Material fatigue limit determination method, system and equipment and computer medium
By generating virtual experimental sequences to process disordered fatigue data, the problem of high cost in determining the fatigue limit of materials is solved, and rapid, low-cost and reliable fatigue limit results are generated.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NAT HIGH SPEED TRAIN QINGDAO TECH INNOVATION CENT
- Filing Date
- 2025-12-31
- Publication Date
- 2026-04-17
AI Technical Summary
In existing technologies, the lack of complete rise and fall sequence information makes it impossible to quickly and cost-effectively determine the fatigue limit of materials, resulting in low utilization of experimental data and high costs.
By acquiring a disordered fatigue test dataset, a virtual test sequence is generated using the backtracking method. The stress-result data are then combined into a virtual test sequence that satisfies the set path conditions according to the rise and fall method rules, thereby generating the fatigue limit result value of the material.
It enables the reuse of disordered fatigue data, quickly and cost-effectively determining the fatigue limit of materials, and ensuring the reliability and robustness of the results.
Smart Images
Figure CN121885044A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of fatigue limit testing technology, and more specifically, to a method, system, device, and computer medium for determining the fatigue limit of materials. Background Technology
[0002] Engineering structures are subjected to various forms of repetitive loads during service, such as the periodic vibrations generated by vehicle movement, the continuous aerodynamic cyclic loads on aircraft wings during flight, and the periodic stresses caused by alternating forces in operating machinery. In the field of metallic materials, such "long-term, repetitive" loading often leads to a failure mode called fatigue failure. Fatigue failure typically occurs at stress levels far below the static strength; therefore, it is essential to accurately determine the fatigue limit of materials in engineering design, for example, through methods such as the rise-fall method.
[0003] However, in real-world engineering environments, data acquisition and management are often complex. On one hand, a large amount of historical fatigue data is accumulated gradually from different projects and batches of tests, recording only the stress level and failure / non-failure results at that time. On the other hand, experimental records may not be fully preserved due to the passage of time, personnel changes, or non-standard management methods, resulting in a large number of datasets that only have "stress + result" information but lack a complete stress-load sequence. In this situation, since the core rule of the stress-load method is: if the current specimen fails, the stress of the next specimen decreases by a fixed step; if the current specimen does not fail, the stress of the next specimen increases by a fixed step; it relies on a strict experimental sequence. Once the sequence information is missing, the stress-load method cannot determine the material fatigue limit. In other words, the stress-load method cannot be used to process datasets with only "stress + result" information but lacking a complete stress-load sequence to determine the fatigue limit. A complete stress-load fatigue experiment must be rearranged and executed, which not only leads to low utilization of the material fatigue test dataset but also increases the cost and time required to determine the material fatigue limit.
[0004] In summary, how to quickly and cost-effectively determine the fatigue limit of materials is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention
[0005] The purpose of this application is to provide a method for determining the fatigue limit of materials, which can, to some extent, solve the technical problem of how to determine the fatigue limit of materials quickly and at low cost. This application also provides a system for determining the fatigue limit of materials, an electronic device, and a computer-readable storage medium.
[0006] To achieve the above objectives, this application provides the following technical solution:
[0007] A method for determining the fatigue limit of a material, comprising:
[0008] Obtain a disordered fatigue test dataset of the material to be analyzed. The disordered fatigue test dataset includes only the stress level and cycle results of the material specimens to be analyzed, and the cycle results include failure or non-failure.
[0009] Determine the initial stress;
[0010] Use the initial stress as the current stress for the current path;
[0011] In the disordered fatigue test dataset, the specimen corresponding to the current stress is added to the current path;
[0012] If the current stress level of the specimen does not fail, the current stress is increased; if the current stress level of the specimen fails, the current stress is decreased.
[0013] Return to the disordered fatigue test dataset and add the specimen corresponding to the current stress to the current path and subsequent steps until a current path that meets the set path conditions is obtained, and use the current path as a virtual test sequence.
[0014] Based on the virtual experimental sequence, fatigue limit results of the material to be analyzed are generated.
[0015] Preferably, adding the specimen corresponding to the current stress to the current path in the disordered fatigue test dataset includes:
[0016] In the disordered fatigue test dataset, the specimens that correspond to the current stress and have not been used are selected as candidate specimens.
[0017] Traverse the candidate samples and add the traversed candidate samples to the current path.
[0018] Preferably, the step of increasing the current stress in response to the stress level of the specimen not failing and decreasing the current stress in response to the stress level of the specimen failing includes:
[0019] Obtain the set stress step size;
[0020] If the current stress level of the specimen does not indicate failure, the stress step size is increased.
[0021] In response to the stress level failure of the specimen corresponding to the current stress, the current stress is reduced by the stress step.
[0022] Preferably, the step of returning to the disordered fatigue test dataset involves adding the specimen corresponding to the current stress to the current path and subsequent paths until a current path that meets the set path conditions is obtained, and using the current path as a virtual test sequence, including:
[0023] Get the set maximum number of generated paths and minimum virtual sequence length values;
[0024] If, in response to the existence of a specimen corresponding to the current stress in the disordered fatigue test dataset, and the length of the current path is less than the maximum number of generated paths, the process returns to the step of adding the specimen corresponding to the current stress to the current path and subsequent paths in the disordered fatigue test dataset.
[0025] In response to the absence of a specimen corresponding to the current stress in the disordered fatigue test dataset, or the length of the current path being equal to the maximum number of generated paths, it is detected whether the length of the current path is greater than or equal to the minimum virtual sequence length value.
[0026] If the length of the current path is greater than or equal to the minimum virtual sequence length value, then the current path is used as a virtual experimental sequence.
[0027] Preferably, generating the fatigue limit result value of the material to be analyzed based on the virtual experimental sequence includes:
[0028] Generate fatigue limit estimates for the virtual experimental sequence;
[0029] All the fatigue limit estimates were used as a statistical sample;
[0030] The statistical sample was analyzed to obtain the fatigue limit result value of the material to be analyzed.
[0031] Preferably, the step of analyzing the statistical sample to obtain the fatigue limit result value of the material to be analyzed includes:
[0032] Generate the probability density function of the statistical sample;
[0033] Determine the mode value of the probability density function;
[0034] The fatigue limit value corresponding to the mode value is taken as the fatigue limit result value of the material to be analyzed.
[0035] Preferably, after taking the fatigue limit value corresponding to the mode value as the fatigue limit result value of the material to be analyzed, the method further includes:
[0036] Based on the statistical sample, an interval analysis was performed on the fatigue limit result value to obtain the interval analysis result;
[0037] Based on the interval analysis results, the stability and confidence level of the fatigue limit results are evaluated.
[0038] A system for determining the fatigue limit of a material, comprising:
[0039] The data acquisition module is used to acquire the disordered fatigue test dataset of the material to be analyzed. The disordered fatigue test dataset only includes the stress level and cycle results of the material specimen to be analyzed. The cycle results include failure or non-failure.
[0040] The initial stress determination module is used to determine the initial stress.
[0041] The current stress determination module is used to use the initial stress as the current stress of the current path;
[0042] The sample addition module is used to add the sample corresponding to the current stress to the current path in the disordered fatigue test dataset;
[0043] The stress update module is used to increase the current stress if the stress level of the specimen corresponding to the current stress has not failed, and to decrease the current stress if the stress level of the specimen corresponding to the current stress has failed.
[0044] The sequence generation module is used to return the steps of adding the specimen corresponding to the current stress to the current path and subsequent steps in the disordered fatigue test dataset until the current path that meets the set path conditions is obtained, and the current path is used as a virtual test sequence.
[0045] The fatigue limit determination module is used to generate fatigue limit result values for the material to be analyzed based on the virtual experimental sequence.
[0046] An electronic device, comprising:
[0047] Memory, used to store computer programs;
[0048] A processor, configured to implement the steps of the material fatigue limit determination method as described above when executing the computer program.
[0049] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of any of the above-described methods for determining the material fatigue limit.
[0050] This application provides a method for determining the fatigue limit of a material. The method involves: acquiring a disordered fatigue test dataset of the material to be analyzed, which includes only the stress level and cycle results of the specimens, with the cycle results indicating failure or non-failure; determining an initial stress; using the initial stress as the current stress for the current path; adding the specimen corresponding to the current stress to the current path from the disordered fatigue test dataset; increasing the current stress if the stress level of the specimen corresponding to the current stress indicates non-failure, and decreasing the current stress if the stress level indicates failure; returning to the previous steps of adding the specimen corresponding to the current stress to the current path from the disordered fatigue test dataset until a current path satisfying the set path conditions is obtained, using the current path as a virtual test sequence; and generating the fatigue limit result value of the material to be analyzed based on the virtual test sequence. In this application, the disordered fatigue test dataset includes the stress level and failure or non-failure results of the material specimen to be analyzed. This can be abstracted into a set of stress-result data pairs. Then, according to the rule of increasing the current stress if the stress level of the specimen corresponding to the current stress is not in failure, and decreasing the current stress if the stress level of the specimen corresponding to the current stress is in failure, the stress-result data pairs in the undisordered fatigue test dataset are combined into a virtual experimental sequence that meets the requirements of the increase / decrease method. This makes the increase / decrease method no longer limited to "sequential complete data" but can be extended to disordered fatigue data through virtual sequences. This realizes the recycling and reuse of existing disordered experimental data resources. If the fatigue limit result value of the material to be analyzed is subsequently generated based on the virtual experimental sequence, there is no need to rearrange and execute a complete increase / decrease method fatigue test, which can quickly and cost-effectively determine the fatigue limit of the material. The material fatigue limit determination system, electronic device, and computer-readable storage medium provided in this application also solve the corresponding technical problems. Attached Figure Description
[0051] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.
[0052] Figure 1 A flowchart illustrating a method for determining the fatigue limit of materials provided in this application embodiment;
[0053] Figure 2 Overall flowchart for the experiment to determine the fatigue limit of materials;
[0054] Figure 3 This is a schematic diagram of the virtual experimental sequence in the experiment;
[0055] Figure 4This is a histogram obtained by statistical analysis of the fatigue limit calculation results of all virtual sequences under a single initial stress.
[0056] Figure 5 The graph shows a comparison of the results of double fitting of the probability density function under a single initial stress.
[0057] Figure 6 A comparison chart of the 95% confidence interval and the highest density interval under a single initial stress.
[0058] Figure 7 This is a schematic diagram of a material fatigue limit determination system provided in an embodiment of this application;
[0059] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application;
[0060] Figure 9 This is another structural schematic diagram of an electronic device provided in an embodiment of this application. Detailed Implementation
[0061] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0062] Please see Figure 1 , Figure 1 This is a flowchart illustrating a method for determining the fatigue limit of materials, as provided in an embodiment of this application.
[0063] This application provides a method for determining the fatigue limit of materials, which may include the following steps:
[0064] Step S101: Obtain the disordered fatigue test dataset of the material to be analyzed. The disordered fatigue test dataset only includes the stress level and cycle results of the material specimen to be analyzed. The cycle results include failure or non-failure.
[0065] In practical applications, an unordered fatigue test dataset of the material to be analyzed can be obtained first. The material to be analyzed can be flexibly determined according to the application scenario, such as metallic materials, corrosive materials, etc. Specifically, the material to be analyzed can be high-performance alloys, advanced composite materials, key aerospace components, etc. The unordered fatigue test dataset only includes the stress level and cycle results of the material specimen. The cycle results can be obtained by comparing the number of cycles with a set life base. The cycle result is either failure or non-failure. In the fatigue test, if the specimen fractures or is damaged before the specified number of cycles, it is recorded as failure. If it fractures or is damaged before the specified number of cycles (e.g., 10 cycles), it is recorded as failure. 6 Or 10 7 If the circuit does not break after (times), it is recorded as not failing (run-out).
[0066] Understandably, the data structure of an unordered fatigue experiment dataset can be flexibly determined according to the application scenario. For example, it can be a list, tree data structure, hierarchical index structure, database structure, etc., and failure / non-failure can be marked by binary variables (0 / 1), categorical variables (Fail / Run-out), logical values based on lifespan thresholds, etc.
[0067] Step S102: Determine the initial stress.
[0068] Step S103: Use the initial stress as the current stress of the current path.
[0069] In practical applications, if the disordered fatigue test dataset is combined according to the principle of the rise and fall method, a virtual test sequence can be obtained to generate fatigue limit result values. In this process, the disordered fatigue test dataset can be combined using the backtracking algorithm. That is, the initial stress can be determined first. The initial stress can be flexibly determined according to actual needs. It can be a single value or multiple different values. The initial stress is used as the current stress of the current path so that the specimen is filled into the current path starting from the initial stress.
[0070] Step S104: In the disordered fatigue test dataset, add the specimen corresponding to the current stress to the current path.
[0071] Step S105: If the stress level of the specimen corresponding to the current stress has not failed, increase the current stress; if the stress level of the specimen corresponding to the current stress has failed, decrease the current stress.
[0072] In practical applications, after setting the initial stress as the current stress, the specimen corresponding to the current stress can be added to the current path in the disordered fatigue test dataset. The specimen corresponding to the current stress refers to the specimen whose stress level is within a set range. The set range can be flexibly adjusted according to actual needs. Then, the current stress needs to be updated in order to add subsequent specimens to the current path. In the process of updating the current stress, the cyclic results need to be applied according to the principle of the rise and fall method. That is, if the stress level of the specimen corresponding to the current stress has not failed, the current stress is increased; if the stress level of the specimen corresponding to the current stress fails, the current stress is decreased.
[0073] In an exemplary embodiment, during the process of adding a specimen corresponding to the current stress to the current path in the disordered fatigue test dataset, unused specimens corresponding to the current stress can be selected as candidate specimens in the disordered fatigue test dataset; the candidate specimens are traversed and added to the current path to exhaustively enumerate all specimens to generate the current path.
[0074] In an exemplary embodiment, in response to the stress level of the specimen corresponding to the current stress not failing, the current stress is increased; in response to the stress level of the specimen corresponding to the current stress failing, the current stress is decreased. During this process, a set stress step size d can be obtained. The stress step size can be determined based on 3%–5% of the estimated fatigue limit of the material properties, such as 5 MPa, 7 MPa, etc. The stress step size directly affects the fluctuation amplitude of the rise and fall path. In response to the stress level of the specimen corresponding to the current stress not failing, the current stress is increased by the stress step size, that is... In response to the stress level failure of the specimen corresponding to the current stress, the current stress is reduced by the stress step size, that is... .
[0075] Step S106: Return to the step of adding the specimen corresponding to the current stress to the current path and subsequent steps in the disordered fatigue test dataset until a current path that meets the set path conditions is obtained, and use the current path as a virtual test sequence.
[0076] In practical applications, after updating the current stress, you can continue to add samples to the current path according to the current stress. That is, you need to return to the disordered fatigue test dataset and add the sample corresponding to the current stress to the current path and the subsequent steps until you get the current path that meets the set path conditions. The current path is then used as a virtual test sequence.
[0077] In an exemplary embodiment, to ensure the validity of the sequence and control the computational scale, the process returns to the step of adding the specimen corresponding to the current stress to the current path and subsequent paths in the disordered fatigue test dataset until a current path that meets the set path conditions is obtained. During the process of using the current path as a virtual test sequence, the maximum number of generated paths can be obtained. and minimum virtual sequence length value If a specimen corresponding to the current stress exists in the disordered fatigue test dataset, and the length of the current path is less than the maximum number of generated paths, then return to the step of adding the specimen corresponding to the current stress to the current path and subsequent steps in the disordered fatigue test dataset; if no specimen corresponding to the current stress exists in the disordered fatigue test dataset, or if the length of the current path is equal to the maximum number of generated paths, then check whether the length of the current path is greater than or equal to the minimum virtual sequence length value; if the length of the current path is greater than or equal to the minimum virtual sequence length value, then use the current path as a virtual test sequence.
[0078] Understandably, the path conditions can be flexibly determined according to the application scenario. For example, it can be based on path probability screening to calculate the probability of occurrence of each path (based on the number of samples under stress) and retain only the paths with high probabilities; or it can be based on heuristic pruning to automatically remove paths that are obviously unlikely to represent the real experimental history; or it can be based on optimization search based on energy function or scoring function, such as "prioritizing the retention of paths closer to the fatigue limit" or "prioritizing the retention of paths that match a certain type of statistical characteristics".
[0079] It should be noted that this application generates virtual experimental sequences by searching for samples using a backtracking method. Alternatively, it can utilize graph traversal algorithms, dynamic programming or state-space search algorithms, Monte Carlo-based random path generation algorithms, and constraint solver-based path generation algorithms to search for samples and generate virtual experimental sequences. Graph traversal algorithms can include depth-first search (DFS), breadth-first search (BFS), and heuristic search (A*, Best-first Search). Dynamic programming or state-space search algorithms treat each stress level as a graph node and "rise" and "fall" as state transitions, searching for all feasible paths using dynamic programming. Monte Carlo-based random path generation algorithms can process samples through random sampling, weighted random transitions (based on sample proportions), and random tree search (MCTS) to generate a large number of possible virtual experimental sequences. Constraint solver-based path generation algorithms can be performed using SAT / SMT solvers or constraint programming (CP), for example, using the rise / fall rule as a hard constraint, unused samples as state variables, and the solver enumerating all virtual experimental sequences that satisfy the constraints.
[0080] Step S107: Generate fatigue limit result values for the material to be analyzed based on the virtual experimental sequence.
[0081] In practical applications, since the virtual experimental sequence is obtained according to the principle of the rise and fall method, it can be processed using the same method to generate fatigue limit values for the material being analyzed. During this process, the virtual sequence can be converted into a unique code (such as a string) and compared to eliminate duplicate paths before generating the fatigue limit values, ensuring the uniqueness of the statistical analysis sample. The rise and fall method can include the Dixon-Mood method, the Karber method, the Robbins-Monro adaptive algorithm, the Probit / Logistic regression model, and Bayesian rise and fall estimation methods, among others.
[0082] In the exemplary embodiment, although this application sorts the specimens in the disordered fatigue test dataset according to the principle of the ascending and descending method to generate a virtual test sequence, this virtual test sequence is not a sequence obtained from actual experiments. Therefore, the fatigue limit result value generated may not be accurate. To address this issue, considering the large number of virtual test sequences generated from the disordered fatigue test dataset, each virtual test sequence can generate a fatigue limit reference value. Statistical analysis of these fatigue limit reference values can weaken the influence of a single abnormal fatigue limit reference value on the overall result. This ensures that the fatigue limit result value reflects the consistent trend of multiple virtual test sequences rather than the randomness of a specific historical path, thereby improving robustness against data noise and outliers and guaranteeing the reliability and robustness of the fatigue limit result value. In other words, during the process of generating the fatigue limit result value of the material to be analyzed based on the virtual test sequence, fatigue limit estimates of the virtual test sequence can be generated; all fatigue limit estimates can be used as statistical samples; and the statistical samples can be analyzed to obtain the fatigue limit result value of the material to be analyzed.
[0083] In specific application scenarios, the method for analyzing statistical samples can be flexibly determined according to needs. For example, it can be analyzed using methods such as KDE, normal fitting, most dense intervals, skewed distributions (e.g., log-normal), gamma distribution, kernel mixture models (GMM), nearest neighbor estimation (kNN density), quantile regression, Bayesian statistical inference, and bootstrap resampling. Among these, Bayesian statistical inference's MCMC samples the posterior distribution of the fatigue limit, and the Highest Posterior Density interval (HPD) results can also provide point estimates and uncertainty quantification; bootstrap resampling of statistical samples can obtain the distribution and confidence intervals.
[0084] In specific application scenarios, during the analysis of statistical samples to obtain the fatigue limit result value of the material to be analyzed, a probability density function of the statistical sample can be generated. For example, a dual probability distribution fitting strategy can be adopted, using kernel density estimation (KDE) and normal distribution fitting methods to calculate the probability density function of the statistical sample respectively; the mode of the probability density function is determined; and the fatigue limit value corresponding to the mode is taken as the fatigue limit result value of the material to be analyzed. Furthermore, after taking the fatigue limit value corresponding to the mode as the fatigue limit result value of the material to be analyzed, interval analysis can be performed on the fatigue limit result value based on the statistical sample, such as performing 95% confidence interval, most dense interval analysis, etc., to obtain the interval analysis results; based on the interval analysis results, the stability and confidence level of the fatigue limit result value can be evaluated. In this way, not only can a smooth probability density function be constructed using kernel density estimation, and its mode be extracted as the most likely fatigue limit, but also indicators such as normal distribution fitting, 95% confidence interval, and densest interval can be used to characterize the uncertainty of the fatigue limit determination results from multiple dimensions such as distribution pattern, central tendency, and dispersion. This allows engineering designers to judge the credibility and risk level of the estimation results based on this, instead of relying solely on a single numerical value for decision-making, thus solving the problem of reliably assessing the fatigue limit from disordered experimental data.
[0085] This application provides a method for determining the fatigue limit of a material. The method involves: acquiring a disordered fatigue test dataset of the material to be analyzed, which includes only the stress level and cycle results of the specimens, with the cycle results indicating failure or non-failure; determining an initial stress; using the initial stress as the current stress for the current path; adding the specimen corresponding to the current stress to the current path from the disordered fatigue test dataset; increasing the current stress if the stress level of the specimen corresponding to the current stress indicates non-failure, and decreasing the current stress if the stress level indicates failure; returning to the previous steps of adding the specimen corresponding to the current stress to the current path from the disordered fatigue test dataset until a current path satisfying the set path conditions is obtained, using the current path as a virtual test sequence; and generating the fatigue limit result value of the material to be analyzed based on the virtual test sequence. In this application, the disordered fatigue test dataset includes the stress levels and failure / non-failure results of the material specimens to be analyzed. This can be abstracted into a set of stress-result data pairs. Then, according to the rule of increasing the current stress if the current stress corresponds to a specimen with no failure, and decreasing the current stress if the current stress corresponds to a specimen with failure, the stress-result data pairs in the undisordered fatigue test dataset are combined into a virtual experimental sequence that meets the requirements of the increase / decrease method. This allows the increase / decrease method to no longer be limited to "sequentially complete data" but to be extended to disordered fatigue data through virtual sequences. This achieves the recycling and reuse of existing disordered experimental data resources. Subsequently, if the fatigue limit result value of the material to be analyzed is generated based on the virtual experimental sequence, there is no need to rearrange and execute a complete increase / decrease method fatigue experiment, enabling rapid and low-cost determination of the material's fatigue limit. Furthermore, this application strictly adheres to the physical rules and standard calculation formulas of the increase / decrease method at its underlying level, ensuring theoretical rigor. At the same time, the path search strategy gives this application the flexibility to handle complex and disordered experimental data, achieving both rigor and flexibility.
[0086] To facilitate understanding of the effectiveness of the material fatigue limit determination method provided in this application, this embodiment uses fatigue test data of a certain type of steel as an example to demonstrate the complete implementation process of this method. The overall process is as follows: Figure 2 As shown in the figure. This dataset contains 30 specimens with stress levels ranging from 185 MPa to 206 MPa, in 7 MPa increments, and the experimental cycle count is set to 10. 7 The implementation process and analysis are as follows.
[0087] Step S1: Data Preparation and Parameter Initialization. Raw Data: Stress vector (stress), cycle count vector (cycles), and initial stress 192 MPa. Data Processing: Generate a result vector (results) based on cycles >= 1e7 (1 = not failed, 0 = failed). Merge stress and results into a data matrix (data). Parameter Settings: Stress step size d = 7 MPa; minimum sequence length 15; maximum number of paths 100,000.
[0088] Step S2: Virtual Experiment Sequence Reconstruction. The backtracking algorithm function is invoked. The algorithm starts with a dataset stress of 192 MPa and searches for unused specimens in the data pool. Assuming the first found specimen is "not failed" (1), the next stress level is 199 MPa. Assuming the second found specimen is also "not failed" (1), the next stress level is 206 MPa. The algorithm recursively searches up to 206 MPa. Assuming the specimen found at 206 MPa is "failed" (0), the next stress level drops back to 192 MPa, and so on. The algorithm traverses all candidate specimens and recursively explores each branch. Finally, when the path length reaches 15 and cannot continue, the path is saved. Through systematic backtracking and searching, under the conditions of this embodiment, after deduplication, 230 unique and valid virtual experiment sequences with a length greater than or equal to 15 are obtained.
[0089] Step S3: Individual Sequence Fatigue Limit Calculation. Traverse the above 230 virtual sequences. For each sequence, call the written function to perform fatigue limit calculation. Internally, this function performs standard Dixon-Mood calculations: determining the stress levels and their indices in the sequence, and counting the failure frequency of each level. Calculate statistics such as A, B, and C, and finally output the fatigue limit of the sequence. and standard deviation .like Figure 3 The image shows the ascending / descending order of a certain sequence. After this step, 230 fatigue limit estimates are obtained. This constitutes the sample for subsequent analysis.
[0090] Step S4: Statistical Inference and Result Verification. Statistical Analysis: Statistically analyze these 230... Simultaneously, the densest interval was found using a function; and a probability density function histogram was generated. Curve fitting was performed using the normal distribution and kernel density, and the mode obtained from the kernel density estimation was finally selected as the final fatigue limit, which was 197.544 MPa. Results verification: The fatigue limit value calculated using the lifting method for this type of steel was 197.5 MPa, which deviates from the result of this method (197.544 MPa) by approximately 0.044 MPa (<0.3%), showing a high degree of consistency and verifying the reliability of this method.
[0091] The above embodiments only processed a single initial stress. To demonstrate the enhanced effect of this method in multi-initial stress aggregation analysis, based on the above experiments, instead of starting only at 192 MPa, all possible initial stress levels (185, 192, 199, 206 MPa) were traversed, and steps S2 and S3 were performed respectively. Data aggregation: The calculation results of all virtual sequences generated under four different initial stresses were aggregated. The samples were merged into a larger total sample set, resulting in 899 samples. Estimated value. Statistical inference and analysis: Perform histogram statistics on this total sample, such as... Figure 4 As shown, the histogram exhibits a smooth, concentrated unimodal distribution, with the densest interval being [197.340, 197.840] MPa, containing approximately 21.03% of the data points. Figure 5 Fitting the probability histogram function. Figure 6 For the 95% confidence interval of the fitted function, both normal estimation and kernel density estimation still yielded good results. Kernel density estimation was selected as the final output, with a fatigue limit of 197.666 MPa, which is still close to the true value of 197.5 MPa. Based on this experiment, it can be seen that this application can further average out the bias of a single starting point by aggregating multiple initiation stresses, making the statistical distribution more representative of all possible experimental histories, and further improving the robustness and confidence of the final fatigue limit result.
[0092] Please see Figure 7 , Figure 7 This is a schematic diagram of a material fatigue limit determination system provided in an embodiment of this application.
[0093] This application provides a material fatigue limit determination system, which may include:
[0094] The data acquisition module 101 is used to acquire the disordered fatigue test dataset of the material to be analyzed. The disordered fatigue test dataset only includes the stress level and cycle results of the material specimen to be analyzed. The cycle results include failure or non-failure.
[0095] Initial stress determination module 102 is used to determine the initial stress;
[0096] The current stress determination module 103 is used to use the initial stress as the current stress of the current path;
[0097] The specimen addition module 104 is used to add the specimen corresponding to the current stress to the current path in the disordered fatigue test dataset;
[0098] The stress update module 105 is used to increase the current stress if the stress level of the specimen corresponding to the current stress has not failed, and to decrease the current stress if the stress level of the specimen corresponding to the current stress has failed.
[0099] The sequence generation module 106 is used to return the steps of adding the specimen corresponding to the current stress to the current path and subsequent steps in the disordered fatigue test dataset until the current path that meets the set path conditions is obtained, and the current path is used as a virtual test sequence.
[0100] The fatigue limit determination module 107 is used to generate fatigue limit result values for the material to be analyzed based on the virtual experimental sequence.
[0101] The present application provides a material fatigue limit determination system, in which the sample addition module can be specifically used to: in a disordered fatigue test dataset, select unused samples corresponding to the current stress as candidate samples; traverse the candidate samples and add the traversed candidate samples to the current path.
[0102] The present application provides a material fatigue limit determination system, wherein the stress update module can be specifically used to: obtain a set stress step size; increase the stress step size of the current stress if the stress level of the specimen corresponding to the current stress has not failed; and decrease the stress step size of the current stress if the stress level of the specimen corresponding to the current stress has failed.
[0103] This application provides a material fatigue limit determination system. The sequence generation module can be specifically used to: obtain a set maximum number of generation paths and a minimum virtual sequence length; responding to the existence of a specimen corresponding to the current stress in the disordered fatigue test dataset, and the length of the current path being less than the maximum number of generation paths, return to the step of adding the specimen corresponding to the current stress to the current path and subsequent steps in the disordered fatigue test dataset; responding to the absence of a specimen corresponding to the current stress in the disordered fatigue test dataset, or the length of the current path being equal to the maximum number of generation paths, detect whether the length of the current path is greater than or equal to the minimum virtual sequence length; responding to the length of the current path being greater than or equal to the minimum virtual sequence length, use the current path as a virtual test sequence.
[0104] This application provides a material fatigue limit determination system. The fatigue limit determination module can be specifically used to: generate fatigue limit estimates of a virtual experimental sequence; use all fatigue limit estimates as statistical samples; and analyze the statistical samples to obtain the fatigue limit result value of the material to be analyzed.
[0105] This application provides a material fatigue limit determination system. The fatigue limit determination module can be specifically used to: generate a probability density function of a statistical sample; determine the mode of the probability density function; and use the fatigue limit value corresponding to the mode value as the fatigue limit result value of the material to be analyzed.
[0106] The material fatigue limit determination system provided in this application embodiment may further include:
[0107] The evaluation module, after the fatigue limit determination module takes the fatigue limit value corresponding to the mode value as the fatigue limit result value of the material to be analyzed, performs interval analysis on the fatigue limit result value based on the statistical sample to obtain the interval analysis result; and evaluates the stability and confidence level of the fatigue limit result value based on the interval analysis result.
[0108] This application also provides an electronic device and a computer-readable storage medium, both of which have the corresponding effects of the material fatigue limit determination method provided in the embodiments of this application. Please refer to... Figure 8 , Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.
[0109] An electronic device provided in this application includes a memory 201 and a processor 202. The memory 201 stores a computer program, and when the processor 202 executes the computer program, it implements the steps of the material fatigue limit determination method described in any of the above embodiments.
[0110] Please see Figure 9 Another electronic device provided in this application embodiment may further include: an input port 203 connected to the processor 202 for transmitting commands input from the outside to the processor 202; a display unit 204 connected to the processor 202 for displaying the processing results of the processor 202 to the outside; and a communication module 205 connected to the processor 202 for enabling communication between the electronic device and the outside. The display unit 204 may be a display panel, a laser scanning display, etc.; the communication method adopted by the communication module 205 includes, but is not limited to, Mobile High-Definition Link (MHL), Universal Serial Bus (USB), High-Definition Multimedia Interface (HDMI), wireless connection: Wireless Fidelity (WiFi), Bluetooth communication technology, Bluetooth Low Energy communication technology, and communication technology based on IEEE 802.11s.
[0111] This application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the steps of the material fatigue limit determination method described in any of the above embodiments.
[0112] The computer-readable storage media involved in this application include random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs (compact disc read-only memory), or any other form of storage media known in the art.
[0113] This application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the material fatigue limit determination method described in any of the above embodiments.
[0114] For descriptions of relevant parts in the material fatigue limit determination system, electronic device, and computer-readable storage medium provided in this application's embodiments, please refer to the detailed description of the corresponding parts in the material fatigue limit determination method provided in this application's embodiments, which will not be repeated here. Furthermore, parts of the technical solutions provided in this application that are consistent with the implementation principles of corresponding technical solutions in the prior art have not been described in detail to avoid excessive elaboration.
[0115] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0116] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for determining the fatigue limit of a material, characterized in that, include: Obtain a disordered fatigue test dataset of the material to be analyzed. The disordered fatigue test dataset includes only the stress level and cycle results of the material specimens to be analyzed, and the cycle results include failure or non-failure. Determine the initial stress; Use the initial stress as the current stress for the current path; In the disordered fatigue test dataset, the specimen corresponding to the current stress is added to the current path; If the current stress level of the specimen does not fail, the current stress is increased; if the current stress level of the specimen fails, the current stress is decreased. Return to the disordered fatigue test dataset and add the specimen corresponding to the current stress to the current path and subsequent steps until a current path that meets the set path conditions is obtained, and use the current path as a virtual test sequence. Based on the virtual experimental sequence, fatigue limit results of the material to be analyzed are generated.
2. The method according to claim 1, characterized in that, Adding the specimen corresponding to the current stress to the current path in the disordered fatigue test dataset includes: In the disordered fatigue test dataset, the specimens that correspond to the current stress and have not been used are selected as candidate specimens. Traverse the candidate samples and add the traversed candidate samples to the current path.
3. The method according to claim 2, characterized in that, The response of increasing the current stress if the stress level of the specimen corresponding to the current stress does not fail, and decreasing the current stress if the stress level of the specimen corresponding to the current stress fails, includes: Obtain the set stress step size; If the current stress level of the specimen does not indicate failure, the stress step size is increased. In response to the stress level failure of the specimen corresponding to the current stress, the current stress is reduced by the stress step.
4. The method according to claim 3, characterized in that, The step of returning to the disordered fatigue test dataset involves adding the specimen corresponding to the current stress to the current path and subsequent paths until a current path that meets the set path conditions is obtained. The current path is then used as a virtual test sequence, including: Get the set maximum number of generated paths and minimum virtual sequence length values; If, in response to the existence of a specimen corresponding to the current stress in the disordered fatigue test dataset, and the length of the current path is less than the maximum number of generated paths, the process returns to the step of adding the specimen corresponding to the current stress to the current path and subsequent paths in the disordered fatigue test dataset. In response to the absence of a specimen corresponding to the current stress in the disordered fatigue test dataset, or the length of the current path being equal to the maximum number of generated paths, it is detected whether the length of the current path is greater than or equal to the minimum virtual sequence length value. If the length of the current path is greater than or equal to the minimum virtual sequence length value, then the current path is used as a virtual experimental sequence.
5. The method according to claim 1, characterized in that, The step of generating fatigue limit result values for the material to be analyzed based on the virtual experimental sequence includes: Generate fatigue limit estimates for the virtual experimental sequence; All the fatigue limit estimates were used as a statistical sample; The statistical sample was analyzed to obtain the fatigue limit result value of the material to be analyzed.
6. The method according to claim 5, characterized in that, The analysis of the statistical sample to obtain the fatigue limit result value of the material to be analyzed includes: Generate the probability density function of the statistical sample; Determine the mode value of the probability density function; The fatigue limit value corresponding to the mode value is taken as the fatigue limit result value of the material to be analyzed.
7. The method according to claim 6, characterized in that, After taking the fatigue limit value corresponding to the mode value as the fatigue limit result value of the material to be analyzed, the method further includes: Based on the statistical sample, an interval analysis was performed on the fatigue limit result value to obtain the interval analysis result; Based on the interval analysis results, the stability and confidence level of the fatigue limit results are evaluated.
8. A system for determining the fatigue limit of a material, characterized in that, include: The data acquisition module is used to acquire the disordered fatigue test dataset of the material to be analyzed. The disordered fatigue test dataset only includes the stress level and cycle results of the material specimen to be analyzed. The cycle results include failure or non-failure. The initial stress determination module is used to determine the initial stress. The current stress determination module is used to use the initial stress as the current stress of the current path; The sample addition module is used to add the sample corresponding to the current stress to the current path in the disordered fatigue test dataset; The stress update module is used to increase the current stress if the stress level of the specimen corresponding to the current stress has not failed, and to decrease the current stress if the stress level of the specimen corresponding to the current stress has failed. The sequence generation module is used to return the steps of adding the specimen corresponding to the current stress to the current path and subsequent steps in the disordered fatigue test dataset until the current path that meets the set path conditions is obtained, and the current path is used as a virtual test sequence. The fatigue limit determination module is used to generate fatigue limit result values for the material to be analyzed based on the virtual experimental sequence.
9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the material fatigue limit determination method as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the material fatigue limit determination method as described in any one of claims 1 to 7.