CN110163511A - A kind of Manufacture quality control method based on association rule mining and fuzzy decision - Google Patents
A kind of Manufacture quality control method based on association rule mining and fuzzy decision Download PDFInfo
- Publication number
- CN110163511A CN110163511A CN201910440621.4A CN201910440621A CN110163511A CN 110163511 A CN110163511 A CN 110163511A CN 201910440621 A CN201910440621 A CN 201910440621A CN 110163511 A CN110163511 A CN 110163511A
- Authority
- CN
- China
- Prior art keywords
- triple
- quality control
- control strategy
- mass property
- affairs
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003908 quality control method Methods 0.000 title claims abstract description 76
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 50
- 238000005065 mining Methods 0.000 title claims abstract description 33
- 238000000034 method Methods 0.000 claims abstract description 120
- 238000012545 processing Methods 0.000 claims abstract description 22
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 15
- 238000013507 mapping Methods 0.000 claims abstract description 15
- 238000004458 analytical method Methods 0.000 claims abstract description 11
- 238000012163 sequencing technique Methods 0.000 claims abstract description 7
- 238000011156 evaluation Methods 0.000 claims description 37
- 239000011159 matrix material Substances 0.000 claims description 14
- 238000011217 control strategy Methods 0.000 claims description 8
- 238000013138 pruning Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 5
- 238000011438 discrete method Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 2
- 238000006467 substitution reaction Methods 0.000 claims 1
- 238000007514 turning Methods 0.000 description 7
- 238000009412 basement excavation Methods 0.000 description 5
- 238000003754 machining Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000005553 drilling Methods 0.000 description 3
- 239000002131 composite material Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000009472 formulation Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000011263 association rule mining technique Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000011058 failure modes and effects analysis Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 230000004083 survival 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
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/02—Computing arrangements based on specific mathematical models using fuzzy logic
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- 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)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Software Systems (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Fuzzy Systems (AREA)
- Operations Research (AREA)
- General Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Linguistics (AREA)
- Mathematical Optimization (AREA)
- Automation & Control Theory (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Computational Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Mathematical Analysis (AREA)
- Artificial Intelligence (AREA)
- Primary Health Care (AREA)
- Manufacturing & Machinery (AREA)
- Algebra (AREA)
- Molecular Biology (AREA)
Abstract
The Manufacture quality control method based on association rule mining and fuzzy decision that the invention discloses a kind of, obtains the qualitative data set of same model part manufacturing process；The mapping of mass property grade is carried out to each mass property in every qualitative data, corresponding quality triple is established with each mass property, so that corresponding to every qualitative data is respectively formed corresponding triple affairs；Triple affairs set D is formed by whole triple affairs again；Mining analysis is carried out to triple transaction set D using Apriori algorithm, to obtain Strong association rule；It is required to make corresponding Quality Control Strategy with processing quality according to Strong association rule；The priority that Quality Control Strategy is judged using fuzzy Decision Making Method determines the sequencing of quality control strategy implement according to priority.The present invention solves the technical issues of being lack of pertinence in the prior art to quality control, can carry out priority decisions to Quality Control Strategy.
Description
Technical field
The invention belongs to be machined manufacturing technology field more particularly to a kind of method for controlling mechanical processing quality.
Background technique
21 century is the century of quality, and quality is to influence one of Business survival and the key element of development, to manufacturing process
In quality controling research be a current popular domain.Both at home and abroad about the research of Manufacture quality control achieve it is many at
Fruit, but its overall monitor for primarily focusing on manufacturing process and prediction lack and targetedly carry out quality control.
Mass property Study on influencing factors content in manufacture course of products is 5M1E, i.e., industrial circle often say people,
Machine, material, method, survey, ring, each factor can resolve into various factors perhaps again.To realize targetedly to manufacturing process
Quality control is carried out, the quality characteristics data for making full use of manufacturing process to generate, the invention proposes be based on association rule mining
With the Manufacture quality control method of fuzzy decision: process quality association of characteristics model is established, from the quality information of manufacturing process
Excavate potential related information between different processes, obtain the key factor for causing quality problems, by abatement crucial effect because
Element promotes product manufacturing quality；Since the Quality Control Strategy of formulation may have multiple, also need actual conditions is combined to carry out preferential
Grade decision, i.e., determine evaluation indice according to process characteristics, establish fuzzy decision model, based on Grey Fuzzy Decision Making method to each
The priority of Quality Control Strategy carries out Analysis of Policy Making.
Summary of the invention
In view of the above shortcomings of the prior art, the present invention provides a kind of manufacture based on association rule mining and fuzzy decision
Method of quality control solves the technical issues of being lack of pertinence in the prior art to quality control, can be to Quality Control Strategy
Carry out priority decisions.
In order to solve the above technical problems, technical scheme is as follows: one kind is based on association rule mining and obscures certainly
The Manufacture quality control method of plan, comprising the following steps:
Step 1: obtaining the qualitative data set of same model part manufacturing process, every quality in qualitative data set
Data include the measured value of mass property corresponding to process required for manufacturing the model part, and every procedure institute is right
The type for the mass property parameter answered is one or more；
Xth qualitative data is expressed asFromIn can know corresponding to the i-th procedure
Jth kind mass property mass property measured value；
Step 2: the mapping of mass property grade being carried out to each mass property in every qualitative data, for each quality
Characteristic establishes corresponding quality triple, so that corresponding to every qualitative data is respectively formed corresponding triple affairs；Again by complete
Portion's triple affairs form triple affairs set D；
Triple affairs corresponding to xth qualitative data are Be byCarry out matter
Flow characteristic grade maps to obtain,Wherein, Pi (x)Indicate the i-th procedure in xth qualitative data；Indicate P in xth qualitative datai (x)Corresponding jth kind mass property,It indicates in xth qualitative data's
Mass property grade；Triple affairs set D={ t1,t2,t3...,tx}；
Step 3: mining analysis is carried out to triple transaction set D using Apriori algorithm, comprising the following steps:
Step 3.1: several frequent item set L are excavated from triple affairs set D according to minimum support minsup；
Step 3.2: each frequent item set L generates all nonvoid subsets, by nonvoid subset X and corresponding supplementary set
(L-X) generation process correlation rule, the process correlation rule refer to that the processing quality between process influences relationship, process association
Rule Expression isX is known as influence factor；(L-X) it is known as being affected factor；
Step 3.3: to meet process execution logic and meet the process correlation rule of min confidence minconf as by force
Correlation rule；
Step 4: requiring to make n Quality Control Strategy according to Strong association rule and processing quality；If n=1 is used
The Quality Control Strategy carries out Manufacture quality control；If n > 1, enters step 5；
Step 5: judging the priority of Quality Control Strategy using fuzzy Decision Making Method, quality control is determined according to priority
The sequencing of strategy implement.
Further, grade classification is carried out to each mass property respectively using wide discrete method, divides width d=dmax/ m,
Wherein, m indicates the total number of grade, dmaxIndicate maximum deviation distance:
dmax=max | amax-anom|,|amin-anom|}；
Wherein, amaxIndicate the permitted maximum value of a certain mass property, aminIndicate the permitted minimum value of mass property,
anomIndicate the nominal value of mass property；
Mass property gradeBy mapping table value:
It obtainsMass property measured value a, | a-xnom| table mass property measured value a and mass property nominal value anom
Distance.
Further, project I1From the nonvoid subset of set I, set I is by triple affairs set D all three
It is obtained after tuple elements de-redundancy；Item Sets I1Support support (I on triple affairs set D1) as follows
It calculates:
In formula, | I1| it indicates to include Item Sets I in triple affairs set D1Triple number of transactions, | D | indicate triple
The triple number of transactions summation of affairs set D；Project of the support greater than minimum support minsup is as frequent episode, frequent episode
Collection be collectively referred to as frequent item set；
All nonvoid subsets of frequent item set L are frequent episode, process correlation ruleConfidence level by such as
Lower formula calculates:
In formula, | X ∪ (L-X) | it indicates in triple affairs set D simultaneously comprising the triple affairs of project X and (L-X)
Number, | X | indicate the triple number of transactions in triple affairs set D comprising project X.
Further, using following steps Mining Frequent Itemsets Based:
Step 3.1.1: the length k=1 of frequent item set is initialized；
Step 3.1.2: candidate's k- item collection C is generatedk, candidate k- item collection CkIt is made of k triple；If k=1, candidate 1- is enabled
Item collection C1For triple affairs set D；If k > 1, by frequent k-1- item collection Lk-1Merged to generate with two stages of beta pruning and be waited
Select k- item collection Ck；
Step 3.1.3: according to minimum support minsup from candidate k- item collection CkIn find out frequent k- item collection Lk；
Step 3.1.4: judge frequent k- item collection LkIt whether is empty set；If so, terminating program；If it is not, entering step
3.1.5；
Step 3.1.5: enabling k=k+1, and returns to step 3.1.2.
Further, fuzzy Decision Making Method the following steps are included:
Step 5.1: establishing evaluation indice U={ U1,U2,U3,U4, wherein U1Indicate equipment economy, U2Indicate artificial
Economy, U3Expression task emergency, U4Presentation technology feasibility；
Step 5.2: fuzzy semantics evaluation being carried out to the evaluation index of each Quality Control Strategy, and with corresponding Factor of Brittleness
Fuzzy evaluation semantic values are substituted, decision matrix is obtained:
In formula, X1,X2,X3…XnRespectively 1 arrives n Quality Control Strategy, the element x of the i-th row jth column in decision matrixij
Indicate in i-th of Quality Control Strategy that the fuzzy semantics of j-th of evaluation index evaluate corresponding Factor of Brittleness, also, 1≤i≤
n,1≤j≤4；
Step 5.3: construction deviation matrix Δ:
The element δ that the i-th row jth arranges in deviation matrix ΔijIndicate deviation relative value:In formula,By such as
Under type value:
As evaluation index UjWhen for positive index, It indicates in decision matrix
The maximum value of j-th of evaluation index；
As evaluation index UjWhen for negative sense index, It indicates in decision matrix
The minimum value of j-th of evaluation index；
Step 5.4: being calculated using weight of the analytic hierarchy process (AHP) to each evaluation index, obtain the weight of each evaluation index
Coefficient constitutes weight coefficient vector W=(w1,w2,w3,w4)；
Step 5.5: the grey for calculating each Quality Control Strategy and desirable quality control strategy about each evaluation index is closed
Contact number, wherein i-th of Quality Control Strategy and grey correlation system of the desirable quality control strategy about j-th of evaluation index
Number γijIt is calculated as follows:
In formula, δijIndicate deviation relative value,ζ indicates resolution ratio；
Step 5.6: calculating the grey relational grade between each Quality Control Strategy and Ideal Control Strategy, wherein i-th
Grey relational grade R (x between Quality Control Strategy and desirable quality control strategyi,x0) it is calculated as follows:
Step 5.7: determining the priority of each Quality Control Strategy according to grey relational grade, held according to priority sequencing
Row Quality Control Strategy carries out Manufacture quality control.
Compared with prior art, the invention has the following advantages:
1, the qualitative data that the present invention makes full use of manufacturing process to generate is input with manufacturing process qualitative data, uses base
Qualitative data is handled in the wide discrete method of Euclidean distance, and forms quality triple element together with process.It will
Data conversion in qualitative data record sheet at triple transaction data set (TDS), based on Apriori algorithm to triple transaction set into
Row mining analysis finds strong incidence relation therein.
2, the present invention has the characteristics that time order and function is suitable for the execution of process in mechanical manufacturing field, calculates Apriori
Method is improved, and during recurrence Frequent Item Sets subset, is deleted the correlation rule for not meeting process execution sequence, is subtracted
The mining analysis of few meaningless rule, promotes operation efficiency.
3, manufacturing process Quality Control Strategy is targetedly proposed according to the Result of Strong association rule, is based on grey
The priority of fuzzy decision theory control strategy.
Detailed description of the invention
Fig. 1 is the overview flow chart of the Manufacture quality control method based on association rule mining and fuzzy decision；
Fig. 2 is mass property grade mapping principle figure；
Fig. 3 is the flow chart in the frequent item set mining stage of Apriori algorithm；
Fig. 4 is the flow chart of Strong association rule excavation phase；
Fig. 5 is schematic diagram of the present embodiment for the frequent item set mining process of Gearmaking Technology.
Specific embodiment
Refering to what is shown in Fig. 1, a kind of Manufacture quality control method based on association rule mining and fuzzy decision includes following
Step:
Step 1: obtaining the qualitative data set of same model part manufacturing process, every quality in qualitative data set
Data include the measured value of mass property corresponding to process required for manufacturing the model part, and every procedure institute is right
The type for the mass property parameter answered is one or more；
Xth qualitative data is expressed asFromIn can know corresponding to the i-th procedure
Jth kind mass property mass property measured value；
Step 2: the mapping of mass property grade being carried out to each mass property in every qualitative data, for each quality
Characteristic establishes corresponding quality triple, so that corresponding to every qualitative data is respectively formed corresponding triple affairs；Again by complete
Portion's triple affairs form triple affairs set D；
Triple affairs corresponding to xth qualitative data are Be byCarry out matter
Flow characteristic grade maps to obtain,Wherein, Pi (x)Indicate the i-th procedure in xth qualitative data；Indicate P in xth qualitative datai (x)Corresponding jth kind mass property,It indicates in xth qualitative data's
Mass property grade；Triple affairs set D={ t1,t2,t3...,tx}；
Step 3: mining analysis is carried out to triple transaction set D using Apriori algorithm, comprising the following steps:
Step 3.1: several frequent item set L are excavated from triple affairs set D according to minimum support minsup；
Step 3.2: each frequent item set L generates all nonvoid subsets, by nonvoid subset X and corresponding supplementary set
(L-X) generation process correlation rule, the process correlation rule refer to that the processing quality between process influences relationship, process association
Rule Expression isX is known as influence factor；(L-X) it is known as being affected factor；
Step 3.3: to meet process execution logic and meet the process correlation rule of min confidence minconf as by force
Correlation rule；
Step 4: requiring to make n Quality Control Strategy according to Strong association rule and processing quality；If n=1 is used
The Quality Control Strategy carries out Manufacture quality control；If n > 1, enters step 5；
Step 5: judging the priority of Quality Control Strategy using fuzzy Decision Making Method, quality control is determined according to priority
The sequencing of strategy implement.
To disclose specifically answering for the Manufacture quality control method proposed in this paper based on association rule mining and fuzzy decision
1000 in the part one month are analyzed by taking the back gear of certain passenger car transmission gear manufacturing enterprise as an example with process
Qualitative data.The manufacturing process of the part mainly include vehicle centre bore, turning, gear hobbing chamfered edge, vehicle burr, drilling, shaving,
The processes such as cleaning, product inspection calculate for convenience of excavating, and carry out according to process execution sequence to process and corresponding mass property
Coding, as shown in table 1.
1 gear manufacture process of table and mass property coding schedule
Grade classification is carried out to each mass property respectively using wide discrete method, divides width d=dmax/ m, wherein m table
Show the total number of grade, dmaxIndicate maximum deviation distance: dmax=max | amax-anom|,|amin-anom|}；
Wherein, amaxIndicate the permitted maximum value of a certain mass property, aminIndicate the permitted minimum value of mass property,
anomIndicate the nominal value of mass property；Take m=5.By taking the hole size of vehicle centre bore process as an example, processing request φ 121 ±
0.04, the part measured value of the mass property is as shown in table 2.Maximum value a in samplemax=121.05, minimum value amin=
120.96, dmax=0.05, divide width d=0.01.It is as shown in table 3 that rate range is then respectively mapped in sample.According in table 3
Credit rating mapping range carries out credit rating mapping to the measured value of the mass property in table 2, and mapping result is as shown in table 4.
2 vehicle centre bore process hole size part measured value of table
3 vehicle centre bore process hole size credit rating mapping range of table
4 vehicle centre bore process hole size credit rating mapping result (part) of table
Qualitative data set is as shown in table 5, and qualitative data compound mapping is triple affairs set D as shown in table 6, reflects
It is as shown in Figure 2 to penetrate principle.
The qualitative data set of 5 manufacturing process of table
6 triple affairs set D of table
Triple affairs corresponding to xth qualitative data are Be byCarry out matter
Flow characteristic grade maps to obtain,Wherein, Pi (x)Indicate the i-th procedure in xth qualitative data；Indicate P in xth qualitative datai (x)Corresponding jth kind mass property,It indicates in xth qualitative data's
Mass property grade；Triple affairs set D={ t1,t2,t3...,tx}。
Get triple affairs set D={ t1,t2,t3...,txAfter, using Apriori algorithm to triple thing
Business collection D carries out mining analysis, and Apriori algorithm belongs to unsupervised learning method.Association rule mining analysis includes two stages:
1. finding frequent item set according to minimum support minsup；2. pass is found in frequent item set according to min confidence minconf
Connection rule.
For Apriori algorithm sheet as existing algorithm, the frequent item set mining stage is identical as existing algorithm, the life of frequent item set
At flow chart refering to what is shown in Fig. 3, the excavation principle of frequent item set is as follows:
If Ci(i=1,2 ..., k) indicates that length is the triple candidate of i, referred to as candidate's i- item collection；Li(i=1,
2 ..., k) indicate that length is the triple frequent item set of i, referred to as frequent i- item collection.Transaction data set (TDS) D is scanned, according to setting
Minimum support minsup finds out frequent 1- item collection L1；By L1Candidate's 2- item collection C are generated with two stages of beta pruning by merging2, root
Frequent 2- item collection L is found out according to minimum support minsup2；By L2Continue to generate candidate 3- item collection C3, by C3Find out frequent 3- item collection
L3；Successively iteration, until frequent k- item collection is empty.Merge and beta pruning is the core procedure for generating Frequent Item Sets, herein
With Lk-1Generate LkFor introduce merge and beta pruning process, k be greater than 1.
Merge: to frequent (k-1)-item collection Lk-1In project using lexcographical order arrange, i.e., for Lk-1In item collection { w1,
w2,…,wk-2,wk-1, meet w1< w2< ... < wk-2< wk-1.By to Lk-1In the only different item of the last one project
Mesh collection, which merges, generates candidate's k- item collection CkIf f1And f2It is Lk-1In Item Sets, meet f1={ i1,i2,…,ik-2,
ik-1, f2={ i1,i2,…,ik-2,i’k-1, ik-1< i 'k-1, then f1And f2It can merge, the result items after merging are f=
{i1,i2,…,ik-2,ik-1,i’k-1}。
Beta pruning: downward closure refers to that there are a frequent item sets, then any nonvoid subset of this item collection is all frequent
Item collection, i.e. there are the item collections of non-frequent subset to be unlikely to be frequent item set in turn.It can be deleted using downward closure
It is unlikely to be the project of frequent item set, saves and calculates the time.For (k-1)-subset of f, judge it whether all in Lk-1If
There is subset not in Lk-1In, then illustrate that f is unlikely to be frequent item set, by it from candidate k- item collection CkMiddle deletion.
But in mechanical manufacturing field, the processing of product is often multi-process, and the execution of process has time order and function suitable
Sequence.Apriori algorithm is concerned with the relevance between rule, executes sequence between process without considering, therefore, this
Invention proposes the process association mining method based on Apriori algorithm for this problem, in recurrence Frequent Item Sets subset
In the process, the correlation rule for not meeting process execution sequence is deleted, the mining analysis of meaningless rule is reduced, promotes operation effect
Rate.
In the frequent item set L of generation, correlation rule is generated by nonvoid subset X and corresponding supplementary set (L-X)The correlation rule for meeting min confidence minconf is the Strong association rule of our searchings, it is contemplated that mechanical
The sequencing that process executes in manufacturing process, the process executed afterwards, which will not generate association to the process first carried out, to be influenced.It is passing
When the Strong association rule for returning discovery frequent item to concentrate, correlation rule is first determined whetherWhether meet process execution to patrol
Volume, if discovery does not meet the excavation terminated when process executes logic to the frequent item, promote excavation speed.Correlation rule generates
Algorithm flow it is as shown in Figure 4.
Support and confidence level are described as follows:
Project I1From the nonvoid subset of set I, set I is by triple element all in triple affairs set D
It is obtained after de-redundancy；Item Sets I1Support support (I on triple affairs set D1) it is calculated as follows:
In formula, | I1| it indicates to include Item Sets I in triple affairs set D1Triple number of transactions, | D | indicate triple
The triple number of transactions summation of affairs set D；Project of the support greater than minimum support minsup is as frequent episode, frequent episode
Collection be collectively referred to as frequent item set；
All nonvoid subsets of frequent item set L are frequent episode, process correlation ruleConfidence level by such as
Lower formula calculates:
In formula, | X ∪ (L-X) | it indicates in triple affairs set D simultaneously comprising the triple affairs of project X and (L-X)
Number, | X | indicate the triple number of transactions in triple affairs set D comprising project X.
For the Gear Processing process in present embodiment, the mining process of frequent item set is as shown in Figure 5.Due to matter
Data set is measured from sampling observation (having inspected the gear of 1000 same models by random samples), a qualitative data is caused often to contain only portion
The measured value of the mass property of operation break-down, the measured value of the mass property not comprising whole processes, therefore triple transaction set D
In be not every triple affairs all include whole processes, but since the item number of total qualitative data is more, from quality
Data acquisition system can cover the measured value of the mass property of every procedure on the whole.
15%, the i.e. minsup=150 that minimum support minsup is total amount of data is arranged in present embodiment, minimum
Confidence level minconf is 0.85.Correlation rule is generated according to Frequent Item Sets, finds the pass for meeting min confidence minconf
Connection rule, the results are shown in Table 7 for the Strong association rule for meeting min confidence of excavation.
7 association rule mining result table of table
From the Result of table 7 it can be concluded that
1. ruleIndicate the gear hobbing chamfered edge when the outer circle jitter parameter of cylindrical turning process is second-rate
The tooth runout quality of process is also poor, illustrates that cylindrical turning process is affected to the processing of gear hobbing chamfering process.The rule
Support is 0.156, confidence level 0.91；
Regular 2. (2,1, I), (3,2, II),Indicate the excircle dimension when cylindrical turning process, gear hobbing
When the precision of the radial composite error of the tooth runout and shaving process of chamfering process is higher, the total cumulative pitch error of finished work-piece
It is smaller.The regular grid DEM is 0.163, confidence level 0.86；
3. ruleWhen indicating that the diameter of bore precision of drilling operating is higher, the inner hole of finished gear is straight
Diameter precision is higher, which is 0.171, confidence level 0.88.
Correlation rule embodies the influence relationship of machining accuracy between process, in order to guarantee the processing matter for being affected process
Amount, it is necessary to which the machining accuracy for influencing process is controlled.The characteristics of present invention is according to mass property grade mapping model, to shadow
The mass property grade for ringing process need to be controlled at III grade or more.According to available three quality of Strong association rule Result
Control strategy: R1: guarantee that the outer circle bounce credit rating of cylindrical turning process is at least III grade；R2: guarantee the outer circle of cylindrical turning process
Size, the tooth runout of gear hobbing chamfering process, shaving process through being at least III grade to the credit rating of composite tolerance；R3: it protects
The diameter of bore credit rating of card drilling operating is at least III grade.
By the manufacturing process quality characteristics data analysis based on association rule mining, different process quality characteristics are searched out
Between Strong association ruleQuality control is formulated for influence factor FactorX according to actual processing
System strategy, to guarantee the quality for being affected factor F actorY.Since the Quality Control Strategy of formulation may have multiple, need to tie
It closes actual conditions and carries out priority decisions.Evaluation indice is determined according to process characteristics, establishes fuzzy decision model, is based on grey
Fuzzy Decision Making Method carries out Analysis of Policy Making to the priority of each Quality Control Strategy.
In mechanical processing process, machining accuracy is to measure the finger of the error degree between processing dimension and theoretical size
Mark, precision is higher, and error is smaller, but processing cost is bigger.Comprehensively consider the factor of equipment, personnel, task, technology etc.,
For Quality Control Strategy, equipment economy (U is proposed1), artificial economy (U2), task emergency (U3), technical feasibility
(U4) 4 evaluation indexes, equipment economy refers to the equipment cost needed for Quality Control Strategy is implemented；What artificial economy referred to
It is cost of labor needed for strategy implement；Task emergency refers to the urgency level of the task of strategy implement；Technical feasibility
Refer to the technical difficulty of strategy implement.This 4 indexs constitute evaluation indice, are indicated with U, evaluation indice U={ U1,
U2,U3,U4}。
Opinion rating collection refers to the fuzzy evaluation set to each evaluation index, is called fuzzy semantics collection again herein.
In conjunction with the actual conditions in mechanical processing process, opinion rating is divided using Pyatyi partitioning, i.e., according to importance journey
Opinion rating is divided into " very big (G) ", " big (H) ", " medium (M) ", " small (L) ", " very little (R) " by degree, indicates that grading is fuzzy with V
Semanteme collection, V={ G, H, M, L, R }.
In order to make Analysis of Policy Making quantification, fuzzy semantics are concentrated using article method in the prior art opinion rating
De-fuzzy processing is carried out, the Factor of Brittleness of each fuzzy semantics is calculated, such as article " Pillay A, wang J.Modified
failure mode and effects analysis using approximate reasoning[J].Reliability
Engineering and System Safety, 2003,79 (1): the described method in 69-85 ".
It is evaluated by domain expert, comprehensively considers each process processing characteristic, processed for present embodiment middle gear
Technique obtains Quality Control Strategy evaluation result, as shown in table 8.
8 Quality Control Strategy decision table of table
Decision matrix is obtained according to the Factor of Brittleness of each evaluation index:
U in 4 evaluation indexes of present embodiment1、U2、U4For negative sense index, U3For positive index.Therefore, it obtains
Standard value vector x0=(0.370,0.370,0.804,0.196).According to deviation relative value calculation formula:It calculates
Each deviation relative value is obtained, deviation matrix is established:
The relative importance of each evaluation index in actual production is considered, using analytic hierarchy process (AHP) Calculation Estimation index
Weight coefficient vector is W=(0.335,0.129,0.421,0.115).
Grey incidence coefficient is calculated according to grey incidence coefficient calculation formula, resolution ratio ζ in present embodiment=
0.5, with γ21Calculating for:
The value of other grey incidence coefficients can be similarly obtained, as shown in table 9.
9 grey incidence coefficient table of table
Grey relational grade is calculated, with R (x1,x0) calculating for:
R(x1,x0)=0.335 × 1+0.129 × 1+0.421 × 0.580+0.115 × 1=0.823
R (x can similarly be acquired2,x0)=0.625, R (x3,x0)=0.650.Due to R (x1,x0) value it is maximum, therefore, it is considered that
Quality Control Strategy R1Highest priority preferentially guarantees the outer circle jitter values precision of cylindrical turning process.
The present invention analyzes the processing characteristic in machining field, from manufacturing process qualitative data, proposes
Method of quality control based on association rule mining and fuzzy decision.Correlation rule based on process quality characteristics data is dug
Pick establishes mapping mould with process actual processing sequence to wide sliding-model control of the quality characteristics data based on Euclidean distance
Type proposes a kind of process association mining method based on Apriori algorithm；It targetedly proposes to manufacture according to Result
Process quality control strategy, the priority based on Grey-fuzzy Theory Decision Control strategy.With certain model gear manufacturing processes matter
Measure data instance, disclose the application process of method, for machine-building process quality control provide a method, have compared with
Good application prospect.
Claims (5)
1. a kind of Manufacture quality control method based on association rule mining and fuzzy decision, which is characterized in that including following step
It is rapid:
Step 1: obtaining the qualitative data set of same model part manufacturing process, every qualitative data in qualitative data set
Comprising the measured value for manufacturing mass property corresponding to process required for the model part, and corresponding to every procedure
The type of mass property parameter is one or more；
Xth qualitative data is expressed asFromIn can know jth corresponding to the i-th procedure
The mass property measured value of kind mass property；
Step 2: the mapping of mass property grade being carried out to each mass property in every qualitative data, for each mass property
Corresponding quality triple is established, so that corresponding to every qualitative data is respectively formed corresponding triple affairs；Again by all three
Tuple affairs form triple affairs set D；
Triple affairs corresponding to xth qualitative data are Be byIt is special to carry out quality
Property grade maps to obtain,Wherein, Pi (x)Indicate the i-th procedure in xth qualitative data；Table
Show P in xth qualitative datai (x)Corresponding jth kind mass property,It indicates in xth qualitative dataQuality it is special
Property grade；Triple affairs set D={ t1,t2,t3...,tx}；
Step 3: mining analysis is carried out to triple transaction set D using Apriori algorithm, comprising the following steps:
Step 3.1: several frequent item set L are excavated from triple affairs set D according to minimum support minsup；
Step 3.2: each frequent item set L generates all nonvoid subsets, by nonvoid subset X and corresponding supplementary set (L-X)
Generation process correlation rule, the process correlation rule refer to that the processing quality between process influences relationship, process correlation rule
It is expressed asX is known as influence factor；(L-X) it is known as being affected factor；
Step 3.3: being associated with using meeting process execution logic and meeting the process correlation rule of min confidence minconf as strong
Rule；
Step 4: requiring to make n Quality Control Strategy according to Strong association rule and processing quality；If n=1, the matter is used
It measures control strategy and carries out Manufacture quality control；If n > 1, enters step 5；
Step 5: judging the priority of Quality Control Strategy using fuzzy Decision Making Method, Quality Control Strategy is determined according to priority
The sequencing of implementation.
2. the Manufacture quality control method according to claim 1 based on association rule mining and fuzzy decision, feature
It is, grade classification is carried out to each mass property respectively using wide discrete method, divides width d=dmax/ m, wherein m expression etc.
The total number of grade, dmaxIndicate maximum deviation distance:
dmax=max | amax-anom|,|amin-anom|}；
Wherein, amaxIndicate the permitted maximum value of a certain mass property, aminIndicate the permitted minimum value of mass property, anom
Indicate the nominal value of mass property；
Mass property gradeBy mapping table value:
It obtainsMass property measured value a, | a-xnom| table mass property measured value a and mass property nominal value anomAway from
From.
3. the Manufacture quality control method according to claim 1 based on association rule mining and fuzzy decision, feature
It is, project I1From the nonvoid subset of set I, set I is by triple element de-redundant all in triple affairs set D
It is obtained after remaining；Item Sets I1Support support (I on triple affairs set D1) it is calculated as follows:
In formula, | I1| it indicates to include Item Sets I in triple affairs set D1Triple number of transactions, | D | indicate triple affairs
The triple number of transactions summation of set D；Project of the support greater than minimum support minsup is as frequent episode, the collection of frequent episode
It is collectively referred to as frequent item set；
All nonvoid subsets of frequent item set L are frequent episode, process correlation ruleConfidence level by following public
Formula calculates:
In formula, | X ∪ (L-X) | the triple number of transactions of project X and (L-X) are indicated in triple affairs set D while including, | X
| indicate the triple number of transactions in triple affairs set D comprising project X.
4. the Manufacture quality control method according to claim 1 based on association rule mining and fuzzy decision, feature
It is, using following steps Mining Frequent Itemsets Based:
Step 3.1.1: the length k=1 of frequent item set is initialized；
Step 3.1.2: candidate's k- item collection C is generatedk, candidate k- item collection CkIt is made of k triple；If k=1, candidate's 1- item collection is enabled
C1For triple affairs set D；If k > 1, by frequent k-1- item collection Lk-1Merged and generates candidate k- with two stages of beta pruning
Item collection Ck；
Step 3.1.3: according to minimum support minsup from candidate k- item collection CkIn find out frequent k- item collection Lk；
Step 3.1.4: judge frequent k- item collection LkIt whether is empty set；If so, terminating program；If it is not, entering step 3.1.5；
Step 3.1.5: enabling k=k+1, and returns to step 3.1.2.
5. the Manufacture quality control method according to claim 1 based on association rule mining and fuzzy decision, feature
Be, fuzzy Decision Making Method the following steps are included:
Step 5.1: establishing evaluation indice U={ U1,U2,U3,U4, wherein U1Indicate equipment economy, U2Indicate artificial economical
Property, U3Expression task emergency, U4Presentation technology feasibility；
Step 5.2: fuzzy semantics evaluation being carried out to the evaluation index of each Quality Control Strategy, and with corresponding Factor of Brittleness substitution
Fuzzy evaluation semantic values, obtain decision matrix:
In formula, X1,X2,X3...XnRespectively 1 arrives n Quality Control Strategy, the element x of the i-th row jth column in decision matrixijTable
Show in i-th of Quality Control Strategy that the fuzzy semantics of j-th of evaluation index evaluate corresponding Factor of Brittleness, also, 1≤i≤n, 1
≤j≤4；
Step 5.3: construction deviation matrix Δ:
The element δ that the i-th row jth arranges in deviation matrix ΔijIndicate deviation relative value:In formula,By such as lower section
Formula value:
As evaluation index UjWhen for positive index, It indicates in decision matrix j-th
The maximum value of evaluation index；
As evaluation index UjWhen for negative sense index, It indicates in decision matrix j-th
The minimum value of evaluation index；
Step 5.4: being calculated using weight of the analytic hierarchy process (AHP) to each evaluation index, obtain the weight system of each evaluation index
Number constitutes weight coefficient vector W=(w1,w2,w3,w4)；
Step 5.5: calculating each Quality Control Strategy and grey correlation system of the desirable quality control strategy about each evaluation index
Number, wherein i-th of Quality Control Strategy and grey incidence coefficient γ of the desirable quality control strategy about j-th of evaluation indexij
It is calculated as follows:
In formula, δijIndicate deviation relative value,ζ indicates resolution ratio；
Step 5.6: calculating the grey relational grade between each Quality Control Strategy and Ideal Control Strategy, wherein i-th of mass
Grey relational grade R (x between control strategy and desirable quality control strategyi,x0) it is calculated as follows:
Step 5.7: determining the priority of each Quality Control Strategy according to grey relational grade, execute matter according to priority sequencing
It measures control strategy and carries out Manufacture quality control.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910440621.4A CN110163511A (en) | 2019-05-24 | 2019-05-24 | A kind of Manufacture quality control method based on association rule mining and fuzzy decision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910440621.4A CN110163511A (en) | 2019-05-24 | 2019-05-24 | A kind of Manufacture quality control method based on association rule mining and fuzzy decision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110163511A true CN110163511A (en) | 2019-08-23 |
Family
ID=67632792
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910440621.4A Pending CN110163511A (en) | 2019-05-24 | 2019-05-24 | A kind of Manufacture quality control method based on association rule mining and fuzzy decision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110163511A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111125082A (en) * | 2019-12-26 | 2020-05-08 | 北京工业大学 | Cement process parameter data analysis method based on association rule mining |
| CN111177862A (en) * | 2019-12-13 | 2020-05-19 | 燕山大学 | Incidence rule-based flatness curve coefficient setting method |
| CN111275332A (en) * | 2020-01-20 | 2020-06-12 | 山东大学 | System and method for evaluating quality of bulldozer blade assembly process based on fuzzy-association rule |
| CN113420375A (en) * | 2021-06-14 | 2021-09-21 | 西北工业大学 | Method for establishing process-quality-unbalance amount association relation model based on Apriori |
| CN117332992A (en) * | 2023-11-24 | 2024-01-02 | 北京国联视讯信息技术股份有限公司 | Collaborative manufacturing method and system for industrial Internet |
| JP7440438B2 (en) | 2021-02-03 | 2024-02-28 | 株式会社日立製作所 | Tolerance analysis system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103268406A (en) * | 2013-05-09 | 2013-08-28 | 湖南大学 | Data mining system and method based on coal mine safety training games |
| CN103341506A (en) * | 2013-07-10 | 2013-10-09 | 鞍钢股份有限公司 | Strip-shaped time series data mining method based on data patterns |
| CN104820716A (en) * | 2015-05-21 | 2015-08-05 | 中国人民解放军海军工程大学 | Equipment reliability evaluation method based on data mining |
| CN106126577A (en) * | 2016-06-17 | 2016-11-16 | 北京理工大学 | A kind of weighted association rules method for digging based on data source Matrix dividing |
| CN106598030A (en) * | 2016-12-22 | 2017-04-26 | 西安理工大学 | Axle temperature correlation analysis method based on data |
-
2019
- 2019-05-24 CN CN201910440621.4A patent/CN110163511A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103268406A (en) * | 2013-05-09 | 2013-08-28 | 湖南大学 | Data mining system and method based on coal mine safety training games |
| CN103341506A (en) * | 2013-07-10 | 2013-10-09 | 鞍钢股份有限公司 | Strip-shaped time series data mining method based on data patterns |
| CN104820716A (en) * | 2015-05-21 | 2015-08-05 | 中国人民解放军海军工程大学 | Equipment reliability evaluation method based on data mining |
| CN106126577A (en) * | 2016-06-17 | 2016-11-16 | 北京理工大学 | A kind of weighted association rules method for digging based on data source Matrix dividing |
| CN106598030A (en) * | 2016-12-22 | 2017-04-26 | 西安理工大学 | Axle temperature correlation analysis method based on data |
Non-Patent Citations (2)
| Title |
|---|
| 马寿福等: "基于关联规则挖掘和模糊决策的质量控制方法", 《中国科技论文在线（时间页）》 * |
| 马寿福等: "基于关联规则挖掘和模糊决策的质量控制方法", 《中国科技论文在线》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111177862A (en) * | 2019-12-13 | 2020-05-19 | 燕山大学 | Incidence rule-based flatness curve coefficient setting method |
| CN111177862B (en) * | 2019-12-13 | 2021-11-16 | 燕山大学 | Incidence rule-based flatness curve coefficient setting method |
| CN111125082A (en) * | 2019-12-26 | 2020-05-08 | 北京工业大学 | Cement process parameter data analysis method based on association rule mining |
| CN111125082B (en) * | 2019-12-26 | 2023-09-22 | 北京工业大学 | Cement process parameter data analysis method based on association rule mining |
| CN111275332A (en) * | 2020-01-20 | 2020-06-12 | 山东大学 | System and method for evaluating quality of bulldozer blade assembly process based on fuzzy-association rule |
| CN111275332B (en) * | 2020-01-20 | 2022-03-18 | 山东大学 | System and method for evaluating quality of bulldozer blade assembly process based on fuzzy-association rule |
| JP7440438B2 (en) | 2021-02-03 | 2024-02-28 | 株式会社日立製作所 | Tolerance analysis system |
| CN113420375A (en) * | 2021-06-14 | 2021-09-21 | 西北工业大学 | Method for establishing process-quality-unbalance amount association relation model based on Apriori |
| CN113420375B (en) * | 2021-06-14 | 2022-05-31 | 西北工业大学 | Method for establishing process-quality-unbalance amount association relation model based on Apriori |
| CN117332992A (en) * | 2023-11-24 | 2024-01-02 | 北京国联视讯信息技术股份有限公司 | Collaborative manufacturing method and system for industrial Internet |
| CN117332992B (en) * | 2023-11-24 | 2024-02-09 | 北京国联视讯信息技术股份有限公司 | Collaborative manufacturing method and system for industrial Internet |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110163511A (en) | A kind of Manufacture quality control method based on association rule mining and fuzzy decision | |
| US7562067B2 (en) | Systems and methods for estimating functional relationships in a database | |
| WO2019001220A1 (en) | Decision tree based turning process energy consumption prediction system and method of numerically controlled lathe | |
| CN111597347A (en) | Knowledge embedded defect report reconstruction method and device | |
| CN104573039A (en) | Keyword search method of relational database | |
| CN104462184A (en) | Large-scale data abnormity recognition method based on bidirectional sampling combination | |
| WO2021128523A1 (en) | Technology readiness level determination method and system based on science and technology big data | |
| KR102465451B1 (en) | Process knowledge push method based on machining characteristics | |
| CN108090303A (en) | A kind of three-dimensional process for machining information reuse method towards marine diesel important Parts | |
| CN111143624B (en) | Land approval surveying and mapping data-oriented adaptive calculation rule base matching method and system | |
| CN116307566A (en) | Dynamic design system for large-scale building construction project construction organization scheme | |
| CN106777237B (en) | A kind of analysis method of surface defect | |
| Nguyen et al. | Grammatical evolution to mine OWL disjointness axioms involving complex concept expressions | |
| Herrera-Viedma et al. | Some remarks on the fuzzy linguistic model based on discrete fuzzy numbers | |
| Saini et al. | Spider bots: Database enhancing and indexing scripts to efficiently convert raw well data into valuable knowledge | |
| CN114386573A (en) | Static sounding soil layer dividing method and system based on multilayer perceptron | |
| Knapp et al. | Material master data quality | |
| CN112948368A (en) | Deep anomaly detection method based on data correlation | |
| CN112395475B (en) | Service provider clustering method based on rapid search and density peak clustering | |
| Zhou et al. | Data mining method based on rough set and fuzzy neural network | |
| CN105138544B (en) | A kind of searching method for remolding logical deduction chain | |
| TWI673723B (en) | Intelligent pre-diagnosis and health management system and method | |
| Pasek | Exploration of rough sets theory use for manufacturing process monitoring | |
| Vellaisamy et al. | Sequential and systematic sampling plans for the Markov‐dependent production process | |
| Veiga | A conceptual framework on biodiversity data quality. |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190823 |
|
| RJ01 | Rejection of invention patent application after publication |