A method for predicting the aging life of rubber

Abstract

The present application relates to the technical field of rubber material, in particular to a rubber aging life prediction method. The rubber aging prediction model parameters are weighted by the rubber natural environment test and the laboratory thermal aging accelerated test; the regression algorithm is used to fit the relationship function of the rubber aging prediction model parameters and the temperature, the parameters in the rubber aging prediction model are determined according to the temperature of the rubber to be measured in the actual storage environment, and finally the aging prediction model of the rubber in the actual storage environment is established. The present application uses the rubber natural environment test data and the laboratory thermal aging accelerated test data as the certain parameter data samples of the rubber aging prediction model, and retains the advantages of the certain parameter method based on the rubber natural environment test data and the certain parameter method based on the laboratory thermal aging accelerated test data; the time span of the rubber aging test is reduced, the test cost is reduced, and the rubber aging prediction model obtained is closer to reality and more accurate.

Description

Technical Field

[0001] This invention relates to the field of rubber materials technology, and in particular to a method for predicting the aging life of rubber. Background Technology

[0002] Rubber is a widely used insulating, sealing, noise-reducing, and vibration-damping material in various types of equipment and products. However, during long-term service, it is prone to performance degradation and even functional failure due to thermo-oxidative aging, posing a significant threat to the normal use of equipment and products. Currently, researchers generally use the Arrhenius model, which aligns the relationship between rubber aging rate and temperature, as the theoretical basis for predicting rubber aging life.

[0003] However, the technical approaches used for modeling rubber aging life vary depending on the data source. One approach is to use linear extrapolation based on accelerated thermal aging test data from laboratories to derive a model of rubber's performance degradation under storage temperature conditions. Another approach is to directly use mathematical modeling based on natural environment test data to obtain a performance degradation model. Once the performance degradation model of the sample is obtained, the predicted lifespan of the sample can be calculated given a known performance degradation threshold.

[0004] Both methods have their limitations, with limited predictive accuracy. While laboratory accelerated thermal aging tests can effectively shorten testing time and save costs, they inevitably differ from the actual service environment of equipment / products. Rubber natural environment tests closely resemble actual service environments, but they also have drawbacks such as long time spans, high testing costs, and limited data sample sizes. Summary of the Invention

[0005] This invention discloses a method for predicting the aging life of rubber. It uses data from both natural environment tests and accelerated thermal aging tests in the laboratory. This solves the problem that the test data obtained from accelerated thermal aging tests in the laboratory deviates from the material performance degradation law under real storage conditions, and also solves the problems of long time span, high cost, and insufficient data samples in natural environment tests of rubber.

[0006] This invention is achieved through the following technical solution: a first modeling step, a second parameter determination step, and a third prediction step; wherein,

[0007] The first modeling step is to establish a rubber aging prediction model;

[0008] The second parameter determination step involves combining natural environment tests and accelerated laboratory thermal aging tests to determine the parameters of the rubber aging prediction model under actual storage conditions for the rubber to be tested.

[0009] The third prediction step involves predicting the storage life of rubber based on the rubber aging prediction model after parameter confirmation.

[0010] The advantage of this embodiment is that the present invention uses both natural environment test data and laboratory accelerated thermal aging test data of rubber as the parameter data sample for the rubber aging prediction model, while retaining the advantages of using natural environment test data and laboratory accelerated thermal aging test data for parameter determination; it reduces the time span of natural environment test of rubber, reduces test costs, and the obtained rubber aging prediction model is closer to the performance degradation law under real storage environment and has a more accurate prediction effect.

[0011] Specifically, in the second step of parameter determination, the method combines natural environment testing of rubber with accelerated thermal aging testing in the laboratory, and the specific method is as follows:

[0012] Based on the temperature of the rubber under test in the actual storage environment, weights are assigned to the parameters of the rubber aging prediction model calculated under different test conditions; the closer the rubber aging test temperature is to the temperature of the rubber under test in the actual storage environment, the higher the weight value of the parameters of the rubber aging prediction model calculated.

[0013] In the second parameter determination step, the parameters of the rubber aging prediction model for the rubber under actual storage conditions are determined. The specific method is as follows:

[0014] Based on the parameters of the fused rubber aging prediction model, a regression algorithm is used to fit the relationship function between the parameters of the rubber aging prediction model and temperature. The parameters in the rubber aging prediction model are then determined based on the temperature of the rubber under actual storage conditions.

[0015] The advantage of this embodiment is that, based on the temperature of the rubber under actual storage conditions, the weights of different rubber aging test parameters are assigned. The closer the test parameters are to the actual storage temperature, the higher their weights. Finally, a regression algorithm is used to determine the parameters in the rubber aging prediction model at the actual storage temperature. The resulting rubber aging prediction model is more accurate in predicting rubber life than the rubber aging prediction model with parameters determined by laboratory accelerated thermal aging tests. At the same time, it does not require a lot of time to conduct a large number of natural environment tests on rubber, thus saving test costs.

[0016] Furthermore, in the first modeling step, the specific formula of the rubber aging prediction model is as follows:

[0017] P = Aexp(-Kτ) α )

[0018] in,

[0019] P is a performance change index, which can be the compression set retention rate, the compression stress relaxation coefficient, or the elongation at break change rate.

[0020] τ represents the aging time;

[0021] K is a constant representing the rate of change of performance, which is temperature-dependent;

[0022] A is a constant;

[0023] α is a constant.

[0024] This embodiment uses the Arrhenius model as the rubber aging prediction model. This model has high prediction accuracy and requires fewer parameters to be confirmed, which can reduce the computational load of parameter confirmation in the rubber aging prediction model and improve the calculation speed.

[0025] Specifically, the constant α is calculated as follows:

[0026] Let X = τ α Y = lnP; a = lnA; b = -K; The rubber aging prediction model is expressed as Y = a + bX;

[0027] Select laboratory accelerated thermal aging test data (P) ij -τ ij If X and Y have a strong linear relationship, α is 1; otherwise, the successive approximation method is used to determine the value of α.

[0028] Rubber aging test data P ij -τ ij , indicating the i-th test temperature T i Different aging times τ i1 ,τ i2 ,τ i3 ,…,τ ij Below, there are different performance change indicators P. i1 ,P i2 ,P i3 ,…P ij (i=1,2,…,m; j=1,2,…,n).

[0029] Furthermore, to determine whether X and Y have a strong linear relationship, at least the following conditions must be met:

[0030] Assuming α is 1, the correlation coefficient r is obtained using the least squares method; then, by referring to the correlation coefficient table with a 99% confidence level and degrees of freedom f = n - 2, the calculated value of r is compared with the value of r. cal Comparison, |r cal |>r

[0031]

[0032] in,

[0033]

[0034]

[0035]

[0036]

[0037] Specifically, the value of α is determined by successive approximation. The specific approximation criterion is that I is minimized when α is accurate to two decimal places.

[0038]

[0039] In the formula, P ij This represents the test value of the performance change index at the j-th test point under the i-th aging test temperature; This represents the predicted value of the performance change index at the j-th test point under the i-th aging test temperature.

[0040] Furthermore, at different temperatures T i The calculation methods for the constant and the performance change rate constant are as follows:

[0041] The least squares method can be used to solve for the coefficients a and b in the equation Y = a + bX, thus obtaining the constant A and the performance change rate constant K.

[0042]

[0043]

[0044] A = exp(a)

[0045] K = -b

[0046] Based on the accelerated thermal aging test data from the laboratory, the different test temperatures T were calculated. i The corresponding acceleration constant A i and the rate constant of acceleration performance change K i ;

[0047] Based on the natural environment test data of rubber, the natural constant A0 and the natural performance change rate constant K0 corresponding to the natural test temperature T0 were calculated.

[0048] Furthermore, by integrating the parameters of the rubber aging prediction model calculated under natural environment tests and accelerated laboratory thermal aging tests, the parameters of the rubber aging prediction model for the rubber under actual storage conditions are determined. The specific method is as follows:

[0049] Using weight matrix Let A be the acceleration constant. i And the natural constant A0, and the natural performance change rate constant K0 and the acceleration performance change rate constant K i Assign weights;

[0050] The specific KT and AT relationships are obtained by solving the convergent weight matrices W and β using the adaptive weighted least squares method.

[0051] Based on the actual storage environment temperature of the rubber to be tested, determine the constants A and the performance change rate constant K in the rubber aging prediction model of the rubber to be tested.

[0052] The advantage of this embodiment is that the weight matrix can effectively reflect the distance relationship between the actual storage temperature and the test temperature. The greater the distance, the lower the weight. The weight matrix is ​​easy to calculate and integrates the parameters of the rubber aging prediction model calculated under the natural environment test and the laboratory accelerated thermal aging test.

[0053] Specifically, in the weight matrix, ω i The calculation formula is:

[0054]

[0055] The advantage of this embodiment is that ω i Using this calculation formula, the effect of the difference between the actual storage temperature and each test temperature on the accuracy of the parameters can be reflected, and the calculation is convenient.

[0056] Specifically, in the weight matrix, ω i The calculation formula is:

[0057]

[0058] in, K′ i and T′ i The constants for performance change and temperature are given after dimensionless processing.

[0059] The advantage of this embodiment is that it takes into account the range of test temperatures under natural environment tests and accelerated laboratory thermal aging tests for rubber, and the rubber aging prediction model obtained according to this weighting calculation formula has higher accuracy.

[0060] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained from the following description and claims. Attached Figure Description

[0061] The accompanying drawings of this invention are described below.

[0062] Figure 1 This is a schematic diagram of the process of the present invention.

[0063] Figure 2 The above is a graph showing experimental data from an example. Detailed Implementation

[0064] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0065] Example 1:

[0066] A method for predicting the aging life of rubber, the specific process of which is as follows:

[0067] S1, First modeling step;

[0068] Establish a rubber aging prediction model:

[0069] P = Aexp(-Kτ) α )

[0070] Where P is a performance change index, which can be the compression set retention rate, the compression stress relaxation coefficient, or the elongation at break rate; τ is the aging time; K is a performance change rate constant, which is related to temperature; A is a constant; and α is a constant.

[0071] In this step, the rubber aging prediction model can also be any other known mathematical model that can be used to predict the aging life of rubber.

[0072] S2, Second parameter confirmation step;

[0073] By combining natural environment tests and laboratory accelerated thermal aging tests on rubber, the parameters of the rubber aging prediction model under actual storage conditions were determined.

[0074] Specifically, combining natural environment tests and laboratory accelerated thermal aging tests on rubber, the specific method is as follows: based on the temperature of the rubber under test in the actual storage environment, weights are assigned to the parameters of the rubber aging prediction model calculated at different test temperatures; the closer the rubber aging test temperature is to the temperature of the rubber under test in the actual storage environment, the higher the weight value of the parameters of the rubber aging prediction model calculated therein.

[0075] The parameters of the rubber aging prediction model for the rubber under actual storage environment are determined by the following method: based on the parameters of the fused rubber aging prediction model, a regression algorithm is used to fit the relationship function between the parameters of the rubber aging prediction model and temperature, and the parameters in the rubber aging prediction model are determined based on the temperature of the rubber under actual storage environment.

[0076] In this embodiment, a weighted linear regression algorithm is used, which combines the advantages of determining parameters through natural environment testing of rubber and accelerated thermal aging testing in the laboratory. It is conceivable that the test data can be directly weighted according to the test temperature to calculate the parameters of the rubber aging prediction model. Alternatively, it is conceivable that any intermediate parameter, whether custom or known, can be used to reflect the temperature difference, and the parameters of the rubber aging prediction model can be calculated using this intermediate parameter.

[0077] S21. The method for calculating the constant α is as follows:

[0078] Let X = τ α Y = lnP; a = lnA; b = -K; The rubber aging prediction model is expressed as Y = a + bX;

[0079] Select laboratory accelerated thermal aging test data (P) ij -τ ij If X and Y have a strong linear relationship, α is 1; otherwise, the successive approximation method is used to determine the value of α.

[0080] Rubber aging test data P ij -τ ij , indicating the i-th test temperature T i Different aging times τ i1 ,τ i2 ,τ i3 ,…,τ ij Below, there are different performance change indicators P. i1 ,P i2 ,P i3 ,…P ij (i=1,2,…,m; j=1,2,…,n).

[0081] To determine whether X and Y have a strong linear relationship, at least the following conditions must be met:

[0082] Assuming α is 1, the correlation coefficient r is obtained using the least squares method; then, by referring to the correlation coefficient table with a 99% confidence level and degrees of freedom f = n - 2, the calculated value of r is compared with the value of r. cal Comparison, |r cal |>r

[0083]

[0084] in,

[0085]

[0086]

[0087]

[0088]

[0089] The value of α is determined by a successive approximation method. The specific approximation criterion is that I is minimized when α is accurate to two decimal places.

[0090]

[0091] In the formula, P ij This represents the test value of the performance change index at the j-th test point under the i-th aging test temperature; This represents the predicted value of the performance change index at the j-th test point under the i-th aging test temperature.

[0092] In this embodiment, the calculation of α does not require combining natural environment tests and laboratory accelerated thermal aging tests of rubber; other conventional methods can also be used for calculation.

[0093] S22, different temperatures T i The calculation methods for constant A and the performance change rate constant K are as follows:

[0094] The least squares method can be used to solve for the coefficients a and b in the equation Y = a + bX, thus obtaining the constant A and the performance change rate constant K.

[0095]

[0096]

[0097] A = exp(a)

[0098] K = -b

[0099] Based on the accelerated thermal aging test data from the laboratory, the different test temperatures T were calculated. i The corresponding acceleration constant A i and the rate constant of acceleration performance change K i ;

[0100] Based on the natural environment test data of rubber, the natural constant A0 and the natural performance change rate constant K0 corresponding to the natural test temperature T0 were calculated.

[0101] In this embodiment, the corresponding constant A is calculated under different temperature conditions. i and the rate constant of performance change K i and the constant A i and the rate constant of performance change K i By associating the data with temperature, it becomes easier to find patterns of temperature change using regression algorithms.

[0102] S23. By integrating the parameters of the rubber aging prediction model obtained from natural environment tests and accelerated laboratory thermal aging tests, the parameters of the rubber aging prediction model for the rubber under actual storage conditions are determined. The specific method is as follows:

[0103] Using weight matrix Let A be the acceleration constant. i And the natural constant A0, and the natural performance change rate constant K0 and the acceleration performance change rate constant K i Assign weights;

[0104] The specific KT and AT relationships are obtained by solving the convergent weight matrices W and β using the adaptive weighted least squares method.

[0105] Based on the actual storage environment temperature of the rubber to be tested, determine the constants A and the performance change rate constant K in the rubber aging prediction model of the rubber to be tested.

[0106] In the weight matrix, ω i The calculation formula is:

[0107]

[0108] In this embodiment, the weight matrix can be adjusted as needed, primarily to reflect the ratio of the difference between the actual storage temperature and the experimental temperature. In this embodiment, a regression algorithm is used to fit the relationship between temperature and parameters; however, other machine learning algorithms or algorithms capable of fitting functions can also be used.

[0109] S3, the third prediction step;

[0110] Based on the rubber aging prediction model with confirmed parameters, the storage life of rubber is predicted.

[0111] In this embodiment, once the parameters of the rubber aging prediction model are determined, the storage life of the rubber can be predicted.

[0112] S4, Fourth Model Evaluation and Error Analysis Steps;

[0113] Calculate a storage temperature Q based on the model. s Below, the aging time is Performance prediction value at time Using predicted values ​​and actual values Model evaluation can be performed by calculating the sum of squared errors (SSE), root mean square error (RMSE), and coefficient of determination (R²). 2 The closer the sum of squared errors and root mean square error are to 0, and the closer the coefficient of determination is to 1, the better the model prediction.

[0114] SSE=Σ(Pi r -P i s ) 2

[0115]

[0116]

[0117] The error at each node is evaluated using relative error erro, and the absolute mean of the error is used. Perform an overall error assessment of the model;

[0118]

[0119]

[0120] This step is optional and its purpose is to evaluate the prediction model and provide data support for subsequent model correction.

[0121] Example 2:

[0122] Step 1: In accordance with GJB 8893-2017 "Test Methods for Natural Environment of Military Equipment" and GB / T 3512-2014, natural environment storage test and laboratory accelerated thermal aging storage test of FS6265 fluorosilicone rubber were carried out under the shed in Jiangjin. The accelerated test temperatures were 90℃ / 100℃ / 110℃ / 120℃, and the compression set retention rate P data at different aging times τ were collected, as shown in Table 1.

[0123] Table 1. Compression set retention rate of samples at different aging times under different test conditions.

[0124]

[0125] Step 2: Select laboratory accelerated thermal aging test data P ij -τ ij Using the successive approximation method as the research object, the parameter α is calculated to be 0.52.

[0126] Step 3: The constant A and the performance change rate constant K under different test conditions were calculated using the least squares method. The results are shown in Table 2.

[0127] Table 2. Constants A and performance change rate constants K under different test conditions

[0128]

[0129] Step 4: Solve for the activation energy E using the Arrhenius model / adaptive weighted linear regression. a and frequency factor Z;

[0130] Where K = Zexp(-E) a / RT), where R is the gas constant, and the results are shown in Table 3.

[0131] Table 3 Activation energy E a and frequency factor Z iteration results

[0132]

[0133] The results show that, after adaptive weighted linear regression, the activation energy E of the sample is... a It is 14.068 kJ·mol -1 The frequency factor Z is 1.217d. -1 According to formula (20), the value of A is 0.974. Therefore, the performance degradation law model of FS6265 fluorosilicone rubber with aging time is determined as follows:

[0134] P = 0.974·exp(-Kτ) 0.52 )

[0135] Where K = 1.217·exp(-1692.128 / T)

[0136] Step 5: For the Jiangjin shed environment, with an average annual temperature of 293K, the performance degradation law model established according to the above formula is P=0.980·exp(-0.00378·τ 0.52 Based on the model, the predicted performance values ​​at different aging times were calculated, and combined with the measured performance values, the model evaluation and error analysis results were obtained, as shown in Table 4.

[0137] Table 4. Model Evaluation and Error Analysis Results

[0138]

[0139] Step 6: Verify the model's prediction performance

[0140] To verify the good predictive performance of the data fusion modeling method proposed in this invention, mathematical model I based solely on laboratory accelerated test data and mathematical model II based solely on natural environment test data were established. The method for establishing model II followed GJB 92.2-86 "Determination of Storage Properties of Vulcanized Rubber by Hot Air Aging Method - Part II: Statistical Methods," while model I was established using Matlab nonlinear fitting tools. The results are as follows:

[0141] Model I, P = 0.977·exp(-0.00232·τ) 0.52 )

[0142] Model II, P = exp(-0.01551·τ)0.30 )

[0143] The three models described above were used to predict the performance values ​​of rubber samples after 720 days of aging, and the errors between the predicted and measured values ​​were compared. The results are shown in Table 5. Figure 2 As shown in the figure. The results demonstrate that the fusion model proposed in this invention has good predictive performance.

[0144] Table 5 Prediction results of different models

[0145]

[0146] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0147] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0148] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0149] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0150] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method of predicting the aging life of a rubber, characterized by, The method comprises: a first modeling step, a second parameter determination step and a third prediction step; wherein, the first modeling step establishes a rubber aging prediction model; the second parameter determination step determines parameters of the rubber aging prediction model of the rubber to be tested under actual storage environment by combining rubber natural environment test data and laboratory thermal aging accelerated test data; the third prediction step predicts the storage life of the rubber according to the rubber aging prediction model after parameter determination. In the first modeling step, the specific formula of the rubber aging prediction model is: P = A exp(-Kτ α ) wherein, P is a performance change index, which can be compression permanent set retention rate, compression stress relaxation coefficient or elongation change rate at break; τ is aging time; K is a performance change rate constant, which is related to temperature; A is a constant; α is a constant; The calculation method of the constant α is: Let X = τ α ; Y = ln P; a = ln A; b = -K; the rubber aging prediction model is expressed as Y = a + bX; Selecting the laboratory thermal aging accelerated test data (P ij -τ ij ), if X and Y meet the strong linear relationship, α takes the value of 1; otherwise, the value of α is determined by successive approximation method; Rubber aging test data P ij -τ ij , represents the i-th test temperature T i , different aging time τ i1 ,τ i2 ,τ i3 ,…,τ ij , there are different performance change indicators P i1 ,P i2 ,P i3 ,…P ij (i=1,2,…,m; j=1,2,…,n); Different temperatures T i The constant A and the performance change rate constant K are calculated as follows. The least square method is used to solve the coefficients a and b in the formula Y=a+bX, and the constant A and the performance change rate constant K are obtained: A=exp(a) K=-b According to the laboratory thermal aging accelerated test data, the different test temperature T i The corresponding acceleration constant A i And the acceleration performance change speed constant K i ; According to the rubber natural environment test data, the natural test temperature T0 corresponds to the natural constant A0 and the natural performance change rate constant K0. The parameters of the rubber aging prediction model calculated by fusing the rubber natural environment test and the laboratory thermal aging accelerated test are determined to determine the parameters of the rubber aging prediction model of the rubber to be tested under actual storage environment, and the specific method is: using a weight matrix for the acceleration constant A i and the natural constant A0, and the natural rate of change of performance constant K0 and the acceleration rate of change of performance constant K i weighting The weight matrix W and β are solved by the adaptive weighted least square method to obtain the specific K-T relationship and A-T relationship; According to the actual storage environment temperature of the rubber to be tested, the constant A and the performance change rate constant K in the rubber aging prediction model of the rubber to be tested are determined.

2. The rubber aging life prediction method according to claim 1, characterized by, In the second parameter determination step, the rubber natural environment test and the laboratory thermal aging accelerated test are combined, and the specific method is: According to the temperature of the rubber to be tested under actual storage environment, the parameters of the rubber aging prediction model calculated under different test temperatures are given weights; The closer the rubber aging test temperature is to the temperature of the rubber to be tested under actual storage environment, the higher the weight value of the parameters of the rubber aging prediction model calculated by it. In the second parameter determination step, the parameters of the rubber aging prediction model of the rubber to be tested under actual storage environment are determined, and the specific method is: According to the fused parameters of the rubber aging prediction model, a regression algorithm is used to fit the relationship function between the parameters of the rubber aging prediction model and the temperature, and the parameters in the rubber aging prediction model are determined according to the temperature of the rubber to be tested under actual storage environment.

3. The rubber aging life prediction method according to claim 1, characterized by, To determine whether X and Y meet the strong linear relationship, at least the following conditions must be met: Set the value of α to 1, and use the least squares method to find the correlation coefficient r; look up the correlation coefficient table for the value of r with a confidence level of 99%, and the value of f = n - 2 degrees of freedom cal Compare, |r cal |>r wherein 4. The rubber aging life prediction method according to claim 1, characterized by, The value of α is determined by successive approximation method, and the approximation criterion is that when α is accurate to two decimal places, I is minimized. In the formula, P ij represents the performance change index test value of the jth test point at the ith aging test temperature; represents the performance change index predicted value of the jth test point at the ith aging test temperature.

5. The rubber aging life prediction method according to claim 1, wherein In the weight matrix, ω i The calculation formula is:

6. The rubber aging life prediction method according to claim 1, wherein In the weight matrix, ω i The calculation formula is: wherein, K' i and T' i are dimensionless performance change constants and temperature.

Images