Method for predicting static property of tire, prediction device, and method for creating prediction model

Abstract

To provide a method that can predict static properties of a tire in a short time.SOLUTION: The present invention relates to a method for predicting static properties of a tire. The method includes: first step S11 of inputting engineering factors of the tire to be predicted; second step S12 of inputting intermediate properties, of the properties of the tire to be predicted, which are influenced by the engineering factors to influence the static properties; third step S13 of inputting a prediction model that can output the static properties of an arbitrary tire when the engineering factors and the intermediate characteristics of the arbitrary tire are input; and fourth step S14 of outputting the static properties of the tire to be predicted by inputting the engineering factors and the intermediate characteristics of the tire to be predicted to the prediction model.SELECTED DRAWING: Figure 8

Description

[Technical field]

[0001] The present invention relates to a method for predicting static characteristics of a tire, a prediction device, and a method for creating a prediction model. [Background technology]

[0002] In recent years, models capable of outputting prediction results based on predetermined input data have been used to make various predictions (see, for example, Patent Document 1 below). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2023-073083 A Summary of the Invention [Problem to be solved by the invention]

[0004] Generally, when predicting static characteristics of a tire, it is necessary to input a large number of design factors that affect the static characteristics into a model, which results in a problem that prediction takes a long time.

[0005] The present invention has been devised in view of the above circumstances, and has as its main object to provide a method capable of predicting the static characteristics of a tire in a short period of time. [Means for solving the problem]

[0006] The present invention is a method for predicting static characteristics of a tire, comprising: a first step of inputting design factors of a tire to be predicted into a computer; a second step of inputting intermediate characteristics, among the characteristics of the tire to be predicted, which are influenced by the design factors and affect the static characteristics, into the computer; a third step of inputting a prediction model capable of outputting static characteristics of an arbitrary tire by inputting the design factors and intermediate characteristics of the tire into the computer; and a fourth step of inputting the design factors and intermediate characteristics of the tire to be predicted into the prediction model, and outputting the static characteristics of the tire to be predicted. Effect of the Invention

[0007] The method for predicting static characteristics of a tire of the present invention employs the above steps, making it possible to predict the static characteristics of a tire in a short period of time. [Brief description of the drawings]

[0008] [Figure 1] 1 is a block diagram showing a computer (a tire static characteristic prediction device) that executes a tire static characteristic prediction method according to the present embodiment. [Diagram 2] FIG. 1 is a cross-sectional view showing an example of a tire. [Diagram 3] 1A is a cross-sectional view illustrating an example of a plunger destructive test, and FIG. 1B is a cross-sectional view illustrating a state in which a tire is mounted on a testing machine. FIG. 1B is a cross-sectional view illustrating a state in which the tire is pressed by the testing machine. [Figure 4] 13 is a graph showing the relationship between the displacement amount of a pressing portion and a load. [Diagram 5] 11 is a flowchart showing the processing steps of a method for creating a prediction model. [Figure 6] Graphs (a) and (b) conceptually show a prediction model (approximate response surface) based on statistical methods. [Figure 7] FIG. 1 is a conceptual diagram of a machine learning-based predictive model. [Figure 8]4 is a flowchart showing an example of a processing procedure of a method for predicting static characteristics of a tire. [Figure 9] 10 is a flowchart showing an example of a processing procedure of a second step. [Figure 10] FIG. 2 is a perspective view showing an example of a tire model and a road surface model. [Figure 11] FIG. 2 is a cross-sectional view showing an example of a tire model. [Figure 12] 1 is a graph showing the relationship between the actual measured value and the predicted value of the fracture energy of the comparative example. [Figure 13] 1 is a graph showing the relationship between the actual measured value and the predicted value of the fracture energy in the examples. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. It should be understood that the drawings include exaggerated expressions and expressions different from the dimensional ratio of the actual structure in order to help understand the contents of the invention. Furthermore, the same or common elements are given the same reference numerals throughout the embodiments, and duplicated explanations are omitted. Furthermore, the specific configurations shown in the embodiments and drawings are for understanding the contents of the present invention, and the present invention is not limited to the specific configurations shown in the drawings.

[0010] In the tire static characteristic prediction method of this embodiment (hereinafter sometimes simply referred to as the "prediction method"), a computer is used to predict the static characteristics of a tire. In this embodiment, a prediction model is used to predict the static characteristics. FIG. 1 is a block diagram showing a computer 1 (tire static characteristic prediction device 1A) that executes the tire static characteristic prediction method of this embodiment.

[0011] [Prediction device for static tire characteristics] The computer 1 of this embodiment has an input section 2 as an input device, an output section 3 as an output device, and a calculation processing device 4 that calculates static characteristics of a tire, etc., and is configured as a tire static characteristic prediction device (hereinafter sometimes simply referred to as a "prediction device") 1A.

[0012] [Input section, output section, calculation processing unit] The input unit 2 may be, for example, a keyboard or a mouse. The output unit 3 may be, for example, a display device or a printer. The arithmetic processing device 4 includes a calculation unit (CPU) 4A that performs various calculations, a storage unit 4B that stores data, programs, etc., and a working memory 4C.

[0013] [Storage] The storage unit 4B is a non-volatile information storage device formed of, for example, a magnetic disk, an optical disk, an SSD, etc. The storage unit 4B includes a data section 5 and a program section 6.

[0014] [Data section] The data section 5 of this embodiment is for storing data and the like necessary for executing the prediction method. The data section 5 of this embodiment includes a first storage section 5A, a second storage section 5B, a third storage section 5C, and a fourth storage section 5D. Furthermore, the data section 5 includes a fifth storage section 5E, a sixth storage section 5F, a seventh storage section 5G, an eighth storage section 5H, and a ninth storage section 5J. Note that the data section 5 is not limited to such an embodiment, and may include other storage sections and the like as necessary, or some of these storage sections may be omitted.

[0015] The first storage unit 5A is for storing design factors of a tire to be predicted. The second storage unit 5B is for storing intermediate characteristics of the tire to be predicted. The third storage unit 5C is for storing a prediction model capable of outputting static characteristics of any tire.

[0016] The fourth storage unit 5D is for storing design factors of multiple types of tires. The fifth storage unit 5E is for storing intermediate characteristics of multiple types of tires. The sixth storage unit 5F is for storing static characteristics of multiple types of tires. These design factors, intermediate characteristics, and static characteristics of multiple types of tires are used to create a prediction model.

[0017] The seventh storage unit 5G is for storing data necessary for performing a computer simulation based on a tire, which will be described later. The data necessary for the computer simulation includes, for example, information about the road surface on which the tire runs, boundary conditions of the computer simulation, and end conditions of the computer simulation. The eighth storage unit 5H is for storing the results of calculations calculated by the computer simulation. The ninth storage unit 5J is for storing static characteristics of the tire to be predicted.

[0018] The program section 6 is a program (application) required for executing the prediction method. The program section 6 is executed by the calculation section 4A.

[0019] The program section 6 of this embodiment includes a static characteristic output section 6A. Furthermore, the program section 6 of this embodiment includes a first design variable input section 6B, a first intermediate characteristic input section 6C, a prediction model input section 6D, a second design variable input section 6E, a second intermediate characteristic input section 6F, a second static specification input section 6G, a prediction model creation section 6H, and an evaluation section 6J. Note that the program section 6 is not limited to such an embodiment, and may include other programs or some of these may be omitted as necessary.

[0020] The static characteristic output unit 6A is for inputting the design factors and intermediate characteristics of the tire to be predicted into a prediction model and outputting the static characteristics of the tire to be predicted. The first design factor input unit 6B is for inputting the design factors of the tire to be predicted. The first intermediate characteristic input unit 6C is for inputting the intermediate characteristics of the tire to be predicted. The prediction model input unit 6D is for inputting a prediction model capable of outputting the static characteristics of any tire.

[0021] The second design factor input unit 6E is for inputting design factors of multiple types of tires. The second intermediate characteristic input unit 6F is for inputting intermediate characteristics of multiple types of tires. The second static characteristic input unit 6G is for inputting static characteristics of multiple types of tires. The prediction model creation unit 6H is for creating a prediction model capable of outputting dynamic characteristics of a prediction target tire (any) from the design factors and static characteristics of the prediction target tire (any). The evaluation unit 6J is for determining whether the static characteristics of the prediction target tire are good or not. Details of the functions of each program unit 6 will be explained in each step of the simulation method described later.

[0022] [tire] 2 is a cross-sectional view showing an example of a tire 11. The tire 11 of this embodiment is, for example, a pneumatic tire for a passenger vehicle. However, the tire 11 is not limited to this embodiment, and may be, for example, a pneumatic tire for a heavy load or a pneumatic tire for a motorcycle.

[0023] The tire 11 of this embodiment is provided with a carcass 16 extending from the tread portion 12 through the sidewall portion 13 to the bead core 15 of the bead portion 14, and a belt layer 17 and a band layer 18 arranged radially outside the carcass 16 and inside the tread portion 12.

[0024] The carcass 16 of the present embodiment extends across a pair of bead portions 14, 14. An inner liner rubber 23 forming the tire inner cavity surface 11i is disposed inside the carcass 16. The inner liner rubber 23 is made of air-impermeable rubber such as butyl rubber, and keeps the air filled in the tire 11 airtight.

[0025] The carcass 16 of this embodiment is composed of at least one carcass ply 16A, one carcass ply in this embodiment. The carcass ply 16A is provided with carcass cords (not shown) arranged at an angle of, for example, 80 to 90 degrees with respect to the tire equator C. The carcass cords are made of organic fiber cords such as aromatic polyamide and rayon.

[0026] The belt layer 17 of this embodiment is disposed on the outer side of the carcass 16 in the tire radial direction and inside the tread portion 12. The belt layer 17 of this embodiment is composed of two belt plies 17A and 17B on the inner side and the outer side in the tire radial direction.

[0027] In the belt plies 17A and 17B of this embodiment, belt cords (not shown) are arranged at an angle of, for example, 10 to 35 degrees with respect to the tire circumferential direction. These belt plies 17A and 17B are overlapped with the belt cords crossing each other. For example, steel, aramid, rayon, or the like can be suitably used as the belt cords.

[0028] The band layer 18 is disposed on the outer side of the belt layer 17 in the tire radial direction. The band layer 18 in this embodiment is composed of one band ply 18A. The band ply 18A has a band cord (not shown) wound in a spiral shape at an angle of, for example, 10 degrees or less, preferably 5 degrees or less, relative to the tire circumferential direction. For the band cord, an organic fiber cord such as a nylon cord is used.

[0029] The tread portion 12 of this embodiment is provided with a tread rubber 19 arranged radially outward of the band layer 18. The tread rubber 19 includes a base rubber 19A and a cap rubber 19B arranged radially outward of the base rubber 19A. The sidewall portion 13 of this embodiment is provided with a sidewall rubber 20 arranged axially outward of the carcass 16.

[0030] The bead portion 14 of this embodiment is provided with a clinch rubber 21 that comes into contact with the rim 26 when the tire 11 is mounted on the rim 26, and a bead apex rubber 22 that extends radially outward from the bead core 15. An insulation rubber (strip apex rubber) 24 is further provided between the bead apex rubber 22 and the carcass 16.

[0031] [Tire design factors] The tire 11 is manufactured based on predetermined design factors. The design factors in this embodiment include the internal pressure (specified air pressure) of the tire 11 and the load (allowable load).

[0032] The internal pressure in this embodiment is specified as a normal internal pressure, but is not particularly limited thereto. The "normal internal pressure" is the air pressure that is determined for each tire by each standard in a standard system including the standard on which the tire 11 is based. Therefore, the normal internal pressure is, for example, the "maximum air pressure" in the case of JATMA, the maximum value listed in the table "TIRE LOAD LIMITS AT VARIOUS COLD INFLATION PRESSURES" in the case of TRA, and the "INFLATION PRESSURE" in the case of ETRTO.

[0033] The load in this embodiment is specified as a normal load, but is not particularly limited to this. The "normal load" is a load that is determined for each tire by each standard in a standard system including the standard on which the tire 11 is based. Therefore, the normal load is, for example, "maximum load capacity" in the case of JATMA, the maximum value listed in the table "TIRE LOAD LIMITS AT VARIOUS COLD INFLATION PRESSURES" in the case of TRA, and "LOAD CAPACITY" in the case of ETRTO.

[0034] In this specification, unless otherwise specified, the dimensions of each part of the tire are indicated as values ​​measured in a normal state. The normal state is a state in which the tire 11 is mounted on a normal rim (rim 26), inflated to the normal internal pressure, and no load is applied. Note that the dimensions of each part of the tire are allowed for normal errors that are inherent in rubber molded products.

[0035] A "genuine rim" is a rim that is determined for each tire by a standard system that includes the standard on which the tire 11 is based. Therefore, a genuine rim is, for example, a "standard rim" for JATMA, a "design rim" for TRA, and a "measuring rim" for ETRTO.

[0036] The design factors of this embodiment include the rim width W1. The rim width W1 is specified as the distance between the inner ends of the pair of rim flange portions 26B, 26B in the tire axial direction. If the rim width W1 is defined in the standard on which the tire 11 is based, for example, the defined value can be used.

[0037] The design factors of this embodiment include the tire size, which can specify the tire section width W2, the tire section height H1, and the aspect ratio (not shown).

[0038] The tire section width W2 may be specified as the maximum axial width of the tire 11 excluding a rim guard (not shown). The tire section height H1 may be specified as the radial distance from a bead base line BL to the radially outermost position of the tread portion 12. The bead base line BL is specified as an axial line passing through a bead diameter position defined by the standard to which the tire 11 is based.

[0039] The aspect ratio (not shown) is the tire section height H1 divided by the tire section width W2, expressed as a percentage. When the tire section width W2, tire section height H1, and aspect ratio are defined by the standard on which the tire 11 is based, for example, the defined values ​​can be adopted.

[0040] The design factors of this embodiment include a case structure, which is specified as the number of carcass plies 16A that constitute the carcass (case) 16.

[0041] The design factors of this embodiment include case strength. The case strength in this embodiment is specified as the breaking strength of the carcass cord (not shown) of the carcass ply 16A. This breaking strength is determined by performing a tensile load test at room temperature (25°C ± 2°C) on the carcass cord sampled from a new tire 11 in accordance with JIS L1017 "Test method for synthetic fiber tire cords," and determining the load at which the carcass cord breaks as the breaking strength.

[0042] The design factors of this embodiment include a case cross-sectional area (not shown). The case cross-sectional area of ​​this embodiment is specified as the cross-sectional area of ​​the carcass ply 16A specified in a meridian section of the tire 11. When a plurality of carcass plies 16A are included, the case cross-sectional area is specified as the total value of the cross-sectional areas of these carcass plies 16A.

[0043] The design factors of this embodiment include a bead apex height H2. The bead apex height H2 of this embodiment is specified as the distance (height) in the tire radial direction from the bead base line BL to the outermost end of the bead apex rubber 22 in the tire radial direction.

[0044] The design factors of this embodiment include the complex modulus E* of the bead apex rubber 22, the complex modulus E* of the cap rubber 19B, the complex modulus E* of the base rubber 19A, the complex modulus E* of the sidewall rubber 20, and the complex modulus E* of the clinch rubber 21. These complex moduli E* are measured in accordance with the provisions of JIS K6394 using a dynamic viscoelasticity measuring device (IPLEXER series) manufactured by GABO under the following measurement conditions. Frequency: 10Hz Initial strain: 5% Dynamic distortion: ±1% Temperature: 30℃ Deformation mode: tension

[0045] The design factors of this embodiment include the number of insulation rubbers 24 and the insulation rubber height H3. The insulation rubber height H3 is specified as the distance (height) in the tire radial direction from the inner end to the outer end of the insulation rubber 24 in the tire radial direction.

[0046] As shown in Fig. 2, the design factors of this embodiment include a first gauge G1, a second gauge G2, a third gauge G3, and a fourth gauge G4. The first gauge G1 is the maximum thickness of the tread rubber 19 at the tire equator C. The second gauge G2 is the maximum rubber thickness of the sidewall rubber 20 at the boundary P2 between the tread rubber 19 and the sidewall rubber 20. The third gauge G3 is the thickness of the sidewall rubber 20 at the tire maximum width position P3. The tire maximum width position P3 is the position where the carcass 16 is the outermost in the tire axial direction. The fourth gauge G4 is the maximum width of the clinch rubber 21.

[0047] The above design factors are examples of design factors of the tire 11, and other design factors may be included.

[0048] [rim] The rim 26 of this embodiment is configured as the above-mentioned regular rim, but is not particularly limited to this. The rim 26 is configured to include a pair of rim seat portions 26A, 26A, a pair of rim flange portions 26B, 26B, and a well portion (not shown) in the same manner as in the conventional rim. The pair of rim seat portions 26A, 26A and the pair of rim flange portions 26B, 26B each have a contact surface 26s, 26s with the tire 11 (bead portion 14, 14).

[0049] The rim 26 of this embodiment is integrally formed with a disk portion 27 (shown in FIG. 3(a)) that extends radially inward from the rim 26. The disk portion 27 is fixed to an axle (tire rotation shaft) (not shown).

[0050] [Static characteristics] The static characteristics indicate the characteristics of the tire 11 when stationary. An example of the static characteristics is fracture energy (hereinafter, sometimes referred to as "fracture energy") obtained by a plunger fracture test. In the present embodiment, the fracture energy is predicted as the static characteristics, but is not limited to such a mode.

[0051] [Plunger Destruction Test] The plunger breaking test is appropriately performed in accordance with, for example, Section 6.1 "Tire strength (breaking energy) test" of JIS D4230. A horizontal plunger tester (hereinafter sometimes referred to as "tester") is used for the plunger breaking test of this embodiment.

[0052] Fig. 3 is a cross-sectional view for explaining an example of a plunger destructive test. Fig. 3(a) shows a state in which a tire 11 is attached to a testing machine 40. Fig. 3(b) shows a state in which the tire 11 is pressed by the testing machine 40.

[0053] The testing machine 40 of this embodiment includes a frame 41, a holding portion 42, a pressing portion 43, and a guide portion 44.

[0054] The frame 41 includes a pair of frame members 41A, 41A. In this embodiment, the pair of frame members 41A, 41A extend upward from a floor surface 45. These frame members 41A, 41A are spaced apart from each other in the horizontal direction.

[0055] The holding portion 42 is for holding the tire 11 to be evaluated. The holding portion 42 in this embodiment is configured to include a pair of arms 42A, 42A for fixing the rim 26 to which the tire 11 is mounted. The rim 26 is not particularly limited as long as it can be fixed by the pair of arms 42A, 42A, and may be the regular rim described above.

[0056] The pair of arms 42A, 42A extend in the horizontal direction. One ends of the pair of arms 42A, 42A are fixed to the pair of frame members 41A, 41A, respectively. The other ends of the pair of arms 42A, 42A are fixed to the rim 26 (disk portion 27), respectively. In this way, the tire 11 can be supported by the frame 41 via the rim 26 and the pair of arms 42A, 42A. At this time, the rotation axis of the tire 11 extends along the horizontal direction.

[0057] The pressing portion 43 is for pressing the tire 11. The pressing portion 43 of the present embodiment includes a main body portion 43A and a contact portion 43B.

[0058] The main body 43A of this embodiment is formed in a generally rectangular block shape when viewed from the front, and is supported by a pair of nuts 44C, 44C of the guide 44, which will be described later. The contact portion 43B of this embodiment is fixed to the upper surface of the main body 43A. This contact portion 43B may be removably fixed to the main body 43A, or may be formed integrally with the main body 43A. An inclined surface 43s inclined with respect to the horizontal direction is formed on the upper portion of the contact portion 43B of this embodiment. The tread portion 12 of the tire 11 is brought into contact with this inclined surface 43s, and the tire 11 is pressed.

[0059] The guide portion 44 is for vertically moving the pressing portion 43. The guide portion 44 in this embodiment is configured as a known ball screw mechanism, and includes a pair of screw shafts 44A, 44A, a pair of drive portions 44B, 44B, and a pair of nut portions 44C, 44C.

[0060] The pair of screw shafts 44A, 44A in this embodiment extend in the vertical direction and are held by the pair of frame members 41A, 41A so as to be rotatable around a vertical axis. The pair of nut portions 44C, 44C in this embodiment are screwed onto the pair of screw shafts 44A, 44A, respectively. These nut portions 44C, 44C are fixed to the main body portion 43A of the pressing portion 43.

[0061] The pair of drive parts 44B, 44B are for rotating the pair of screw shafts 44A, 44A around a vertical axis. By rotating the pair of screw shafts 44A, 44A around the vertical axis, the pair of nut parts 44C, 44C can be moved in the up and down direction.

[0062] Next, the procedure of the plunge destructive test (hereinafter sometimes referred to as "the test") of this embodiment will be described. In the test of this embodiment, first, as shown in FIG. 3(a), a tire 11 is attached to a testing machine 40. In this embodiment, the tire 11 to be evaluated is mounted on a rim 26, and the tire is inflated to internal pressure (normal internal pressure). Then, the rim-mounted tire 11 is fixed to a pair of arms 42A, 42A of a holding portion 42. At this time, the tire 11 is positioned above the pressing portion 43.

[0063] Next, in the test of this embodiment, the pressing portion 43 is moved upward by the guide portion 44. As a result, the tire 11 is pressed by the abutment portion 43B.

[0064] In this embodiment, the inclined surface 43s of the contact portion 43B presses the tread portion 12 of the tire 11, causing the sidewall portion 13 on one side in the tire axial direction (the right side in FIG. 3(b)) to deform relatively greatly. This can cause a pinch cut due to breakage of the carcass cord (not shown).

[0065] The testing machine 40 of this embodiment measures the vertical displacement (movement distance) of the pressing part 43 and the load (pressing force) acting on the pressing part 43. Fig. 4 is a graph showing the relationship between the displacement [m] of the pressing part and the load [N].

[0066] In this embodiment, in the graph shown in Fig. 4, the displacement P of the pressing portion 43 from when the inclined surface 43s of the pressing portion 43 abuts against the tread portion 12 until a pinch cut occurs in the tire 11 and the load F acting on the pressing portion 43 are specified. Then, the fracture energy E [N m] is calculated using the following formula. E=P×F÷2

[0067] In the test of this embodiment, it is possible to evaluate the durability against pinch cut by obtaining the above-mentioned fracture energy E. A larger fracture energy E indicates a higher durability against pinch cut.

[0068] In this embodiment, a horizontal plunger test machine (test machine 40 shown in FIG. 3) is used, which, unlike a conventional plunger test machine (not shown), can relatively greatly deform the sidewall portion 13 on one side in the tire axial direction (the right side in FIG. 3). This allows the durability against pinch cuts that occur when the tire 11 runs over a curb or the like to be evaluated with high accuracy.

[0069] Incidentally, in order to carry out a plunger destructive test on a tire 11 at the design stage, it is necessary to manufacture (prototype) the tire 11 at that design stage. For this reason, the plunger destructive test requires a lot of time and cost. It is also possible to use a computer simulation based on the finite element method to simulate the plunger destructive test without manufacturing the tire 11 at the design stage. However, the finite element method has difficulty in calculating large deformations and destruction of the tire 11 that occur in the plunger destructive test, and therefore it is difficult to accurately calculate the destruction energy.

[0070] In the prediction method of this embodiment, by using a prediction model described later, the static characteristics (fracture energy) of the tire 11A to be predicted (shown in FIG. 2) can be output without having to prototype the tire 11A to be predicted and perform a plunger destructive test, as in the conventional method. Furthermore, in the prediction method of this embodiment, there is no need to simulate a plunger destructive test using a computer simulation based on the finite element method. This makes it possible to predict the static characteristics of the tire 11A to be predicted in a short time. Details of such a prediction model and a method of predicting static characteristics using the prediction model will be described later.

[0071] [How to create a forecast model] In this embodiment, before the static characteristics of the tire 11A to be predicted are predicted (the prediction method is implemented), a prediction model used for predicting the static characteristics is created. Note that, when the prediction method is implemented, if the prediction model has already been created, the implementation of a method for creating a prediction model (hereinafter, sometimes referred to as the "creation method") described below may be omitted.

[0072] In the creation method of this embodiment, a prediction model capable of outputting static characteristics of a tire to be predicted is created using training data including design factors of each of a plurality of types of tires 11 (shown in FIG. 2). In general, in order to create a prediction model capable of predicting the static characteristics of the tire 11, training data including a large number of design factors (explanatory variables) that affect the static characteristics is required. Examples of the large number of design factors include those mentioned above. If the number of such explanatory variables increases, a problem occurs in that a large amount of training data is required to create a prediction model.

[0073] Furthermore, when there is a large amount of training data, not only does it take a lot of time to create a predictive model, but there is also the problem of overfitting, which means that the model deviates significantly from the trends originally suggested by the training data.

[0074] As a result of extensive research, the inventors have found that it is possible to reduce the number of design factors required for predicting the static characteristics by using the intermediate characteristics of the tire 11. The intermediate characteristics are characteristics that are influenced by the design factors of the tire 11 and that affect the static characteristics of the tire 11, among various characteristics that the tire 11 has.

[0075] The intermediate characteristic is not particularly limited as long as it has the above-mentioned characteristics (i.e., characteristics that are influenced by the design factors of the tire 11 and affect the static characteristics of the tire 11). The intermediate characteristic of the present embodiment preferably includes at least one of a vertical spring constant and a lateral spring constant. Such a vertical spring constant and a lateral spring constant are influenced (closely related) by multiple design factors among the above-mentioned design factors and can affect the static characteristics of the tire 11.

[0076] As described above, in this embodiment, the fracture energy obtained by the plunger fracture test (testing machine 40 shown in FIG. 3) is predicted as the static characteristic. In this case, it is preferable to adopt the vertical spring constant, which mainly affects the fracture energy, as the intermediate characteristic out of the vertical spring constant and the horizontal spring constant.

[0077] In the creation method of this embodiment, the number of explanatory variables (design factors) can be reduced by including intermediate characteristics in the teacher data (explanatory variables) used to create a prediction model. This reduces the amount of teacher data required to create a prediction model, shortening the time required to create the prediction model and suppressing overlearning. Figure 5 is a flowchart showing the processing steps of the prediction model creation method.

[0078] [Enter design factors for multiple types of tires] In the production method of this embodiment, the design factors of each of a plurality of types of tires 11 (shown in FIG. 2) are inputted into a computer 1 (shown in FIG. 1) (step S1).

[0079] In step S1 of this embodiment, first, the second design factor input section 6E included in the program section 6 shown in Fig. 1 is loaded into the working memory 4C. The second design factor input section 6E is a program for inputting the design factors of each of a plurality of types of tires 11 (shown in Fig. 2). The second design factor input section 6E is executed by the calculation section 4A, whereby the computer 1 can function as a means for inputting the design factors of each of a plurality of types of tires.

[0080] In step S1 of this embodiment, the design factors are identified for multiple types of tires 11 (shown in FIG. 2) that differ in at least one of the above design factors. For example, design data (including CAD data) of each tire 11 is used to identify the design factors. This makes it possible to easily identify the design factors. Note that measurement results of the tires 11 may also be used to identify the design factors. These design factors are used as explanatory variables of the teacher data of the prediction model. In addition, from the viewpoint of creating a prediction model that can accurately predict static characteristics, it is preferable that the multiple types of tires 11 are in the same category as the tire to be predicted (for example, pneumatic tires for passenger cars). In addition, from the viewpoint of creating a prediction model with high prediction accuracy in a short time, it is preferable that the number (number of types) of tires 11 for which design factors are identified is, for example, 10 to 100.

[0081] As described above, in the creation method of this embodiment, the explanatory variables of the teacher data include the intermediate characteristics of the tire 11, and design factors closely related to the intermediate characteristics (which can be replaced by the intermediate characteristics) are omitted, thereby reducing the number of design factors used as explanatory variables. When the vertical spring constant is used as the intermediate characteristic as in this embodiment, the design factors which can be replaced by the vertical spring constant (hereinafter sometimes referred to as "alternative design factors") include, for example, the rim width W1, tire section width W2, and aspect ratio (not shown) shown in FIG. 2.

[0082] Further, the alternative design factors include a case cross-sectional area (not shown) and a bead apex height H2. Further, the alternative design factors include a complex modulus E* of the bead apex rubber 22, a complex modulus E* of the cap rubber 19B, a complex modulus E* of the base rubber 19A, a complex modulus E* of the sidewall rubber 20, and a complex modulus E* of the clinch rubber 21. Further, the alternative design factors include the number of insulation rubbers 24, an insulation rubber height H3, a first gauge G1, a second gauge G2, a third gauge G3, and a fourth gauge G4.

[0083] Among the above design factors, the tire cross-sectional height H1, the case structure (the number of carcass plies 16A), and the case strength can also affect the vertical spring constant. However, these design factors have a greater impact (contribution) on the predicted static characteristics (fracture energy) than the above alternative design factors. For this reason, in order to create a prediction model with high prediction accuracy, it is preferable to include these design factors in the explanatory variables of the training data.

[0084] In step S1 of this embodiment, the tire section height H1, the case structure (the number of carcass plies 16A), and the case strength are specified as design factors of the multiple types of tires 11. These design factors are stored in the fourth storage unit 5D (computer 1) by the second design factor input unit 6E shown in FIG.

[0085] [Enter intermediate characteristics of multiple types of tires] Next, in the production method of this embodiment, the intermediate characteristics of each of a plurality of types of tires 11 (shown in FIG. 2) are inputted into a computer (shown in FIG. 1) (step S2).

[0086] In step S2 of this embodiment, first, the second intermediate characteristic input section 6F included in the program section 6 shown in Fig. 1 is loaded into the working memory 4C. The second intermediate characteristic input section 6F is a program for inputting intermediate characteristics of multiple types of tires. The second intermediate characteristic input section 6F is executed by the calculation section 4A, so that the computer 1 can function as a means for inputting the respective intermediate characteristics of multiple types of tires.

[0087] As described above, the intermediate characteristic of this embodiment employs a vertical spring constant. This vertical spring constant can be obtained, as in the past, by using a known static testing machine (not shown) as the ratio of vertical load / lateral deflection when a vertical load is applied to the tire 11 in a normal state. The magnitude of the vertical load can be set appropriately according to the category of the tire 11, for example. Note that, when a lateral spring constant is included in the intermediate characteristic, the lateral spring constant can be obtained by using a conventional static testing machine as the ratio of lateral load / lateral deflection when a lateral load is applied to the tire 11 in a normal state.

[0088] In step S2 of this embodiment, a known static testing machine (not shown) is used to measure the vertical load and lateral deflection of each of the multiple types of tires 11. Then, the second intermediate characteristic input unit 6F shown in Fig. 1 calculates the ratio of the vertical load / lateral deflection based on the measurement results, thereby obtaining the vertical spring constants. These vertical spring constants are stored in the fifth storage unit 5E (computer 1) shown in Fig. 1 as intermediate characteristics of each of the multiple types of tires 11, for example.

[0089] [Enter static properties of multiple types of tires] Next, in the production method of this embodiment, the static characteristics of each of a plurality of types of tires 11 (shown in FIG. 2) are inputted into a computer (shown in FIG. 1) (step S3).

[0090] In step S3 of this embodiment, first, the second static characteristic input section 6G included in the program section 6 shown in Fig. 1 is loaded into the working memory 4C. The second static characteristic input section 6G is a program for inputting static characteristics of multiple types of tires 11 (shown in Fig. 2). The second intermediate characteristic input section 6F is executed by the calculation section 4A, whereby the computer 1 can function as a means for inputting the respective static characteristics of the multiple types of tires 11.

[0091] As described above, in this embodiment, the above-mentioned fracture energy is adopted as the static characteristic. Therefore, in step S3 of this embodiment, first, a plunger fracture test is performed on each of the multiple types of tires 11 using the testing machine 40 shown in FIG. 3. In this plunger fracture test, as shown in FIG. 4, the displacement amount (movement distance) P of the pressing part 43 until the pinch cut occurs and the load (pressing force) F acting on the pressing part 43 are measured for each of the multiple types of tires 11. Then, based on these measured values ​​and the above formula, the second static specification input unit 6G shown in FIG. 1 calculates the fracture energy E of each of the multiple types of tires 11. These fracture energies E are stored in the sixth storage unit 5F (computer 1) as static characteristics of the multiple types of tires 11.

[0092] [Create a forecast model] Next, in the creation method of this embodiment, the computer 1 (shown in FIG. 1) creates a prediction model (step S4). In step S4 of this embodiment, a prediction model capable of outputting the static characteristics of the tire 11A to be predicted from the design factors and intermediate characteristics of the tire 11A to be predicted is created using the design factors, intermediate characteristics, and static characteristics of multiple types of tires 11 as teacher data.

[0093] 1, in step S4 of this embodiment, first, the design factors of the multiple types of tires 11 inputted in the fourth storage unit 5D and the intermediate characteristics of the multiple types of tires 11 inputted in the fifth storage unit 5E are read into the working memory 4C. Next, the static characteristics of the multiple types of tires 11 inputted in the sixth storage unit 5F are read into the working memory 4C.

[0094] Next, a prediction model creation unit 6H included in the program unit 6 is loaded into the working memory 4C. The prediction model creation unit 6H is a program for creating a prediction model. The prediction model creation unit 6H is executed by the calculation unit 4A, thereby allowing the computer 1 to function as a means for creating a prediction model.

[0095] The prediction model can be created as appropriate as long as it can output the static characteristics of the tire 11A to be predicted from the design factors and intermediate characteristics of the tire 11A to be predicted. The prediction model of the present embodiment is created based on a statistical method or machine learning.

[0096] First, a prediction model based on a statistical method will be described. Examples of the statistical method include the Kriging method and Bayesian linear regression. The prediction model based on the statistical method is created as an approximate response surface (approximate response function).

[0097] Fig. 6 is a graph conceptually showing a prediction model (approximate response surface) 31 based on a statistical method. In Fig. 6(a), the case structure and vertical spring constant are shown in the foreground. In Fig. 6(b), the case strength and tire section height, which are design factors, are shown in the foreground. Such a prediction model (approximate response surface) 31 can accurately express the relationship between the input (in this example, the design factors and intermediate characteristics of the tire 11) and the output (in this example, the static characteristics of the tire 11) even if there is a strong nonlinear relationship.

[0098] In this embodiment, for example, teacher data (design factors, intermediate characteristics, and static characteristics of multiple types of tires 11) are substituted into commercially available computer software (for example, MATLAB ("MATLAB" is a registered trademark) made by The MathWorks, Inc., modeFRONTIER made by ESTECO, etc.). This creates a prediction model 31 in which the design factors and intermediate characteristics of the tire 11 are explanatory variables and the static characteristics of the tire 11 are objective variables.

[0099] In such a prediction model 31, for example, design factors (in this example, tire section height, case structure, and case strength) and intermediate characteristics (in this example, vertical spring constant) of any tire 11 can be input as explanatory variables. In this way, the prediction model 31 can obtain the static characteristics (in this example, fracture energy) of any tire 11 by complementing them with the design factors, intermediate characteristics, and static characteristics of multiple known types of tires 11. The obtained static characteristics of any tire 11 can then be predicted (output) as objective variables.

[0100] Next, a prediction model based on machine learning will be described. The prediction model based on machine learning can be generated by, for example, deep learning using artificial intelligence (AI). FIG. 7 is a conceptual diagram of a prediction model 31 based on machine learning.

[0101] The machine learning-based prediction model 31 is defined by an input layer 32, an output layer 33, and an intermediate layer (hidden layer) .

[0102] The input layer 32 of this embodiment includes a plurality of nodes 35. The design factors and intermediate characteristics of the tire 11 shown in FIG. 2 can be input as explanatory variables to these nodes 35. As described above, the design factors of this embodiment include the tire section height H1, the case structure (the number of carcass plies 16A), and the case strength. In addition, the intermediate characteristics of this embodiment include the vertical spring constant.

[0103] The output layer 33 of this embodiment is composed of one node 35. From this node 35, the static characteristic of the tire 11 can be output as the objective variable. The static characteristic in this embodiment is the above-mentioned fracture energy.

[0104] The intermediate layer 34 of this embodiment includes a combination of a plurality of nodes (neurons) 35 arranged in a multi-stage hierarchy and optimized weighting coefficients 36 (parameters). Each node 35 is connected by the weighting coefficients 36. Such an intermediate layer 34 is called a neural network. That is, the prediction model 31 of this embodiment includes a neural network.

[0105] The prediction model 31 of this embodiment is trained by adjusting the weighting coefficient 36 of each node 35 so as to reduce the difference between the output for the input and the true output (teaching data). Such a training method is called backpropagation. Each node 35 and each weighting coefficient 36 are determined.

[0106] In this embodiment, for a plurality of types of tires 11, the weighting coefficients 36 are adjusted so as to reduce the difference between the predicted results of the static characteristics output from the output layer 33 in response to input of the design factors and intermediate characteristics to the input layer 32 and the actual static characteristics. In this way, a prediction model 31 is created in which the design factors and intermediate characteristics of the tire 11 are explanatory variables and the static characteristics of the tire 11 are objective variables.

[0107] In the prediction model 31, for example, design factors and intermediate characteristics of any tire 11 are input as explanatory variables to an input layer 32. Based on these explanatory variables, static characteristics of any tire 11 predicted via an intermediate layer (neural network) 34 can be output as objective variables from an output layer 33. Machine learning software (for example, modeFRONTIER manufactured by ESTECO) is used to create the prediction model 31 based on machine learning.

[0108] In this embodiment, the explanatory variables of the prediction model 31 include intermediate characteristics, so that the number of design factors (explanatory variables) can be reduced. As a result, in this embodiment, it is possible to reduce the amount of training data required to create a prediction model 31 with high prediction accuracy. In addition, by reducing the amount of training data, it is possible to shorten the time required to create the prediction model 31. Furthermore, by reducing the amount of training data, it is possible to suppress the occurrence of overlearning, which is the case when the training data deviates significantly from the trends originally suggested by the training data, and it is possible to predict the static characteristics of the tire 11 with high accuracy.

[0109] In step S4 of the present embodiment, a prediction model 31 (shown in FIG. 6) is created based on a statistical method (Kriging method), but a prediction model 31 (shown in FIG. 7) may be created based on machine learning. The prediction model 31 is stored in the third storage unit 5C (computer 1) shown in FIG. 1.

[0110] [Method for predicting static characteristics of a tire (first embodiment)] Next, a method for predicting static characteristics of a tire according to the present embodiment will be described with reference to a flowchart shown in FIG.

[0111] [Enter the design parameters of the tire to be predicted (first step)] In the prediction method of this embodiment, first, the design factors of the tire 11A to be predicted, shown in FIG. 2, are input to the computer 1 (shown in FIG. 1) (first step S11).

[0112] In the first step S11 of this embodiment, first, a first design factor input section 6B included in the program section 6 shown in Fig. 1 is loaded into the working memory 4C. The first design factor input section 6B is a program for inputting design factors of a tire 11A (shown in Fig. 2) to be predicted. The first design factor input section 6B is executed by the calculation section 4A, whereby the computer 1 can function as a means for inputting design factors of the tire 11A to be predicted.

[0113] The tire 11A to be predicted is not limited to a tire that is actually manufactured, and may be a tire in the design stage (unmanufactured) as long as the static characteristics can be predicted.

[0114] In the first step S11 of this embodiment, among the multiple design factors of the tire 11A to be predicted, the same design factors as those of the multiple types of tires 11 input in step S1 of the production method shown in FIG. 5 are specified. That is, in this embodiment, the tire section height H1, the case structure (the number of carcass plies 16A), and the case strength are specified as the design factors of the tire 11A to be predicted. For example, design data (including CAD data) of the tire 11A to be predicted is used to specify these design factors. This makes it possible to easily specify the design factors for the tire 11A to be predicted that is in the design stage. Note that, when the tire 11A to be predicted actually exists (has been manufactured), the design factors may be specified using the measurement results of the tire 11A to be predicted. The specified design factors are input to the first storage unit 5A (computer 1) by the first design factor input unit 6B shown in FIG. 1.

[0115] [Enter intermediate characteristics of the tire to be predicted (2nd step)] Next, in the prediction method of this embodiment, among the characteristics of the tire 11A to be predicted shown in FIG. 2, intermediate characteristics that are influenced by design factors and affect static characteristics are input to the computer 1 (shown in FIG. 1) (second step S12).

[0116] In the second step S12 of this embodiment, first, a first intermediate characteristic input section 6C included in the program section 6 shown in Fig. 1 is loaded into the working memory 4C. The first intermediate characteristic input section 6C is for inputting intermediate characteristics of the tire 11A (shown in Fig. 2) to be predicted. The first intermediate characteristic input section 6C is executed by the calculation section 4A, whereby the computer 1 can function as a means for inputting intermediate characteristics of the tire 11A to be predicted.

[0117] As described above, the vertical spring constant is used for the intermediate characteristic of this embodiment. When the tire 11A to be predicted actually exists (has been manufactured), the vertical load and the lateral deflection when the vertical load is applied to the tire 11A to be predicted, which is in a normal state, are measured using a known static testing machine (not shown) as in the conventional case. The magnitude of the vertical load can be set appropriately according to the category of the tire 11, for example. Then, the first intermediate characteristic input unit 6C shown in FIG. 1 can calculate the ratio of the vertical load / lateral deflection to obtain the vertical spring constant. The intermediate characteristic is stored in the second storage unit 5B (computer 1).

[0118] On the other hand, when the tire 11A (shown in FIG. 2) to be predicted is in the design stage (unmanufactured), the intermediate characteristics cannot be directly acquired. For this reason, the second step S12 of this embodiment includes a step in which the first intermediate characteristic input unit 6C shown in FIG. 1 executes a computer simulation based on the tire 11A to be predicted, and acquires the intermediate characteristics of the tire 11A to be predicted. FIG. 9 is a flowchart showing an example of the processing procedure of the second step S12.

[0119] [Enter tire model] In the second step S12 of this embodiment, first, a tire model in which a tire 11A (shown in FIG. 2) to be predicted is modeled is input to a computer 1 (shown in FIG. 1) (step S21). In step S21 of this embodiment, the tire model is modeled based on the design factors of the tire 11A to be predicted. Such modeling is performed, for example, according to the procedure described in a patent document (JP Patent Publication No. 2017-126272).

[0120] Fig. 10 is a perspective view showing an example of a tire model 28 and a road surface model 29. Fig. 11 is a cross-sectional view showing an example of a tire model 28. Note that a tread pattern, an element F(i) shown in Fig. 11, and the like are omitted in Fig. 10.

[0121] In step S21 of this embodiment, the tire 11A (shown in FIG. 2) to be predicted is discretized (modeled) with a finite number of elements F(i) (i=1, 2, ...) as shown in Fig. 11. As a result, in step S21, a tire model 28 is set.

[0122] The element F(i) can be handled by a numerical analysis method. For example, the finite element method, the finite volume method, the difference method, or the boundary element method (in this embodiment, the finite element method) can be appropriately adopted as the numerical analysis method. For example, a three-dimensional tetrahedral solid element, a pentahedral solid element, or a hexahedral solid element is used as the element F(i).

[0123] Each element F(i) includes a plurality of nodes 37. Each element F(i) is defined with numerical data such as an element number, a node 37 number, coordinate values ​​of the node 37, and material properties (e.g., density, Young's modulus, damping coefficient, thermal conductivity, and heat transfer coefficient). The tire model 28 is input to the eighth storage unit 5H (computer 1) shown in FIG. 1.

[0124] [Enter road surface model] Next, in a second step S12 of this embodiment, a road surface model 29 (shown in FIG. 10) that models the road surface is input to the computer 1 (shown in FIG. 1) (step S22). In step S22 of this embodiment, the road surface model 29 is modeled based on information (e.g., contour data, etc.) about the road surface input to the seventh storage unit 5G (shown in FIG. 1). The modeling of the road surface model 29 is performed based on the same procedure as in the patent document (JP Patent Publication No. 2017-126272).

[0125] In step S22 of this embodiment, based on information about the road surface (not shown), the road surface is discretized using a finite number of elements G(i) (i=1, 2, ...) that can be handled by a numerical analysis method (in this embodiment, the finite element method). As a result, in step S22, a road surface model 29 that models the road surface is set. The elements G(i) are defined as rigid plane elements that are defined to be undeformable. The elements G(i) are provided with a plurality of nodes 38. Furthermore, for the elements G(i), numerical data such as element numbers and coordinate values ​​of the nodes 38 are defined. The road surface model 29 is input to the eighth storage unit 5H (computer 1) shown in FIG. 1.

[0126] [Define Boundary Conditions] Next, in the second step S12 of this embodiment, boundary conditions for the simulation are defined in the computer 1 (shown in FIG. 1) (step S23). The boundary conditions include, for example, the internal pressure condition of the tire model 28 shown in FIGS. 10 and 11, the vertical load L, the camber angle, the slip angle, and the friction coefficient between the tire model 28 and the road surface model 29. These boundary conditions are stored in the seventh storage unit 5G (computer 1) shown in FIG. 1 based on the specifications of the tire 11A to be predicted, etc.

[0127] [Calculate tire model after internal pressure filling] Next, in a second step S12 of this embodiment, the computer 1 (shown in FIG. 1) calculates the tire model 28 after internal pressure filling (step S24). In step S24 of this embodiment, the tire model 28 after internal pressure filling is calculated based on the internal pressure condition (e.g., normal internal pressure) inputted in the seventh storage unit 5G (shown in FIG. 1).

[0128] In step S24, first, as shown in Fig. 11, the bead portions 28c, 28c of the tire model 28 are constrained by a rim model 30 in which the rim 26 (shown in Fig. 2) is modeled. Furthermore, in step S24, the deformation of the tire model 28 is calculated based on a uniformly distributed load w that corresponds to the internal pressure condition. As a result, the tire model 28 after internal pressure filling is calculated.

[0129] In the deformation calculation of the tire model 28, a mass matrix, a stiffness matrix, and a damping matrix of each element F(i) are created based on the shape and material properties of each element F(i). Furthermore, these matrices are combined to create a matrix of the entire system. Then, the computer 1 (first intermediate characteristic input unit 6C) shown in FIG. 1 creates an equation of motion by applying the various conditions, and performs deformation calculation of the tire model 28 for each unit time T(x) (x=0, 1, ...) of the simulation. Such deformation calculation (including rolling calculation, etc., described later) can be calculated using commercially available finite element analysis application software such as LS-DYNA manufactured by LSTC. The unit time T(x) can be set appropriately depending on the required simulation accuracy.

[0130] Calculate intermediate properties Next, in a second step S12 of this embodiment, the computer 1 (shown in FIG. 1) calculates intermediate characteristics of the tire model 28 (step S25). In this embodiment, the static characteristics of the tire model 28 to be predicted are calculated based on the tire model 28 after internal pressure filling (shown in FIG. 10).

[0131] In step S25 of this embodiment, a vertical spring constant can be calculated by applying a vertical load to the tire model 28 after internal pressure filling shown in Fig. 10 and Fig. 11. Also, a lateral load can be applied to the tire model 28 after internal pressure filling to calculate a lateral spring constant. The magnitudes of the vertical load and the lateral load can be set as appropriate depending on the category of the tire 11, for example.

[0132] In step S25 of this embodiment, a vertical spring constant is acquired as the intermediate characteristic of the tire 11A to be predicted. In this case, the vertical load L on the tire model 28 may be applied by the road surface model 29. The intermediate characteristic is stored in the second storage unit 5B (computer 1) shown in FIG.

[0133] In this manner, in the second step S12 of the present embodiment, a computer simulation is performed based on the tire 11A to be predicted, thereby making it possible to obtain intermediate characteristics of the tire 11A to be predicted. This makes it possible to obtain the intermediate characteristics of the tire 11A to be predicted in a short time and at low cost without actually manufacturing the tire 11A to be predicted (shown in FIG. 2).

[0134] [Enter the prediction model (Step 3)] Next, in the prediction method of this embodiment, the prediction model 31 (shown in FIG. 6) is input to the computer 1 (third step S13). As described above, the prediction model 31 is capable of outputting the static characteristics of any tire 11 by inputting the design factors and intermediate characteristics of any tire 11.

[0135] In the third step S13 of this embodiment, first, a prediction model input unit 6D included in the program unit 6 shown in Fig. 1 is loaded into the working memory 4C. The prediction model input unit 6D is a program for inputting a prediction model 31 (shown in Fig. 6) capable of outputting static characteristics of an arbitrary tire 11 to the computer 1. The prediction model input unit 6D is executed by the calculation unit 4A, whereby the computer 1 can function as a means for inputting the prediction model 31.

[0136] In the case where the prediction model 31 has been created prior to the prediction method as in this embodiment, the prediction model 31 is used. On the other hand, in the case where the prediction model 31 has not yet been created, the third step S13 may include a step of creating the prediction model 31 based on machine learning or a statistical method. This prediction model 31 can be created based on the processing procedure of the creation method shown in Fig. 5. The prediction model 31 is stored in the third storage unit 5C (computer 1) shown in Fig. 1.

[0137] [Output static characteristics of the tire to be predicted (4th step)] Next, in the prediction method of this embodiment, the computer 1 (shown in FIG. 1) outputs the static characteristics of the tire 11A (shown in FIG. 2) to be predicted (fourth step S14). In the fourth step S14 of this embodiment, the design factors and intermediate characteristics of the tire 11A to be predicted are input to a prediction model 31 (shown in FIG. 6). As a result, the static characteristics of the tire 11A to be predicted (in this example, the fracture energy obtained by a plunger fracture test) are output.

[0138] In the fourth step S14 of this embodiment, first, as shown in Fig. 1, the prediction model 31 (shown in Fig. 6) inputted to the third storage unit 5C is read into the working memory 4C. Next, the design factors of the tire 11A (shown in Fig. 2) to be predicted inputted to the first storage unit 5A and the intermediate characteristics of the tire 11A to be predicted inputted to the second storage unit 5B are read into the working memory 4C.

[0139] Next, in a fourth step S14 of this embodiment, a static characteristic output unit 6A included in the program unit 6 shown in Fig. 1 is loaded into the working memory 4C. The static characteristic output unit 6A inputs design factors and intermediate characteristics of the tire 11A to be predicted into a prediction model 31 (shown in Fig. 6) to output the static characteristics of the tire 11A to be predicted. By executing this static characteristic output unit 6A by the calculation unit 4A, the computer 1 can function as a means for outputting the static characteristics of the tire 11A to be predicted.

[0140] In the fourth step S14 of this embodiment, the design factors and intermediate characteristics of the tire 11A (shown in FIG. 2) to be predicted are input as explanatory variables to the prediction model 31 (shown in FIG. 6). As a result, in the fourth step S14, the static characteristics of the tire 11A to be predicted (the above-mentioned fracture energy in this example) are output as objective variables by complementing the design factors, intermediate characteristics, and static characteristics of multiple types of known tires 11 (shown in FIG. 2).

[0141] The output static characteristics of the tire 11A to be predicted are output from an output unit 3 (shown in FIG. 1) such as a display device or a printer. Furthermore, in this embodiment, the static characteristics of the tire 11A to be predicted are input to a ninth storage unit 5J (shown in FIG. 1).

[0142] [Functions of the method and device for predicting static tire characteristics] In the prediction method (prediction device 1A) of this embodiment, the explanatory variables of the prediction model 31 (shown in FIG. 6) include intermediate characteristics, so that the number of explanatory variables (design factors) can be reduced. This can improve the prediction accuracy of the static characteristics by the prediction model 31 (highly maintaining robustness against unknown data different from the teacher data), while reducing the amount of teacher data required to create the prediction model 31 having high prediction accuracy. In addition, by reducing the amount of teacher data, the creation time of the prediction model 31 is shortened, so that the static characteristics of the tire 11A (shown in FIG. 2) to be predicted can be output in a short time. Furthermore, by reducing the amount of teacher data, it is possible to suppress the occurrence of overlearning, in which the teacher data deviates significantly from the tendency originally suggested by the teacher data, and the static characteristics of the tire 11A to be predicted can be predicted with high accuracy.

[0143] Moreover, in the prediction method of this embodiment, intermediate characteristics of the tire 11A (shown in FIG. 2) to be predicted are acquired by computer simulation. Furthermore, static characteristics of the tire 11A to be predicted are output from a prediction model 31 (shown in FIG. 6). Therefore, in order to acquire intermediate characteristics and static characteristics of the tire 11A (shown in FIG. 2) to be predicted, it is not necessary to actually manufacture (prototype) the tire 11A to be predicted. Furthermore, it is not necessary to perform a computer simulation based on the plunger destructive test shown in FIG. 3 to acquire the static characteristics of the tire 11A to be predicted. Therefore, in the prediction method of this embodiment, the static characteristics of the tire 11A to be predicted can be output in a short time and at low cost.

[0144] [Evaluate the static characteristics of the tire to be predicted (step 5)] Next, in the prediction method of this embodiment, the computer 1 (shown in FIG. 1) evaluates whether the static characteristics of the tire 11A (shown in FIG. 2) to be predicted are good or not (fifth step S15). In the fifth step S15 of this embodiment, first, the static characteristics of the tire 11A to be predicted inputted into the ninth storage unit 5J shown in FIG. 1 are read into the working memory 4C. Furthermore, the evaluation unit 6J included in the program unit 6 is read into the working memory 4C. The evaluation unit 6J is a program for determining whether the static characteristics of the tire 11A to be predicted are good or not. The evaluation unit 6J is executed by the calculation unit 4A, so that the computer 1 can function as a means for evaluating the static characteristics of the tire 11A to be predicted.

[0145] Whether the static characteristics are good or not can be evaluated appropriately. In this embodiment, when the static characteristics (in this example, the breaking energy) are equal to or greater than a predetermined threshold, the static characteristics are determined to be good. The threshold can be set appropriately according to, for example, the durability against pinch cut required for the tire 11A (shown in FIG. 2) to be predicted.

[0146] In the fifth step S15, if it is determined that the static characteristics of the tire 11A (shown in FIG. 2) to be predicted are good ("Yes" in the fifth step S15), the tire 11A to be predicted is manufactured based on the design factors of the tire 11A to be predicted (sixth step S16). On the other hand, if it is determined that the static characteristics of the tire 11A to be predicted are not good ("No" in the fifth step S15), at least one of the design factors of the tire 11A to be predicted is changed (seventh step S17), and the first step S11 to the fifth step S15 are performed again. As a result, the prediction method of this embodiment makes it possible to reliably design and manufacture a tire 11 with good static characteristics.

[0147] Although a particularly preferred embodiment of the present invention has been described in detail above, the present invention is not limited to the illustrated embodiment and can be modified and carried out in various forms. EXAMPLES

[0148] Prior to predicting the static characteristics of a tire, a prediction model for the static characteristics of a tire was created based on the processing procedure shown in FIG. 5 (Example).

[0149] In the embodiment, first, the design factors, intermediate characteristics, and static characteristics of multiple types of tires, each of which has at least one different design factor, are input to a computer. Next, in the embodiment, a prediction model capable of outputting the static characteristics of a tire to be predicted from the design factors and intermediate characteristics of the tire to be predicted is created using the design factors, intermediate characteristics, and static characteristics of the multiple types of tires as training data.

[0150] Next, in the embodiment, in order to perform a blind test, the design factors and intermediate characteristics of multiple types of tires were input to the prediction model based on the processing procedure shown in Fig. 8. Then, the static characteristics of the multiple types of tires were predicted. Furthermore, in the embodiment, the design factors and intermediate characteristics (new data) of a new tire different from the training data were input to the prediction model as the design factors and intermediate characteristics of the tire to be predicted. Then, the static characteristics of the new tires were predicted.

[0151] For comparison, a prediction model was created that could output the static characteristics of the tire to be predicted from the design factors of the tire to be predicted, using only the design factors and static characteristics of multiple types of tires as training data (comparative example).

[0152] Next, in the comparative example, in order to conduct a blind test, the design factors of multiple types of tires were input into the prediction model. Then, the static characteristics of the multiple types of tires were predicted. Furthermore, in the comparative example, the design factors (unknown data) of a new tire were input into the prediction model as the design factors of the tire to be predicted. Then, the static characteristics of the new tires were predicted. The common specifications, etc. are as follows: Tire sizes: 205 / 55R16, 235 / 60R18 New tires (prediction target data): Approximately 100 units Static properties: Destruction Energy Explanatory variables in the example (total: 4): Design factors: Tire section height Case structure (number of carcass plies) Case is strong Intermediate characteristics: Vertical spring constant Explanatory variables for comparison (total: 19): Design factors: internal pressure load Rim width W1 Tire size Case structure (number of carcass plies) Case is strong Case cross-sectional area Bead apex height Bead apex rubber complex modulus E* Cap rubber complex modulus E* Complex modulus of elasticity E* of base rubber Sidewall rubber complex modulus E* Complex modulus E* of clinch rubber Number of insulation rubber sheets Insulation Rubber Height 1st gauge G1~4th gauge G4

[0153] Fig. 12 is a graph showing the relationship between the measured and predicted values ​​of the fracture energy of the comparative example. Fig. 13 is a graph showing the relationship between the measured and predicted values ​​of the fracture energy of the example. In Figs. 12 and 13, the fracture energies of multiple types of tires used as training data are displayed as "□". In addition, the fracture energies of new data (data to be predicted) different from the training data are displayed as "△".

[0154] The coefficient of determination R2 between the predicted and measured values ​​of the fracture energy of the teacher data was 0.9598 in the example, while it was 0.9731 in the comparative example. Thus, since the coefficient of determination R2 was almost the same in the example and the comparative example, the prediction accuracy of the fracture energy of the teacher data was about the same in the example and the comparative example.

[0155] On the other hand, the coefficient of determination R2 between the predicted value and the actual measured value of the fracture energy of new data (prediction target data) different from the teacher data was 0.8788 in the example, whereas it was 0.7834 in the comparative example.

[0156] In addition, the mean absolute error (MAE), which is the average of the absolute value of the difference between the predicted and measured fracture energy of the new data, was 60.343 for the example, while it was 74.165 for the comparative example. Furthermore, the root mean squared error (RMSE), which is the average of the squared value of the difference between the predicted and measured fracture energy of the new data, was 86.828 for the example, while it was 116.06 for the comparative example. Note that the smaller the MAE and RMSE, the higher the prediction accuracy.

[0157] In the embodiment, the coefficient of determination R2 is larger than in the comparative example, and furthermore, the MAE and RMSE are smaller. Therefore, in the embodiment, the prediction accuracy of new data (data to be predicted) is improved compared to the comparative example. Thus, in the embodiment, the number of explanatory variables can be reduced by including intermediate characteristics in the explanatory variables of the prediction model. Furthermore, in the embodiment, the prediction accuracy of the dynamic performance of the tire can be improved (high robustness against unknown data can be maintained) with a small amount of training data (less than 50).

[0158] In this way, in the Example, unlike the Comparative Example, the intermediate characteristics were used, so that the number of design factors required for predicting the static characteristics could be reduced. As a result, the Example was able to predict the static characteristics of the tire in a short time and with high accuracy compared to the Comparative Example.

[0159] [Note] The present invention includes the following aspects.

[0160] [Invention 1] 1. A method for predicting static properties of a tire, comprising: A first step of inputting design parameters of a tire to be predicted into a computer; A second step of inputting intermediate characteristics, which are influenced by the design factors and affect the static characteristics, of the tire characteristics to be predicted into the computer; A third step of inputting a prediction model capable of outputting static characteristics of an arbitrary tire by inputting design factors and intermediate characteristics of the arbitrary tire into the computer; and a fourth step in which the computer inputs the design factors and the intermediate characteristics of the tire to be predicted into the prediction model, and outputs the static characteristics of the tire to be predicted. A method for predicting static tire properties. [Invention 2] The method for predicting static characteristics of a tire according to the first aspect of the present invention, wherein the first step includes a step of acquiring the design factors of the tire to be predicted based on design data of the tire to be predicted. [Invention 3] 3. The method for predicting static characteristics of a tire according to claim 1 or 2, wherein the second step includes a step of performing a computer simulation based on the tire to be predicted, and acquiring the intermediate characteristics. [Invention 4] The method for predicting static characteristics of a tire according to any one of claims 1 to 3, wherein the third step includes a step of creating the prediction model based on a statistical method or machine learning. [Invention 5] 5. The method for predicting static characteristics of a tire according to any one of claims 1 to 4, wherein the intermediate characteristics include at least one of a vertical spring constant and a lateral spring constant. [Invention 6] A method for predicting static properties of a tire according to the present invention 5, wherein the static properties include fracture energy obtained by a plunger fracture test. [Invention 7] 1. An apparatus for predicting static properties of a tire, comprising: A first storage unit that stores design factors of a tire to be predicted; a second storage unit configured to store intermediate characteristics that are influenced by the design factors and affect the static characteristics among the characteristics of the tire to be predicted; a third storage unit configured to store a prediction model capable of outputting static characteristics of an arbitrary tire by inputting design factors and intermediate characteristics of the arbitrary tire; a static characteristic output unit that inputs the design factors and the intermediate characteristics of the tire to be predicted into the prediction model and outputs the static characteristics of the tire to be predicted, A device for predicting static tire characteristics. [Invention 8] 1. A method for creating a predictive model of static properties of a tire, comprising: inputting design factors for each of a plurality of types of tires into a computer; inputting intermediate characteristics, which are influenced by the design factors and affect the static characteristics, of the respective characteristics of the plurality of types of tires into the computer; inputting the static characteristics of each of the plurality of types of tires into the computer; and creating the prediction model capable of outputting the static characteristics of a tire to be predicted from the design factors and the intermediate characteristics of the tire to be predicted, by using the design factors, the intermediate characteristics, and the static characteristics of the plurality of types of tires as training data, by the computer. How to create a predictive model. [Explanation of symbols]

[0161] S11 1st process S12 2nd process S13 3rd process S14 4th process

Claims

1. 1. A method for predicting static properties of a tire, comprising: A first step of inputting design parameters of a tire to be predicted into a computer; a second step of inputting intermediate characteristics, which are influenced by the design factors and affect the static characteristics, of the tire characteristics to be predicted into the computer; a third step of inputting a prediction model capable of outputting static characteristics of an arbitrary tire by inputting design factors and intermediate characteristics of the arbitrary tire into the computer; and a fourth step in which the computer inputs the design factors and the intermediate characteristics of the tire to be predicted into the prediction model, and outputs the static characteristics of the tire to be predicted. A method for predicting static tire properties.

2. 2. The method for predicting static characteristics of a tire according to claim 1, wherein the first step includes a step of acquiring the design factors of the tire to be predicted based on design data of the tire to be predicted.

3. 2. The method for predicting static characteristics of a tire according to claim 1, wherein the second step includes a step of performing a computer simulation based on the tire to be predicted, to obtain the intermediate characteristics.

4. The method for predicting static characteristics of a tire according to claim 1 , wherein the third step includes a step of creating the prediction model based on a statistical method or machine learning.

5. The method for predicting static characteristics of a tire according to claim 1 , wherein the intermediate characteristics include at least one of a longitudinal spring constant and a lateral spring constant.

6. The method for predicting static properties of a tire according to claim 5 , wherein the static properties include a fracture energy obtained by a plunger fracture test.

7. 1. An apparatus for predicting static properties of a tire, comprising: A first storage unit that stores design factors of a tire to be predicted; a second storage unit configured to store intermediate characteristics, which are influenced by the design factors and affect the static characteristics, among the characteristics of the tire to be predicted; a third storage unit configured to store a prediction model capable of outputting static characteristics of an arbitrary tire by inputting design factors and intermediate characteristics of the arbitrary tire; a static characteristic output unit that inputs the design factors and the intermediate characteristics of the tire to be predicted into the prediction model and outputs the static characteristics of the tire to be predicted, A device for predicting static tire characteristics.

8. 1. A method for creating a predictive model of static properties of a tire, comprising: inputting design factors for each of a plurality of types of tires into a computer; inputting intermediate characteristics, which are influenced by the design factors and affect the static characteristics, of the respective characteristics of the plurality of types of tires into the computer; inputting the static characteristics of each of the plurality of types of tires into the computer; and creating the prediction model capable of outputting the static characteristics of a tire to be predicted from the design factors and the intermediate characteristics of the tire to be predicted, by using the design factors, the intermediate characteristics, and the static characteristics of the plurality of types of tires as teacher data, by the computer. How to create a predictive model.

Images