CN112269818B - Equipment parameter root cause positioning method, system, device and medium - Google Patents
Equipment parameter root cause positioning method, system, device and medium Download PDFInfo
- Publication number
- CN112269818B CN112269818B CN202011336989.5A CN202011336989A CN112269818B CN 112269818 B CN112269818 B CN 112269818B CN 202011336989 A CN202011336989 A CN 202011336989A CN 112269818 B CN112269818 B CN 112269818B
- Authority
- CN
- China
- Prior art keywords
- data
- analysis data
- analysis
- equipment
- attribute
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000004458 analytical method Methods 0.000 claims abstract description 109
- 238000012545 processing Methods 0.000 claims abstract description 16
- 238000012937 correction Methods 0.000 claims description 31
- 230000001419 dependent effect Effects 0.000 claims description 19
- 238000012163 sequencing technique Methods 0.000 claims description 18
- 238000004590 computer program Methods 0.000 claims description 17
- 239000011159 matrix material Substances 0.000 claims description 16
- 238000012217 deletion Methods 0.000 claims description 10
- 230000037430 deletion Effects 0.000 claims description 10
- 238000003860 storage Methods 0.000 claims description 8
- 238000012098 association analyses Methods 0.000 claims description 4
- 238000005192 partition Methods 0.000 claims description 4
- 230000009466 transformation Effects 0.000 claims description 4
- 238000009825 accumulation Methods 0.000 claims description 3
- 238000011282 treatment Methods 0.000 claims description 3
- 230000004807 localization Effects 0.000 claims 1
- 238000012847 principal component analysis method Methods 0.000 claims 1
- 238000004519 manufacturing process Methods 0.000 abstract description 11
- 230000007547 defect Effects 0.000 abstract description 6
- 238000000513 principal component analysis Methods 0.000 abstract description 6
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 239000011521 glass Substances 0.000 description 13
- 230000000694 effects Effects 0.000 description 8
- 230000008569 process Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000012216 screening Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000005816 glass manufacturing process Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/2228—Indexing structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- 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
-
- 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)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Probability & Statistics with Applications (AREA)
- Business, Economics & Management (AREA)
- Algebra (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Operations Research (AREA)
- Manufacturing & Machinery (AREA)
- Fuzzy Systems (AREA)
- Computing Systems (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- General Factory Administration (AREA)
Abstract
The invention discloses a method, a system, a device and a medium for positioning the root cause of equipment parameters, which relate to the technical field of intelligent manufacturing and artificial intelligence, and the method is based on actual performance value data of the equipment parameters, comprehensively adopts a method combining principal component analysis and mean value clustering to identify suspicious equipment, and uses a discriminant analysis method to position the suspicious parameters of the equipment; the invention not only makes up the defect that the existing method cannot effectively analyze suspicious parameters of single equipment, but also greatly reduces the time cost of the traditional analysis method, realizes quick root cause positioning, reduces the burden of manually processing a large amount of resume parameter data, improves the positioning efficiency of bad root causes and reduces the production cost.
Description
Technical Field
The invention relates to the technical field of intelligent manufacturing and artificial intelligence, in particular to a method, a system, a device and a medium for positioning equipment parameter root causes.
Background
In the glass production process, the process equipment (the process equipment refers to equipment needed to be used in the product processing and manufacturing process) can automatically keep the actual state values of the corresponding parameters in the glass manufacturing process, and for a large amount of glass in the same process, the parameter set values of the process equipment are kept consistent, but fluctuation ranges of different degrees can possibly lead to the production of bad glass. In the prior art, the common effect of multiple devices on the occurrence of faults is comprehensively analyzed on the basis of a regression method such as XGboost, logistic regression and the like on the assumption that a glass sample is sufficiently large and a fault sample is sufficient.
However, there may be a large number of defects in the data recorded during the actual production process, which makes it not effective to directly apply the conventional machine learning method to identify the fluctuation of the parameters. In addition, most common defects are that partial actual state values of parameters in equipment parameter data occur successively, and the actual performances of glass in different parallel processing equipment are probably different although parameter set values are consistent, and finally a data wide table of all glass samples cannot be formed, so that the traditional machine learning method only can consider the data collected on a single equipment. The problem that the actual state value of the equipment parameter which occurs successively corresponds to the same sample tag dependent variable is also solved, so that the generated bad effect caused by the follow-up equipment parameter fluctuation directly interferes with the correlation analysis of the previous equipment parameter fluctuation to the bad effect, and the effect of the traditional method for identifying the parameter fluctuation is also affected.
Disclosure of Invention
The invention provides a method, a system, a device and a medium for positioning root causes of equipment parameters. The invention not only makes up the defect that the existing method cannot effectively analyze suspicious parameters of single equipment, but also greatly reduces the time cost of the traditional analysis method, realizes quick root cause searching, reduces the burden of manually processing a large amount of resume parameter data, improves the positioning efficiency of bad root causes and reduces the production cost.
In order to achieve the above object, in one aspect, the present invention provides a method for positioning a root cause of a parameter of a device, a process device for a product comprising a device A 1 To apparatus A n N is an integer greater than or equal to 2, the method comprising:
for device a 1 To apparatus A n Obtaining the corresponding association strength value and correction of each deviceThe positive predictive label is used for sorting all the obtained association strength values in a descending order, obtaining correction predictive label data corresponding to a plurality of bits of equipment before sorting, and marking the correction predictive label data as fifth analysis data;
aiming at fifth analysis data, sequentially analyzing each device, extracting second analysis data attribute column data and correction prediction label data of the corresponding device as independent variables and classification dependent variables respectively, analyzing independent variable linear combination of optimal classification dependent variables to obtain combination coefficients, marking the combination coefficients as weight coefficients of device parameters, and sequencing according to the descending order of absolute values of the weight coefficients;
and integrating the combined coefficients by taking the equipment parameters as indexes to form an equipment parameter root cause positioning table, taking the product of the association strength value corresponding to each equipment in the equipment parameter root cause positioning table and the absolute value of the weight coefficient as a sorting field, sorting according to descending order to obtain a sorting result, and obtaining the equipment parameter root cause positioning result based on the sorting result.
For device a 1 To apparatus A n The following treatments were respectively carried out:
for device a i I is more than or equal to 1 and less than or equal to n, and the equipment A is provided with i The recorded actual values of the parameters are arranged to form first analysis data by taking the product names as indexes and the parameter names as attribute names;
normalizing the attribute data in the first analysis data to obtain second analysis data;
the data of the attribute column in the second analysis data are taken out, the principal component in the attribute column data is obtained, the principal component interpretation variance is sequenced according to the principal component interpretation variance, and candidate principal components are selected from the principal components based on the principal component sequencing result and the principal component accumulation interpretation variance duty ratio;
performing linear transformation on the input second analysis data by using the coefficient matrix of the candidate principal component to obtain analysis matrix data, wherein the analysis matrix data is marked as third analysis data;
for the third analysis data, clustering to form 2 categories by using a clustering algorithm, and outputting and recording a clustering label of each product sample to obtain fourth analysis data;
taking out the label data of the product, which are finally detected to be good and bad, and carrying out association analysis on fourth analysis data by taking the product name as an index, grouping according to the clustering labels of the product samples, respectively counting bad occupation ratios, and outputting the maximum bad occupation ratio in the grouping result to obtain an association strength value;
and (3) adjusting the cluster label with the largest reject ratio and the actual reject ratio to be 1 and the rest to be 0 to obtain the correction prediction label.
Preferably, the present invention is directed to device a i I is more than or equal to 1 and less than or equal to n, and the equipment A is provided with i The recorded actual values of the parameters are used as indexes, the names of the parameters are used as attribute names to form a wide table, attribute columns with the deletion rate of the wide table exceeding a preset percentage and attribute columns with the value of the attribute columns being constant and the standard deviation being 0 are removed, the reserved attribute columns are subjected to interpolation of the deletion value by the median of the attribute to obtain first analysis data, wherein the method requires that the data does not have the deletion value, if the deletion needs to be interpolated, if the deletion rate is too high, the interpolation deletion value distorts the data misleading analysis result, and the data needs to be removed; in addition, the constant value attribute does not provide any information, and the complexity of increasing the data dimension by the aid of the evidence is eliminated.
Preferably, the method centers the first analysis data according to the attribute mean and divides the first analysis data by the standard deviation to obtain second analysis data. The method firstly uses principal component analysis, requires data standardization and prevents the influence of individual attribute dimension on analysis results.
Preferably, the method obtains principal components in the attribute column data using principal component analysis. The original data has high parameter attribute dimension, high noise and collinearity, and can interfere subsequent cluster analysis, so that the main components are required to be extracted first to obtain main useful information.
The specific steps or modes for acquiring the principal components in the attribute column data by using principal component analysis are as follows:
1. calculating a covariance matrix of the second analysis data;
2. performing feature decomposition of the covariance matrix, wherein the feature vector is a principal component coefficient, and the feature value is recorded as a principal component interpretation variance;
3. and taking principal components with the cumulative sum of the orders exceeding 80% according to the descending order of interpretation variance.
Preferably, the method uses a clustering algorithm to cluster the third analysis data to form 2 categories, and in this embodiment, a Kmeans clustering algorithm is used, but the method is not limited to a Kmeans clustering algorithm, such as spectral clustering, systematic clustering, and the like.
Preferably, the method uses linear discriminant analysis to give linear combinations of independent variables of the optimal partition classification dependent variables to obtain the combination coefficients. The method is expected to find independent variables capable of identifying the classified dependent variables and contribution effects of the independent variables, and the discriminant analysis and integration meet requirements.
The method for obtaining the combination coefficient by using the independent variable linear combination of the optimal classification dependent variable given by the linear discriminant analysis specifically comprises the following steps: and setting the undetermined coefficient as weight, carrying out weighted summation on independent variables, counting the intra-group variance and the inter-group variance according to two classification groups, and dividing the intra-group variance by the inter-group variance to take the value to minimize as an optimization target, so as to obtain an optimization problem solution, namely a solution of the undetermined coefficient.
The invention also provides a system for positioning the root cause of the equipment parameters, and the processing equipment of the product comprises equipment A 1 To apparatus A n N is an integer greater than or equal to 2, the system comprising:
a first analysis data obtaining unit to be processed for the device A 1 To apparatus A n Obtaining an association strength value and a correction prediction label corresponding to each device, sorting all the obtained association strength values in a descending order, obtaining correction prediction label data corresponding to a plurality of devices before sorting, and marking the correction prediction label data as first to-be-processed analysis data;
a second analysis data obtaining unit for obtaining analysis data for the device A i I is more than or equal to 1 and less than or equal to n, and the equipment A is provided with i Taking the recorded actual value of the parameter as an index, taking the parameter name as an attribute name, and carrying out standardization processing on the attribute data to obtain second analysis data to be processed;
a weight coefficient obtaining and sorting unit for device parameters, which is used for sequentially analyzing each device for the first to-be-processed analysis data, extracting attribute column data and correction prediction label data in the second to-be-processed analysis data corresponding to the device as independent variables and classification dependent variables respectively, analyzing independent variable linear combination of optimal partition classification dependent variables to obtain combination coefficients, wherein the combination coefficients are recorded as weight coefficients of equipment parameters, and are sorted according to descending order of absolute values of the weight coefficients;
the device parameter root cause positioning result obtaining unit is used for integrating the combination coefficients by taking the device parameters as indexes to form a device parameter root cause positioning table, taking the product of the association strength value corresponding to each device in the device parameter root cause positioning table and the absolute value of the weight coefficient as a sequencing field, sequencing the device parameter root cause positioning table according to descending order to obtain a sequencing result, and obtaining the device parameter root cause positioning result based on the sequencing result.
The invention also provides a device for positioning the root cause of the equipment parameter, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the steps of the method for positioning the root cause of the equipment parameter when executing the computer program.
The invention also provides a computer readable storage medium storing a computer program which when executed by a processor implements the steps of the device parameter root cause positioning method.
The one or more technical schemes provided by the invention have at least the following technical effects or advantages:
generally, the production resume equipment parameter data has the characteristics of large data volume, dispersed effective information, complex flow and the like, and the traditional manual search analysis and investigation mode has low efficiency. The invention can form an automatic searching and matching step, all the most suspicious root causes are listed, and the most suspicious root causes are sorted in descending order according to the degree of suspicion of equipment parameters, so that technicians are assisted to directly verify from the most suspicious root causes, and the most suspicious root causes are positioned as much as possible at the highest speed.
The invention overcomes the defects of slow speed and unsatisfactory effect of the traditional automatic analysis method on the analysis of the equipment parameter data, not only eliminates the interference of the follow-up equipment parameters on the analysis of poor correlation, but also can easily expand and consider the combined action of different equipment parameters by combining with the path analysis, and has the advantages of practicability and reliability.
Drawings
The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention;
FIG. 1 is a schematic diagram of a method for positioning the root cause of a parameter of a device according to the present invention;
FIG. 2 is a schematic diagram of the components of the system for locating the root cause of a parameter of a device according to the present invention.
Detailed Description
In order that the above-recited objects, features and advantages of the present invention will be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description. In addition, the embodiments of the present invention and the features in the embodiments may be combined with each other without collision.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than within the scope of the description, and the scope of the invention is therefore not limited to the specific embodiments disclosed below.
It will be understood that the terms "a" and "an" should be interpreted as referring to "at least one" or "one or more," i.e., in one embodiment, the number of elements may be one, while in another embodiment, the number of elements may be plural, and the term "a" should not be interpreted as limiting the number.
Example 1
The invention will be specifically described below by taking a product as glass as an example, and in practical application, the product can be other types of products, and the invention is not limited to the specific types of products.
Referring to fig. 1, the invention provides a method for positioning root causes of equipment parameters by comprehensively applying a principal component clustering method and a discriminant analysis method.
The first embodiment of the invention provides a method for positioning the root cause of equipment parameters, and the manufacturing equipment of the product comprises equipment A 1 To apparatus A n N is an integer greater than or equal to 2, the method comprising:
for device a 1 To apparatus A n Obtaining an association strength value and a correction prediction label corresponding to each device, sorting all the obtained association strength values in a descending order, obtaining correction prediction label data corresponding to a plurality of devices before sorting, and marking the correction prediction label data as first to-be-processed analysis data;
for device a i I is more than or equal to 1 and less than or equal to n, and the equipment A is provided with i Taking the recorded actual value of the parameter as an index, taking the parameter name as an attribute name, and carrying out standardization processing on the attribute data to obtain second analysis data to be processed;
for the first to-be-processed analysis data, sequentially analyzing each device, extracting attribute column data and correction prediction tag data in the second to-be-processed analysis data corresponding to the device as independent variables and classification dependent variables respectively, analyzing independent variable linear combination of optimal division classification dependent variables to obtain combination coefficients, recording the combination coefficients as weight coefficients of device parameters, and sequencing according to the descending order of absolute values of the weight coefficients;
and integrating the combined coefficients by taking the equipment parameters as indexes to form an equipment parameter root cause positioning table, taking the product of the association strength value corresponding to each equipment in the equipment parameter root cause positioning table and the absolute value of the weight coefficient as a sorting field, sorting according to descending order to obtain a sorting result, and obtaining the equipment parameter root cause positioning result based on the sorting result.
The method comprises the following specific implementation steps:
Step1:
assuming analysis processing equipment A, the actual values of parameters recorded by the equipment are indexed by glass names, the parameter names are arranged by attribute names to form a wide table, attribute columns with the deletion rate exceeding 30% and constant attribute columns are removed, the reserved attribute columns are interpolated by the deletion value of the median of the attribute, the result table is shown as an analysis table 1, wherein the percentage of the removed attribute columns can be flexibly adjusted according to actual needs, and the embodiment of the invention is not particularly limited.
Step2:
The analysis table 1 is normalized by the attribute mean value and divided by the standard deviation, and the result is expressed as analysis 2.
Step3:
The data of the attribute column in the analysis table 2 is taken out, principal components are found by using principal component analysis and are sorted in descending order of interpretation variance, and candidate principal components which are the principal components in front and whose accumulated variance interpretation ratio exceeds 80% are selected. And (3) performing linear transformation on the input analysis data by using the screened principal component coefficient matrix to obtain a scoring matrix, wherein matrix table data are recorded as analysis data 3. The selection percentage of the candidate principal components can be flexibly adjusted according to actual needs, and the embodiment of the invention is not particularly limited. The principal component coefficient matrix is a matrix formed by obtaining a feature vector column corresponding to a principal component according to a pre-description screening method, the screening is a post-screening result which is formed by sorting the principal component according to an interpretation variance descending order and accumulating the condition that the interpretation variance ratio exceeds 80% after the principal component analysis, and the screening is performed to extract main information in an original high-dimensional variable and avoid noise and co-linearity interference.
Step4:
Taking out the analysis data 3, clustering the analysis data into 2 categories by using a Kmeans clustering algorithm, outputting and recording a clustering label of each glass sample, and recording the clustering label as analysis data 4; other clustering algorithms can be selected for clustering in practical application, and the embodiment does not limit specific clustering modes and algorithms.
Step5:
And (3) taking out the label data of the final detected good and bad glass, taking the glass name as index association analysis data 4, grouping according to the clustering labels of the glass samples, respectively counting the bad occupation ratio, outputting the maximum bad occupation ratio in the grouping result, and marking the maximum bad occupation ratio as an association strength value.
Step6:
The cluster label with the largest bad proportion and the actual bad is adjusted to be 1, the rest is adjusted to be 0, and the cluster label is recorded as a correction prediction label
Step7:
And (3) executing the operations of the steps 1-6 on each device, outputting corresponding association strength values, sorting the association strength values in descending order, enabling the device with the earlier sorting to be more suspicious, and acquiring correction prediction label data corresponding to the device with the earlier sorting, and recording the correction prediction label data as analysis data 5.
Step8:
And (5) extracting analysis data 5, sequentially analyzing each device, extracting attribute column data and correction prediction label data of an analysis table 2 of the corresponding device, respectively serving as independent variables and classification dependent variables, using linear discriminant analysis to give independent variable linear combination of optimal classification dependent variables, marking the combination coefficients as weight coefficients of device parameters, and sorting according to an absolute value descending order, wherein the more previous parameters are suspicious.
Step9:
And integrating the combined coefficients into a table by taking the equipment parameters as indexes, calculating the product of the correlation strength value and the absolute value of the weight coefficient as a sorting field, sorting in descending order, and assisting in positioning the root cause of the equipment parameters with bad glass when the former equipment parameters are more suspicious.
The method can form an automatic searching and matching step, all the most suspicious root causes are listed, and the most suspicious root causes are sorted in descending order according to the degree of suspicion of equipment parameters, so that technicians are assisted to directly verify from the most suspicious root causes, and the most suspicious root causes are positioned as much as possible at the highest speed.
The method overcomes the defects of slow speed and unsatisfactory effect of the traditional automatic analysis method on the analysis of the equipment parameter data, eliminates the interference of the follow-up equipment parameters on the analysis of poor correlation, can be combined with the path analysis to easily expand and consider the combined action of different equipment parameters, and has the advantages of practicability and reliability.
Example two
Referring to fig. 2, a second embodiment of the present invention provides an equipment parameter root cause positioning system, and a product manufacturing apparatus includes an equipment a 1 To apparatus A n N is an integer greater than or equal to 2, the system comprising:
a first analysis data obtaining unit to be processed for the device A 1 To apparatus A n Obtaining the corresponding association strength value and correction prediction label of each device, and obtainingAll the obtained association strength values are sorted in descending order, correction prediction tag data corresponding to a plurality of bits of equipment before sorting is obtained, and the correction prediction tag data is recorded as first analysis data to be processed;
a second analysis data obtaining unit for obtaining analysis data for the device A i I is more than or equal to 1 and less than or equal to n, and the equipment A is provided with i Taking the recorded actual value of the parameter as an index, taking the parameter name as an attribute name, and carrying out standardization processing on the attribute data to obtain second analysis data to be processed;
a weight coefficient obtaining and sorting unit for device parameters, which is used for sequentially analyzing each device for the first to-be-processed analysis data, extracting attribute column data and correction prediction label data in the second to-be-processed analysis data corresponding to the device as independent variables and classification dependent variables respectively, analyzing independent variable linear combination of optimal partition classification dependent variables to obtain combination coefficients, wherein the combination coefficients are recorded as weight coefficients of equipment parameters, and are sorted according to descending order of absolute values of the weight coefficients;
the device parameter root cause positioning result obtaining unit is used for integrating the combination coefficients by taking the device parameters as indexes to form a device parameter root cause positioning table, taking the product of the association strength value corresponding to each device in the device parameter root cause positioning table and the absolute value of the weight coefficient as a sequencing field, sequencing the device parameter root cause positioning table according to descending order to obtain a sequencing result, and obtaining the device parameter root cause positioning result based on the sequencing result.
Example III
The third embodiment of the invention provides a device parameter root cause positioning device, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the steps of the device parameter root cause positioning method when executing the computer program.
The processor may be a central processing unit (CPU, central Processing Unit), other general purpose processors, digital signal processors (digital signal processor), application specific integrated circuits (Application Specific Integrated Circuit), off-the-shelf programmable gate arrays (Fieldprogrammable gate array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may be used to store the computer program and/or the module, and the processor may implement various functions of the inventive device parameter root cause locating apparatus by running or executing the data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, application programs required for at least one function (such as a sound playing function, an image playing function, etc.), and the like. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart memory card, secure digital card, flash memory card, at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device.
Example IV
A fourth embodiment of the present invention provides a computer readable storage medium storing a computer program, where the computer program when executed by a processor implements the steps of the device parameter root cause positioning method.
The device parameter root cause location means may be stored in a computer readable storage medium if implemented in the form of a software functional unit and sold or used as a stand alone product. Based on such understanding that the present invention implements all or part of the flow of the method of the above-described embodiments, the steps of each method embodiment described above may also be implemented by a computer program stored in a computer readable storage medium, where the computer program when executed by a processor. Wherein the computer program comprises computer program code, object code forms, executable files, or some intermediate forms, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash disk, a removable hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, a point carrier signal, a telecommunication signal, a software distribution medium, and the like. It should be noted that the content of the computer readable medium can be appropriately increased or decreased according to the requirements of the legislation and the patent practice in the jurisdiction.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (9)
1. A method for positioning the root cause of equipment parameters is characterized in that the processing equipment of a product comprises equipment A 1 To apparatus A n N is an integer greater than or equal to 2, the method comprising:
for device a 1 To apparatus A n Obtaining an association strength value and a correction prediction label corresponding to each device, sorting all the obtained association strength values in a descending order, obtaining correction prediction label data corresponding to a plurality of devices before sorting, and marking the correction prediction label data as first to-be-processed analysis data;
for device a i I is more than or equal to 1 and less than or equal to n, and the equipment A is provided with i Taking the recorded actual value of the parameter as an index, taking the parameter name as an attribute name, and carrying out standardization processing on the attribute data to obtain second analysis data to be processed;
for the first to-be-processed analysis data, sequentially analyzing each device, extracting attribute column data and correction prediction tag data in the second to-be-processed analysis data corresponding to the device as independent variables and classification dependent variables respectively, analyzing independent variable linear combination of optimal division classification dependent variables to obtain combination coefficients, recording the combination coefficients as weight coefficients of device parameters, and sequencing according to the descending order of absolute values of the weight coefficients;
integrating the combined coefficients by taking the equipment parameters as indexes to form an equipment parameter root cause positioning table, taking the product of the correlation strength value corresponding to each equipment in the equipment parameter root cause positioning table and the absolute value of the weight coefficient as a sorting field, sorting according to descending order to obtain a sorting result, and obtaining an equipment parameter root cause positioning result based on the sorting result;
for device a 1 To apparatus A n The following treatments were respectively carried out:
for device a i Device A is set up i The recorded actual values of the parameters are arranged to form first analysis data by taking the product names as indexes and the parameter names as attribute names;
normalizing the attribute data in the first analysis data to obtain second analysis data;
the data of the attribute column in the second analysis data are taken out, the principal component in the attribute column data is obtained, the principal component interpretation variance is sequenced according to the principal component interpretation variance, and candidate principal components are selected from the principal components based on the principal component sequencing result and the principal component accumulation interpretation variance duty ratio;
performing linear transformation on the input second analysis data by using the coefficient matrix of the candidate principal component to obtain analysis matrix data, wherein the analysis matrix data is marked as third analysis data;
for the third analysis data, clustering to form 2 categories by using a clustering algorithm, and outputting and recording a clustering label of each product sample to obtain fourth analysis data;
taking out the label data of the product, which are finally detected to be good and bad, and carrying out association analysis on fourth analysis data by taking the product name as an index, grouping according to the clustering labels of the product samples, respectively counting bad occupation ratios, and outputting the maximum bad occupation ratio in the grouping result to obtain an association strength value;
and (3) adjusting the cluster label with the largest reject ratio and the actual reject ratio to be 1 and the rest to be 0 to obtain the correction prediction label.
2. A device parameter root cause positioning method according to claim 1, characterized in that for device a i Device A is set up i The actual value of the recorded parameter is indexed by the product nameAnd (3) arranging the parameter names into a wide table for the attribute names, removing attribute columns with the deletion rate exceeding a preset percentage of the wide table and attribute columns with the values of the attribute columns being constant and the standard deviation being 0, and interpolating the deletion values of the reserved attribute columns by using the median of the attributes to obtain first analysis data.
3. The method for positioning a root cause of a parameter of a device according to claim 1, wherein the method centers the first analysis data according to the attribute mean and divides the first analysis data by a standard deviation to obtain the second analysis data.
4. The method of claim 1, wherein the principal component analysis method is used to obtain principal components in the attribute column data.
5. The method of claim 1, wherein the method uses Kmeans clustering algorithm to cluster into 2 categories for the third analysis data.
6. The method of claim 1, wherein the method uses linear discriminant analysis to provide linear combinations of independent variables of the optimal partitioned classification dependent variables to obtain the combined coefficients.
7. An equipment parameter root cause positioning system is characterized in that the processing equipment of a product comprises equipment A 1 To apparatus A n N is an integer greater than or equal to 2, the system comprising:
a first analysis data obtaining unit to be processed for the device A 1 To apparatus A n Obtaining an association strength value and a correction prediction label corresponding to each device, sorting all the obtained association strength values in a descending order, obtaining correction prediction label data corresponding to a plurality of devices before sorting, and marking the correction prediction label data as first to-be-processed analysis data;
a second analysis data obtaining unit for obtaining analysis data for the device A i ，1≤i≤n, device A i Taking the recorded actual value of the parameter as an index, taking the parameter name as an attribute name, and carrying out standardization processing on the attribute data to obtain second analysis data to be processed;
a weight coefficient obtaining and sorting unit for device parameters, which is used for sequentially analyzing each device for the first to-be-processed analysis data, extracting attribute column data and correction prediction label data in the second to-be-processed analysis data corresponding to the device as independent variables and classification dependent variables respectively, analyzing independent variable linear combination of optimal partition classification dependent variables to obtain combination coefficients, wherein the combination coefficients are recorded as weight coefficients of equipment parameters, and are sorted according to descending order of absolute values of the weight coefficients;
the device parameter root cause positioning result obtaining unit is used for integrating the combination coefficients by taking the device parameters as indexes to form a device parameter root cause positioning table, taking the product of the association strength value corresponding to each device in the device parameter root cause positioning table and the absolute value of the weight coefficient as a sequencing field, sequencing the combination coefficients according to descending order to obtain a sequencing result, and obtaining the device parameter root cause positioning result based on the sequencing result;
a processing unit for aiming at the equipment A 1 To apparatus A n The following treatments were respectively carried out:
for device a i Device A is set up i The recorded actual values of the parameters are arranged to form first analysis data by taking the product names as indexes and the parameter names as attribute names;
normalizing the attribute data in the first analysis data to obtain second analysis data;
the data of the attribute column in the second analysis data are taken out, the principal component in the attribute column data is obtained, the principal component interpretation variance is sequenced according to the principal component interpretation variance, and candidate principal components are selected from the principal components based on the principal component sequencing result and the principal component accumulation interpretation variance duty ratio;
performing linear transformation on the input second analysis data by using the coefficient matrix of the candidate principal component to obtain analysis matrix data, wherein the analysis matrix data is marked as third analysis data;
for the third analysis data, clustering to form 2 categories by using a clustering algorithm, and outputting and recording a clustering label of each product sample to obtain fourth analysis data;
taking out the label data of the product, which are finally detected to be good and bad, and carrying out association analysis on fourth analysis data by taking the product name as an index, grouping according to the clustering labels of the product samples, respectively counting bad occupation ratios, and outputting the maximum bad occupation ratio in the grouping result to obtain an association strength value;
and (3) adjusting the cluster label with the largest reject ratio and the actual reject ratio to be 1 and the rest to be 0 to obtain the correction prediction label.
8. A device parameter root cause positioning apparatus comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of a device parameter root cause positioning method according to any of claims 1-6 when the computer program is executed by the processor.
9. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the steps of a device parameter root cause localization method as claimed in any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011336989.5A CN112269818B (en) | 2020-11-25 | 2020-11-25 | Equipment parameter root cause positioning method, system, device and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011336989.5A CN112269818B (en) | 2020-11-25 | 2020-11-25 | Equipment parameter root cause positioning method, system, device and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112269818A CN112269818A (en) | 2021-01-26 |
| CN112269818B true CN112269818B (en) | 2023-11-21 |
Family
ID=74340414
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011336989.5A Active CN112269818B (en) | 2020-11-25 | 2020-11-25 | Equipment parameter root cause positioning method, system, device and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112269818B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE112021001736T5 (en) * | 2021-03-26 | 2023-01-05 | Beijing Zhongxiangying Technology Co., Ltd. | DATA PROCESSING METHODS AND DEVICES, ELECTRONIC DEVICE AND STORAGE MEDIA |
| CN113552856B (en) * | 2021-09-22 | 2021-12-10 | 成都数之联科技有限公司 | Process parameter root factor positioning method and related device |
| CN114239323B (en) * | 2022-02-22 | 2022-04-26 | 成都数之联科技股份有限公司 | Root cause positioning method, device, equipment, medium and product for production abnormity |
| CN114444986B (en) * | 2022-04-11 | 2022-06-03 | 成都数之联科技股份有限公司 | Product analysis method, system, device and medium |
| CN114780298B (en) * | 2022-06-16 | 2022-09-06 | 深圳市慧为智能科技股份有限公司 | File data processing method and device, computer terminal and storage medium |
| CN117194963B (en) * | 2023-11-02 | 2024-02-09 | 合肥喆塔科技有限公司 | Industrial FDC quality root cause analysis method, device and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA3156908A1 (en) * | 2015-01-06 | 2016-07-14 | David Burton | Mobile wearable monitoring systems |
| CN107577460A (en) * | 2017-08-29 | 2018-01-12 | 苏州优圣美智能系统有限公司 | A kind of method from unstructured data extraction structural data |
| CN110888755A (en) * | 2019-11-15 | 2020-03-17 | 亚信科技（中国）有限公司 | Method and device for searching abnormal root node of micro-service system |
| CN111082966A (en) * | 2019-11-01 | 2020-04-28 | 平安科技（深圳）有限公司 | Positioning method and device based on batch alarm events, electronic equipment and medium |
| CN111159645A (en) * | 2019-12-19 | 2020-05-15 | 成都数之联科技有限公司 | Bad root cause positioning method based on product production record and parameters |
| CN111193605A (en) * | 2019-08-28 | 2020-05-22 | 腾讯科技（深圳）有限公司 | Fault positioning method and device and storage medium |
| CN111935761A (en) * | 2020-07-15 | 2020-11-13 | 山东建筑大学 | Method and system for automatically positioning quality difference root cause of mobile communication network |
| CN111932394A (en) * | 2020-08-10 | 2020-11-13 | 成都数之联科技有限公司 | Bad root cause path analysis method and system based on association rule mining |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150339573A1 (en) * | 2013-09-30 | 2015-11-26 | Manyworlds, Inc. | Self-Referential Semantic-based Method, System, and Device |
-
2020
- 2020-11-25 CN CN202011336989.5A patent/CN112269818B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA3156908A1 (en) * | 2015-01-06 | 2016-07-14 | David Burton | Mobile wearable monitoring systems |
| CN107577460A (en) * | 2017-08-29 | 2018-01-12 | 苏州优圣美智能系统有限公司 | A kind of method from unstructured data extraction structural data |
| CN111193605A (en) * | 2019-08-28 | 2020-05-22 | 腾讯科技（深圳）有限公司 | Fault positioning method and device and storage medium |
| CN111082966A (en) * | 2019-11-01 | 2020-04-28 | 平安科技（深圳）有限公司 | Positioning method and device based on batch alarm events, electronic equipment and medium |
| CN110888755A (en) * | 2019-11-15 | 2020-03-17 | 亚信科技（中国）有限公司 | Method and device for searching abnormal root node of micro-service system |
| CN111159645A (en) * | 2019-12-19 | 2020-05-15 | 成都数之联科技有限公司 | Bad root cause positioning method based on product production record and parameters |
| CN111935761A (en) * | 2020-07-15 | 2020-11-13 | 山东建筑大学 | Method and system for automatically positioning quality difference root cause of mobile communication network |
| CN111932394A (en) * | 2020-08-10 | 2020-11-13 | 成都数之联科技有限公司 | Bad root cause path analysis method and system based on association rule mining |
Non-Patent Citations (3)
| Title |
|---|
| semi-autonomous labeling of unstructured maintenance log data for diagnostic root cause analysis;Michael E等;APMs 2017 international conference advances in production management systems;1-10 * |
| 基于大规模网络日志的模板提取研究;崔元;张琢;;计算机科学;第44卷(第S2期);448-452 * |
| 基于根因分析的电力监控系统日志大数据处理方法;卢巍等;浙江电力;第38卷(第12期);70-75 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112269818A (en) | 2021-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112269818B (en) | Equipment parameter root cause positioning method, system, device and medium | |
| US20040122859A1 (en) | System to identify a wafer manufacturing problem and method therefor | |
| CN111444247A (en) | KPI (Key performance indicator) -based root cause positioning method and device and storage medium | |
| CN108960269B (en) | Feature acquisition method and device for data set and computing equipment | |
| CN111932394B (en) | Bad root cause path analysis method and system based on association rule mining | |
| CN111860576A (en) | Endometrium tumor classification labeling method based on random forest | |
| CN106055613A (en) | Cleaning method for data classification and training databases based on mixed norm | |
| CN112417202B (en) | Content screening method and device | |
| CN111860642A (en) | Unbalanced sample classification method and device | |
| CN114756420A (en) | Fault prediction method and related device | |
| CN114385465A (en) | Fault prediction method, equipment and storage medium | |
| CN111488520B (en) | Crop planting type recommendation information processing device, method and storage medium | |
| CN112182140B (en) | Information input method, device, computer equipment and medium combining RPA and AI | |
| CN114116528A (en) | Memory access address prediction method and device, storage medium and electronic equipment | |
| CN113705625A (en) | Method and device for identifying abnormal life guarantee application families and electronic equipment | |
| CN114077663A (en) | Application log analysis method and device | |
| CN1783092A (en) | Data analysis device and data analysis method | |
| CN113836784B (en) | Apple identification system and method based on information fusion technology | |
| CN116821087A (en) | Power transmission line fault database construction method, device, terminal and storage medium | |
| US11386340B2 (en) | Method and apparatus for performing block retrieval on block to be processed of urine sediment image | |
| CN108062563A (en) | A kind of representative sample based on classification equilibrium finds method | |
| CN113268494B (en) | Method and device for processing database statement to be optimized | |
| CN113569106B (en) | CAN data identification method, device and equipment | |
| CN116228484B (en) | Course combination method and device based on quantum clustering algorithm | |
| CN112084896B (en) | Combined recursion weighted spatial filtering method based on Lp/q-mixed norm |
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 | ||
| CB02 | Change of applicant information |
Address after: 610042 No. 270, floor 2, No. 8, Jinxiu street, Wuhou District, Chengdu, Sichuan Applicant after: Chengdu shuzhilian Technology Co.,Ltd. Address before: No.2, floor 4, building 1, Jule road crossing, Section 1, West 1st ring road, Wuhou District, Chengdu City, Sichuan Province 610041 Applicant before: CHENGDU SHUZHILIAN TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20210126 Assignee: Chengdu Haijixian Intelligent Technology Co.,Ltd. Assignor: Chengdu shuzhilian Technology Co.,Ltd. Contract record no.: X2024510000011 Denomination of invention: A method, system, device, and medium for root cause localization of equipment parameters Granted publication date: 20231121 License type: Common License Record date: 20240717 |
|
| EE01 | Entry into force of recordation of patent licensing contract |