:

[0085] ① Input structure symmetry determination module: After receiving design parameters, the system automatically calculates the left-right symmetry index (such as left-right difference / average value) of key parameters such as tire body positioning points, belt layer offset, and triangular rubber mass distribution. If the symmetry is lower than a preset threshold such as 90% (e.g., left-right difference of tire body positioning points > 1.5mm), it is marked as "asymmetric structure"; ② Dynamic simulation strategy switching: For asymmetric structures, the system automatically calls a lightweight three-dimensional axisymmetric extended model (only densifying the mesh in local asymmetric areas) instead of forcing the use of a pure two-dimensional model, ensuring the physical accuracy of key samples; ③ Deviation compensation training mechanism: The prediction difference between the two-dimensional model and the three-dimensional extended model is used as a "correction factor" input into the training set, so that the target model can automatically identify and correct the inherent systematic errors of the two-dimensional model during learning. This mechanism ensures overall efficiency while implementing precise compensation for high-risk structures.

[0086] In some embodiments of this application, the simulated dimensional information includes the simulated inflatable cross-sectional width and the simulated inflatable outer diameter. Based on this, the initial model can be trained in the following ways to obtain the target model: the initial model is trained based on the simulated inflatable cross-sectional width to obtain a first target model for predicting the inflatable cross-sectional width; the initial model is trained based on the simulated inflatable outer diameter to obtain a second target model for predicting the inflatable outer diameter.

[0087] In some embodiments of this application, the following steps may also be performed: obtaining the predicted demand of the tire to be predicted; and selecting at least one of the first target model and the second target model for prediction based on the predicted demand.

[0088] Specifically, an independent supervised training strategy can be adopted, allowing each model to focus on a single physical response, significantly improving prediction accuracy. After training, when design engineers complete the design of a new tire scheme on the platform, the system pops up a prediction panel, allowing users to select the desired prediction target: "Predict only section width," "Predict only outer diameter," or "Predict both simultaneously." If the user only selects "Predict section width," the system only calls the first target model, outputs the S-ISW prediction value, and prompts whether it exceeds the limit (e.g., the standard range 445–452mm). If the user is concerned about vehicle passability, only the second target model is called to obtain the outer diameter to check the wheel arch clearance. If a comprehensive evaluation is required, both models are called in parallel, and the two results are output simultaneously. This mechanism achieves on-demand prediction and optimal resource allocation, avoids invalid calculations, and keeps the prediction response time stable within 0.03 seconds, greatly improving the efficiency of the design closed loop.

[0089] Through steps S202 to S204 above, the geometric design parameters of the tire in its uninflated state are extracted and input into a target model with the ability to learn the inflation deformation law. The target size information after inflation is automatically output, achieving the goal of quickly obtaining accurate inflation size without human experience intervention and physical simulation experiments. This realizes the technical effect of efficient and automated prediction of tire inflation section width, and solves the technical problem that the prediction of tire inflation section width in related technologies relies on the experience judgment of designers and repeated finite element simulation verification, resulting in a long prediction cycle and low accuracy.

[0090] Figure 3 This is a schematic diagram of the overall process of a method for determining tire inflation size according to an embodiment of this application, as shown below. Figure 3 As shown, in some embodiments of this application, the following steps can be performed:

[0091] S302: Obtain historical geometric design parameters.

[0092] The system collects complete sets of structural and operating parameters from previously validated tire design schemes. Through a database interface integrated into the parametric design platform, the system automatically extracts all design variables corresponding to tire design schemes that have undergone finite element simulation or experimental verification in historical projects. These variables cover a total of 220 key input factors, including geometric parameters (such as mold cross-section width, tread rubber thickness, sidewall rubber thickness, triangular rubber dimensions, tire carcass cord positioning point coordinates, belt layer angle and width), material properties (such as rubber modulus, cord linear density and elastic modulus), and operating conditions (such as rim specifications and inflation pressure).

[0093] To ensure data traceability and consistency, the following steps can be performed: First, perform parameter integrity screening to remove abnormal records with missing key variables or inconsistent units; second, implement parameter standardization to normalize all continuous variables to a unified dimension, and convert textual variables (such as tire cord material models) into binary vector representations through one-hot encoding to eliminate semantic ambiguity; finally, perform specification-based clustering based on tire specifications (such as 11R22.5, 12R22.5, etc.), classifying tire parameters of different specifications into independent sub-databases to avoid cross-specification parameter interference and provide a high-quality input sample set with clear structure and consistent semantics for subsequent model training.

[0094] S304: Determine simulation dimension information.

[0095] Based on each set of historical design parameters obtained from S302, the parametric modeling engine built into the simulation platform is automatically invoked to generate a complete and topologically consistent two-dimensional cross-sectional model of the tire. The geometric dimensions and spatial distribution of each component in the model (tread rubber, sidewall rubber, carcass cord layer, belt layer, triangular rubber, and bead) are dynamically driven by the input parameters to ensure that the geometric structure strictly matches the design intent.

[0096] Furthermore, the system automatically generates high-precision axisymmetric finite element meshes and assigns corresponding constitutive models to different components based on a material property library (e.g., the rubber compound uses the Mooney-Rivlin hyperelastic constitutive model, and the cord uses an orthotropic linear elastic model), achieving accurate simulation of the material's nonlinear behavior. In boundary condition settings, the system automatically identifies the inner surface of the tire as the inflation pressure application surface and applies radially uniform pressure to simulate the tire's inflation state on the rim; the outer surface serves as the deformation response monitoring surface.

[0097] After the simulation calculation is completed, the system uses an automated post-processing program to extract the extreme coordinates of the outer surface contour and accurately calculate two key geometric response values: the simulated inflatable cross-section width (maximum width in the horizontal direction) and the simulated inflatable outer diameter (maximum height in the vertical direction). This process does not require 3D modeling, significantly reducing computational costs. The single simulation cycle is controlled within 15 minutes, supporting large-scale parameter sweeping and generating tens of thousands of design-performance corresponding samples to form a physical reality label library for model training.

[0098] S306: Model training.

[0099] Using the simulation dimensional information obtained from S304 as supervision labels, and taking the simulation inflatable cross-sectional width and simulation inflatable outer diameter as independent output targets, two parallel training processes are initiated: First, the first target model is trained with 220-dimensional standardized design parameters as input and the simulation inflatable cross-sectional width as output; second, the second target model is trained with the same input parameters as input and the simulation inflatable outer diameter as output.

[0100] During training, the system uses 80% of the samples for model learning and 20% for cross-validation. It compares the prediction performance of mainstream machine learning algorithms such as random forest, gradient boosting decision tree (XGBoost), support vector regression, and multilayer perceptron. The final architecture of each target model is selected based on the mean absolute error (MAE) and coefficient of determination (R²) as the optimization indicators.

[0101] To improve model robustness, the system implements a simulation confidence weighting strategy before training: low-confidence samples caused by coarse mesh, missing material parameters, or convergence anomalies are automatically assigned lower training weights to suppress their interference with model parameter updates. At the same time, during training, the system outputs the normalized importance weights of each design factor to the target output, forming an interpretable ranking table of influence factors, providing a quantitative basis for design optimization.

[0102] For example, during the model training phase, a list of key influencing factors and their respective percentages for tire inflation section width can be derived. This list provides a reference for product optimization, allowing engineers to adjust design schemes accordingly based on the ranking of influencing factors and design principles. Table 1 shows the design factors and their percentages of influence in the tire inflation section width prediction model.

[0103] Table 1: Percentage of factors influencing tire inflation section width.

[0104]

[0105] As can be seen from Table 1, the design cross-sectional width of the mold has a significant impact on the tire inflation cross-sectional width, accounting for 0.73. This means that adjusting the mold cross-sectional width has a very obvious effect, which is consistent with the results of actual experience. It should be noted that if the mold is already locked, it is difficult to modify some parameters. In this case, engineers can search for design factors downwards from Table 1. For example, adjusting the angle of the second belt layer also has a significant impact on the inflation cross-sectional width. Engineers can also try changing the tire body positioning point or the width of the third belt layer, the thickness of the triangular rubber, and other related parameters to achieve rapid adjustment of the tire inflation cross-sectional width performance.

[0106] S308: Obtain real-time geometric design parameters.

[0107] After engineers complete the geometric modeling of a new tire design in the parametric design platform, the system automatically reads all 220 design variables included in the current design scheme through the platform's built-in real-time parameter synchronization interface. These variables include the latest configurations such as tire body positioning point offset, belt layer angle, rubber modulus, and air pressure settings. This process does not require manual export or input, ensuring that the input data is completely consistent with the current design intent.

[0108] To maintain data consistency with the training phase, the system automatically performs the same preprocessing procedures on real-time parameters as on historical data: unit standardization, one-hot encoding of text variables, and specification matching verification, generating standardized input vectors that are perfectly aligned with the training sample space. This mechanism realizes "design as data," integrating the originally fragmented "design-simulation-verification" process into a continuous closed loop, providing real, complete, and compliant input conditions for subsequent model predictions.

[0109] S310: Model Analysis.

[0110] Based on the prediction target selected by the engineer in the interactive interface (such as "predict only the cross-sectional width", "predict only the outer diameter", or "predict both at the same time"), the system automatically calls the first target model, the second target model, or both in parallel to perform the prediction, with a response time that is stable within 0.03 seconds.

[0111] The prediction results are presented visually, including the predicted value, the design tolerance range (e.g., the standard range of cross-sectional width 445–452 mm), and over-limit warning signs. If the prediction result exceeds the design range, the system automatically activates the interpretability optimization guidance module. Based on the design factor influence weight ranking table output in stage S306, it recommends the adjustable variables that have the most significant impact on the target parameters to the engineer (e.g., "It is recommended to increase the angle of the second belt layer by 2.5°" or "Reduce the offset of tire body positioning point 1 by 0.8 mm"). The system also simulates the trend of the adjusted prediction value in real time, forming a closed-loop feedback mechanism of "prediction-diagnosis-recommendation-repregnancy".

[0112] The above process completely changes the traditional design model that relies on experience and trial and error, enabling engineers to predict performance and optimize accurately in the early stages of the solution, significantly reducing the cost of later mold modifications and physical testing.

[0113] Figure 4 This is a finite element simulation flowchart of a method for determining tire inflation size according to an embodiment of this application, as shown below. Figure 4 As shown, in some embodiments of this application, finite element simulation includes the following steps:

[0114] S402: Parametric design platform generates tire design schemes.

[0115] The parametric design platform is not a general-purpose CAD system. Instead, it is based on the engineering knowledge system of tire structural design and pre-embeds the topological association rules and dimensional ratios between various tire components. These rules include the coupling relationship between the tire carcass cord positioning points and the sidewall rubber thickness, the matching logic between the belt layer width and the tread width, and the support requirements of the triangular rubber geometry for bead stiffness. When a set of design parameters (such as mold cross-section width, tread rubber thickness, tire carcass cord angle, number of belt layers, etc.) are input, the platform automatically and dynamically adjusts the geometric dimensions and spatial coordinates of each component according to the preset design rule library, generating a two-dimensional axisymmetric tire cross-section model that is structurally complete, component-continuous, and free of geometric conflicts.

[0116] S404: The simulation platform generates finite element meshes.

[0117] After receiving the two-dimensional cross-sectional geometry of the tire from the parametric design platform, the simulation platform automatically invokes a high-precision mesh generation algorithm. Based on the component material properties and expected stress distribution, it implements differentiated mesh density control for different regions: in stress concentration areas (such as the contact area between the bead and rim, the end points of the belt layer, and the root of the triangular rubber), the mesh is densified to capture local strain gradients; in uniform deformation areas (such as the center of the tread and the main body of the sidewall), a relatively sparse mesh is used to improve computational efficiency. After mesh generation, the system automatically divides the finite element elements into multiple independent sets (such as the carcass cord layer set, belt layer set, tread rubber set, triangular rubber set, etc.) according to the component topology, and assigns the corresponding constitutive models and physical parameters to each set based on the built-in material property library. For example, the rubber component is assigned a Mooney-Rivlin or Ogden hyperelastic model, with parameters derived from standard test data; the cord component is assigned an orthotropic linear elastic model, considering its axial and circumferential stiffness differences; and the bead wire adopts a linear elastic isotropic model.

[0118] S406: Simulation Analysis.

[0119] Based on a pre-meshed and material-assigned two-dimensional axisymmetric model, the system applies precise physical boundary conditions: First, the inner surface of the tire is identified as the surface under inflation pressure, and uniform radial surface pressure is applied to simulate the tire inflation process on a standard rim. Second, the rim boundary is used as a rigid constraint surface to restrict the radial displacement degrees of freedom of the inner surface nodes, ensuring a realistic reproduction of the tire-rim assembly relationship. Subsequently, the system activates a nonlinear statics solver, considering multiple physical effects such as large material deformation, contact nonlinearity (bead-rim interface slippage), and geometric nonlinearity (cord stretching and rubber compression), iteratively calculating the overall deformation field of the tire under inflation pressure. After the solution is completed, the system automatically extracts the displacement results of the outer surface nodes, reconstructs the tire's outer contour curve using a high-precision contour fitting algorithm, and accurately calculates its maximum horizontal span as the simulated inflation section width and its maximum vertical height as the simulated inflation outer diameter.

[0120] Figure 5 This is a schematic diagram illustrating the prediction accuracy of a method for determining tire inflation size according to an embodiment of this application, as shown in the diagram. Figure 5 As shown, in Figure 5In this model, True Cr Values ​​refers to the actual tire stiffness values, Predicted Cr Values ​​refers to the tire stiffness values ​​predicted by the model, Test Set refers to the dataset used to evaluate model performance, MAE (Mean Absolute Error) refers to the mean absolute error, RMSE (Root Mean Square Error) refers to the root mean square error, R² (Coefficient of Determination) refers to the coefficient of determination, also known as the "proportion of explained variance," Test Points refers to the data points in the test set (i.e., a set of input-output pairs used to evaluate model performance), and Ideal Prediction refers to the prediction under ideal conditions, i.e., a perfect prediction. As an example, the model was validated using 346 sets of data. The closer the data is to the diagonal, the better the accuracy; a smaller MAE is better, and an R² closer to 1 is better. The model data shows that the average prediction error is 0.011 mm.

[0121] Figure 6 This is a structural diagram of a tire inflation size determination device according to an embodiment of this application, as shown below. Figure 6 As shown, the device includes:

[0122] The acquisition module 602 is used to acquire the geometric design parameters of the tire to be predicted, wherein the geometric design parameters include the initial size information of the tire to be predicted in the uninflated state;

[0123] Analysis module 604 is used to analyze geometric design parameters using a target model to obtain target size information of the tire to be predicted. The target model includes a learning model for learning the changes in the initial size information of the tire to be predicted under inflation conditions.

[0124] It should be noted that, Figure 6 The tire inflation size determination device shown is used to perform... Figure 2 The method for determining tire inflation size shown is therefore Figure 2 The explanations and instructions regarding the method for determining tire inflation size also apply to... Figure 6 The device for determining tire inflation size shown will not be described in detail here.

[0125] This application also provides an electronic device, which includes a memory and a processor. The memory is used to store program instructions, and the processor is connected to the memory to execute the steps of the method for determining tire inflation size in various embodiments of this application.

[0126] This application also provides a non-volatile storage medium including a stored computer program, wherein the device containing the non-volatile storage medium executes the steps of the tire inflation size determination method in various embodiments of this application by running the computer program.

[0127] This application also provides a computer program product, including computer instructions that, when executed by a processor, implement the steps of the tire inflation size determination method in various embodiments of this application.

[0128] This application also provides a computer program that, when executed by a processor, implements the steps of the tire inflation size determination method in various embodiments of this application.

[0129] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0130] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0131] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0132] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0133] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0134] If the integrated unit is implemented as 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 this application, in essence, 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. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0135] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for determining tire inflation size, characterized in that, include: Obtain the geometric design parameters of the tire to be predicted, wherein the geometric design parameters include the initial size information of the tire to be predicted in an uninflated state; The geometric design parameters are analyzed using a target model to obtain the target size information of the tire to be predicted. The target model includes a learning model for learning the changes in the initial size information of the tire to be predicted under inflation conditions.

2. The method according to claim 1, characterized in that, The initial size information includes the mold cross-sectional width of the tire to be predicted; The geometric design parameters are analyzed using a target model to obtain the target size information of the tire to be predicted, including: The first target model is used to predict the geometric design parameters to obtain the section width increment of the tire to be predicted. The first target model is trained based on historical geometric design parameters and corresponding historical section widths. The section width increment is used to characterize the degree of deformation of the tire to be predicted as it expands to the left and right sides under the action of internal pressure. The inflation section width of the tire to be predicted is determined based on the mold section width and the section width increment, and the inflation section width is used as the first target size information.

3. The method according to claim 1, characterized in that, The initial size information includes the initial outer diameter of the tire to be predicted; The geometric design parameters are analyzed using a target model to obtain the target size information of the tire to be predicted, including: The second target model is used to predict the geometric design parameters to obtain the inflated outer diameter of the tire to be predicted. The second target model is trained based on historical geometric design parameters and corresponding historical outer diameters. The inflated outer diameter is used to characterize the deformation behavior of the tire to be predicted as it expands to the upper and lower sides under the action of internal pressure. The outer diameter of the inflatable structure is used as the target size information.

4. The method according to claim 1, characterized in that, The target model is trained in the following way: Obtain historical geometric design parameters for multiple tires; Determine the geometric model corresponding to the historical geometric design parameters, wherein the geometric model is used to characterize the parametric geometric profile of the tire cross section; The geometric model is converted into a simulation model, and the simulation model is analyzed to obtain simulation dimension information corresponding to the historical geometric design parameters. The simulation dimension information is used to quantitatively represent the deformation behavior of the tire in the inflated state. The initial model is trained using the historical geometric design parameters and the simulation size information to obtain the target model.

5. The method according to claim 4, characterized in that, The simulation dimensional information includes the simulated inflatable cross-sectional width and the simulated inflatable outer diameter; the initial model is trained using the historical geometric design parameters and the simulation dimensional information to obtain the target model, including: The initial model is trained based on the simulated inflatable cross-sectional width to obtain a first target model for predicting the inflatable cross-sectional width. The initial model is trained based on the simulated inflatable outer diameter to obtain a second target model for predicting the inflatable outer diameter.

6. The method according to claim 5, characterized in that, The method further includes: Obtain the predicted demand for the tire to be predicted; Based on the prediction requirements, at least one of the first target model and the second target model is selected for prediction.

7. The method according to claim 4, characterized in that, The simulation model includes a two-dimensional tire model; simulation analysis is performed on the simulation model to obtain simulation dimensional information corresponding to the historical geometric design parameters, including: Determine the inner and outer surfaces of the tire corresponding to the two-dimensional tire model; Radial surface pressure is applied to the inner surface of the tire to simulate the tire inflation process, and the simulated size information of the two-dimensional tire model after inflation is extracted based on the outer surface of the tire.

8. A device for determining tire inflation size, characterized in that, include: The acquisition module is used to acquire the geometric design parameters of the tire to be predicted, wherein the geometric design parameters include the initial size information of the tire to be predicted in an uninflated state; The analysis module is used to analyze the geometric design parameters using a target model to obtain the target size information of the tire to be predicted. The target model includes a learning model for learning the changes in the initial size information of the tire to be predicted under the inflation state.

9. An electronic device, characterized in that, include: A memory and a processor, wherein the memory is used to store program instructions; the processor is connected to the memory and is used to execute the method for determining tire inflation size according to any one of claims 1 to 7.

10. A non-volatile storage medium, characterized in that, The non-volatile storage medium includes a stored computer program, wherein the device containing the non-volatile storage medium executes the method for determining tire inflation size according to any one of claims 1 to 7 by running the computer program.

11. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the method for determining tire inflation size as described in any one of claims 1 to 7.

Images