CN117371626A - Casting quality prediction method, device and medium - Google Patents
Casting quality prediction method, device and medium Download PDFInfo
- Publication number
- CN117371626A CN117371626A CN202311677068.9A CN202311677068A CN117371626A CN 117371626 A CN117371626 A CN 117371626A CN 202311677068 A CN202311677068 A CN 202311677068A CN 117371626 A CN117371626 A CN 117371626A
- Authority
- CN
- China
- Prior art keywords
- performance
- prediction model
- data
- candidate input
- test
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005266 casting Methods 0.000 title claims abstract description 51
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000012360 testing method Methods 0.000 claims abstract description 111
- 230000001066 destructive effect Effects 0.000 claims abstract description 10
- 238000012549 training Methods 0.000 claims description 60
- 238000004519 manufacturing process Methods 0.000 claims description 51
- 238000005070 sampling Methods 0.000 claims description 29
- 238000012216 screening Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 230000000295 complement effect Effects 0.000 claims description 5
- 238000003860 storage Methods 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 13
- 238000011056 performance test Methods 0.000 abstract description 8
- 238000001514 detection method Methods 0.000 description 11
- 239000000463 material Substances 0.000 description 11
- 230000008569 process Effects 0.000 description 11
- 238000004891 communication Methods 0.000 description 10
- 230000007547 defect Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000004512 die casting Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 238000005457 optimization Methods 0.000 description 4
- 230000005236 sound signal Effects 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000012952 Resampling Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000001816 cooling Methods 0.000 description 2
- 238000005336 cracking Methods 0.000 description 2
- 230000006378 damage Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000005461 lubrication Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000007431 microscopic evaluation Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 239000004576 sand Substances 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000000153 supplemental effect Effects 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 238000013475 authorization Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 239000013078 crystal Substances 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000009658 destructive testing Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000009661 fatigue test Methods 0.000 description 1
- 238000011990 functional testing Methods 0.000 description 1
- 238000007542 hardness measurement Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 239000010985 leather Substances 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000009659 non-destructive testing Methods 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000002893 slag Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
- 238000001931 thermography Methods 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Manufacturing & Machinery (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- General Factory Administration (AREA)
Abstract
The disclosure relates to the technical field of electric digital data processing, in particular to a casting quality prediction method, a casting quality prediction device and a casting quality prediction medium. The method comprises the following steps: obtaining a test result of a first performance index of a target piece; and inputting the test result of the first performance index into a performance prediction model to obtain a predicted value of the second performance index of the target piece. Wherein the first performance level is tested for less destructive than the second performance level. Thus, according to the performance prediction model, the performance index which can be obtained in practice only by the high-destructive test can be predicted by performing the low-destructive test on the produced target piece. Therefore, the target parts damaged by the performance test are reduced while the performance index is accurately predicted, and the performance test cost is reduced.
Description
Technical Field
The disclosure relates to the technical field of electric digital data processing, in particular to a casting quality prediction method, a casting quality prediction device and a casting quality prediction medium.
Background
In industrial production, in order to find preferred process parameters of products, accessories, or to develop new materials, prediction and optimization can be performed by a machine learning method in addition to actual production tests. In the machine learning method, a model can be built, and the performance index of a product or a fitting is predicted according to production process parameters and material components, so that the optimization speed is greatly improved, and the research and development cost is reduced.
Disclosure of Invention
To overcome the problems in the related art, the present disclosure provides a casting quality prediction method, apparatus, and medium.
According to a first aspect of embodiments of the present disclosure, there is provided a casting quality prediction method, including:
obtaining a test result of a first performance index of a target piece;
inputting the test result of the first performance index into a performance prediction model to obtain a predicted value of a second performance index of the target piece;
wherein the first performance level is tested for less destructive than the second performance level.
Optionally, the inputting the test result of the first performance index into a performance prediction model to obtain a predicted value of the second performance index of the target piece includes:
acquiring technological parameters of the target piece in the production process;
and inputting the test result of the first performance index and the technological parameters of the target piece in the production process into the performance prediction model to obtain a predicted value of a second performance index of the target piece.
Optionally, the input dimension of the performance prediction model is determined by:
acquiring sampling data of a plurality of candidate input dimensions of the performance prediction model and target output corresponding to the performance prediction model, wherein the target output is test data of a second performance index of the target piece;
determining a correlation between the target output and each of the candidate input dimensions from the sampling data and the test data;
screening the candidate input dimensions according to the correlation corresponding to the candidate input dimensions to obtain new candidate input dimensions;
and re-executing the steps of acquiring the sampling data of a plurality of candidate input dimensions of the performance prediction model and the target output corresponding to the performance prediction model until the step of screening the candidate input dimensions according to the correlation corresponding to the candidate input dimensions to acquire new candidate input dimensions until the test data reach the production target.
Optionally, the filtering the candidate input dimensions according to the correlation corresponding to the candidate input dimensions to obtain new candidate input dimensions includes:
and taking the candidate input dimension with the highest correlation and the complementary input dimension as new candidate input dimensions.
Optionally, after the acquiring the sampling data of the plurality of candidate input dimensions of the performance prediction model and the target output corresponding to the performance prediction model, the method further includes:
adding the sampling data and the test data to a training data set;
and training the performance prediction model according to the training data set.
Optionally, the training the performance prediction model according to the training data set includes:
and training the performance prediction model according to the training data set and the weight of the training data.
Optionally, the more recently the training data is added to the training data set, the greater the weight of the training data.
Optionally, said determining a correlation between said target output and each of said candidate input dimensions from said sampling data and said test data comprises:
and determining the correlation between the target output and each candidate input dimension according to the sampling data, the testing data and the weight of the sampling data.
Optionally, the closer the time of acquisition of the sampled data, the greater the weight of the sampled data.
According to a second aspect of embodiments of the present disclosure, there is provided a casting quality prediction apparatus comprising:
the acquisition module is configured to acquire a test result of a first performance index of the target piece;
the input module is configured to input a test result of the first performance index into a performance prediction model to obtain a predicted value of a second performance index of the target piece;
wherein the first performance level is tested for less destructive than the second performance level.
According to a third aspect of embodiments of the present disclosure, there is provided a casting quality prediction apparatus comprising:
a first processor;
a memory for storing first processor-executable instructions;
wherein the first processor is configured to:
obtaining a test result of a first performance index of a target piece;
inputting the test result of the first performance index into a performance prediction model to obtain a predicted value of a second performance index of the target piece;
wherein the first performance level is tested for less destructive than the second performance level.
According to a fourth aspect of embodiments of the present disclosure, there is provided a computer readable storage medium having stored thereon computer program instructions which, when executed by a second processor, implement the steps of the casting quality prediction method provided by the first aspect of the present disclosure.
The technical scheme provided by the embodiment of the disclosure can comprise the following beneficial effects:
the input of the performance prediction model comprises a first performance indicator of the target piece with low test destructiveness, and the output of the performance prediction model comprises a second performance indicator of the target piece with high test destructiveness. Thus, according to the performance prediction model, the performance index which can be obtained in practice only by the high-destructive test can be predicted by performing the low-destructive test on the produced target piece. Therefore, the target parts damaged by the performance test are reduced while the performance index is accurately predicted, and the performance test cost is reduced.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure.
FIG. 1 is a flow chart illustrating a casting quality prediction method according to an exemplary embodiment.
FIG. 2 is a flow chart illustrating a casting quality prediction method according to another exemplary embodiment.
FIG. 3 is a block diagram illustrating a casting quality prediction apparatus according to an exemplary embodiment.
FIG. 4 is a block diagram illustrating an apparatus for casting quality prediction according to an exemplary embodiment.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present disclosure as detailed in the accompanying claims.
It should be noted that, all actions for acquiring signals, information or data in the present disclosure are performed under the condition of conforming to the corresponding data protection rule policy of the country of the location and obtaining the authorization given by the owner of the corresponding device.
FIG. 1 is a flowchart illustrating a casting quality prediction method according to an exemplary embodiment, as shown in FIG. 1, including steps S101-S102.
In step S101, a test result of a first performance index of a target piece is obtained.
In step S102, the test result of the first performance index is input into a performance prediction model to obtain a predicted value of the second performance index of the target piece. Wherein the destructiveness of the test first performance index is lower than the destructiveness of the test second performance index.
The object is a target casting, which is a product or a fitting to be die-cast, for example a large casting in a vehicle. Casting quality is used to characterize the characteristics of a casting that meet a specified need, or the ability to achieve a predetermined purpose or use, otherwise known as the performance of the casting. Casting quality may be characterized by the dimensions, shape, surface imperfections and wear, internal structural imperfections, strength, hardness, toughness, chemical composition, corrosion resistance, durability, life, etc. of the casting.
The production test can be carried out on the target piece according to the technological parameters and components selected by expert experience, and the first performance index of the produced target piece is tested.
The performance prediction model may be pre-trained. The input of the performance prediction model comprises a first performance indicator of the target piece with low test destructiveness, and the output of the performance prediction model comprises a second performance indicator of the target piece with high test destructiveness. The low-destructive test includes a non-destructive test. Nondestructive testing refers to the evaluation and detection of a material, component, or whole device (e.g., whole vehicle) by means of observation, measurement, analysis, etc., without damaging the test object.
The first performance index may include internal defect detection, external defect detection, and geometric detection with low test destructiveness. Wherein the internal defects may include detection of:
air holes: during casting, air holes may be generated in the casting;
slag inclusion: the integrity and performance of the casting are affected by impurities in the casting process;
cold partition: discontinuities due to non-uniformity in cooling rate;
decarburized layer: the loss of carbon due to the reaction of the casting surface with oxygen.
Appearance defects may include detection of the following defects:
cracking: cracks are serious appearance defects, which can lead to failure of parts;
a leather shell: surface irregularities or slight irregularities;
cold cracking: cracks generated during cooling;
sand hole: sand grains fixed on the casting.
The geometric detection may include detection of the following parameters:
dimensional tolerances: checking whether the casting meets the designed size requirement;
shape deviation: checking whether the shape of the casting is consistent with the design drawing;
flatness, roundness, straightness: the basic geometry of the casting should be within a certain tolerance.
The above internal defects can be inspected by using an X-ray or ultrasonic technique or the like. Appearance defects may be detected by means of visual inspection by a technician, light irradiation detection, photography and image analysis, infrared thermal imaging, coating agents and fluorescent materials, color difference detection, laser scanning, film thickness measurement, and the like. Geometric detection may be detected by measurement tools and equipment. The metallographic structure of the vehicle material can be subjected to microscopic analysis through metallographic and microscopic analysis, the structure and the characteristics of the structure are evaluated, and the vehicle environment adaptability test is performed by testing the adaptability of the whole vehicle under different climates and road conditions under different environment conditions.
Destructive testing refers to bringing a material, component, or whole piece of equipment (e.g., a whole vehicle) into a state of destruction by applying a load or condition in the test to obtain performance information thereof before and after the destruction. The following are destructive test categories relating to large castings as part of a complete vehicle:
collision test: performing various collision tests such as a frontal collision, a side collision, etc., evaluating the collision safety and occupant protection ability of the vehicle;
fatigue test: applying a cyclic load, simulating long-time use and conditions under various road conditions, and evaluating the durability and the service life of the whole vehicle;
strength and hardness testing: applying a static or dynamic load to evaluate the strength and hardness of the vehicle materials and components;
functional and performance testing: performing various functional tests such as dynamic performance, suspension performance and the like to evaluate the performance of the whole vehicle;
and (3) simulation test of the whole vehicle: and (3) performing virtual test by using the whole vehicle simulator, and simulating different working conditions and situations so as to predict the performance and response of the whole vehicle.
The various detected destructive levels may be pre-ordered, sorted. The input dimension and the output of the performance prediction model are determined in advance, and training is performed according to the existing data.
Because the first performance index and the second performance index both belong to the production results of the target piece, the performance prediction model in the scheme can be utilized to predict one part of production results according to the other part of production results, and the application of machine learning is enriched. And according to the performance prediction model, the performance index which can be obtained in practice only by the high-destructive test can be predicted and obtained by performing the low-destructive test on the produced target piece. Therefore, the target parts damaged by the performance test are reduced while the performance index is accurately predicted, and the performance test cost is reduced.
In yet another embodiment, inputting the test result of the first performance index into the performance prediction model to obtain a predicted value of the second performance index of the target piece includes:
acquiring technological parameters of a target piece in the production process;
and inputting the test result of the first performance index and the technological parameters of the target piece in the production process into a performance prediction model to obtain a predicted value of the second performance index of the target piece.
That is, the input of the performance prediction model in this embodiment includes, in addition to the first performance index, the process parameters of the target part during the production process. The process parameters include, for example, material composition, die casting temperature, die casting speed, die casting pressure, material lubrication parameters, design and efficiency of the casting system, microstructure and crystal size of the material, heat treatment process after die casting, etc.
Because the input of the performance prediction model comprises the technological parameters in the production process, the predicted value of the predicted second performance index is more accurate.
In yet another embodiment, the input dimension of the performance prediction model is determined by:
acquiring sampling data of a plurality of candidate input dimensions of the performance prediction model and target output corresponding to the performance prediction model, wherein the target output is test data of a second performance index of a target piece;
determining a correlation between the target output and each candidate input dimension according to the sampling data and the test data;
screening the candidate input dimensions according to the correlation corresponding to the candidate input dimensions to obtain new candidate input dimensions;
and re-executing the steps of acquiring the sampling data of a plurality of candidate input dimensions of the performance prediction model and the target output corresponding to the performance prediction model until the step of screening the candidate input dimensions according to the correlation corresponding to the candidate input dimensions to acquire new candidate input dimensions until the test data reach the production target.
In this embodiment, the determination of the input dimension may be a process of multiple iterations. The first round may be preceded by an experienced expert to determine candidate input dimensions. And then, through an actual production test, after the target piece is manufactured by using the candidate input dimension determined by an expert, detecting the target piece to obtain a first performance index and a second performance index.
For each candidate input dimension, its relevance to the output may be determined. For example, the correlation may be obtained by fitting by modeling. The candidate input dimension with higher correlation can be reserved, the candidate input dimension with lower correlation is abandoned, and the next round of production test is carried out according to the new candidate input dimension.
And obtaining a batch of production data after each round of production test, and supplementing the production data into a database storing all the production data to update the database. According to the updated data in the database, the correlation of the current candidate input dimension can be determined again, and then the next round of production test is carried out according to the new candidate input dimension. Thus, through multiple production tests and updating of candidate input dimensions, candidate input dimensions with high correlation with output can be screened out. And through a plurality of production tests, if the test data reach the production target, the corresponding process parameters can be determined as target process parameters, namely the process parameters which can meet the production target are found.
In the embodiment, through multiple production tests and iteration of candidate input dimensions, the process parameters which can meet the production targets are quickly found while the appropriate model input dimensions are found, so that process parameter optimization and model optimization are synchronously performed.
In yet another embodiment, filtering the candidate input dimensions according to the correlation corresponding to the candidate input dimensions to obtain new candidate input dimensions includes: and taking the candidate input dimension with the highest correlation and the complementary input dimension as new candidate input dimensions.
The most relevant candidate input dimensions may include a predetermined proportion of the candidate input dimensions, e.g., taking the top 50% of the candidate input dimensions that have the highest relevance. Alternatively, a predetermined number of candidate input dimensions may be included, for example, taking the top 50 of the candidate input dimensions that have the highest correlation.
The additional input dimensions may be those that the technician considers to be selected based on experience or historical test data, but are not included in the candidate input dimensions that have the highest correlation. The technician may enter additional input dimensions at each round of production trial. The supplemental input dimensions may include input dimensions that were previously determined to be candidate input dimensions, which were later discarded due to insufficient relevance, or may include factors that were not previously determined to be candidate input dimensions.
In the embodiment, according to the combination of an algorithm and manpower, new candidate input dimensions with high correlation are screened out, so that iteration is more efficient.
In yet another embodiment, after obtaining the sampled data for the plurality of candidate input dimensions of the performance prediction model and the target output corresponding to the performance prediction model, further comprising:
adding the sampling data and the test data into a training data set;
and training the performance prediction model according to the training data set.
That is, the sampled data and the test data obtained in each production test are integrated into a data set, and the performance prediction model can be trained once after each iteration. Although the candidate input dimensions for each iteration may not be exactly the same, over multiple iterations, with additional input dimensions entered by the technician, the input dimensions of the performance prediction model may be a subset of the candidate input dimensions in previous or subsequent iterations, and the test data at each iteration may be available for use with the test data at other iterations.
For example, a certain material lubrication parameter, at iteration round i, is discarded because its relevance as determined by the algorithm is low and is also not listed in the complementary input dimension. At iteration round i+5, the correlation determined by its algorithm is higher, or is listed in the supplemental input dimension and retrieved into the candidate input dimension. When training the performance prediction model, the data set containing all historical data is used for training, and all data containing the current (iterative) candidate input dimension is used for training, so that the test data is fully utilized.
FIG. 2 is a flow chart illustrating a casting quality prediction method according to another exemplary embodiment. As shown in fig. 2, the steps in the solid line box may be performed by a computer, and the steps in the broken line box may be performed manually.
In yet another embodiment, training the performance prediction model from the training dataset includes: and training the performance prediction model according to the training data set and the weight of the training data.
The training data set may include all of the historical test data obtained for each round of production testing. Each training data may have its own weight. For training, the data may be combined with its weights. The weights may be determined by the skilled artisan after each production run based on other considerations in the production process. Other considerations may include, for example, the proficiency of workers, the number of workers, the stability of the equipment, etc. during the manufacturing process. The higher the proficiency of the workers, the more workers and the more stable the equipment at the time of the test, the greater the weight of the data obtained from the wheel production test. The model may be trained, for example, by adding weighting factors in a resampling manner.
In the embodiment, the weight factors can be added to the training data obtained through the test to train the model, so that the trained performance prediction model is more accurate.
In yet another embodiment, the more recently training data is added to the training data set, the more heavily the training data.
The sampling data and the test data after each round of production test are used as training data obtained by the round of production test. After each round of production trial, training data is added to the training data set. The time at which each round of training data is added to the training data set may be the time at which each round of production test is completed. The time for adding training data of the latter round of the iteration to the training data set is shorter than the time for adding training data of the former round to the training data set. The training data with more recent time can be considered to have higher credibility, so that the more the weight is set, and the more accurate the training performance prediction model is.
In yet another embodiment, determining a correlation between the target output and each candidate input dimension from the sampled data and the test data includes: and determining the correlation between the target output and each candidate input dimension according to the sampling data, the test data and the weight of the sampling data.
That is, consideration of the weights of the sampled data may also be added in determining the correlation for each candidate input dimension. During each iteration, the correlation may be calculated using all of the historical sample data and test data obtained from all of the previous production runs. As described above, in each production process, there may be some external factors (e.g., proficiency of workers, the number of workers, the stability of equipment, etc. in the production process) that affect the reliability of the data. Therefore, the weight of the sampling data at each production run may be set based on these external factors. The weights of the sampled data may be the same or different for the same round of production trial. The correlation corresponding to the candidate input dimension can also be calculated by adding the factors of the sampled data weights in a resampling mode, so that the determined correlation is more accurate.
In yet another embodiment, the more recent the time that the sampled data is acquired, the greater the weight of the sampled data.
After each round of production test, the sampling data and the testing data corresponding to the round can be obtained. The time at which each round of corresponding sample data is acquired may be the time at which the round of production test is completed. The time of the latter cycle of sample data acquisition of the iteration is more recent than the time of the former cycle of sample data acquisition. The more recently sampled data may be considered to have a higher confidence, and thus the greater its weight is set, the more accurate the correlation is determined.
Based on the same inventive concept, the present disclosure also provides a casting quality prediction apparatus. FIG. 3 is a block diagram illustrating a casting quality prediction apparatus according to an exemplary embodiment. As shown in fig. 3, the casting quality prediction apparatus 300 includes an acquisition module 301 and an input module 302.
The acquisition module 301 is configured to acquire a test result of the first performance index of the target piece.
The input module 302 is configured to input the test result of the first performance index into the performance prediction model to obtain a predicted value of the second performance index of the target piece. Wherein the destructiveness of the test first performance index is lower than the destructiveness of the test second performance index.
Optionally, the input module 302 includes an acquisition sub-module and an input sub-module.
The acquisition sub-module is configured to acquire process parameters of the target part during the production process.
The input sub-module is configured to input a test result of the first performance index and a technological parameter of the target piece in the production process into the performance prediction model to obtain a predicted value of a second performance index of the target piece.
Optionally, the casting quality prediction apparatus 300 further includes a determination module.
The determination module is configured to determine an input dimension of the performance prediction model by:
acquiring sampling data of a plurality of candidate input dimensions of the performance prediction model and target output corresponding to the performance prediction model, wherein the target output is test data of a second performance index of a target piece;
determining a correlation between the target output and each candidate input dimension according to the sampling data and the test data;
screening the candidate input dimensions according to the correlation corresponding to the candidate input dimensions to obtain new candidate input dimensions;
and re-executing the steps of acquiring the sampling data of a plurality of candidate input dimensions of the performance prediction model and the target output corresponding to the performance prediction model until the step of screening the candidate input dimensions according to the correlation corresponding to the candidate input dimensions to acquire new candidate input dimensions until the test data reach the production target.
Optionally, the determination module is further configured to use the most relevant candidate input dimension and the complementary input dimension as new candidate input dimensions.
Optionally, the determining module is further configured to: after acquiring sample data for a plurality of candidate input dimensions of the performance prediction model and target outputs corresponding to the performance prediction model:
adding the sampling data and the test data into a training data set;
and training the performance prediction model according to the training data set.
Optionally, the determining module is further configured to train the performance prediction model according to the training data set and weights of the training data.
Optionally, the more recently training data is added to the training data set, the greater the weight of the training data.
Optionally, the determining module is further configured to: and determining the correlation between the target output and each candidate input dimension according to the sampling data, the test data and the weight of the sampling data.
Alternatively, the closer the time of acquisition of the sampled data, the greater the weight of the sampled data.
The specific manner in which the various modules perform the operations in the apparatus of the above embodiments have been described in detail in connection with the embodiments of the method, and will not be described in detail herein.
By the technical scheme, the input of the performance prediction model comprises a first performance index of the target piece with low testing destructiveness, and the output of the performance prediction model comprises a second performance index of the target piece with high testing destructiveness. Thus, according to the performance prediction model, the performance index which can be obtained in practice only by the high-destructive test can be predicted by performing the low-destructive test on the produced target piece. Therefore, the target parts damaged by the performance test are reduced while the performance index is accurately predicted, and the performance test cost is reduced.
The present disclosure also provides a casting quality prediction apparatus including a first processor and a memory for storing instructions executable by the first processor. Wherein the first processor is configured to:
obtaining a test result of a first performance index of a target piece;
and inputting the test result of the first performance index into a performance prediction model to obtain a predicted value of the second performance index of the target piece. Wherein the destructiveness of the test first performance index is lower than the destructiveness of the test second performance index.
The present disclosure also provides a computer readable storage medium having stored thereon computer program instructions which, when executed by a second processor, implement the steps of the casting quality prediction method provided by the present disclosure.
FIG. 4 is a block diagram illustrating an apparatus 800 for casting quality prediction according to an exemplary embodiment. For example, apparatus 800 may be a mobile phone, computer, digital broadcast terminal, messaging device, game console, tablet device, medical device, exercise device, personal digital assistant, or the like.
Referring to fig. 4, apparatus 800 may include one or more of the following components: a processing component 802, a memory 804, a power component 806, a multimedia component 808, an audio component 810, an input/output interface 812, a sensor component 814, and a communication component 816.
The processing component 802 generally controls overall operation of the apparatus 800, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing component 802 may include one or more third processors 820 to execute instructions to perform all or part of the steps of the methods described above. Further, the processing component 802 can include one or more modules that facilitate interactions between the processing component 802 and other components. For example, the processing component 802 can include a multimedia module to facilitate interaction between the multimedia component 808 and the processing component 802.
The memory 804 is configured to store various types of data to support operations at the apparatus 800. Examples of such data include instructions for any application or method operating on the device 800, contact data, phonebook data, messages, pictures, videos, and the like. The memory 804 may be implemented by any type or combination of volatile or nonvolatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disk.
The power supply component 806 provides power to the various components of the device 800. The power components 806 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the device 800.
The multimedia component 808 includes a screen between the device 800 and the user that provides an output interface. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensor may sense not only the boundary of a touch or slide action, but also the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front camera and/or a rear camera. The front camera and/or the rear camera may receive external multimedia data when the apparatus 800 is in an operational mode, such as a photographing mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have focal length and optical zoom capabilities.
The audio component 810 is configured to output and/or input audio signals. For example, the audio component 810 includes a Microphone (MIC) configured to receive external audio signals when the device 800 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may be further stored in the memory 804 or transmitted via the communication component 816. In some embodiments, audio component 810 further includes a speaker for outputting audio signals.
Input/output interface 812 provides an interface between processing component 802 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to: homepage button, volume button, start button, and lock button.
The sensor assembly 814 includes one or more sensors for providing status assessment of various aspects of the apparatus 800. For example, the sensor assembly 814 may detect an on/off state of the device 800, a relative positioning of the components, such as a display and keypad of the device 800, the sensor assembly 814 may also detect a change in position of the device 800 or a component of the device 800, the presence or absence of user contact with the device 800, an orientation or acceleration/deceleration of the device 800, and a change in temperature of the device 800. The sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscopic sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.
The communication component 816 is configured to facilitate communication between the apparatus 800 and other devices, either in a wired or wireless manner. The device 800 may access a wireless network based on a communication standard, such as WiFi,2G or 3G, or a combination thereof. In one exemplary embodiment, the communication component 816 receives broadcast signals or broadcast related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, the communication component 816 further includes a Near Field Communication (NFC) module to facilitate short range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, ultra Wideband (UWB) technology, bluetooth (BT) technology, and other technologies.
In an exemplary embodiment, the apparatus 800 may be implemented by one or more Application Specific Integrated Circuits (ASICs), digital Signal Processors (DSPs), digital Signal Processing Devices (DSPDs), programmable Logic Devices (PLDs), field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic elements for executing the methods described above.
In an exemplary embodiment, a non-transitory computer readable storage medium is also provided, such as memory 804 including instructions executable by third processor 820 of apparatus 800 to perform the above-described method. For example, the non-transitory computer readable storage medium may be ROM, random Access Memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.
The apparatus may be a stand-alone electronic device or may be part of a stand-alone electronic device, for example, in one embodiment, the apparatus may be an integrated circuit (Integrated Circuit, IC) or a chip, where the integrated circuit may be an IC or may be a collection of ICs; the chip may include, but is not limited to, the following: GPU (Graphics Processing Unit, graphics processor), CPU (Central Processing Unit ), FPGA (Field Programmable Gate Array, programmable logic array), DSP (Digital Signal Processor ), ASIC (Application Specific Integrated Circuit, application specific integrated circuit), SOC (System on Chip, SOC, system on Chip or System on Chip), etc. The integrated circuits or chips described above may be used to execute executable instructions (or code) to implement the casting quality prediction methods described above. The executable instructions may be stored on the integrated circuit or chip or may be retrieved from another device or apparatus, such as the integrated circuit or chip including a processor, memory, and interface for communicating with other devices. The executable instructions may be stored in the memory, which when executed by the processor, implement the casting quality prediction method described above; alternatively, the integrated circuit or chip may receive executable instructions via the interface and transmit them to the processor for execution to implement the casting quality prediction method described above.
In another exemplary embodiment, a computer program product is also provided, comprising a computer program executable by a programmable apparatus, the computer program having code portions for performing the above-described casting quality prediction method when executed by the programmable apparatus.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure. This disclosure is intended to cover any adaptations, uses, or adaptations of the disclosure following the general principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It is to be understood that the present disclosure is not limited to the precise arrangements and instrumentalities shown in the drawings, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (12)
1. A method for predicting casting quality, comprising:
obtaining a test result of a first performance index of a target piece;
inputting the test result of the first performance index into a performance prediction model to obtain a predicted value of a second performance index of the target piece; wherein the first performance level is tested for less destructive than the second performance level.
2. The method of claim 1, wherein inputting the test results of the first performance level into a performance prediction model yields a predicted value of a second performance level for the target piece, comprising:
acquiring technological parameters of the target piece in the production process;
and inputting the test result of the first performance index and the technological parameters of the target piece in the production process into the performance prediction model to obtain a predicted value of a second performance index of the target piece.
3. The method of claim 2, wherein the input dimension of the performance prediction model is determined by:
acquiring sampling data of a plurality of candidate input dimensions of the performance prediction model and target output corresponding to the performance prediction model, wherein the target output is test data of a second performance index of the target piece;
determining a correlation between the target output and each of the candidate input dimensions from the sampling data and the test data;
screening the candidate input dimensions according to the correlation corresponding to the candidate input dimensions to obtain new candidate input dimensions;
and re-executing the steps of acquiring the sampling data of a plurality of candidate input dimensions of the performance prediction model and the target output corresponding to the performance prediction model until the step of screening the candidate input dimensions according to the correlation corresponding to the candidate input dimensions to acquire new candidate input dimensions until the test data reach the production target.
4. A method according to claim 3, wherein said filtering the candidate input dimensions according to the correlation corresponding to the candidate input dimensions to obtain new candidate input dimensions comprises:
and taking the candidate input dimension with the highest correlation and the complementary input dimension as new candidate input dimensions.
5. A method according to claim 3, further comprising, after said obtaining sample data for a plurality of candidate input dimensions of said performance prediction model and a target output for said performance prediction model,:
adding the sampling data and the test data to a training data set;
and training the performance prediction model according to the training data set.
6. The method of claim 5, wherein training the performance prediction model from the training dataset comprises:
and training the performance prediction model according to the training data set and the weight of the training data.
7. The method of claim 6, wherein the training data is weighted more heavily the more recently the training data is added to the training data set.
8. A method according to claim 3, wherein said determining a correlation between said target output and each of said candidate input dimensions from said sample data and said test data comprises:
and determining the correlation between the target output and each candidate input dimension according to the sampling data, the testing data and the weight of the sampling data.
9. The method of claim 8, wherein the closer in time the sampled data is acquired, the greater the weight of the sampled data.
10. A casting quality prediction apparatus, comprising:
the acquisition module is configured to acquire a test result of a first performance index of the target piece;
the input module is configured to input a test result of the first performance index into a performance prediction model to obtain a predicted value of a second performance index of the target piece;
wherein the first performance level is tested for less destructive than the second performance level.
11. A casting quality prediction apparatus, comprising:
a first processor;
a memory for storing first processor-executable instructions;
wherein the first processor is configured to:
obtaining a test result of a first performance index of a target piece;
inputting the test result of the first performance index into a performance prediction model to obtain a predicted value of a second performance index of the target piece;
wherein the first performance level is tested for less destructive than the second performance level.
12. A computer readable storage medium having stored thereon computer program instructions, which when executed by a second processor, implement the steps of the method of any of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311677068.9A CN117371626A (en) | 2023-12-07 | 2023-12-07 | Casting quality prediction method, device and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311677068.9A CN117371626A (en) | 2023-12-07 | 2023-12-07 | Casting quality prediction method, device and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117371626A true CN117371626A (en) | 2024-01-09 |
Family
ID=89395066
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311677068.9A Pending CN117371626A (en) | 2023-12-07 | 2023-12-07 | Casting quality prediction method, device and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117371626A (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100292966A1 (en) * | 2009-05-12 | 2010-11-18 | Gm Global Technology Oeprations, Inc. | Methods of predicting residual stresses and distortion in quenched aluminum castings |
| CN105183928A (en) * | 2014-06-04 | 2015-12-23 | 通用汽车环球科技运作有限责任公司 | Quick Analysis Of Residual Stress And Distortion In Cast Aluminum Components |
| CN110059738A (en) * | 2019-04-11 | 2019-07-26 | 清华大学深圳研究生院 | A kind of method for early warning and system of quality of die casting |
| CN110261249A (en) * | 2019-07-01 | 2019-09-20 | 广西玉柴机器股份有限公司 | A method of evaluation and test castings of gray cast iron endurance limit under pull compression |
| CN111476348A (en) * | 2020-03-06 | 2020-07-31 | 南京钢铁股份有限公司 | Casting blank quality forecasting method |
| CN113486457A (en) * | 2021-06-04 | 2021-10-08 | 宁波海天金属成型设备有限公司 | Die casting defect prediction and diagnosis system |
| CN114486515A (en) * | 2021-12-15 | 2022-05-13 | 中国科学院金属研究所 | Creep graphite cast iron fatigue strength prediction method based on microstructure and tensile property |
| CN115455602A (en) * | 2022-09-23 | 2022-12-09 | 重庆长安汽车股份有限公司 | Method and system for predicting fatigue life of aluminum casting and electronic equipment |
| CN115985416A (en) * | 2021-10-13 | 2023-04-18 | 上海交通大学 | Method and system for predicting mechanical property of as-cast aluminum-silicon alloy |
| CN116242700A (en) * | 2022-12-30 | 2023-06-09 | 中国科学院金属研究所 | Vermicular cast iron fatigue strength prediction method based on microstructure content and tensile strength |
| CN117036797A (en) * | 2023-08-03 | 2023-11-10 | 大连理工大学 | Continuous casting billet longitudinal crack prediction method based on feature extraction and random forest classification |
-
2023
- 2023-12-07 CN CN202311677068.9A patent/CN117371626A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100292966A1 (en) * | 2009-05-12 | 2010-11-18 | Gm Global Technology Oeprations, Inc. | Methods of predicting residual stresses and distortion in quenched aluminum castings |
| CN105183928A (en) * | 2014-06-04 | 2015-12-23 | 通用汽车环球科技运作有限责任公司 | Quick Analysis Of Residual Stress And Distortion In Cast Aluminum Components |
| CN110059738A (en) * | 2019-04-11 | 2019-07-26 | 清华大学深圳研究生院 | A kind of method for early warning and system of quality of die casting |
| CN110261249A (en) * | 2019-07-01 | 2019-09-20 | 广西玉柴机器股份有限公司 | A method of evaluation and test castings of gray cast iron endurance limit under pull compression |
| CN111476348A (en) * | 2020-03-06 | 2020-07-31 | 南京钢铁股份有限公司 | Casting blank quality forecasting method |
| CN113486457A (en) * | 2021-06-04 | 2021-10-08 | 宁波海天金属成型设备有限公司 | Die casting defect prediction and diagnosis system |
| CN115985416A (en) * | 2021-10-13 | 2023-04-18 | 上海交通大学 | Method and system for predicting mechanical property of as-cast aluminum-silicon alloy |
| CN114486515A (en) * | 2021-12-15 | 2022-05-13 | 中国科学院金属研究所 | Creep graphite cast iron fatigue strength prediction method based on microstructure and tensile property |
| CN115455602A (en) * | 2022-09-23 | 2022-12-09 | 重庆长安汽车股份有限公司 | Method and system for predicting fatigue life of aluminum casting and electronic equipment |
| CN116242700A (en) * | 2022-12-30 | 2023-06-09 | 中国科学院金属研究所 | Vermicular cast iron fatigue strength prediction method based on microstructure content and tensile strength |
| CN117036797A (en) * | 2023-08-03 | 2023-11-10 | 大连理工大学 | Continuous casting billet longitudinal crack prediction method based on feature extraction and random forest classification |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11276158B2 (en) | Method and apparatus for inspecting corrosion defect of ladle | |
| CN108830837A (en) | A kind of method and apparatus for detecting ladle corrosion defect | |
| CN110726724A (en) | Defect detection method, system and device | |
| EP3901827B1 (en) | Image processing method and apparatus based on super network, intelligent device and computer storage medium | |
| CN116863286B (en) | Double-flow target detection method and model building method thereof | |
| CN107895119A (en) | Program installation packet inspection method, device and electronic equipment | |
| CN112884751B (en) | Device and method for synchronously measuring deformation and temperature by using ultraviolet fluorescent mark | |
| CN117371626A (en) | Casting quality prediction method, device and medium | |
| CN111047049B (en) | Method, device and medium for processing multimedia data based on machine learning model | |
| CN112800338A (en) | Method and device for recommending get-off point, electronic equipment and computer program product | |
| CN116721724A (en) | Alloy performance prediction method and device, storage medium, electronic equipment and chip | |
| EP4068202A1 (en) | Method and repair apparatus for repairing flexible panel, device and storage medium | |
| CN117371341B (en) | Process parameter determining method, target piece manufacturing device and medium | |
| CN114782845A (en) | Method, device, equipment and medium for identifying new cracks of building after blasting | |
| CN115859632B (en) | On-line source analysis model uncertainty adjustment method, device and readable storage medium | |
| CN112817870A (en) | Software testing method, device and medium | |
| CN112036507B (en) | Training method and device of image recognition model, storage medium and electronic equipment | |
| CN112733141B (en) | Information processing method and device | |
| CN113377673B (en) | Method, device and equipment for predicting test forward workload duty cycle of software test | |
| CN111382061A (en) | Test method, test device, test medium and electronic equipment | |
| CN117972437B (en) | Regional building earthquake damage prediction method and system aiming at complex terrain geological conditions | |
| CN117373582B (en) | Material composition prediction method and device based on simulation experiment model and storage medium | |
| CN117830286A (en) | Weld surface defect detection method and device, electronic equipment and storage medium | |
| CN116703874A (en) | Target detection method, device and storage medium | |
| CN116484228A (en) | Model training method, process determining method, device, electronic equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |