Test Screening System and Method under Point Transition

Abstract

This invention discloses a test screening system and method based on point transitions, relating to the field of data processing technology. The system includes: extracting structural features of the test object, identifying test points, and generating K test points; mining the frequency of performance defects in mineral castings to generate K frequency values; determining defect coefficients and generating a test point extraction wheel; interacting with M test items; extracting from the test point extraction wheel to obtain a set of M test points, and performing point transition authentication. If authentication is successful, the system performs performance tests on the mineral castings using the M test point sets for the M test items, obtaining M test result sets, and performing centralized screening and analysis to obtain M project test results, which are then used as the target performance test results. This invention solves the technical problem of low reliability of test results in existing technologies for performance testing of mineral castings, achieving the technical effect of improving the accuracy of mineral casting performance testing.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically to a test and screening system and method for point transitions. Background Technology

[0002] With the rapid development of the manufacturing industry, mineral castings have been widely used in many industrial fields. Due to their superior mechanical properties and durability, mineral castings are widely used in construction, automotive, aerospace, and machinery manufacturing industries. However, with the continuous expansion of production scale and the increasing complexity of casting processes, the quality of a batch of mineral castings varies during production, often resulting in significant differences in performance test data, making it difficult to guarantee the reliability and consistency of test results. Traditional methods are proving inadequate in addressing these problems. Summary of the Invention

[0003] This application provides a test screening system and method under point transition, which is used to address the technical problem of low reliability of test results in the performance testing of mineral castings in the prior art.

[0004] In view of the above problems, this application provides a test screening system and method under point transition.

[0005] A first aspect of this application provides a test screening system for point transitions, the system comprising:

[0006] The test point identification module extracts the structural features of the test object in the target test task and identifies test points based on these features, generating K test points. A performance defect frequency mining module mines the frequency of performance defects in mineral castings based on the structural features of the test object and the K test points, generating K frequency values. A defect coefficient determination module determines the defect coefficient of the K test points based on the K frequency values, generating a test point extraction wheel. A target test task interaction module allows interaction with M test items in the target test task. The test items include M test samples; a test point acquisition module, which performs multiple extractions of the test point extraction wheel based on the M test samples to obtain a set of M test points; a point transition authentication module, which traverses the set of M test points to perform point transition authentication. If the authentication is successful, the mineral casting performance of the M test items is tested based on the set of M test points to obtain a set of M test results; and a centralized screening and analysis module, which performs centralized screening and analysis on the set of M test results to obtain M project test results, which are used as the target performance test results for the target test task.

[0007] A second aspect of this application provides a test screening method under point transitions, the method comprising:

[0008] Extract the structural features of the test object in the target test task, and identify test points based on the structural features to generate K test points. Based on the structural features and the K test points, mine the frequency of performance defects in the mineral casting to generate K frequency values. Determine the defect coefficients of the K test points based on the K frequency values ​​to generate a test point extraction wheel. Interact with M test items in the target test task, where each M test item includes M test samples. Extract test points multiple times from the test point extraction wheel based on the M test sample quantities to obtain a set of M test points. Iterate through the set of M test points for point transition authentication. If authentication is successful, perform mineral casting performance tests on the M test points to obtain a set of M test results. Perform centralized screening and analysis on the set of M test results to obtain M project test results, which are then used as the target performance test results for the target test task.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] This application extracts the structural features of the test object in the target testing task, identifies test points based on these features, and generates K test points. Based on the structural features and the K test points, it mines the frequency of performance defects in mineral castings, generating K frequency values. It determines the defect coefficients of the K test points based on the K frequency values, generating a test point extraction wheel. It interacts with M test items in the target testing task, where each M test item includes M test samples. It extracts test points multiple times from the test point extraction wheel based on the M test sample quantities, obtaining a set of M test points. It iterates through the M test point sets for point transition authentication. If authentication is successful, it performs mineral casting performance tests on the M test points, obtaining a set of M test results. It then performs centralized screening and analysis on the M test results sets to obtain M project test results, which are used as the target performance test results for the target testing task. This invention addresses the technical problem of low reliability of test results in the performance testing of mineral castings in existing technologies. By extracting structural features, mining the frequency of defects, determining the defect coefficient, extracting test points, verifying point transitions, and conducting centralized screening analysis, it achieves the technical effect of improving the accuracy of performance testing of mineral castings. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A schematic diagram of the test screening system structure under point transition provided in an embodiment of this application;

[0013] Figure 2 This is a schematic diagram of the test screening method under point transition provided in the embodiments of this application.

[0014] Explanation of reference numerals in the attached diagram: 11 Test point identification module, 12 Performance defect frequency mining module, 13 Defect coefficient determination module, 14 Target test task interaction module, 15 Test point acquisition module, 16 Point transition authentication module, 17 Centralized screening and analysis module. Detailed Implementation

[0015] This application provides a test screening system and method based on point transitions, which addresses the technical problem of low reliability of test results in the performance testing of mineral castings in existing technologies. By extracting structural features, mining the frequency of defects, determining the defect coefficient, extracting test points, verifying point transitions, and conducting centralized screening analysis, the technical effect of improving the accuracy of performance testing of mineral castings is achieved.

[0016] 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 a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0017] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products, or devices.

[0018] Example 1, as Figure 1 As shown in the embodiment of this application, a test screening system under point transition is provided, the system comprising:

[0019] Test point identification module 11 extracts the structural features of the test object of the target test task and identifies test points based on the structural features of the test object to generate K test points.

[0020] Furthermore, in the system provided in the application embodiment, the test point identification module 11 is also used for:

[0021] The target test task is feature extracted according to the preset structural indicators to obtain the structural features of the test object; a structural feature-test point mapping relationship is constructed; the K test points are generated by searching in the structural feature-test point mapping relationship using the structural features of the test object as the index.

[0022] Furthermore, the preset structural indicators include grain structure, external dimensions, shape characteristics, wall thickness distribution, and surface hardness.

[0023] In this embodiment, features of the target test task are first extracted based on preset structural indicators to obtain the structural features of the test object. These preset structural indicators include grain structure, external dimensions, shape features, wall thickness distribution, and surface hardness.

[0024] Next, CT scanning technology was used to acquire structural data of the test object. High-resolution 3D tomographic images were obtained using a CT scanner, such as the YXLON CTCompact. The acquired CT scan data was preprocessed, with Gaussian filtering used to remove noise, followed by calibration and alignment. Then, the Canny edge detection algorithm was used to segment the preprocessed CT images, extracting key structural features of the test object. These features include grain edges, external contours, and internal structural details. Morphological operations, such as erosion and dilation, were then used to further process the extracted structural features to obtain clear morphological characteristics. ImageJ software was then used to calculate the geometric features of the test object, such as length, width, height, and thickness. Simultaneously, the surface hardness of the test object was measured using a microhardness tester, and VGStudio Max software was used to perform 3D reconstruction and volume rendering of the CT data, analyzing the internal structural features of the test object. This process yielded the structural characteristics of the test object.

[0025] After obtaining the structural features of the test object, a mapping relationship between structural features and test points is established. A decision tree model is used to learn the association between these features and test points from historical data. Specifically, a large amount of historical data containing known test points and corresponding structural features is collected to train the decision tree model. The input is the structural features, and the output is the potential test points. After training, the decision tree model can predict the corresponding test points based on the input structural features. The trained decision tree model is used to construct the structural feature-test point mapping relationship. During actual testing, when new structural features of the test object are extracted, these features are input into the trained decision tree model, and the model outputs the corresponding test points.

[0026] Finally, using the structural features of the test object as an index, the structural feature-test point mapping relationship is searched. Based on the mapping relationship, the test points that best match these structural features are retrieved, generating a preliminary set of test points. However, the preliminary set of test points contains redundant and unevenly distributed points. Therefore, the K-means clustering algorithm is applied to filter and optimize the preliminary set of test points. Through K-means clustering, the test points are reasonably grouped, redundant points are removed, and the test points are ensured to be evenly distributed and cover key areas. Through the above steps, the structural features of the test object are extracted, and K test points are generated based on these features.

[0027] The performance defect frequency mining module 12 mines the frequency of performance defects in mineral castings based on the structural features of the test object and the K test points, and generates K frequency values.

[0028] In this embodiment, data containing the structural features and corresponding defect information of the test object is first collected from historical test data. This data is stored in a database and includes the structural features of the test object and defect information for each test point, such as cracks, holes, and stress concentration areas. To ensure data integrity and accuracy, data cleaning is performed to remove missing and outlier values. Then, statistical analysis methods are used to process and analyze the cleaned data. Specifically, the frequency of defects at each test point is calculated by counting the number of times defects occur at each test point in the historical data. For example, using frequency statistics, the total number of times defects occur at each test point in historical tests is counted to obtain the frequency of each test point.

[0029] To delve deeper into the correlation between structural features and defect frequency, frequent pattern mining algorithms, such as the Apriori algorithm, are used. By converting the structural features of the test object and the defect information of the test points into a data format suitable for the Apriori algorithm, frequent pattern mining is performed to discover which structural features have significant correlations with defect frequency. Specifically, historical data is first converted into a data format suitable for frequent pattern mining. For example, the defect information of each test object and its corresponding test point is converted into a transaction set, where each transaction includes the structural features of the test object and the test point where defects occur. Then, the Apriori algorithm is used to mine frequent patterns in these transaction sets, identifying frequently occurring combinations of structural features and test points. After mining frequent patterns, association rules are further generated. These rules describe the correlation between structural features and defect frequency. For example, an association rule might state that "if the grain structure has a certain morphology, certain test points are more likely to have crack defects." By filtering association rules related to defect frequency, a corresponding frequency count is generated for each test point. K frequent counts are generated through the above process.

[0030] The defect coefficient determination module 13 determines the defect coefficient of the K test points based on the K frequency counts and generates a test point extraction wheel.

[0031] Furthermore, in the system provided in the application embodiment, the defect coefficient determination module 13 is also used for:

[0032] The ratio of each of the K frequent occurrences to the sum of the K frequent occurrences is taken as the K defect coefficients; the K defect coefficients are multiplied by the area of ​​the wheel to obtain the area of ​​the K test point wheel; the wheel is divided into regions based on the area of ​​the K test point wheel to generate a test point extraction wheel, wherein the K divided regions in the test point extraction wheel have K test point identifiers.

[0033] In this embodiment, the defect coefficient of each test point is first calculated based on the K frequent occurrences generated in the previous step. The defect coefficient represents the frequency of defects occurring at each test point relative to other test points. Specifically, the defect coefficient of each test point is obtained by counting the frequent occurrences of each test point and dividing it by the total frequent occurrences.

[0034] Next, the defect coefficient of each test point is multiplied by the total area of ​​the wheel to obtain the area of ​​each test point on the wheel. Assuming the total area of ​​the wheel is 1, then the area of ​​each test point on the wheel is... ,in, Let be the area of ​​the i-th test point on the roulette wheel. Let be the defect coefficient of the i-th test point.

[0035] Then, based on the area of ​​the wheel at each test point, the wheel is divided into regions to generate a test point extraction wheel. First, the cumulative area of ​​each test point is calculated. The cumulative area represents the total area of ​​the current test point and all previous test points on the wheel. The formula for calculating the cumulative area is as follows: , of which, cumulative area This represents the cumulative area of ​​the i-th test point. This represents the area of ​​the roulette wheel at the j-th test point.

[0036] The roulette wheel is divided into regions based on the cumulative area, with each region corresponding to a test point. For example, assuming there are three test points with cumulative areas of 0.3, 0.8, and 1.0, the regions on the roulette wheel for these three points are 0 to 0.3, 0.3 to 0.8, and 0.8 to 1.0, respectively. The corresponding test point is marked within each region so that the specific test point can be identified when drawing from the roulette wheel. In actual testing, a test point drawing roulette wheel is used to randomly draw test points, and the probability of drawing is determined based on the defect coefficient of each test point. Specifically, a random number r between 0 and 1 is generated to determine which region it falls into. The region on the roulette wheel where the random number r falls determines the drawn test point. For example, if the random number r is 0.35, and the cumulative areas are 0.2, 0.5, and 1, then r falls into the region between 0.2 and 0.5, so the corresponding test point is the second test point.

[0037] For example, suppose there are 3 test points with frequencies of 3, 5, and 2 respectively, for a total frequency of 10. The calculation steps are as follows: First, determine the defect coefficient: the defect coefficient for the first test point is 3 / 10 = 0.3, the defect coefficient for the second test point is 5 / 10 = 0.5, and the defect coefficient for the third test point is 2 / 10 = 0.2. Then, calculate the wheel area: the wheel area for the first test point is 0.3, the wheel area for the second test point is 0.5, and the wheel area for the third test point is 0.2. Next, calculate the cumulative area: the cumulative area for the first test point is 0.3, the cumulative area for the second test point is 0.8, and the cumulative area for the third test point is 1.0. Finally, divide the area and label it: the wheel area is divided into zones 0 to 0.3 for the first test point, 0.3 to 0.8 for the second test point, and 0.8 to 1.0 for the third test point.

[0038] Through the above steps, the defect coefficient determination module determines the defect coefficient of K test points based on K frequent occurrences and generates a test point extraction wheel.

[0039] The target test task interaction module 14 interacts with M test items in the target test task, wherein the M test items include M test sample quantities.

[0040] In this embodiment of the application, before the target test task begins, the user inputs or selects the detailed information of the test task through the system interface. This information includes the overall goal of the test task, the specific items to be tested, i.e., M test items, and the number of test samples required for each test item, i.e., M test sample quantities.

[0041] Through interaction, the target test task interaction module obtains M test items in the target test task, where the M test items include M test sample numbers.

[0042] The test point acquisition module 15 performs multiple extractions of the test point extraction wheel according to the number of M test samples to obtain a set of M test points.

[0043] In this embodiment, the test point selection roulette wheel is used multiple times based on the number of test samples for each test item. For each test item, the corresponding number of test samples is obtained, and the roulette wheel selection is performed. Each time a random number r between 0 and 1 is generated, the selected test point is determined by comparing the random number r with the cumulative area of ​​each region on the roulette wheel.

[0044] Specifically, there are M test items, and the number of test samples for each test item is as follows: , , A total of M test sites need to be selected. Each test item is processed sequentially, and the number of test samples is selected according to its test sample quantity.

[0045] For each test item, N samplings are performed based on the number of test samples. During each sampling, a random number r between 0 and 1 is generated. By comparing this random number r with the cumulative area of ​​each region of the roulette wheel, the sampling test point is determined and recorded to form a set of test points.

[0046] For example, a certain test item requires 3 test samples. The test point extraction wheel has been divided into different areas, representing different test points. Assuming the generated random numbers are 0.2, 0.7, and 0.9, the extracted test points are determined by comparing these random numbers with the cumulative area of ​​the wheel's regions, and these extracted test points are recorded in the test point set.

[0047] Through the above steps, the test point acquisition module performs multiple extractions of the test point extraction wheel based on the number of M test samples to obtain a set of M test points.

[0048] The point transition authentication module 16 traverses the set of M test points to perform point transition authentication. If the authentication is successful, the mineral casting performance test of the M test items is performed based on the set of M test points to obtain a set of M test results.

[0049] In this embodiment, each set of test points is first traversed to ensure that each set effectively represents the structural characteristics of the test object. The purpose of point transition authentication is to verify whether the selected test points can reflect the frequency and distribution of defects in the test object, thereby ensuring the representativeness and validity of the test results. During the traversal of M test point sets, statistical analysis and algorithmic verification methods are used to perform point transition authentication on each set. During point transition authentication, the difference between the distribution coefficient of the test point set and the previously determined defect coefficient is calculated, and the difference is compared with a preset coefficient difference threshold. Authentication passes when the number of point coefficient differences meets a preset condition; otherwise, authentication fails.

[0050] If certification is successful, the performance of the mineral castings will be tested for M test items based on a set of M test points, resulting in M ​​sets of test results. Specifically, based on the certified test point set, appropriate testing equipment and samples will be prepared, ensuring the testing environment and conditions meet the requirements. Then, the performance of the mineral castings will be tested. According to the test methods and standards for each test item, performance tests will be conducted on the selected test points, and the test results will be recorded. Finally, the test results will be collected and summarized to form M sets of test results. Each set of test results includes specific measured values ​​for different test items, such as hardness, tensile strength, and microstructural characteristics.

[0051] Furthermore, in the system provided in the application embodiment, the point transition authentication module 16 is also used for:

[0052] The M test point sets are aggregated for identical points to obtain M test point quantity sets, where each test point quantity set includes K test quantities of K test points participating in the corresponding test item; M test point distribution coefficient sets are calculated based on the M test point quantity sets; the difference set between the M test point distribution coefficient sets and the M point coefficient difference set of the K defect coefficients is calculated; when the number of differences in the M point coefficient difference set that meet the preset coefficient difference threshold is less than or equal to the preset difference number, the certification fails, and the test point extraction wheel is extracted multiple times according to the M test sample quantity, and the M test point sets are updated according to the extraction results.

[0053] In this embodiment, firstly, identical test point sets are aggregated to obtain M test point quantity sets. Each test point quantity set contains K test quantities of K test points participating in the corresponding test item. By aggregating the test point sets, the frequency of occurrence of each test point is counted to calculate its distribution and coefficients later. Next, M test point distribution coefficient sets are calculated based on the M test point quantity sets. The distribution coefficient represents the relative frequency of each test point in the set. By dividing the frequency of occurrence of each test point by the total number of test points, the distribution coefficient of each test point in the test item is obtained, reflecting the frequency of the test point. Then, the set of point coefficient differences between the M test point distribution coefficient sets and the previously determined K defect coefficients is calculated. The point coefficient differences are used to measure the difference between the test point sets and the expected defect distribution. For each test point, the difference between its distribution coefficient and the corresponding defect coefficient is calculated to form the point coefficient difference set.

[0054] Next, the set of point coefficient difference values ​​is compared with a preset coefficient difference threshold. If the number of differences in the point coefficient difference value set that satisfy the preset coefficient difference threshold is less than or equal to the preset number of differences, the certification fails. If the certification fails, the test point extraction and point set update are performed again. Specifically, the test point extraction wheel is used multiple times based on the number of M test samples, and the set of M test points is updated based on the extraction results. Through these steps, the point transition certification module traverses the set of M test points, performs point transition certification, and ensures that the selected test points effectively represent the defect distribution of the test object.

[0055] The centralized screening and analysis module 17 performs centralized screening and analysis on the M test result sets to obtain M project test results, and uses the M project test results as the target performance test results of the target test task.

[0056] In this embodiment, when performing centralized screening and analysis on M test result sets, statistical analysis methods, such as box plots, are used for preliminary data screening and outlier detection. For example, box plots are used to calculate interquartile ranges to identify and remove outliers outside the normal range. After preliminary processing, the data for each test item is analyzed centrally. Statistical analysis software is used to perform detailed data analysis, including calculating statistical indicators such as mean, standard deviation, and coefficient of variation. To ensure the accuracy of the data analysis, further screening and verification are performed. For example, the quartile range method is used to divide the data into four intervals and calculate the first and third quartiles. By removing outliers outside the Q1 and Q3 ranges, more accurate test data is obtained.

[0057] Then, the final data for each test item are summarized and compared to generate M test results. Each test result represents the overall performance index of the corresponding test item, such as average hardness, average tensile strength, etc. Finally, the M test results are used as the target performance test results for the target test task.

[0058] Furthermore, in the system provided in the application embodiment, the centralized screening and analysis module 17 is also used for:

[0059] A first test result set is extracted from the M test result sets, wherein the first test result set includes multiple first test results, and the number of first test results in the first test result set is consistent with the number of corresponding first test samples; the interquartile range of the multiple first test results is calculated using the box plot method to determine the first test interquartile range; when the first test interquartile range is less than or equal to a preset value, the mean of the multiple first test results is calculated to obtain the first item test result; the M test result sets are then centrally screened and analyzed to obtain the M item test results.

[0060] In this embodiment, a first test result set is first randomly selected from M test result sets. This set includes multiple first test results, each with the same number of test samples. Next, the interquartile range of these first test results is calculated using a box plot. The box plot is an effective statistical analysis tool used to detect data distribution and identify outliers. By calculating the first and third quartiles of the data, the interquartile range, i.e., the distance between Q1 and Q3, is determined.

[0061] After calculating the interquartile range (ICM), check if the ICM of the first test is less than or equal to a preset value. If the ICM is less than or equal to the preset value, it indicates that the data dispersion is small and the test results are relatively stable. In this case, the mean of multiple first test results is calculated to obtain the first item test result. If the ICM of the first test is greater than the preset value, it indicates that the data dispersion is large, and there may be outliers or uneven data distribution. In this case, further data filtering and processing are performed to remove outliers that exceed the ranges of Q1 and Q3, thereby ensuring the accuracy and representativeness of the final test results.

[0062] After completing the above steps, all test results are centrally screened and analyzed. Statistical analysis software is used to perform detailed data analysis, including calculating statistical indicators such as mean, standard deviation, and coefficient of variation. Through these statistical analyses, the central tendency and dispersion of the data are assessed, and the stability and consistency of the test results are determined. After statistical analysis of all test results, M test results are generated. These results represent the overall performance indicators of the corresponding test items, such as average hardness and average tensile strength.

[0063] Furthermore, in the system provided in the application embodiment, the centralized screening and analysis module 17 is also used for:

[0064] When the interquartile range of the first test is greater than the preset value, the first quartile and the third quartile of the plurality of first test results are extracted; outliers are removed from the plurality of first test result sets based on the first quartile and the third quartile to obtain a first set of filtered test results; the mean of the first set of filtered test results is calculated to obtain the test result of the first item.

[0065] In this embodiment, when the first test interquartile range (IQR) is greater than a preset value, the first quartile (Q1) and the third quartile (Q3) of the first test result set are extracted. Next, based on the first quartile and the third quartile, outlier removal is performed on multiple first test result sets. Outliers are defined as values ​​lower than Q1 - 1.5IQR or higher than Q3 + 1.5IQR, where IQR is Q3 - Q1. Test results outside these ranges are removed to obtain the first filtered test result set.

[0066] After removing outliers, the mean of the first set of screening test results is calculated. The first test result is obtained by summing all the values ​​in the first set of screening test results and dividing the sum by the number of data points.

[0067] In summary, the embodiments of this application have at least the following technical effects:

[0068] This application extracts the structural features of the test object in the target testing task, identifies test points based on these features, and generates K test points. Based on the structural features and the K test points, it mines the frequency of performance defects in mineral castings, generating K frequency values. It determines the defect coefficients of the K test points based on the K frequency values, generating a test point extraction wheel. It interacts with M test items in the target testing task, where each M test item includes M test samples. It extracts test points multiple times from the test point extraction wheel based on the M test sample quantities, obtaining a set of M test points. It iterates through the M test point sets for point transition authentication. If authentication is successful, it performs mineral casting performance tests on the M test points, obtaining a set of M test results. It then performs centralized screening and analysis on the M test results sets to obtain M project test results, which are used as the target performance test results for the target testing task. This invention addresses the technical problem of low reliability of test results in the performance testing of mineral castings in existing technologies. By extracting structural features, mining the frequency of defects, determining the defect coefficient, extracting test points, verifying point transitions, and conducting centralized screening analysis, it achieves the technical effect of improving the accuracy of performance testing of mineral castings.

[0069] Example 2, based on the same inventive concept as the test screening system under point transition in the previous examples, such as... Figure 2 As shown in the embodiment of this application, a test screening method under point transition is provided, the method including:

[0070] Extract the structural features of the test object in the target test task, and identify test points based on the structural features to generate K test points. Based on the structural features and the K test points, mine the frequency of performance defects in the mineral casting to generate K frequency values. Determine the defect coefficients of the K test points based on the K frequency values ​​to generate a test point extraction wheel. Interact with M test items in the target test task, where each M test item includes M test samples. Extract test points multiple times from the test point extraction wheel based on the M test sample quantities to obtain a set of M test points. Iterate through the set of M test points for point transition authentication. If authentication is successful, perform mineral casting performance tests on the M test points to obtain a set of M test results. Perform centralized screening and analysis on the set of M test results to obtain M project test results, which are then used as the target performance test results for the target test task.

[0071] Furthermore, based on the K frequency values, the defect coefficients of the K test points are determined, and a test point extraction wheel is generated, including:

[0072] The ratio of each of the K frequent occurrences to the sum of the K frequent occurrences is taken as the K defect coefficients; the K defect coefficients are multiplied by the area of ​​the wheel to obtain the area of ​​the K test point wheel; the wheel is divided into regions based on the area of ​​the K test point wheel to generate a test point extraction wheel, wherein the K divided regions in the test point extraction wheel have K test point identifiers.

[0073] Furthermore, the structural features of the test object in the target test task are extracted, and test points are identified based on the structural features of the test object to generate K test points, including:

[0074] The target test task is feature extracted according to the preset structural indicators to obtain the structural features of the test object; a structural feature-test point mapping relationship is constructed; the K test points are generated by searching in the structural feature-test point mapping relationship using the structural features of the test object as the index.

[0075] Furthermore, the preset structural indicators include grain structure, external dimensions, shape characteristics, wall thickness distribution, and surface hardness.

[0076] Furthermore, point transition authentication is performed by traversing the set of M test points, including:

[0077] The M test point sets are aggregated for identical points to obtain M test point quantity sets, where each test point quantity set includes K test quantities of K test points participating in the corresponding test item; M test point distribution coefficient sets are calculated based on the M test point quantity sets; the difference set between the M test point distribution coefficient sets and the M point coefficient difference set of the K defect coefficients is calculated; when the number of differences in the M point coefficient difference set that meet the preset coefficient difference threshold is less than or equal to the preset difference number, the certification fails, and the test point extraction wheel is extracted multiple times according to the M test sample quantity, and the M test point sets are updated according to the extraction results.

[0078] Furthermore, the M test result sets are centrally filtered and analyzed to obtain M project test results, including:

[0079] A first test result set is extracted from the M test result sets, wherein the first test result set includes multiple first test results, and the number of first test results in the first test result set is consistent with the number of corresponding first test samples; the interquartile range of the multiple first test results is calculated using the box plot method to determine the first test interquartile range; when the first test interquartile range is less than or equal to a preset value, the mean of the multiple first test results is calculated to obtain the first item test result; the M test result sets are then centrally screened and analyzed to obtain the M item test results.

[0080] Furthermore, the method also includes:

[0081] When the interquartile range of the first test is greater than the preset value, the first quartile and the third quartile of the plurality of first test results are extracted; outliers are removed from the plurality of first test result sets based on the first quartile and the third quartile to obtain a first set of filtered test results; the mean of the first set of filtered test results is calculated to obtain the test result of the first item.

[0082] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0083] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0084] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A test screening system under point transition, characterized in that, The system comprises: A test point position recognition module extracts test object structure features of a target test task and performs test point position recognition based on the test object structure features to generate K test point positions; A performance defect frequency mining module mines mineral casting performance defect frequencies based on the test object structure features and the K test point positions to generate K frequency counts; A defect coefficient determination module determines defect coefficients for the K test point positions based on the K frequency counts to generate a test point position extraction wheel; A target test task interaction module interacts M test items in a target test task, wherein the M test items include M test sample quantities; A test point position acquisition module extracts the test point position extraction wheel multiple times based on the M test sample quantities to obtain M test point position sets; A point position transition authentication module performs point position transition authentication by traversing the M test point position sets, and if the authentication is passed, performs mineral casting performance tests for the M test items based on the M test point position sets to obtain M test result sets; A centralized screening analysis module performs centralized screening analysis on the M test result sets to obtain M project test results, and takes the M project test results as target performance test results of the target test task; Traversing the M test point position sets to perform point position transition authentication further comprises: Aggregating the M test point position sets to obtain M test point quantity sets, wherein each test point quantity set includes K test quantities of K test point positions participating in a corresponding test item; Calculating M test point distribution coefficient sets based on the M test point quantity sets; Calculating M point coefficient difference sets between the M test point distribution coefficient sets and the K defect coefficients; For each test point, calculating the difference between its distribution coefficient and the corresponding defect coefficient to form a point coefficient difference set; When the number of differences in the M point coefficient difference sets that meet the preset coefficient difference threshold is less than or equal to the preset difference number, the authentication is not passed, the test point position extraction wheel is extracted multiple times based on the M test sample quantities, and the M test point position sets are updated based on the extraction results.

2. The test screening system under point transition as claimed in claim 1, wherein, Determining defect coefficients for the K test point positions based on the K frequency counts to generate a test point position extraction wheel further comprises: Taking the ratio of each of the K frequency counts to the sum of the K frequency counts as a K defect coefficient; Multiplying the K defect coefficients by the wheel area to obtain K test point wheel areas; Dividing the wheel into regions based on the K test point wheel areas to generate a test point position extraction wheel, wherein the K divided regions in the test point position extraction wheel have K test point identifiers.

3. The test screening system of point transitions according to claim 1, characterized in that The method comprises the following steps: extracting a test object structure feature of a target test task, and identifying a test point based on the test object structure feature to generate K test points; and performing feature extraction on the target test task according to a preset structure index to obtain the test object structure feature; constructing a structure feature-test point mapping relationship; indexing the test object structure feature to search in the structure feature-test point mapping relationship to generate the K test points.

4. The test screening system of point transitions according to claim 3, characterized in that The preset structure index comprises a grain structure, an external size, a shape feature, a wall thickness distribution and a surface hardness.

5. The test screening system of point transitions according to claim 1, characterized in that The method further comprises the following steps: extracting a first test result set from the M test result sets, wherein the first test result set comprises a plurality of first test results, and the number of first test results in the first test result set is consistent with the number of corresponding first test samples; performing quartile range calculation on the plurality of first test results by using a box plot method to determine a first test quartile range; when the first test quartile range is less than or equal to a preset value, performing mean value calculation on the plurality of first test results to obtain a first project test result; and performing centralized screening analysis on the M test result sets to obtain the M project test results. The centralized screening analysis module further comprises the following steps: when the first test quartile range is greater than the preset value, extracting a first quartile and a third quartile of the plurality of first test results; performing outlier elimination on the plurality of first test result sets based on the first quartile and the third quartile to obtain a first screening test result set; and performing mean value calculation on the first screening test result set to obtain the first project test result. The method is performed by the test screening system in the point transition according to any one of claims 1 to 6, and comprises the following steps: extracting a test object structure feature of a target test task, and identifying a test point based on the test object structure feature to generate K test points; mining a performance defect frequency of a mineral casting based on the test object structure feature and the K test points to generate K frequency numbers; 6. The test screening system of point transitions according to claim 5, characterized in that determining a defect coefficient of the K test points based on the K frequency numbers to generate a test point extraction wheel; interacting with M test items in a target test task, wherein the M test items comprise M test sample numbers; extracting the test point extraction wheel a plurality of times according to the M test sample numbers respectively to obtain M test point sets; performing point transition authentication on the M test point sets, and if the authentication is passed, performing mineral casting performance testing on the M test items based on the M test point sets to obtain M test result sets; 7. The test screening method under point transition, characterized in that, performing centralized screening analysis on the M test result sets to obtain M project test results, and taking the M project test results as target performance test results of the target test task. ​ ​ ​ ​ ​ ​ ​

Images