Tire test analysis and reliability evaluation method based on damage evolution model

Abstract

The invention provides a tire test analysis and reliability evaluation method based on a damage evolution model, and the method comprises the steps: obtaining data: obtaining 12 mechanical channel data of forces and displacements of a tire in three directions and moments and rotation angles of three axes; identifying a periodic tread jitter value and a sidewall movement value of the tire as deformation channel data; acquiring a tire temperature field as temperature channel data; data analysis: analyzing the reliability of the tire based on the mechanical signal, the visual deformation signal and the temperature field signal; and reliability prediction: judging the tire damage probability based on reliability analysis results of the mechanical signal, the visual deformation signal and the temperature field signal.

Description

Technical Field

[0001] The present invention belongs to the technical field of tires, and in particular relates to a tire test analysis and reliability evaluation method based on a damage evolution model. Background Art

[0002] Aircraft and automobile tires face the risk of tire blowout during use, which can easily cause loss of life and property. Therefore, accurate test prediction of tire reliability is of great significance. Existing studies have shown that tire blowouts during tests are mainly caused by damage to the internal cords or interfaces of the tire body, and these damages are difficult to produce visible changes on the outer surface of the tire, which makes it difficult to accurately predict the risk of tire blowouts. When conducting traditional tire tests in accordance with current standards, test personnel are required to closely inspect the tire status and complete operations such as replenishing tire pressure at regular intervals. If a sudden blowout occurs at this time, it will cause a major safety accident; on the other hand, even if there are no test personnel around when the tire blows out, the huge impact force generated by the blowout can easily damage the test bench, sensor system and other devices, which will significantly reduce the service life of the test instrument in the long run.

[0003] Therefore, a prediction method is needed that can accurately predict the reliability of the tire in the test, determine whether the tire is abnormal, and predict the possibility of tire blowout in the future. Summary of the invention

[0004] In order to solve the above problems, the present invention provides a tire test analysis and reliability evaluation method based on a damage evolution model. The method can predict tire blowout based on the internal structure of the tire and the principle of mechanical damage inversion. According to the six-channel mechanics and deformation information in the tire test data, a mechanical model is used to establish the tire damage evolution law, determine whether the tire reliability is abnormal, and predict the possibility of tire blowout in the future.

[0005] The specific scheme of the present invention is: a tire test analysis and reliability evaluation method based on a damage evolution model, the method comprising: Acquire data, including: acquire 12 mechanical channel data including force and displacement in three directions of the tire and torque and rotation angle of three axes; identify the periodic tread vibration value and sidewall movement value of the tire as deformation channel data; acquire the tire temperature field as temperature channel data; Data analysis, including: tire reliability analysis based on mechanical signals, visual deformation signals, and temperature field signals; The tire reliability analysis based on mechanical signals includes: using the force and displacement in three directions and the torque and angle information of three axes to monitor the changes of the mechanical stability of the tire in real time, and obtaining the comprehensive stress state of the tire through mechanical model inversion or combining fast Fourier transform to analyze the overall load spectrum of the tire, and instantly judging the mechanical state of the tire; The tire reliability analysis based on visual deformation signals includes: using the periodic tread jitter value and sidewall movement value of the tire to detect the deformation of the tire during movement, and judging the probability of tire damage based on the real-time detection of tire deformation; The tire reliability analysis based on the temperature field signal includes: judging the tire status according to the local temperature of the tire; Reliability prediction, including: judging the probability of tire damage based on reliability analysis results of mechanical signals, visual deformation signals, and temperature field signals.

[0006] Beneficial effects of the present invention: 1. The tire test analysis and reliability evaluation method based on the damage evolution model provided by the present invention utilizes the forces in three directions and the torque information of three axes in the tire test data to monitor the changes in the mechanical stability of the tire in real time during the test, and obtains the comprehensive stress state of the tire through the inversion of the mechanical model, so as to instantly judge whether the mechanical state of the tire is abnormal; 2. The tire test analysis and reliability evaluation method based on the damage evolution model provided by the present invention utilizes the six-channel information in the tire test data and combines it with the fast Fourier transform to analyze the tire's overall load spectrum, and uses the tire spectrum response as one of the indicators for tire blowout prediction, making the tire reliability prediction more accurate; 3. The tire test analysis and reliability evaluation method based on the damage evolution model provided by the present invention uses computer vision recognition technology to quickly and intuitively detect the deformation of the tire during movement by locking the characteristic information such as the tire surface and tire body contour in multiple directions, and judges the possibility of tire blowout based on the real-time detection of tire deformation; 4. The tire test analysis and reliability evaluation method based on the damage evolution model provided by the present invention, combined with the tire test big data formed by the above three points, uses the principle of probability to accurately calculate the possibility of tire blowout in the future, thereby minimizing the tire test workload and shortening the tire R&D cycle while ensuring safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 A tire test data acquisition system provided by the present invention; Figure 2 A schematic diagram of six directional components of a tire provided by the present invention; Figure 3This is a schematic diagram of the test data and nonlinear curve fitting results provided by the present invention; Figure 4 A schematic diagram of the tread visual recognition preprocessing provided by the present invention; Figure 5 A schematic diagram of the preprocessing of the visual recognition of the tire sidewall provided by the present invention; Figure 6 A schematic diagram of the test data analysis and tire reliability prediction process provided by the present invention.

[0008] Reference numerals: 1-six-channel sensor; 2-tire; 21-longitudinal groove on tread; 22-tread marking line; 23-sidewall marking line; 24-tread profile; 3-image acquisition device; 4-test loading device; 5-rigid contact end; 7-thermal imaging system; 6-area; 61-raw data analysis curve A; 62-nonlinear fitting curve; 63-raw data analysis curve B. DETAILED DESCRIPTION

[0009] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in each embodiment of the present invention described below can be combined with each other as long as they do not conflict with each other. To achieve the above-mentioned purpose, the present invention adopts the following technical scheme.

[0010] It should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components in a certain specific posture. If the specific posture changes, the directional indication will also change accordingly.

[0011] In the first aspect, the present invention provides a tire test reliability prediction method based on mechanical signals, using a six-channel sensor installed on the tire shaft or lever arm to output signals of six channels at a certain frequency, wherein the force signals include: Fx, Fy, Fz, Mx, My, Mz, and the displacement signals include: dx, dy, dz, θx, θy, θz. The x-axis points to the heading of the tire, the y-axis points to the lateral direction of the tire, and the z-axis points to the radial direction of the tire. F represents the magnitude of the force along the three axes, M represents the magnitude of the torque rotating around the three axes, d represents the magnitude of the displacement along the three axes, and θ represents the magnitude of the rotation angle around the three axes. Since the tire will have internal damage accumulation and evolution during the test, the data of the six channels will also change accordingly as the test proceeds. The data of the six channels are fully recorded and analyzed in real time based on the tire damage evolution theoretical model. The analysis method includes but is not limited to: data discreteness analysis, Fourier transform, etc., combined with the tire test big data, to determine whether the tire is abnormal and predict its reliability.

[0012] In the second aspect, the present invention provides a tire test analysis and reliability evaluation method based on a damage evolution model based on visual deformation signals. It uses optical measurement and computer vision recognition technology to track characteristic information such as the tire surface and tire body contour in multiple directions, and quickly and automatically calculates the external deformation and motion morphology parameters of the tire during movement. Also based on the tire damage evolution model, it judges the tire reliability and predicts whether it will have a tire blowout.

[0013] In the third aspect, the present invention provides a tire test analysis and reliability evaluation method based on a damage evolution model based on thermal imaging signals, which uses an infrared thermal imaging system to track and monitor the temperature field information on the tire surface, quickly identify local heat accumulation on the tire surface, and combine the tire structure to estimate and predict the possibility of tire blowout.

[0014] The specific technical solution is: Figure 1 A schematic diagram of a tire test data acquisition system provided by the present invention includes: The six-channel sensor 1 is rigidly connected to the test loading device 4, and has the ability to output the six forces or moments Fx, Fy, Fz, Mx, My and Mz sensed by the sensor, and the ability to output the six directional displacements or rotation angles dx, dy, dz, θx, θy and θz. The output signals are used as 12 mechanical channel data and are transmitted to the external storage terminal (i.e., computer) in real time by wired or wireless means. The accuracy and range of the sensor are selected according to the load required for the test; the output frequency should not be lower than 20Hz.

[0015] The tire 2 is the tire to be tested, and its axis is connected to the test loading device 4 and can rotate around the central axis.

[0016] Image acquisition equipment 3 can be placed in several directions facing the tire in the heading, sideways, and radial directions. The image resolution should be as high as possible, not less than 2K; the image acquisition frame rate should be not less than 60fps. If conditions permit, a high-speed camera can be used and equipped with a light source, or an infrared camera can be added to collect the tire temperature field.

[0017] The test loading device 4 is used to apply radial force, deflection angle and other working conditions to the tire 2.

[0018] The rigid contact end 5 is used to cooperate with the test loading device 4 to apply a corresponding test load to the tire 2 .

[0019] The thermal imaging system 7 is used to collect the temperature field information of the tire 2 surface; and obtain the tire temperature field as temperature channel data.

[0020] In one embodiment of the present invention, the tire generates a large amount of heat during the test due to the self-heating property of rubber. For a good tire, its temperature field during a normal rolling test should be uniform in the circumferential direction; at a certain moment, the tire has uneven damage inside during the test, resulting in local heat accumulation. When the temperature of this part reaches a temperature sufficient to cause the tire material to be damaged (such as the melting point of rubber) due to heat accumulation, the tire will inevitably burst. Therefore, the thermal imaging system can be used to issue an early warning before the local temperature of the tire reaches the material damage temperature, stop the test in time, and cool the tire.

[0021] Figure 2 The schematic diagram of the six-channel mechanical components of the tire provided by the present invention is mainly intended to standardize the description of the six-directional components in the present invention, wherein: Fx represents the tread force of the tire; Fy represents the lateral force of the tire; Fz represents the radial force of the tire; Mx represents the moment of the tire around the tread axis; My represents the moment of the tire around the lateral axis; Mz represents the moment of the tire around the radial axis. In addition, the displacement components of the tire in the six channels correspond to the mechanical components, dx corresponds to Fx, dy corresponds to Fy, dz corresponds to Fz, θx corresponds to Mx, θy corresponds to My, and θz corresponds to Mz.

[0022] Figure 3 The schematic diagram of the test data and nonlinear curve fitting results provided by the present invention, wherein region 6 includes original data analysis curve A 61 and nonlinear fitting curve 62. Original data analysis curve A 61 is obtained by processing the six channel data output by the six-channel sensor 1 with a certain mathematical method and mechanical inversion. Nonlinear fitting curve 62 is obtained by performing nonlinear fitting on the original data analysis curve A 61 using classical iterative regression algorithms such as Levenberg-Marquardt, and its form includes but is not limited to: trigonometric function, power function polynomial, Boltzmann function, etc.

[0023] In one embodiment of the present invention, a tire 2 is subjected to a long-term continuous rolling test, during which the six-channel sensor 1 continuously outputs data collected by the six channels. Since the mechanical properties of the tire 2 itself will decline to a certain extent after a long period of rolling, the data of the six channels will also change to a certain extent as time goes by. In one possible case, when the tire 2 produces the same radial deformation dz, Fz gradually decreases over time, that is, the radial stiffness Kz=Fz / dz of the tire decreases. When the radial stiffness Kz decreases to a certain extent, the tire 2 is damaged. At this time, if the time from the start of the test to the damage of the tire 2 is divided into several sections, and the Kz data in each period is averaged, a result similar to the following will be obtained. Figure 3 The original data analysis curve B63 shown in the figure shows a trend curve. For tires of the same type, the reliability of the tires can be predicted using the same curve rule. When the Kz value of the tested tire 2 drops to a dangerous level obtained through accumulation of big data experience, an alarm is triggered and the test is stopped.

[0024] In another embodiment of the present invention, a tire 2 is subjected to an intermittent high-load rolling test. As time goes by, the tire 2 generates damage that cannot be directly observed, causing fluctuations in multiple mechanical performance indicators of the tire 2. In one possible case, the cord inside one side of the tire 2 breaks, causing the tire to shake in the y direction. At this time, the values ​​of the six-channel sensor fluctuate to a certain extent. Therefore, the variance of the six component force signals is calculated within each period of t seconds to evaluate its stability. The variance σ calculation formula is: (1) In the formula, N is the number of samples, Xi is the sample value, and μ is the sample average. The data curve after the above processing is as follows: Figure 3 As shown in the original data analysis curve A61, the original data analysis curve A61 is analyzed using the correlation function , preferably the Boltzmann function The nonlinear fitting curve 62 is obtained by fitting, where a, b, and c are constants. The nonlinear fitting curve 62 is the damage evolution model of the tire under the working condition. For the same type of damage form of the tire under the same working condition, the value of the nonlinear fitting curve 62 is used. Make a prediction to determine the value of the nonlinear fitting curve 62 Whether the critical value is reached. The above critical values ​​are obtained through multiple tests. Each working condition and each type of tire corresponds to a set of critical values. During the test, the parameters and curve values ​​reached when the tire blows out are recorded. The distribution law of the values ​​obtained from multiple tests conforms to the normal distribution. Then the normal distribution law can be used to calculate the possibility of a blowout.

[0025] In another embodiment of the present invention, when a tire 2 is undergoing an intermittent high-load rolling test, the data of the six-channel sensor can be Fourier transformed to obtain its spectrum curve, and the characteristic value in the spectrum curve is taken to obtain a new raw data analysis curve A61, and the above process is repeated to achieve accurate early warning. The expression of Fourier transform is: (2) Where ω represents frequency, t represents time, θ(t) is the spectral density function, and function Θ(ω) is the Fourier transform of θ(t).

[0026] Figure 4 , Figure 5 The schematic diagrams of the tread visual recognition preprocessing and the sidewall visual recognition preprocessing provided by the present invention respectively include: Longitudinal grooves 21 on the tread; The tread marking line 22 is a stable marking line drawn on the tire tread and having a certain width around the circumference of the tire. The marking line can be in the form of a solid line or a dotted line. The sidewall marking line 23 is a circumferential curve drawn on the sidewall of the tire. It should be as round as possible and concentric with the wheel axle. The width of the marking line 23 should also be kept stable. It can be in the form of a solid line or a dotted line. The number of the above-mentioned marking lines is not unique, and they can also be replaced by marking points, speckles, etc. with a certain degree of recognition; Tread profile 24.

[0027] In a possible embodiment, the present invention arranges an image acquisition device 3 on the heading direction and the side of the tire 2 respectively. When the tire 2 is subjected to a rolling test, the viewing angles seen by the two image acquisition devices 3 are the same as those of the tire 2. Figure 4 , Figure 5 In the initial state, the tread marking line 22 and the sidewall marking line 23 on the tire 2 have regular shapes and are a stable curve or straight line in the field of view of the image acquisition device 3; as the performance of the tire 2 gradually deteriorates, the dynamic stability of the tire 2 also changes. In the field of view of the image acquisition device 3, the tread marking line 22 begins to periodically shake left and right, and the shape of the sidewall marking line 23 also moves periodically. These two periodic shaking values ​​are recorded as and , and The size can be converted according to the image pixels and shooting distance. Identify the periodic tread vibration value and sidewall movement value of the tire as deformation channel data; and When the value reaches a certain value, it indicates that the internal damage of the tire 2 has accumulated to a certain extent, and a tire blowout will occur soon. and The damage theory critical value is corrected, where and The calculation formula is: , x0, y0 are the coordinates of the feature point before deformation, x1, y1 are the coordinates of the feature point after deformation. The tire damage probability W can be calculated using the following formula: (3) In the formula, is obtained through fitting and The feature prediction function of The total number of tires that reach the theoretical critical value; m is The total number of tires that burst after reaching the theoretical critical value. The probability calculation method here uses the normal distribution function Calculation, the probability calculation method here is also applicable to the six-channel data processing method described above.

[0028] Figure 6 The present invention provides a schematic diagram of the test data analysis and tire reliability prediction process. The left frame contains the original test data of the tire 2 collected by the six-channel sensor 1 or the image acquisition device 3 during the test. These data are then analyzed and processed in real time, and converted into parameters that can represent the state of the tire 2 using the various processing methods described above. Finally, based on a large amount of test data, a probabilistic relationship between these parameters and tire reliability is established to achieve accurate prediction of tire reliability.

[0029] In a possible embodiment, for a certain type of tire 2, a certain amount of test data has been accumulated according to the above method, including variances, prediction functions, and the like of 12 channels, namely, Fx, Fy, Fz, Mx, My, Mz, dx, dy, dz, θx, θy, and θz. , and the prediction function of the tire surface marking jitter value , each prediction function , Each corresponds to a critical value or critical range. However, for any tire, not every channel's prediction function reaches the critical value or critical range before a tire blows out. Therefore, it is necessary to multiply the signal of each channel by a weighting constant E for weighted processing according to different types of tires and working conditions. The method for determining the weighting constant E is as follows: For the kth channel, the critical value of its prediction function in a certain experiment is (x) or , the quotient of the prediction function critical value of the channel and the sum of the prediction function critical values ​​of all other channels is the weighted constant corresponding to the channel : or (4) Then, the data signal evolution characteristics under different working conditions can be integrated to summarize a multi-channel tire test analysis and reliability evaluation method based on the damage evolution model applicable to various test conditions, which can be summarized as the following formula (5). The meaning of formula (5) is that based on the mechanical signals and visual signals mentioned in the previous embodiments, the probability relationship between these channel signals and tire reliability is obtained respectively. For different types of tires, such as bias tires or radial tires, the relationship between tire reliability and these channels is different, so it is necessary to weight the signal of each channel; at the same time, the corresponding reliability change of each tire when facing different working conditions is different. Different working conditions are functions of test loading time and load size, so here it is also necessary to weight different working conditions, and then the tire blowout possibility prediction function W is obtained: (5) Where M is the total number of tires that reach the theoretical critical value; m is the total number of tires that burst after reaching the theoretical critical value; q is the total number of operating conditions that the tire has passed through during the test; and p is the total number of data signal outputs, including 24 nonlinear fitting curves obtained by fitting 12 force\displacement signal channels (Fx, Fy, Fz, Mx, My, Mz, dx, dy, dz, θx, θy, θz). , several marking line jitter values ​​Δ obtained from several visual signal channels n , the abnormal temperature value T obtained by the temperature field signal channel, F represents the load on the tire in each working condition, S represents the time the tire experiences in each working condition in the test, is the weighted constant for each operating condition's contribution to tire blowout, It is the weighted constant of the prediction capability of each channel under each working condition, which can be continuously revised according to the test big data. Refers to the characteristic prediction function used in each working condition, including the prediction function mentioned above , The remaining parameters have the same meanings as in formula (3).

[0030] In another embodiment of the present invention, a method for evaluating tire reliability through temperature field is provided: an infrared camera 7 is added to collect the temperature field of tire 2. When there is no damage inside the tire 2, the temperature field of the tire 2 should be uniformly distributed at 360° along its circumference. As the tire test proceeds, local damage occurs inside the tire 2, which is manifested as abnormal accumulation of local heat in the area corresponding to the damage position on the outer surface of the tire 2. The highest abnormal temperature collected is recorded as temperature T. Previously, the lowest melting point t of the rubber used in the tire needs to be determined in advance. When T=0.9t, an early warning needs to be issued. The temperature field data and the damage probability W calculated by formula (5) are used in parallel to judge the tire reliability. When the obtained temperature field reaches 90% of the material melting point, that is, T=0.9t, an early warning is issued. At this time, even if the damage probability W does not reach the early warning value, it is necessary to stop loading the tire; if the temperature field T<0.9t, but the damage probability W reaches 0.9, it is still necessary to stop loading the tire. It should be noted that the damage probability of 0.9 and the temperature reaching 90% of the melting point described here can be appropriately adjusted according to the test needs to ensure the safety of the test.

Claims

1. A tire test analysis and reliability evaluation method based on a damage evolution model, characterized in that: The method comprises: Acquire data, including: acquire 12 mechanical channel data including force and displacement in three directions of the tire and torque and rotation angle of three axes; identify the periodic tread vibration value and sidewall movement value of the tire as deformation channel data; acquire the tire temperature field as temperature channel data; Data analysis, including: tire reliability analysis based on mechanical signals, visual deformation signals, and temperature field signals; The tire reliability analysis based on mechanical signals includes: using the force and displacement in three directions and the torque and angle information of three axes to monitor the changes of the mechanical stability of the tire in real time, and obtaining the comprehensive stress state of the tire through mechanical model inversion or combining fast Fourier transform to analyze the overall load spectrum of the tire, and instantly judging the mechanical state of the tire; The tire reliability analysis based on visual deformation signals includes: using the periodic tread jitter value and sidewall movement value of the tire to detect the deformation of the tire during movement, and judging the probability of tire damage based on the real-time detection of tire deformation; The tire reliability analysis based on the temperature field signal includes: judging the tire status according to the local temperature of the tire; Reliability prediction, including: judging the probability of tire damage based on reliability analysis results of mechanical signals, visual deformation signals, and temperature field signals.

2. The method according to claim 1, characterized in that The forces in three directions and the torque information of three axes are Fx, Fy, Fz, Mx, My, and Mz respectively; the corresponding displacements and rotation angles in each direction are dx, dy, dz, θx, θy, and θz, where Fx represents the heading force of the tire; Fy represents the lateral force of the tire; Fz represents the radial force of the tire; Mx represents the torque of the tire around the heading axis; My represents the torque of the tire around the lateral axis; Mz represents the torque of the tire around the radial axis. The displacement components in the six channels correspond to the mechanical components. The displacement dx corresponds to the heading force Fx, dy corresponds to the lateral force Fy, dz corresponds to the radial force Fz, the rotation angle θx corresponds to the heading axis torque Mx, the rotation angle θy corresponds to the lateral axis torque My, and the rotation angle θz corresponds to the radial axis torque Mz; several periodic tread jitter values ​​or sidewall movement values ​​are recorded as Δ n ; Local abnormal temperature signal T on the tire surface.

3. The method according to claim 2, characterized in that The comprehensive stress state of the tire obtained by inversion of the mechanical model includes: calculating the variance of the force, displacement in three directions and the torque and angle of rotation of three axes in each period of t seconds. The calculation formula of the variance σ is: （1） Where N is the number of samples, is the sample value, μ is the sample average, and the original data analysis curves are obtained respectively, and the nonlinear fitting curve is obtained by fitting with the correlation function , the nonlinear fitting curve is the damage evolution model of the tire under this working condition; It is the characteristic prediction function based on mechanical signal analysis.

4. The method according to claim 3, characterized in that: The related function forms include but are not limited to the Boltzmann function. The specific formula of the Boltzmann function is: , where a, b, c are constants.

5. The method according to claim 3, characterized in that: The overall tire load spectrum is analyzed by fast Fourier transform, including: performing Fourier transform on the data of force, displacement in three directions and torque and rotation angle of three axes to obtain their spectrum curves respectively, taking the characteristic values ​​in the spectrum curve to obtain a new original data analysis curve, and repeating the nonlinear curve fitting process. The expression of the Fourier transform used is: （2） Where ω represents frequency, t represents time, θ(t) is the spectral density function, and function Θ(ω) is the Fourier transform of θ(t).

6. The method according to claim 3, characterized in that Several periodic tread vibration values ​​or sidewall movement values ​​are recorded as Δ n , and the way to obtain it is: Tread markings are marks drawn on the tire tread with a uniform width around the circumference of the tire; Marking sidewall marks are marks drawn on the sidewall of the tire, around the circumference of the tire side, concentric with the wheel axle, and with a uniform width; Several image acquisition devices are arranged on the heading direction and side of the tire. When the tire rolls, each image acquisition device collects images. The marking lines gradually degrade with the tire performance. In the field of view of the image acquisition device, the tread marking line shakes periodically, and the shape of the sidewall marking line also moves periodically. The periodic shaking value or movement value is recorded as n ; After fitting, we get n Feature prediction function .

7. The method according to claim 6, characterized in that The fitted n Feature prediction function The fitting methods include but are not limited to the Boltzmann function. The specific formula of the Boltzmann function is: , where a, b, c are constants.

8. The method according to claim 6, characterized in that The number of marks is 1 or more, in the form of marking lines, marking points, or speckles.

9. The method according to claim 6, characterized in that According to different types of tires and working conditions, the signal of each channel is multiplied by the weighting constant E for weighted processing. The method for determining the weighting constant E is as follows: For the kth channel, the critical value of the prediction function is (x) or , the weight constant corresponding to the channel for: or (4).

10. The method according to claim 2, characterized in that The temperature field monitoring method is as follows: an infrared camera is added to the image acquisition device to collect the tire temperature field, and the highest abnormal temperature collected is recorded as temperature T.

11. The method according to claim 9, characterized in that Based on the reliability analysis results of mechanical signals, visual deformation signals, and temperature field signals, judgment is made, including: the probability of failure of multi-channel tires under various test conditions for: （5） Wherein, M is the total number of tires that reach the theoretical critical value; m is the total number of tires that burst after reaching the theoretical critical value; q represents the total number of working conditions that the tire has passed; p represents the total number of data signal outputs, including the nonlinear fitting curve obtained by fitting the 12 force\displacement signal channels, namely Fx, Fy, Fz, Mx, My, Mz, dx, dy, dz, θx, θy, θz. , several marking line jitter values ​​Δ obtained from several visual signal channels n ; F represents the load on the tire in each working condition, S represents the time the tire experiences in each working condition in the test, is the weighted function of each working condition's contribution to tire blowout, is the weighted constant of the prediction capability of each channel under each working condition, Refers to the feature prediction function of each channel data, including , .

12. The method according to claim 11, characterized in that Judging the tire damage probability based on the reliability analysis results of the mechanical signal, the visual deformation signal, and the temperature field signal also includes: based on the abnormal temperature value T obtained by the temperature field signal channel, calculating the temperature field data damage probability W and judging the tire reliability in parallel. When the obtained temperature field reaches 90% of the material melting point, that is, T=0.9t, an early warning is issued. At this time, even if the damage probability W has not reached the early warning value, loading on the tire is stopped; if the temperature field T<0.9t, but the damage probability W reaches 0.9, loading on the tire is stopped.

Images