CN117669164A - Method for optimizing fatigue life data - Google Patents
Method for optimizing fatigue life data Download PDFInfo
- Publication number
- CN117669164A CN117669164A CN202311554508.1A CN202311554508A CN117669164A CN 117669164 A CN117669164 A CN 117669164A CN 202311554508 A CN202311554508 A CN 202311554508A CN 117669164 A CN117669164 A CN 117669164A
- Authority
- CN
- China
- Prior art keywords
- data
- fatigue life
- fusion
- stress level
- stress
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 23
- 230000004927 fusion Effects 0.000 claims abstract description 106
- 238000009826 distribution Methods 0.000 claims abstract description 22
- 230000002159 abnormal effect Effects 0.000 claims abstract description 10
- 239000007769 metal material Substances 0.000 claims abstract description 8
- 238000012360 testing method Methods 0.000 claims description 16
- 238000009661 fatigue test Methods 0.000 claims description 7
- 238000007499 fusion processing Methods 0.000 claims description 7
- 238000013441 quality evaluation Methods 0.000 claims description 7
- 239000002184 metal Substances 0.000 claims description 3
- 102100037651 AP-2 complex subunit sigma Human genes 0.000 claims description 2
- 101000806914 Homo sapiens AP-2 complex subunit sigma Proteins 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 4
- 238000005516 engineering process Methods 0.000 abstract description 3
- 238000005457 optimization Methods 0.000 description 9
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 206010043417 Therapeutic response unexpected Diseases 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Investigating Strength Of Materials By Application Of Mechanical Stress (AREA)
Abstract
The invention belongs to the technology of optimizing the quality of fatigue life data of a small sample of a metal material and eliminating experimental abnormal data, and relates to a method for optimizing the fatigue life data. The invention realizes the expansion of small sample data based on the consistency principle, and assumes that when the data quality is consistent, the fatigue life data fused to the same stress level has the same distribution. Therefore, the quality of the data can be checked by calculating the distribution difference of the large sample data obtained based on the fusion of the fatigue life data of different small samples. By repeatedly eliminating different data and carrying out fatigue life data fusion, the result finds that the data of the same stress level can be more similar to the fused data of the same stress level under different conditions, which indicates that the data quality is improved after the data is eliminated, and the effect of optimizing the data quality is achieved.
Description
Technical Field
The invention belongs to the technology of optimizing the quality of fatigue life data of a small sample of a metal material and eliminating experimental abnormal data, and relates to a method for optimizing the fatigue life data.
Background
In practical engineering applications, fatigue is a common damaging behavior. In order to prevent the serious economic loss caused by the damage behavior, a large number of fatigue test tests are required, but due to the high cost of the fatigue test, a small number of tests are often carried out, and then the fatigue life is predicted by a model prediction or other data processing means, so that the fatigue life prediction with high quality is realized, and not only a good-effect prediction model is required, but also good-quality fatigue life test data are required. However, the fatigue life has great discreteness, so that the judgment of the quality of the fatigue life data is difficult, whether the experimental abnormal data exist or not is difficult to determine only by a small amount of experimental data, and when the data expansion or prediction is performed on the basis of the small amount of experimental data, a great prediction error can be caused by only one experimental data abnormality. Therefore, in order to avoid a large experimental error in the experiment, it is necessary to examine the fatigue life experimental data, so that high-quality fatigue life data prediction can be realized.
The data optimization method utilizes a data fusion method under different stress conditions proposed by Xie Liyang and the like to convert small sample data into large sample data through a consistency principle. And based on the consistency principle, it can be found that when the fatigue life data meets normal distribution, under the condition of the same data quality, different large sample data under the same stress level obtained by utilizing different small sample data have the same distribution condition, so that the data quality can be judged according to the difference between different large sample data under the same stress, when the large difference exists, the data deviating from the overall rule exists in the small sample data, the effect of data optimization can be achieved by eliminating the data, and the method searches experimental data with larger error and eliminates the experimental data through repeated iterative test, thereby realizing the data quality optimization.
Disclosure of Invention
The invention aims to provide a method for optimizing the quality of metal material fatigue life data, which has high feasibility. The method realizes the expansion of small sample data based on the consistency principle, and assumes that when the data quality is consistent, the fatigue life data fused to the same stress level has the same distribution. Therefore, the quality of the data can be checked by calculating the distribution difference of the large sample data obtained based on the fusion of the fatigue life data of different small samples. By repeatedly eliminating different data and carrying out fatigue life data fusion, the result finds that the data of the same stress level can be more similar to the fused data of the same stress level under different conditions, which indicates that the data quality is improved after the data is eliminated, and the effect of optimizing the data quality is achieved.
The invention is realized by the following technical scheme: a method of optimizing fatigue life data, comprising the steps of:
step 1: selecting at least 3 stress levels above the fatigue limit for performing a metal material fatigue test, and obtaining at least 3 groups of fatigue life data under each group of stress levels;
step 2: performing preliminary data quality evaluation on each group of fatigue life data, and eliminating obvious experimental abnormal data;
step 3: performing data fusion on the fatigue life data obtained in the step 2, and determining the stress level of the fused data;
step 4: combining the stress level determined in the step 3 with the stress level determined in the step 1 respectively, combining three stress levels to form 2 groups, and fusing fatigue life data of each group under different stress levels; and calculating the mean value mu 1, mu 2 and the standard deviation sigma 1, sigma 2;
step 5: under the principle of consistency checking, checking the quality of the fusion data of the output fusion data;
step 6: repeating steps 3 to 5; realizing the fatigue data fusion under different stress levels;
step 7: and (3) eliminating any experimental data participating in data fusion in the step (4), repeating the steps (4) to (6) to form new fusion data, respectively calculating the mean value and standard deviation of the new fusion data and the original fusion data, judging whether the error between the mean value and standard deviation of the new fusion data and the original fusion data is larger than a preset error d, eliminating the data if the error is larger than the preset error d, otherwise, repeating the step (7).
The step 4 specifically comprises the following steps:
step 4-1: the relationship between the metal fatigue life distribution and the stress is proposed as follows:
σ i =σ 1 +k(s 1 -s i ) (1)
σ i logarithmic fatigue life standard deviation at the i-th stress level, i=1, 2;
σ 1 the standard deviation of logarithmic fatigue life under the stress level of the fusion direction;
s i is the ith stress level size;
s 1 stress level magnitude in the selected fusion direction;
k is a coefficient to be determined, and the range of the k value of the coefficient to be determined is 0-1;
step 4-2: fatigue life distribution assumptions are presented: the logarithm of fatigue life under each stress level obeys normal distribution;
step 4-3: providing fatigue life relations under different stress levels;
step 4-4: calculating the average value mu of the test fatigue life data logarithm under each stress level;
step 4-5: setting the k value range of the coefficient to be determined in the step 4-1;
step 4-6: setting convergence conditions in the data fusion process:
step 4-7: dividing a range of k values of the coefficient to be determined into cells, and calculating whether the fusion result meets a convergence condition under the condition of each k value;
step 4-8: and outputting the fusion data meeting the convergence condition.
The fatigue life relationship at different stress levels is the consistency principle.
The k value range of the 4-5 predetermined coefficient is 0-1.
The fatigue life relationship under different stress levels in the step 4-3 is as follows:
N kj fatigue life for j samples converted to k stress levels; n (N) ij Fatigue life for j samples at i stress level; sigma (sigma) k The standard deviation of logarithmic fatigue life at k stress levels; sigma (sigma) i The standard deviation of logarithmic fatigue life at i stress level; mu (mu) k Is the logarithmic fatigue life average at k stress levels; mu (mu) i Is the logarithmic fatigue life average at i stress levels.
The 4-1 fusion direction is as follows: and determining the stress level above the fatigue limit as the fusion direction.
The convergence condition in the fusion process of the step 4-6 is as follows:
Δ=Abs(σ 1 -σ right )<1×10 -6 (3)。
the quality of the optimized fatigue life data is determined according to the relationship between the error between the mean value and the standard deviation of the new fusion data and the original fusion data and the preset error.
The invention has the technical effects that:
according to the invention, data fusion is carried out on the small sample data based on the consistency principle to form large sample data, and the small sample data is removed through iteration, so that the optimization of the quality of the small sample data is realized. The result shows that the experimental data deviating from the overall data distribution can be simply determined by the method, so that poor data are removed to realize fatigue life data optimization, and the practical fatigue life prediction of engineering is of great significance.
Drawings
FIG. 1 is a flow chart of data quality optimization according to the present invention;
FIG. 2, different large sample data distributions at 250MPa;
fig. 3, different large sample data distributions at 650 MPa.
Detailed Description
The invention is further illustrated by the following examples in conjunction with the accompanying drawings:
a method of optimizing fatigue life data, comprising the steps of:
step 1: selecting at least 3 stress levels above the fatigue limit for performing a metal material fatigue test, and obtaining at least 3 groups of fatigue life data under each group of stress levels;
step 2: performing preliminary data quality evaluation on each group of test data, eliminating obvious experimental abnormal data, and ensuring the reliability of fatigue life data;
step 3: determining the stress level of the fusion data;
step 4: carrying out fatigue life data fusion under different stress levels;
the specific process of the step 4 is as follows:
step 4-1: the relationship between the metal fatigue life distribution and the stress is proposed as follows:
σ i =σ 1 +k(s 1 -s i ) (4)
σ i logarithmic fatigue life standard deviation at the i-th stress level, i=1, 2;
σ 1 the standard deviation of logarithmic fatigue life under the stress level of the fusion direction;
s i is the ith stress level size;
s 1 stress level magnitude in the selected fusion direction;
k is a coefficient to be determined;
step 4-2: fatigue life distribution assumptions are presented: the logarithm of fatigue life under each stress level obeys normal distribution;
step 4-3: providing fatigue life relations under different stress levels, namely a consistency principle;
N kj fatigue life for j samples converted to k stress levels;
N ij fatigue life for j samples at i stress level;
σ k the standard deviation of logarithmic fatigue life at k stress levels;
σ i the standard deviation of logarithmic fatigue life at i stress level;
μ k is the logarithmic fatigue life average at k stress levels;
μ i is the logarithmic fatigue life average at i stress level;
step 4-4: calculating the average value mu of the test fatigue life data logarithm under each stress level;
step 4-5: setting the k value range of the coefficient to be determined in the step 4-1;
step 4-6: setting convergence conditions in the fusion process of the number:
Δ=Abs(σ 1 -σ right )<10 -6 (6)
σ right calculating the standard deviation of logarithmic fatigue life under the fusion stress level after data fusion;
step 4-7: dividing a range of k values of the coefficient to be determined into cells, and calculating whether the fusion result meets a convergence condition under the condition of each k value;
step 4-8: outputting the fusion data meeting the convergence condition;
step 5: checking the quality of fusion data of fatigue life data fusion under the consistency principle;
step 6: repeating steps 3 to 5; realizing the fatigue data fusion under different stress levels;
step 7: and (3) eliminating one experimental data participating in data fusion in the step (4), repeating the steps (4) to (6) to form new fusion data, respectively calculating the mean value and standard deviation of the new fusion data and the original fusion data, judging whether the error between the mean value and standard deviation of the new fusion data and the original fusion data is larger than a preset error d, eliminating the data if the error is larger than the preset error d, otherwise, repeating the step (7). As shown in the flow chart of fig. 1.
Example 1
The invention is realized by the following technology:
step 1: selecting at least 3 stress levels above the fatigue limit for performing a metal material fatigue test, and obtaining at least 3 groups of fatigue life data under each group of stress levels; case 1 the following experimental data were selected and are shown in Table 1
TABLE 1 fatigue life experimental data
Step 2: performing preliminary data quality evaluation on each group of test data, eliminating obvious experimental abnormal data, and ensuring the reliability of fatigue life data; preliminary data quality evaluation is carried out on each group of test data in the table 1, and no obvious experimental abnormal data are found;
step 3: determining the stress level of the fusion data to be 250MPa;
step 4: and selecting data under the stress levels of 250, 300 and 350MPa for data fusion, and selecting data under the stress levels of 250, 400 and 500MPa for data fusion. Respectively obtaining fusion data 1 and fusion data 2 under 250MPa, and calculating standard deviation errors and mean values between the fusion data 1 and the fusion data 2; the detailed steps for realizing data fusion are shown in the steps 4-1 to 4-8
Step 4-1: the basic theoretical formula for data fusion is as follows:
σ i =σ 1 +k(s 1 -s i ) (7)
σ i logarithmic fatigue life standard deviation at the i-th stress level, i=1, 2;
σ 1 logarithmic fatigue life at stress level for the fusion directionStandard deviation;
s i is the ith stress level size;
s 1 stress level magnitude in the selected fusion direction;
k is a coefficient to be determined;
step 4-2: assuming that the fatigue life logarithm of each stress level is under 250MPa-500MPa in the table 1, and obeying normal distribution;
step 4-3: providing fatigue life relations under different stress levels, namely a consistency principle;
N kj fatigue life for j samples converted to k stress levels;
N ij fatigue life for j samples at i stress level;
σ k the standard deviation of logarithmic fatigue life at k stress levels;
σ i the standard deviation of logarithmic fatigue life at i stress level;
μ k is the logarithmic fatigue life average at k stress levels;
μ i is the logarithmic fatigue life average at i stress level;
step 4-4: the average μ of the log of the test fatigue life data at each stress level was calculated and the results are shown in table 2:
TABLE 2 logarithmic fatigue life mean
Step 4-5: setting the range of k values of the coefficient to be determined in the step 4-1 to be 0-1 multiplied by 10-5;
step 4-6: setting convergence conditions in the fusion process of the number:
Δ=Abs(σ 1 -σ right )<1×10 -6 (9)
σ right computing fusion for post-fusion dataStandard deviation of logarithmic fatigue life at stress level;
step 4-7: calculating whether the fusion result meets the convergence condition under the condition of each k value:
step 4-8: outputting the fusion data meeting the convergence condition;
step 5: checking the quality of fusion data of fatigue life data fusion under the consistency principle;
step 6: repeating steps 3 to 5; realizing the fatigue data fusion under different stress levels;
step 7: removing one experimental data participating in data fusion in the step 4, repeating the steps 4 to 6 to form new fusion data, respectively calculating the mean value and standard deviation of the new fusion data and the original fusion data, judging whether the error between the mean value and standard deviation of the new fusion data and the original fusion data is larger than a preset error d, removing the data if yes, otherwise, repeating the step 7;
step 8: outputting the fusion data meeting the convergence condition and drawing a life distribution image, as shown in fig. 2;
when the set error d=5%, 171168 experimental data under 300MPa are removed, data under 250, 300 and 350MPa stress levels are selected for data fusion, and new experimental data are obtained through removing 171168 in steps 1-7, and are shown in table 3.
TABLE 3 fatigue life test data after optimization
In the method, fatigue life experimental data in the table 1 are screened by adopting a method for optimizing the fatigue life, and the 171168 data in the table 1 under 300MPa is finally determined to be poor in quality by carrying out iterative elimination check on different data, so that the data quality optimization is realized by eliminating the data.
Example 2
Step 1: selecting at least 3 stress levels above the fatigue limit for performing a metal material fatigue test, and obtaining at least 3 groups of fatigue life data under each group of stress levels; case 2 the following experimental data were selected and are shown in Table 4
Table 4 fatigue life experimental data
Step 2: performing preliminary data quality evaluation on each group of test data, eliminating obvious experimental abnormal data, and ensuring the reliability of fatigue life data; preliminary data quality evaluation is carried out on each group of test data in table 4, and no obvious abnormal test data are found;
step 3: determining the stress level of the fusion data to be 650MPa;
step 4: data under 650, 800 and 1000MPa stress levels are selected for data fusion, fusion data 1 and fusion data 2 under 650MPa are obtained, and standard deviation errors and mean values between the fusion data 1 and the fusion data 2 are calculated; the detailed steps for realizing data fusion are shown in the steps 4-1 to 4-8
Step 4-1: the basic theoretical formula for data fusion is as follows:
σ i =σ 1 +k(s 1 -s i ) (1)
σ i logarithmic fatigue life standard deviation at the i-th stress level, i=1, 2;
σ 1 the standard deviation of logarithmic fatigue life under the stress level of the fusion direction;
s i is the ith stress level size;
s 1 stress level magnitude in the selected fusion direction;
k is a coefficient to be determined;
step 4-2: assuming the fatigue life logarithm of each stress level under 650, 800 and 1000MPa in the table 1, obeying normal distribution;
step 4-3: providing fatigue life relations under different stress levels, namely a consistency principle;
N kj fatigue life for j samples converted to k stress levels;
N ij fatigue life for j samples at i stress level;
σ k the standard deviation of logarithmic fatigue life at k stress levels;
σ i the standard deviation of logarithmic fatigue life at i stress level;
μ k is the logarithmic fatigue life average at k stress levels;
μ i is the logarithmic fatigue life average at i stress level;
step 4-4: the average μ of the log of the test fatigue life data at each stress level was calculated and the results are shown in table 5:
TABLE 5 logarithmic fatigue life mean
Step 4-5: setting the range of k values of the coefficient to be determined in the step 4-1 to 0-1 multiplied by 10 -5 ;
Step 4-6: setting convergence conditions in the fusion process of the number:
Δ=Abs(σ 1 -σ right )<1×10 -6 (3)
σ right calculating the standard deviation of logarithmic fatigue life under the fusion stress level after data fusion;
step 4-7: calculating whether the fusion result meets the convergence condition under the condition of each k value;
step 4-8: outputting the fusion data meeting the convergence condition;
step 5: checking the quality of fusion data of fatigue life data fusion under the consistency principle;
step 6: repeating steps 3 to 5; realizing the fatigue data fusion under different stress levels;
step 7: removing one experimental data participating in data fusion in the step 4, repeating the steps 4 to 6 to form new fusion data, respectively calculating the mean value and standard deviation of the new fusion data and the original fusion data, judging whether the error between the mean value and standard deviation of the new fusion data and the original fusion data is larger than a preset error d, removing the data if yes, otherwise, repeating the step 7;
step 8: outputting the fusion data meeting the convergence condition and drawing a life distribution image, as shown in fig. 3;
when the setting error is d=5%, the 55037 experiment data under 1000MPa is searched out in the steps 1-7, and new data are obtained by eliminating.
TABLE 6 fatigue life experimental data
In the example, fatigue life experimental data in table 4 are screened by adopting a method for optimizing the fatigue life data, and the 55037 data quality in table 4 under 1000MPa is finally determined to be poor by carrying out iterative elimination check on different data, so that the data quality optimization is realized by eliminating the data.
Claims (8)
1. A method of optimizing fatigue life data, comprising the steps of:
step 1: selecting at least 3 stress levels above the fatigue limit for performing a metal material fatigue test, and obtaining at least 3 groups of fatigue life data under each group of stress levels;
step 2: performing preliminary data quality evaluation on each group of fatigue life data, and eliminating obvious experimental abnormal data;
step 3: performing data fusion on the fatigue life data obtained in the step 2, and determining the stress level of the fused data;
step 4: combining the stress level determined in the step 3 with the stress level determined in the step 1 respectively, combining three stress levels to form 2 groups, and fusing fatigue life data of each group under different stress levels; and calculating the mean value mu 1, mu 2 and the standard deviation sigma 1, sigma 2;
step 5: under the principle of consistency checking, checking the quality of the fusion data of the output fusion data;
step 6: repeating steps 3 to 5; realizing the fatigue data fusion under different stress levels;
step 7: and (3) eliminating any experimental data participating in data fusion in the step (4), repeating the steps (4) to (6) to form new fusion data, respectively calculating the mean value and standard deviation of the new fusion data and the original fusion data, judging whether the error between the mean value and standard deviation of the new fusion data and the original fusion data is larger than a preset error d, eliminating the data if the error is larger than the preset error d, otherwise, repeating the step (7).
2. The method for optimizing fatigue life data according to claim 1, wherein the step 4 is specifically:
step 4-1: the relationship between the metal fatigue life distribution and the stress is proposed as follows:
σ i =σ 1 +k(s 1 -s i ) (1)
σ i logarithmic fatigue life standard deviation at the i-th stress level, i=1, 2;
σ 1 the standard deviation of logarithmic fatigue life under the stress level of the fusion direction;
s i is the ith stress level size;
s 1 stress level magnitude in the selected fusion direction;
k is a coefficient to be determined, and the range of the k value of the coefficient to be determined is 0-1;
step 4-2: fatigue life distribution assumptions are presented: the logarithm of fatigue life under each stress level obeys normal distribution;
step 4-3: providing fatigue life relations under different stress levels;
step 4-4: calculating the average value mu of the test fatigue life data logarithm under each stress level;
step 4-5: setting the k value range of the coefficient to be determined in the step 4-1;
step 4-6: setting convergence conditions in the data fusion process:
step 4-7: dividing a range of k values of the coefficient to be determined into cells, and calculating whether the fusion result meets a convergence condition under the condition of each k value;
step 4-8: and outputting the fusion data meeting the convergence condition.
3. A method of optimizing fatigue life data according to claim 2, wherein the fatigue life relationship at different stress levels is a consistency principle.
4. A method of optimizing fatigue life data according to claim 2, wherein: the k value range of the 4-5 predetermined coefficient is 0-1.
5. A method of optimizing fatigue life data as claimed in claim 1, wherein: the fatigue life relationship under different stress levels in the step 4-3 is as follows:
N kj fatigue life for j samples converted to k stress levels; n (N) ij Fatigue life for j samples at i stress level; sigma (sigma) k The standard deviation of logarithmic fatigue life at k stress levels; sigma (sigma) i The standard deviation of logarithmic fatigue life at i stress level; mu (mu) k Is the logarithmic fatigue life average at k stress levels; mu (mu) i Is the logarithmic fatigue life average at i stress levels.
6. A method of optimizing fatigue life data according to claim 1, wherein: the 4-1 fusion direction is as follows: and determining the stress level above the fatigue limit as the fusion direction.
7. A method of optimizing fatigue life data as claimed in claim 1, wherein: the convergence condition in the fusion process of the step 4-6 is as follows:
Δ=Abs(σ 1 -σ right )<1×10 -6 (3)。
8. the method of optimizing fatigue life data according to claim 1, wherein the quality of the optimized fatigue life data is determined based on determining a relationship between an error between a mean and a standard deviation of new and original fusion data and a preset error.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311554508.1A CN117669164A (en) | 2023-11-21 | 2023-11-21 | Method for optimizing fatigue life data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311554508.1A CN117669164A (en) | 2023-11-21 | 2023-11-21 | Method for optimizing fatigue life data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117669164A true CN117669164A (en) | 2024-03-08 |
Family
ID=90085508
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311554508.1A Pending CN117669164A (en) | 2023-11-21 | 2023-11-21 | Method for optimizing fatigue life data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117669164A (en) |
-
2023
- 2023-11-21 CN CN202311554508.1A patent/CN117669164A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109977491B (en) | Degradation modeling and service life prediction method under condition of impact damage restorable | |
CN108536017B (en) | Random distribution interconnection system cooperation fault-tolerant control method based on dynamic feedback control | |
CN113567795B (en) | Step-back stress acceleration reliability test method based on Weibull distribution | |
CN108132914B (en) | High-confidence statistical inference and confirmation method for failure rate of limited number of product performances | |
CN116305522A (en) | Digital twin-based aircraft structure reliability simulation test model calibration method, device and medium | |
Lai et al. | Robust design of assembly parameters on membrane electrode assembly pressure distribution | |
CN116187838A (en) | Quality evaluation method, system and device for power equipment and storage medium | |
CN117669164A (en) | Method for optimizing fatigue life data | |
WO2019019429A1 (en) | Anomaly detection method, device and apparatus for virtual machine, and storage medium | |
CN113761755A (en) | Accelerated life analysis method considering cognitive uncertainty under temperature and humidity dual stress | |
CN112529209A (en) | Model training method, device and computer readable storage medium | |
CN110348167B (en) | Determination of metal detail fatigue rated strength cut-off value DFRcutoffIs a probabilistic model method | |
CN116469448B (en) | Flash memory particle screening method and device | |
CN113536490B (en) | Sucker rod fatigue life prediction method and system | |
CN115563824A (en) | Method for predicting performance of dual-phase material composite tube based on machine learning | |
CN114791553A (en) | Chip failure parameter prediction method | |
CN108320099B (en) | Industrial analysis and calculation element analysis method based on large sample and utilizing coal | |
CN112231886A (en) | Method for predicting failure time of components | |
CN107766884B (en) | Bayes fusion evaluation method based on representative point optimization | |
CN109325289B (en) | Method for estimating reliability parameters of electronic part | |
CN109870257B (en) | Method for predicting distribution of quenching residual stress in thickness direction of plate | |
CN112445632A (en) | HPC reliability evaluation method based on fault data modeling | |
CN111241614A (en) | Engineering structure load inversion method based on condition generation confrontation network model | |
CN112765034B (en) | Software defect prediction method based on neural network | |
CN113593651B (en) | Conductive silver paste component design and performance prediction method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |