CN117131364A - Rolling bearing process detection integration method and system - Google Patents
Rolling bearing process detection integration method and system Download PDFInfo
- Publication number
- CN117131364A CN117131364A CN202311227226.0A CN202311227226A CN117131364A CN 117131364 A CN117131364 A CN 117131364A CN 202311227226 A CN202311227226 A CN 202311227226A CN 117131364 A CN117131364 A CN 117131364A
- Authority
- CN
- China
- Prior art keywords
- bearing
- detection
- process detection
- constraint
- node
- 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.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 553
- 238000000034 method Methods 0.000 title claims abstract description 367
- 230000008569 process Effects 0.000 title claims abstract description 335
- 230000010354 integration Effects 0.000 title claims abstract description 34
- 238000005096 rolling process Methods 0.000 title claims abstract description 22
- 238000012545 processing Methods 0.000 claims abstract description 264
- 238000003754 machining Methods 0.000 claims abstract description 140
- 238000005516 engineering process Methods 0.000 claims abstract description 37
- 238000005065 mining Methods 0.000 claims abstract description 32
- 230000008878 coupling Effects 0.000 claims description 34
- 238000010168 coupling process Methods 0.000 claims description 34
- 238000005859 coupling reaction Methods 0.000 claims description 34
- 238000007726 management method Methods 0.000 claims description 19
- 238000013507 mapping Methods 0.000 claims description 14
- 238000012216 screening Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000004140 cleaning Methods 0.000 claims description 8
- 238000012795 verification Methods 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 7
- 238000007689 inspection Methods 0.000 claims description 6
- 238000009412 basement excavation Methods 0.000 claims description 4
- 230000004927 fusion Effects 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000007417 hierarchical cluster analysis Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 5
- 230000000875 corresponding effect Effects 0.000 description 37
- 238000004519 manufacturing process Methods 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 3
- 238000012797 qualification Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000000227 grinding Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 239000000306 component Substances 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005538 encapsulation Methods 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000001845 vibrational spectrum Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The application discloses a process detection integration method and a process detection integration system for a rolling bearing, belonging to the field of intelligent detection, wherein the process detection integration method comprises the following steps: obtaining a bearing product to be produced; performing processing technology matching on bearing type characteristics to obtain bearing processing technology information; obtaining bearing machining sequence constraint, and executing characteristic constraint of bearing machining process information according to the bearing machining sequence constraint to generate a bearing machining topological network; based on the first bearing machining precision constraint, node process detection feature mining is carried out on bearing machining nodes, and a machining node process detection feature chain is generated, so that a first bearing process detection integrated network is generated; and performing process detection of the bearing product to be produced. The application solves the technical problems of incomplete and inaccurate process detection caused by the lack of integral planning and standardization of the bearing process detection in the prior art, achieves the technical effects of realizing the whole process and standardized process detection integration of the bearing and improving the accuracy and the comprehensiveness of the process detection.
Description
Technical Field
The application relates to the field of intelligent detection, in particular to a process detection integration method and system for a rolling bearing.
Background
With the development of modern industry, mechanical equipment evolves to high-speed, accurate and intelligent directions, and higher requirements are put on the processing quality of mechanical core components. Rolling bearings are used as key basic components essential for various mechanical equipment, and quality control in processing and manufacturing of the rolling bearings is directly related to the performance and service life of mechanical products. However, the existing bearing process detection technology is carried out in a segmented mode, detection among different processing segments is incomplete and inaccurate, and the overall processing quality of the bearing cannot be estimated.
Disclosure of Invention
The application provides a process detection integration method and system for a rolling bearing, and aims to solve the technical problems of incomplete and inaccurate process detection caused by lack of integral planning and standardization of bearing process detection in the prior art.
In view of the above problems, the present application provides a process detection integration method and system for a rolling bearing.
In a first aspect of the disclosure, a process detection integration method for a rolling bearing is provided, the method comprising: obtaining a first bearing product to be produced, wherein the first bearing product to be produced has a first bearing type characteristic and a first bearing machining precision constraint which correspond to the marks; performing processing technology matching on the first bearing type characteristic according to a multidimensional bearing processing source to obtain first bearing processing technology information, wherein the first bearing processing technology information comprises N bearing processing nodes, and N is a positive integer greater than 1; obtaining bearing machining sequence constraint, and executing characteristic constraint of first bearing machining process information according to the bearing machining sequence constraint to generate a bearing machining topological network, wherein the bearing machining topological network comprises a first bearing machining node and a second bearing machining node … nth bearing machining node, and N belongs to N; based on the first bearing machining precision constraint, carrying out node process detection feature mining on a first bearing machining node according to a multidimensional bearing machining source to generate a first machining node process detection feature chain, wherein the first machining node process detection feature chain comprises a plurality of process detection feature nodes corresponding to the first bearing machining node; continuously excavating node process detection characteristics of the nth bearing processing node of the second bearing processing node … according to the multidimensional bearing processing source based on the first bearing processing precision constraint to generate a second processing node process detection characteristic chain … nth processing node process detection characteristic chain; integrating the first processing node process detection feature chain, the second processing node process detection feature chain … and the nth processing node process detection feature chain to generate a first bearing process detection integrated network; and sending the first bearing process detection integrated network to a bearing processing management end, and executing process detection of the first bearing product to be produced by the bearing processing management end according to the first bearing process detection integrated network.
In another aspect of the present disclosure, a process inspection integrated system for a rolling bearing is provided, the system comprising: the bearing product acquisition module is used for acquiring a first bearing product to be produced, and the first bearing product to be produced has a first bearing type characteristic and a first bearing machining precision constraint of corresponding marks; the processing technology matching module is used for carrying out processing technology matching on the first bearing type characteristics according to the multidimensional bearing processing source to obtain first bearing processing technology information, wherein the first bearing processing technology information comprises N bearing processing nodes, and N is a positive integer greater than 1; the process characteristic constraint module is used for obtaining bearing machining sequence constraint, executing characteristic constraint of first bearing machining process information according to the bearing machining sequence constraint, and generating a bearing machining topological network, wherein the bearing machining topological network comprises a first bearing machining node and a second bearing machining node … nth bearing machining node, and N belongs to N; the detection feature mining module is used for mining node process detection features of the first bearing processing nodes according to the multidimensional bearing processing source based on the first bearing processing precision constraint to generate a first processing node process detection feature chain, wherein the first processing node process detection feature chain comprises a plurality of process detection feature nodes corresponding to the first bearing processing nodes; the other feature mining module is used for continuously mining node process detection features of the nth bearing processing node of the second bearing processing node … according to the multidimensional bearing processing source based on the first bearing processing precision constraint to generate a second processing node process detection feature chain … nth processing node process detection feature chain; the detection integrated network module is used for integrating the first processing node process detection characteristic chain, the second processing node process detection characteristic chain … and the nth processing node process detection characteristic chain to generate a first bearing process detection integrated network; the product process detection module is used for sending the first bearing process detection integrated network to the bearing processing management end, and the bearing processing management end executes process detection of the first bearing product to be produced according to the first bearing process detection integrated network.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
the bearing type characteristics and the bearing machining precision constraint of the first bearing product to be produced are obtained, so that the personalized information of the bearing required to be subjected to process detection is obtained; according to the bearing type characteristics, matching the processing technology to obtain bearing processing technology information, including bearing processing nodes, so as to realize personalized processing scheme formulation aiming at different bearing types; obtaining bearing processing sequence constraint, generating a bearing processing topological network, planning a bearing processing process, and defining the processing sequence and association relation of each procedure; performing process detection feature mining on each bearing processing node to generate a standardized feature chain, so as to realize standardization and intellectualization of process detection; integrating each process detection characteristic chain to form a process detection integrated network, so as to realize the process detection of the whole bearing flow; the integrated bearing processing management end is sent and used for executing process detection and completing the technical scheme of closed-loop control of detection, so that the technical problems of incomplete and inaccurate process detection caused by lack of integral planning and standardization in the process detection of the bearing in the prior art are solved, the technical effects of realizing the whole-flow and standardized process detection integration of the bearing and improving the accuracy and comprehensiveness of the process detection are achieved.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
Drawings
FIG. 1 is a schematic flow chart of a process detection integration method for a rolling bearing according to an embodiment of the application;
FIG. 2 is a schematic flow chart of a process detection feature chain for generating a first processing node in a process detection integration method for a rolling bearing according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of a process detection integrated system for a rolling bearing according to an embodiment of the present application.
Reference numerals illustrate: the device comprises a bearing product acquisition module 11, a processing technology matching module 12, a technology characteristic constraint module 13, a detection characteristic mining module 14, a rest characteristic mining module 15, a detection integrated network module 16 and a product technology detection module 17.
Detailed Description
The technical scheme provided by the application has the following overall thought:
the embodiment of the application provides a process detection integration method and system for a rolling bearing. Firstly, the characteristic information of a single bearing product is acquired, and the processing technology of the bearing product is accurately matched according to the personalized information, so that individuation is realized. And then constructing a topological network of the bearing processing technology, and defining the processing sequence and association relation of each working procedure to realize the overall planning and systemization of the technology detection. And carrying out standardized process detection feature extraction on each process node to form a standardized detection chain. The detection chains of the processes are integrated to generate a detection network, and the transition from dispersion to integration is completed. And finally, the detection network is applied to closed-loop control, so that the active control of detection is realized, and the accuracy, the comprehensiveness and the automation level of process detection are improved.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Example 1
As shown in fig. 1, an embodiment of the present application provides a process detection integration method for a rolling bearing, including:
obtaining a first bearing product to be produced, wherein the first bearing product to be produced has a first bearing type characteristic and a first bearing machining precision constraint which correspond to the marks;
in an embodiment of the present application, the first bearing type feature refers to a feature parameter that distinguishes between different bearing products, including but not limited to inner ring diameter, outer ring diameter, bearing type number, rolling element type, guide groove type, etc. The first bearing machining precision constraint refers to precision requirements, including dimensional precision and rotational precision, set forth for bearing machining quality.
First, a bearing product characteristic database is established, wherein the database prestores a plurality of reference bearing product examples, and each reference bearing product example comprises bearing type characteristics such as bearing type number, inner and outer diameter sizes, guide groove shape and the like, and a corresponding bearing machining precision level. Then, input information of a first bearing product to be produced is received, the input information comprises data such as product description, model specification and the like, and a corresponding reference bearing product example is searched and matched in a bearing product characteristic database according to the input information of the first bearing product to be produced. When there is a matching reference bearing product instance, extracting bearing type features of the reference instance as first bearing type features and extracting a precision level of the reference instance as a first bearing machining precision constraint.
Performing processing technology matching on the first bearing type characteristic according to a multidimensional bearing processing source to obtain first bearing processing technology information, wherein the first bearing processing technology information comprises N bearing processing nodes, and N is a positive integer greater than 1;
in the embodiment of the application, the multidimensional bearing processing source refers to various bearing processing parameters, process schemes and other information provided by a plurality of bearing processing factories.
Firstly, inquiring a multidimensional bearing processing source, and acquiring processing technology information provided by each bearing processing manufacturer for bearing products matched with the characteristics of a first bearing type, wherein the information comprises N bearing processing nodes, and N is a positive integer greater than 1. The bearing processing node refers to each key working procedure in the bearing processing process, such as inner diameter rough grinding, outer diameter rough grinding and the like. And then, by analyzing the processing node information provided by each bearing processing manufacturer, selecting a common bearing processing technology suitable for the first bearing type characteristic, and generating first bearing processing technology information which comprises N common bearing processing nodes.
And the processing technology matching is carried out on the first bearing type characteristics by inquiring the multidimensional bearing processing source, so that the first bearing processing technology information containing N bearing processing nodes is finally obtained, and the production of a bearing product with better quality is facilitated.
Obtaining bearing machining sequence constraint, and executing characteristic constraint of the first bearing machining process information according to the bearing machining sequence constraint to generate a bearing machining topological network, wherein the bearing machining topological network comprises a first bearing machining node and a second bearing machining node … nth bearing machining node, and N belongs to N;
in the embodiment of the application, the bearing processing sequence constraint refers to a processing sequence constraint condition among processing nodes of the bearing; feature constraints refer to limitations in ordering of the processing nodes in the bearing processing process information.
Firstly, a machining sequence constraint rule of N bearing machining nodes in first bearing machining process information, namely bearing machining sequence constraint, is obtained. And then, according to the machining sequence constraint rules, sequencing N shaft machining nodes in the first bearing machining process information, generating a bearing machining node sequencing result meeting the machining sequence requirement, and realizing feature constraint on the bearing machining process information. And finally, based on the ordered bearing processing node results, carrying out visual expression on the sequence association among the bearing processing nodes to form a bearing processing topological network. The topology network comprises a first bearing machining node and a second bearing machining node.
The bearing processing nodes are sequentially ordered by acquiring the bearing processing sequence constraint, and a bearing processing topological network meeting the sequence requirement is generated, so that a foundation is laid for the follow-up accurate process detection.
Performing node process detection feature mining on the first bearing processing node according to the multidimensional bearing processing source based on the first bearing processing precision constraint to generate a first processing node process detection feature chain, wherein the first processing node process detection feature chain comprises a plurality of process detection feature nodes corresponding to the first bearing processing node;
in the embodiment of the application, the node process detection feature mining refers to process detection data analysis of a bearing processing node to obtain process detection features; the process detection feature chain is a detection feature structure formed by linking a plurality of process detection feature chains.
Firstly, inquiring historical process detection data corresponding to a first bearing processing node in a multidimensional bearing processing source based on the processing precision constraint requirement of a first bearing product. These process test data are then analyzed to extract a plurality of process test features that are reflective of the quality characteristics of the first bearing processing node. Next, the process detection features are organized in the form of nodes and are chained in sequence to form a process detection feature chain corresponding to the first bearing processing node, i.e., a first processing node process detection feature chain.
And the process detection characteristic chain of the first bearing processing node is obtained through node process detection characteristic excavation under the guidance of precision constraint, and characteristic support is provided for process detection.
Continuously excavating node process detection characteristics of the nth bearing processing node of the second bearing processing node … according to the multidimensional bearing processing source based on the first bearing processing precision constraint to generate a second processing node process detection characteristic chain … nth processing node process detection characteristic chain;
in the embodiment of the application, after the process detection feature of the first bearing processing node is mined, similar operations are continued for the second bearing processing node to the nth bearing processing node. Firstly, inquiring historical process detection data corresponding to other bearing processing nodes in a multidimensional bearing processing source based on the processing precision constraint requirement of a first bearing product. Then, feature extraction is performed on the detection data of the nodes to obtain a plurality of process detection features reflecting the processing quality of the process detection features. And finally, carrying out chain organization on the process detection characteristics of each bearing processing node to form a process detection characteristic chain corresponding to the second processing node to the nth processing node, and providing support for subsequent process detection integration.
Integrating the first processing node process detection feature chain, the second processing node process detection feature chain … and the nth processing node process detection feature chain to generate a first bearing process detection integrated network;
in the embodiment of the application, the process detection integrated network refers to a bearing process detection network formed by integrating and combining process detection characteristic chains of a plurality of bearing processing nodes.
And loading process detection characteristic chains corresponding to the first bearing processing node to the nth bearing processing node, uniformly integrating the scattered node process detection characteristic chains, establishing logic association among nodes according to the topological structure of the bearing processing process, constructing an integral bearing process detection characteristic network, and being capable of predicting and integrating the production quality of the bearing and providing support for lean production.
And sending the first bearing process detection integrated network to a bearing processing management end, wherein the bearing processing management end executes process detection of the first bearing product to be produced according to the first bearing process detection integrated network.
In the embodiment of the application, the bearing processing management end refers to a computing platform for realizing bearing processing monitoring and quality prediction. And after the process detection integrated network of the first bearing product to be produced is generated, the network model is sent to a bearing processing management end through a standard communication interface. The bearing processing management end receives the process detection integrated network and loads the process detection integrated network. When a first bearing product to be produced enters an actual processing stage, the bearing processing management end collects online bearing processing process data, the online process data is input into a loaded first bearing process detection integrated network, a network model is operated, the processing quality prediction and diagnosis of the first bearing product are executed, the intelligent process detection of the whole process is completed, and the intelligent prediction, diagnosis and control of the bearing product processing process are realized.
Further, as shown in fig. 2, the embodiment of the present application further includes:
taking the first bearing processing node as a retrieval constraint, and collecting process detection item records of the multidimensional bearing processing source according to the retrieval constraint to obtain a first node process detection item record library;
performing decoupling degree cleaning on the first node process detection item record library to obtain a plurality of node process detection items corresponding to the first bearing processing node;
traversing the plurality of node process detection items based on the first bearing processing precision constraint to perform joint excavation of a detection scheme and a normal detection interval, and obtaining a plurality of item detection constraint sources corresponding to the plurality of node process detection items;
and generating a plurality of process detection feature nodes based on the plurality of node process detection items and the plurality of item detection constraint sources, and storing the plurality of process detection feature nodes in a chain manner to obtain the first processing node process detection feature chain.
In a possible implementation manner, first, a first bearing processing node, namely a bearing processing procedure node to be subjected to process detection feature mining, is determined, the first bearing processing node is used as a search keyword, a multi-dimensional bearing processing source database is queried, all historical process detection item records of the bearing processing node, namely collected sample data of various detection indexes under different processing conditions, are screened out from query results, and a first node process detection item record library is formed. And secondly, analyzing and processing sample detection items in the first node process detection item record library, eliminating redundant related detection items, and obtaining a plurality of reduced node process detection items corresponding to the first bearing processing nodes. And then traversing a plurality of simplified node process detection items based on the first bearing machining precision constraint, and carrying out detection scheme design and normal interval determination to obtain detection constraint rules corresponding to each detection item, namely an item detection constraint source.
And then, according to a plurality of node process detection items corresponding to the first bearing processing node, referring to corresponding item detection constraint sources, extracting a plurality of process detection characteristic parameters capable of representing the processing quality of the node. The process sensing feature parameters are packaged into a plurality of process sensing feature nodes, each node containing a feature parameter. And then determining the logic association sequence among the feature nodes, and linearly organizing the feature nodes to form a chain structure from input features to output results, thereby completing the generation of the process detection feature chain of the first processing node.
Further, the embodiment of the application further comprises:
the first node process detection item record library comprises a plurality of sample detection items corresponding to the first bearing processing node;
performing coincidence cleaning on the plurality of sample detection items to obtain a plurality of non-coincidence sample detection items;
obtaining a plurality of item detection features corresponding to the plurality of non-coincident sample detection items;
performing pairwise coupling degree analysis on the plurality of item detection features to obtain a plurality of item feature coupling degrees;
obtaining a preset project characteristic coupling degree, and screening the project characteristic coupling degrees according to the preset project characteristic coupling degree to obtain a plurality of project characteristic strong coupling degrees which are larger than/equal to the preset project characteristic coupling degree;
And carrying out data fusion on the plurality of non-coincident sample detection items based on the plurality of item feature decoupling degrees to generate the plurality of node process detection items.
In a preferred embodiment, the first node process test item record library covers various online and offline monitoring indicators of the current bearing manufacturing process, including a plurality of sample test items collected under different process conditions. Secondly, loading a first node process detection item record library, wherein the record library comprises a plurality of sample detection items, traversing all the sample detection items, and judging whether repetition exists among the sample items. For example, the two detection items of the bearing temperature rise curve and the bearing temperature real-time data are highly repeatable. And deleting the detected repeated sample items, only reserving one representative sample item, and cleaning the repeated sample detection items to obtain a plurality of non-coincident sample detection items.
Second, based on the obtained plurality of non-repeating sample detection items, the process detection features presented by each sample item are analyzed. For example, the sample detection item of the bearing vibration spectrum includes vibration peak value, harmonic content and the like, and reflects dynamic characteristics in the bearing rotation process. And traversing all non-repeated sample detection items, extracting corresponding characteristic parameters of the non-repeated sample detection items to form a plurality of item detection characteristics, wherein different sample detection items possibly have the same detection purpose, and therefore a plurality of detection characteristics corresponding to the same detection purpose possibly appear. Then, for a plurality of item detection features, two item detection features are constructed as a group, and all possible combinations are calculated in a traversal manner, and the combinations form a feature pair set. Then, for each feature pair, analyzing the internal association degree between the two features to obtain quantitative feature coupling degrees, thereby obtaining a plurality of item feature coupling degrees.
And then, determining a preset project feature coupling degree according to experience, traversing the obtained plurality of project feature coupling degrees as a criterion for judging whether the feature combinations are strongly related, and comparing the obtained plurality of project feature coupling degrees with the preset project feature coupling degree. If the item feature coupling degree of the feature pair is larger than or equal to the preset item feature coupling degree, judging that the feature combination has a strong association relationship and belongs to the strong coupling feature pair. Otherwise, if the correlation of the characteristics is weaker, outputting the non-coincident sample detection item as a node process detection item. And reserving all feature pairs with the item feature coupling degree larger than or equal to the preset item feature coupling degree as a plurality of item feature strong coupling degrees.
And then, combining the two sample items into a new item or reserving one item with better detection effect for the sample item with the item characteristic strong coupling degree. And directly outputting non-coincident sample detection items which are not strongly correlated as node process detection items. Thereby obtaining a plurality of node process detection items.
Further, the embodiment of the application further comprises:
obtaining an ith node process detection item based on the plurality of node process detection items, wherein i is a positive integer, i is more than or equal to 1 and less than or equal to P, and P is the total number of the plurality of node process detection items;
Performing confidence detection scheme screening based on the ith node process detection items to obtain an ith item confidence detection scheme;
based on the first bearing machining precision constraint, performing normal upper limit calibration of the ith node process detection project according to the ith project confidence detection scheme to obtain an ith normal upper limit;
based on the first bearing machining precision constraint, performing normal lower limit calibration of the ith node process detection project according to the ith project confidence detection scheme to obtain an ith normal lower limit;
generating an ith normal detection interval based on the ith normal lower limit and the ith normal upper limit;
constructing an ith detection early warning operator corresponding to the ith node process detection project, wherein the ith detection early warning operator is used for generating an ith detection early warning signal when an ith scheme detection value of the ith project confidence detection scheme does not meet the ith normal detection interval;
integrating the ith item confidence detection scheme, the ith normal detection interval and the ith detection early warning operator, generating an ith item detection constraint source corresponding to the ith node process detection item, and adding the ith item detection constraint source to the multiple item detection constraint sources.
In a preferred embodiment, first, according to the obtained multiple node process detection items after redundancy removal, an item iteration index i is set, which takes a positive integer from 1 to the total number P of node detection items. In each iteration, an ith node process detection item is selected from a plurality of detection items and is used as a current processing object. Secondly, acquiring an ith node process detection item of the current iteration, inquiring historical sample data of the detection item, collecting various detection schemes which are used by the detection item, calculating indexes such as test precision and stability of the detection schemes, obtaining confidence coefficient of each scheme, selecting one detection scheme with the highest confidence coefficient, namely the best precision and reliability, from the candidate schemes, and obtaining the ith node process detection item as a final confidence detection scheme.
Then, based on the first bearing machining precision constraint, loading an ith item confidence detection scheme, applying the detection scheme to calibrate a historical qualified sample of the ith detection item, and mining a sample maximum value reflecting good machining quality as an ith normal upper limit. And meanwhile, the confidence detection scheme of the ith item is applied, the historical qualified sample of the ith detection item is calibrated, and the minimum value of the sample reflecting good processing quality is mined and used as the ith normal upper limit. Then, a closed interval is constructed by taking the ith normal lower limit and the ith normal upper limit as interval boundaries, and the closed interval is used for representing the normal allowed floating range of the ith node process detection item.
And setting a logic judgment rule to trigger abnormality when the real-time monitoring signal of the ith item confidence detection scheme exceeds an ith normal detection interval and generate an ith detection early warning signal, so that the construction of an ith detection early warning operator corresponding to the ith node process detection item is realized. And then, collecting an ith item confidence detection scheme, an ith normal detection interval and an ith detection early warning operator which are generated for the ith node process detection item, uniformly carrying out data fusion and encapsulation to form an integral detection constraint model for the ith node process detection item, namely an ith item detection constraint source, and finally adding the ith item detection constraint source into a plurality of item constraint sources.
Further, the embodiment of the application further comprises:
sample detection scheme collection is carried out based on the ith node process detection item, and a sample detection scheme library is obtained;
traversing the sample detection scheme library to calculate confidence coefficients, and obtaining confidence coefficients of a plurality of sample schemes;
maximum value screening is carried out based on the confidence levels of the sample schemes, so that the confidence level of the optimal sample scheme is obtained;
and matching the sample detection scheme library based on the optimal sample scheme confidence, and generating the ith item confidence detection scheme.
In a preferred embodiment, first, an i-node process detection item for which a confidence detection scheme is currently required to be screened is determined, various detection schemes related to the detection item are searched in historical sample data, and the detection schemes are collected to construct a sample detection scheme library of the i-node process detection item. Secondly, traversing a sample detection scheme in the sample detection scheme, calculating indexes such as detection precision, stability, reliability and the like of the sample detection scheme, quantitatively reflecting the test efficiency of the detection scheme, normalizing all calculated indexes, and obtaining the comprehensive confidence coefficient of the sample scheme, thereby obtaining the confidence coefficients of a plurality of sample schemes. Then, loading a plurality of sample scheme confidence degrees, traversing to find the maximum value in the confidence degrees, and taking the maximum value as the optimal sample scheme confidence degree. And then, screening the matched detection scheme from the sample detection scheme library by using the optimal sample scheme confidence, and determining the detection scheme as the ith item confidence detection scheme.
Further, the embodiment of the application further comprises:
acquiring detection qualified records of the ith item confidence detection scheme based on the multidimensional bearing processing source, and acquiring a plurality of processing source-scheme detection qualified records;
Traversing a plurality of processing sources, namely, the project detection qualified records, and extracting the maximum value to obtain the project detection qualified maximum value records;
setting a qualified maximum clustering deviation threshold based on the first bearing machining precision constraint;
traversing the scheme detection qualified maximum value record to perform pairwise difference value calculation to obtain a plurality of qualified maximum value deviations;
performing hierarchical clustering analysis on the scheme detection qualified maximum value records according to the qualified maximum value cluster deviation based on the qualified maximum value cluster deviation threshold value to obtain a plurality of detection qualified maximum value cluster results;
based on the plurality of detection qualified maximum clustering results, a plurality of clustering feature quantities are obtained, and the plurality of clustering feature quantities are subjected to duty ratio calculation to obtain a plurality of clustering weight coefficients;
and carrying out weighted average calculation on the clustering results of the plurality of detection qualified maximum values based on the plurality of clustering weight coefficients, and generating the ith normal upper limit.
In a preferred embodiment, first, based on a multidimensional bearing processing source, using an i-th item confidence detection scheme as a search keyword, searching the processing source for a detection qualified record using the detection scheme, including a specific bearing processing source and a detection qualified result of the bearing source when using the i-th detection scheme, and obtaining a plurality of processing source-scheme detection qualified records. And secondly, traversing and accessing the qualified records of the multi-go processing source-scheme detection, checking the detection values in each record one by one, and storing the maximum value of the detection values in each record, namely if the maximum detection value of the current record is larger than the stored maximum value, updating the result by using the current maximum value. And accessing all the qualified detection records through traversal, recording the maximum detection value of the qualified detection records, and forming a scheme qualified detection maximum value record. Meanwhile, based on the first bearing machining precision constraint, a set qualification maximum clustering deviation threshold is determined, bearing precision requirements are considered by the threshold, and acceptable deviations of bearings of different machining sources on qualification detection values are fully reflected.
After obtaining the maximum value record of the scheme detection qualification, traversing the maximum value record, and taking out every two maximum values one by one to calculate the difference value. And calculating specific deviation among the maximum values to obtain a plurality of qualified maximum value deviations, reflecting the distribution condition of the maximum detection values of different bearing processing sources under the same detection scheme, and providing basic data support for subsequent cluster analysis. And then, based on preset qualified maximum value cluster deviation thresholds of different layers, taking out a plurality of qualified maximum value deviations in sequence as a cluster distance threshold, and comparing the plurality of qualified maximum value deviations with the cluster distance threshold. When one deviation value is smaller than or equal to the distance threshold value, the corresponding two maximum values are judged to belong to one cluster. Through traversing all the deviation values, comparing the magnitude relation with the distance threshold value, the clustering division is realized, and a plurality of detection qualified maximum clustering results are obtained.
And then, counting the maximum value record quantity contained in each clustering result after a plurality of detection qualified maximum value clustering results are obtained, namely obtaining the characteristic quantity of each cluster, and obtaining a plurality of cluster characteristic quantities. Then, the number of the cluster features is added to obtain the total number. And then, the proportion of the number of the clustering features to the total number of the features is calculated, namely the duty ratio of each cluster is obtained, and a plurality of cluster weight coefficients are obtained. After a plurality of detection qualified maximum value clustering results and corresponding clustering weight coefficients are obtained, the clustering center values of the detection qualified maximum value clustering results are multiplied by the corresponding weight coefficients, and weighting operation is carried out, so that a normal upper limit value of an ith node process detection item is obtained and is used as an ith normal upper limit.
Further, the embodiment of the application further comprises:
obtaining a second bearing product to be produced, wherein the second bearing product to be produced has second bearing type characteristics and second bearing machining precision constraints corresponding to the marks;
comparing the second bearing type characteristic with the first bearing type characteristic to obtain type characteristic similarity;
judging whether the type feature similarity meets a preset type feature similarity constraint;
if the type feature similarity meets the preset type feature similarity constraint, generating a machining precision verification instruction, and comparing the second bearing machining precision constraint with the first bearing machining precision constraint according to the machining precision verification instruction to obtain precision constraint consistency;
judging whether the precision constraint consistency degree meets precision consistency degree constraint;
if the precision constraint consistency meets the precision consistency constraint, generating a bearing process detection mapping instruction;
and mapping the first bearing process detection integration network into a second bearing process detection integration network based on the bearing process detection mapping instruction, and executing process detection of the second bearing product to be produced based on the second bearing process detection integration network.
In a preferred embodiment, first, when a second bearing product to be produced is obtained, a second bearing type feature and a second bearing machining accuracy constraint of the bearing product are obtained. And secondly, comparing the first bearing type characteristics with the second bearing type characteristics one by one to obtain type characteristic similarity, reflecting the consistency degree of the two bearing products on the type characteristics, wherein the high similarity means that the two products are relatively similar in type, and the two bearing products are expected to share a process detection network.
And after the similarity of the type features of the two bearing products is obtained, judging whether the similarity meets the preset type feature similarity constraint. The type feature similarity constraint means that the two bearing product types have similarity threshold requirements close enough, and the similarity constraint is determined through empirical data. If the type feature similarity is greater than or equal to the preset type feature similarity constraint, judging that the two product types have enough similarity consistency, and generating a machining precision verification instruction. And then, comparing the second bearing machining precision constraint with the first bearing machining precision constraint according to the machining precision verification instruction, and judging whether the two precision constraints are consistent or not to obtain precision constraint consistency. And then, judging whether the obtained precision constraint consistency meets precision consistency constraint, wherein the precision consistency constraint is set according to the process detection requirement, and if the precision constraint consistency meets the precision consistency constraint, indicating that the type characteristics and the precision constraint of the two bearing products are verified to meet the requirement, indicating that the two products have sufficient conditions for sharing the detection network. At this time, a bearing process detection mapping instruction is generated to realize rapid mapping of the two-bearing detection network.
After the bearing process detection mapping instruction is obtained, the quick mapping of the two bearing process detection integrated networks is executed, the structure, the detection elements and the like of the first bearing process detection integrated network are directly copied to generate the second bearing process detection integrated network, one-to-one mapping of the structure and the content of the two networks is realized, and the efficient automatic construction of the second bearing process detection integrated network is realized. Based on the generated second bearing process detection integrated network, process detection can be performed on the actual second bearing product to be produced, and the deployment efficiency of the process detection integrated network is improved.
In summary, the process detection integration method for the rolling bearing provided by the embodiment of the application has the following technical effects:
and obtaining a first bearing product to be produced, wherein the first bearing product to be produced has a first bearing type characteristic and a first bearing machining precision constraint which correspond to the marks, and the first bearing type characteristic and the first bearing machining precision constraint are used for guiding the personalized implementation of the subsequent process detection. And carrying out processing technology matching on the first bearing type characteristic according to the multidimensional bearing processing source to obtain first bearing processing technology information, wherein the first bearing processing technology information comprises N bearing processing nodes, N is a positive integer greater than 1, and the N is used for establishing a personalized technology detection scheme. And obtaining the bearing processing sequence constraint, executing the characteristic constraint of the first bearing processing process information according to the bearing processing sequence constraint, generating a bearing processing topological network, and defining the sequence and the association relation among the working procedures to realize the overall planning and systemization of process detection. Based on the first bearing machining precision constraint, carrying out node process detection feature mining on a first bearing machining node according to a multidimensional bearing machining source to generate a first machining node process detection feature chain, wherein the first machining node process detection feature chain comprises a plurality of process detection feature nodes corresponding to the first bearing machining node; based on the first bearing machining precision constraint, according to the multidimensional bearing machining source, node process detection feature mining is continuously carried out on the nth bearing machining node of the second bearing machining node …, a second machining node process detection feature chain … nth machining node process detection feature chain is generated, process detection feature mining is carried out on each bearing machining node, a standardized feature chain is generated, and standardization of process detection is achieved. And integrating the first processing node process detection characteristic chain, the second processing node process detection characteristic chain … and the nth processing node process detection characteristic chain to generate a first bearing process detection integrated network so as to realize integrated detection of the whole process. And sending the first bearing process detection integrated network to a bearing processing management end, and executing the process detection of the first bearing product to be produced by the bearing processing management end according to the first bearing process detection integrated network, so that the accuracy and the comprehensiveness of the process detection are improved.
Example two
Based on the same inventive concept as the process detection integration method of a rolling bearing in the foregoing embodiment, as shown in fig. 3, an embodiment of the present application provides a process detection integration system of a rolling bearing, including:
a bearing product acquisition module 11, configured to acquire a first bearing product to be produced, where the first bearing product to be produced has a first bearing type feature and a first bearing machining precision constraint that are identified correspondingly;
a machining process matching module 12, configured to perform machining process matching on the first bearing type feature according to a multidimensional bearing machining source, to obtain first bearing machining process information, where the first bearing machining process information includes N bearing machining nodes, and N is a positive integer greater than 1;
the process feature constraint module 13 is configured to obtain a bearing machining sequence constraint, and execute feature constraint of the first bearing machining process information according to the bearing machining sequence constraint to generate a bearing machining topology network, where the bearing machining topology network includes a first bearing machining node, a second bearing machining node …, and N is N;
the detection feature mining module 14 is configured to perform node process detection feature mining on the first bearing processing node according to the multidimensional bearing processing source based on the first bearing processing precision constraint, and generate a first processing node process detection feature chain, where the first processing node process detection feature chain includes a plurality of process detection feature nodes corresponding to the first bearing processing node;
The other feature mining module 15 is configured to continuously perform node process detection feature mining on the n-th bearing processing node of the second bearing processing node … according to the multidimensional bearing processing source based on the first bearing processing precision constraint, and generate a second processing node process detection feature chain … n-th processing node process detection feature chain;
the detection integration network module 16 is configured to integrate the first processing node process detection feature chain, the second processing node process detection feature chain …, and the nth processing node process detection feature chain to generate a first bearing process detection integration network;
and the product process detection module 17 is configured to send the first bearing process detection integrated network to a bearing processing management end, where the bearing processing management end performs process detection of the first bearing product to be produced according to the first bearing process detection integrated network.
Further, the detection feature mining module 14 includes the following execution steps:
taking the first bearing processing node as a retrieval constraint, and collecting process detection item records of the multidimensional bearing processing source according to the retrieval constraint to obtain a first node process detection item record library;
Performing decoupling degree cleaning on the first node process detection item record library to obtain a plurality of node process detection items corresponding to the first bearing processing node;
traversing the plurality of node process detection items based on the first bearing processing precision constraint to perform joint excavation of a detection scheme and a normal detection interval, and obtaining a plurality of item detection constraint sources corresponding to the plurality of node process detection items;
and generating a plurality of process detection feature nodes based on the plurality of node process detection items and the plurality of item detection constraint sources, and storing the plurality of process detection feature nodes in a chain manner to obtain the first processing node process detection feature chain.
Further, the detection feature mining module 14 further includes the following execution steps:
the first node process detection item record library comprises a plurality of sample detection items corresponding to the first bearing processing node;
performing coincidence cleaning on the plurality of sample detection items to obtain a plurality of non-coincidence sample detection items;
obtaining a plurality of item detection features corresponding to the plurality of non-coincident sample detection items;
performing pairwise coupling degree analysis on the plurality of item detection features to obtain a plurality of item feature coupling degrees;
Obtaining a preset project characteristic coupling degree, and screening the project characteristic coupling degrees according to the preset project characteristic coupling degree to obtain a plurality of project characteristic strong coupling degrees which are larger than/equal to the preset project characteristic coupling degree;
and carrying out data fusion on the plurality of non-coincident sample detection items based on the plurality of item feature decoupling degrees to generate the plurality of node process detection items.
Further, the detection feature mining module 14 further includes the following execution steps:
obtaining an ith node process detection item based on the plurality of node process detection items, wherein i is a positive integer, i is more than or equal to 1 and less than or equal to P, and P is the total number of the plurality of node process detection items;
performing confidence detection scheme screening based on the ith node process detection items to obtain an ith item confidence detection scheme;
based on the first bearing machining precision constraint, performing normal upper limit calibration of the ith node process detection project according to the ith project confidence detection scheme to obtain an ith normal upper limit;
based on the first bearing machining precision constraint, performing normal lower limit calibration of the ith node process detection project according to the ith project confidence detection scheme to obtain an ith normal lower limit;
Generating an ith normal detection interval based on the ith normal lower limit and the ith normal upper limit;
constructing an ith detection early warning operator corresponding to the ith node process detection project, wherein the ith detection early warning operator is used for generating an ith detection early warning signal when an ith scheme detection value of the ith project confidence detection scheme does not meet the ith normal detection interval;
integrating the ith item confidence detection scheme, the ith normal detection interval and the ith detection early warning operator, generating an ith item detection constraint source corresponding to the ith node process detection item, and adding the ith item detection constraint source to the multiple item detection constraint sources.
Further, the detection feature mining module 14 further includes the following execution steps:
sample detection scheme collection is carried out based on the ith node process detection item, and a sample detection scheme library is obtained;
traversing the sample detection scheme library to calculate confidence coefficients, and obtaining confidence coefficients of a plurality of sample schemes;
maximum value screening is carried out based on the confidence levels of the sample schemes, so that the confidence level of the optimal sample scheme is obtained;
and matching the sample detection scheme library based on the optimal sample scheme confidence, and generating the ith item confidence detection scheme.
Further, the detection feature mining module 14 further includes the following execution steps:
acquiring detection qualified records of the ith item confidence detection scheme based on the multidimensional bearing processing source, and acquiring a plurality of processing source-scheme detection qualified records;
traversing a plurality of processing sources, namely, the project detection qualified records, and extracting the maximum value to obtain the project detection qualified maximum value records;
setting a qualified maximum clustering deviation threshold based on the first bearing machining precision constraint;
traversing the scheme detection qualified maximum value record to perform pairwise difference value calculation to obtain a plurality of qualified maximum value deviations;
performing hierarchical clustering analysis on the scheme detection qualified maximum value records according to the qualified maximum value cluster deviation based on the qualified maximum value cluster deviation threshold value to obtain a plurality of detection qualified maximum value cluster results;
based on the plurality of detection qualified maximum clustering results, a plurality of clustering feature quantities are obtained, and the plurality of clustering feature quantities are subjected to duty ratio calculation to obtain a plurality of clustering weight coefficients;
and carrying out weighted average calculation on the clustering results of the plurality of detection qualified maximum values based on the plurality of clustering weight coefficients, and generating the ith normal upper limit.
Further, the embodiment of the application also comprises a bearing comparison detection module, which comprises the following execution steps:
obtaining a second bearing product to be produced, wherein the second bearing product to be produced has second bearing type characteristics and second bearing machining precision constraints corresponding to the marks;
comparing the second bearing type characteristic with the first bearing type characteristic to obtain type characteristic similarity;
judging whether the type feature similarity meets a preset type feature similarity constraint;
if the type feature similarity meets the preset type feature similarity constraint, generating a machining precision verification instruction, and comparing the second bearing machining precision constraint with the first bearing machining precision constraint according to the machining precision verification instruction to obtain precision constraint consistency;
judging whether the precision constraint consistency degree meets precision consistency degree constraint;
if the precision constraint consistency meets the precision consistency constraint, generating a bearing process detection mapping instruction;
and mapping the first bearing process detection integration network into a second bearing process detection integration network based on the bearing process detection mapping instruction, and executing process detection of the second bearing product to be produced based on the second bearing process detection integration network.
Any of the steps of the methods described above may be stored as computer instructions or programs in a non-limiting computer memory and may be called by a non-limiting computer processor to identify any method for implementing an embodiment of the present application, without unnecessary limitations.
Further, the first or second element may not only represent a sequential relationship, but may also represent a particular concept, and/or may be selected individually or in whole among a plurality of elements. It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the scope of the application. Thus, the present application is intended to include such modifications and alterations insofar as they come within the scope of the application or the equivalents thereof.
Claims (8)
1. A process inspection integration method of a rolling bearing, the method comprising:
obtaining a first bearing product to be produced, wherein the first bearing product to be produced has a first bearing type characteristic and a first bearing machining precision constraint which correspond to the marks;
performing processing technology matching on the first bearing type characteristic according to a multidimensional bearing processing source to obtain first bearing processing technology information, wherein the first bearing processing technology information comprises N bearing processing nodes, and N is a positive integer greater than 1;
Obtaining bearing machining sequence constraint, and executing characteristic constraint of the first bearing machining process information according to the bearing machining sequence constraint to generate a bearing machining topological network, wherein the bearing machining topological network comprises a first bearing machining node and a second bearing machining node … nth bearing machining node, and N belongs to N;
performing node process detection feature mining on the first bearing processing node according to the multidimensional bearing processing source based on the first bearing processing precision constraint to generate a first processing node process detection feature chain, wherein the first processing node process detection feature chain comprises a plurality of process detection feature nodes corresponding to the first bearing processing node;
continuously excavating node process detection characteristics of the nth bearing processing node of the second bearing processing node … according to the multidimensional bearing processing source based on the first bearing processing precision constraint to generate a second processing node process detection characteristic chain … nth processing node process detection characteristic chain;
integrating the first processing node process detection feature chain, the second processing node process detection feature chain … and the nth processing node process detection feature chain to generate a first bearing process detection integrated network;
And sending the first bearing process detection integrated network to a bearing processing management end, wherein the bearing processing management end executes process detection of the first bearing product to be produced according to the first bearing process detection integrated network.
2. The method of claim 1, wherein performing node process detection feature mining on the first bearing process node from the multi-dimensional bearing process source based on the first bearing process accuracy constraint, generating a first process node process detection feature chain, comprising:
taking the first bearing processing node as a retrieval constraint, and collecting process detection item records of the multidimensional bearing processing source according to the retrieval constraint to obtain a first node process detection item record library;
performing decoupling degree cleaning on the first node process detection item record library to obtain a plurality of node process detection items corresponding to the first bearing processing node;
traversing the plurality of node process detection items based on the first bearing processing precision constraint to perform joint excavation of a detection scheme and a normal detection interval, and obtaining a plurality of item detection constraint sources corresponding to the plurality of node process detection items;
And generating a plurality of process detection feature nodes based on the plurality of node process detection items and the plurality of item detection constraint sources, and storing the plurality of process detection feature nodes in a chain manner to obtain the first processing node process detection feature chain.
3. The method of claim 2, wherein performing a decoupling degree cleaning on the first node process test item record library comprises:
the first node process detection item record library comprises a plurality of sample detection items corresponding to the first bearing processing node;
performing coincidence cleaning on the plurality of sample detection items to obtain a plurality of non-coincidence sample detection items;
obtaining a plurality of item detection features corresponding to the plurality of non-coincident sample detection items;
performing pairwise coupling degree analysis on the plurality of item detection features to obtain a plurality of item feature coupling degrees;
obtaining a preset project characteristic coupling degree, and screening the project characteristic coupling degrees according to the preset project characteristic coupling degree to obtain a plurality of project characteristic strong coupling degrees which are larger than/equal to the preset project characteristic coupling degree;
and carrying out data fusion on the plurality of non-coincident sample detection items based on the plurality of item feature decoupling degrees to generate the plurality of node process detection items.
4. The method of claim 2, wherein traversing the plurality of node process detection items for joint mining of detection schemes and normal detection intervals based on the first bearing machining precision constraint, obtaining a plurality of item detection constraint sources corresponding to the plurality of node process detection items, comprises:
obtaining an ith node process detection item based on the plurality of node process detection items, wherein i is a positive integer, i is more than or equal to 1 and less than or equal to P, and P is the total number of the plurality of node process detection items;
performing confidence detection scheme screening based on the ith node process detection items to obtain an ith item confidence detection scheme;
based on the first bearing machining precision constraint, performing normal upper limit calibration of the ith node process detection project according to the ith project confidence detection scheme to obtain an ith normal upper limit;
based on the first bearing machining precision constraint, performing normal lower limit calibration of the ith node process detection project according to the ith project confidence detection scheme to obtain an ith normal lower limit;
generating an ith normal detection interval based on the ith normal lower limit and the ith normal upper limit;
constructing an ith detection early warning operator corresponding to the ith node process detection project, wherein the ith detection early warning operator is used for generating an ith detection early warning signal when an ith scheme detection value of the ith project confidence detection scheme does not meet the ith normal detection interval;
Integrating the ith item confidence detection scheme, the ith normal detection interval and the ith detection early warning operator, generating an ith item detection constraint source corresponding to the ith node process detection item, and adding the ith item detection constraint source to the multiple item detection constraint sources.
5. The method of claim 4, wherein performing a confidence detection scheme screening based on the ith node process detection item to obtain an ith item confidence detection scheme comprises:
sample detection scheme collection is carried out based on the ith node process detection item, and a sample detection scheme library is obtained;
traversing the sample detection scheme library to calculate confidence coefficients, and obtaining confidence coefficients of a plurality of sample schemes;
maximum value screening is carried out based on the confidence levels of the sample schemes, so that the confidence level of the optimal sample scheme is obtained;
and matching the sample detection scheme library based on the optimal sample scheme confidence, and generating the ith item confidence detection scheme.
6. The method of claim 4, wherein performing a normal upper bound calibration of the ith node process inspection item according to the ith item confidence inspection scheme based on the first bearing machining accuracy constraint, obtaining an ith normal upper bound, comprises:
Acquiring detection qualified records of the ith item confidence detection scheme based on the multidimensional bearing processing source, and acquiring a plurality of processing source-scheme detection qualified records;
traversing a plurality of processing sources, namely, the project detection qualified records, and extracting the maximum value to obtain the project detection qualified maximum value records;
setting a qualified maximum clustering deviation threshold based on the first bearing machining precision constraint;
traversing the scheme detection qualified maximum value record to perform pairwise difference value calculation to obtain a plurality of qualified maximum value deviations;
performing hierarchical clustering analysis on the scheme detection qualified maximum value records according to the qualified maximum value cluster deviation based on the qualified maximum value cluster deviation threshold value to obtain a plurality of detection qualified maximum value cluster results;
based on the plurality of detection qualified maximum clustering results, a plurality of clustering feature quantities are obtained, and the plurality of clustering feature quantities are subjected to duty ratio calculation to obtain a plurality of clustering weight coefficients;
and carrying out weighted average calculation on the clustering results of the plurality of detection qualified maximum values based on the plurality of clustering weight coefficients, and generating the ith normal upper limit.
7. The method of claim 1, wherein the method comprises:
Obtaining a second bearing product to be produced, wherein the second bearing product to be produced has second bearing type characteristics and second bearing machining precision constraints corresponding to the marks;
comparing the second bearing type characteristic with the first bearing type characteristic to obtain type characteristic similarity;
judging whether the type feature similarity meets a preset type feature similarity constraint;
if the type feature similarity meets the preset type feature similarity constraint, generating a machining precision verification instruction, and comparing the second bearing machining precision constraint with the first bearing machining precision constraint according to the machining precision verification instruction to obtain precision constraint consistency;
judging whether the precision constraint consistency degree meets precision consistency degree constraint;
if the precision constraint consistency meets the precision consistency constraint, generating a bearing process detection mapping instruction;
and mapping the first bearing process detection integration network into a second bearing process detection integration network based on the bearing process detection mapping instruction, and executing process detection of the second bearing product to be produced based on the second bearing process detection integration network.
8. A process inspection integration system for a rolling bearing, for implementing a process inspection integration method for a rolling bearing according to any one of claims 1 to 7, said system comprising:
the bearing product acquisition module is used for acquiring a first bearing product to be produced, and the first bearing product to be produced is provided with a first bearing type feature and a first bearing machining precision constraint which correspond to the identification;
the processing technology matching module is used for carrying out processing technology matching on the first bearing type characteristics according to a multidimensional bearing processing source to obtain first bearing processing technology information, wherein the first bearing processing technology information comprises N bearing processing nodes, and N is a positive integer greater than 1;
the process characteristic constraint module is used for obtaining bearing machining sequence constraint, executing characteristic constraint of the first bearing machining process information according to the bearing machining sequence constraint and generating a bearing machining topological network, wherein the bearing machining topological network comprises a first bearing machining node and a second bearing machining node … nth bearing machining node, and N belongs to N;
The detection feature mining module is used for mining node process detection features of the first bearing processing nodes according to the multidimensional bearing processing source based on the first bearing processing precision constraint to generate a first processing node process detection feature chain, wherein the first processing node process detection feature chain comprises a plurality of process detection feature nodes corresponding to the first bearing processing nodes;
the other feature mining module is used for continuously mining the node process detection feature of the nth bearing processing node of the second bearing processing node … according to the multidimensional bearing processing source based on the first bearing processing precision constraint to generate a second processing node process detection feature chain … nth processing node process detection feature chain;
the detection integrated network module is used for integrating the first processing node process detection characteristic chain, the second processing node process detection characteristic chain … and the nth processing node process detection characteristic chain to generate a first bearing process detection integrated network;
the product process detection module is used for sending the first bearing process detection integrated network to a bearing processing management end, and the bearing processing management end executes process detection of the first bearing product to be produced according to the first bearing process detection integrated network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311227226.0A CN117131364B (en) | 2023-09-22 | 2023-09-22 | Rolling bearing process detection integration method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311227226.0A CN117131364B (en) | 2023-09-22 | 2023-09-22 | Rolling bearing process detection integration method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117131364A true CN117131364A (en) | 2023-11-28 |
| CN117131364B CN117131364B (en) | 2024-02-09 |
Family
ID=88856436
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311227226.0A Active CN117131364B (en) | 2023-09-22 | 2023-09-22 | Rolling bearing process detection integration method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117131364B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117436771A (en) * | 2023-12-20 | 2024-01-23 | 天津撒布浪斯探测仪器有限公司 | Intelligent on-line quality monitoring and optimizing control system for vermicular cast iron products |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021035638A1 (en) * | 2019-08-29 | 2021-03-04 | 亿可能源科技（上海）有限公司 | Fault diagnosis method and system for rotary mechanical device, and storage medium |
| WO2021047021A1 (en) * | 2019-09-09 | 2021-03-18 | 平安科技（深圳）有限公司 | Information mining method and apparatus, device, and storage medium |
| CN116300765A (en) * | 2023-04-10 | 2023-06-23 | 嘉善法兰克尼亚电磁兼容有限公司 | Intelligent monitoring method and system for wave-absorbing plate production and processing process |
-
2023
- 2023-09-22 CN CN202311227226.0A patent/CN117131364B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021035638A1 (en) * | 2019-08-29 | 2021-03-04 | 亿可能源科技（上海）有限公司 | Fault diagnosis method and system for rotary mechanical device, and storage medium |
| WO2021047021A1 (en) * | 2019-09-09 | 2021-03-18 | 平安科技（深圳）有限公司 | Information mining method and apparatus, device, and storage medium |
| CN116300765A (en) * | 2023-04-10 | 2023-06-23 | 嘉善法兰克尼亚电磁兼容有限公司 | Intelligent monitoring method and system for wave-absorbing plate production and processing process |
Non-Patent Citations (1)
| Title |
|---|
| 赵乾坤;万小金;徐增丙;王凯;李清蕾;: "基于集成软竞争ART的滚动轴承性能退化趋势预测", 机械传动, no. 01, 15 January 2018 (2018-01-15) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117436771A (en) * | 2023-12-20 | 2024-01-23 | 天津撒布浪斯探测仪器有限公司 | Intelligent on-line quality monitoring and optimizing control system for vermicular cast iron products |
| CN117436771B (en) * | 2023-12-20 | 2024-03-08 | 天津撒布浪斯探测仪器有限公司 | Intelligent on-line quality monitoring and optimizing control system for vermicular cast iron products |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117131364B (en) | 2024-02-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101917006B1 (en) | Semiconductor Manufacturing Yield Prediction System and Method based on Machine Learning | |
| CN117131364B (en) | Rolling bearing process detection integration method and system | |
| CN107490964B (en) | Rotating machinery fault feature reduction method based on feature evidence discretization | |
| CN106682350B (en) | Three-dimensional model-based multi-attribute decision quality detection method | |
| CN113092981B (en) | Wafer data detection method and system, storage medium and test parameter adjustment method | |
| CN111027615B (en) | Middleware fault early warning method and system based on machine learning | |
| CN112257963B (en) | Defect prediction method and device based on spaceflight software defect data distribution outlier | |
| CN112528519A (en) | Method, system, readable medium and electronic device for engine quality early warning service | |
| CN113454552A (en) | Sensor metering data integration | |
| CN112508053A (en) | Intelligent diagnosis method, device, equipment and medium based on integrated learning framework | |
| CN109263271B (en) | Printing equipment detection and analysis method based on big data | |
| CN114118224A (en) | Neural network-based system-wide remote measurement parameter anomaly detection system | |
| CN111199361A (en) | Electric power information system health assessment method and system based on fuzzy reasoning theory | |
| CN116467674B (en) | Intelligent fault processing fusion updating system and method for power distribution network | |
| CN114064932A (en) | Data integration and mining method and device for curved surface part milling process system | |
| CN116450399A (en) | Fault diagnosis and root cause positioning method for micro service system | |
| CN107122907B (en) | Method for analyzing symbolized quality characteristics of mechanical and electrical products and tracing fault reasons | |
| CN116399630A (en) | Operation monitoring management method and system based on equipment working condition | |
| Diallo et al. | Data-based fault diagnosis model using a Bayesian causal analysis framework | |
| CN115345256B (en) | Industrial product testing system applied to intelligent manufacturing | |
| CN115469585B (en) | Hydropower unit state monitoring method and system based on big data | |
| CN110275875B (en) | Method and apparatus for providing an instantiated industrial semantic model for an industrial infrastructure | |
| CN117572837B (en) | Intelligent power plant AI active operation and maintenance method and system | |
| CN117150097B (en) | Automatic matching method for law enforcement checklist | |
| Soller et al. | Analysis of common prediction models for a fuzzy connected source target production based on time dependent significance. |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |