CN108664752A - A kind of process parameter optimizing method based on process rule and big data analysis technology - Google Patents
A kind of process parameter optimizing method based on process rule and big data analysis technology Download PDFInfo
- Publication number
- CN108664752A CN108664752A CN201810499633.XA CN201810499633A CN108664752A CN 108664752 A CN108664752 A CN 108664752A CN 201810499633 A CN201810499633 A CN 201810499633A CN 108664752 A CN108664752 A CN 108664752A
- Authority
- CN
- China
- Prior art keywords
- rule
- decision
- attribute
- decision table
- index
- 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
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000005516 engineering process Methods 0.000 title claims abstract description 11
- 238000007405 data analysis Methods 0.000 title claims abstract description 10
- 229910000831 Steel Inorganic materials 0.000 claims abstract description 21
- 239000010959 steel Substances 0.000 claims abstract description 21
- 238000002360 preparation method Methods 0.000 claims abstract description 15
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000004519 manufacturing process Methods 0.000 claims description 21
- 230000015572 biosynthetic process Effects 0.000 claims description 4
- 229910052799 carbon Inorganic materials 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 2
- 238000005457 optimization Methods 0.000 abstract description 7
- 230000008901 benefit Effects 0.000 abstract description 3
- 238000005097 cold rolling Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000000137 annealing Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005266 casting Methods 0.000 description 1
- 238000010924 continuous production Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000005098 hot rolling Methods 0.000 description 1
- 210000001161 mammalian embryo Anatomy 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000004886 process control Methods 0.000 description 1
- 238000000275 quality assurance Methods 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Husbandry (AREA)
- Mining & Mineral Resources (AREA)
- Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Agronomy & Crop Science (AREA)
- General Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Marine Sciences & Fisheries (AREA)
- Primary Health Care (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Factory Administration (AREA)
Abstract
The invention discloses a kind of whole process product quality optimization method based on process rule and big data analysis technology, method and step include：It builds flat steel product preparation process index storehouse and builds decision table, discretization of decision table processing, attribute reduction and technological compensa tion rule generate, rule query and feedback compensation, Policy Updates.The present invention has many advantages, such as that principle is simple, error rate is low, scalability is strong, strong robustness.
Description
Technical field
The invention belongs to whole process intelligence preparing technical field, and in particular to be based on process rule and big data analysis
Process parameter optimizing method.
Background technology
Flat steel product preparation process is that flat steel product preparation process is an extremely complex industrial process, is related to converter, connects
The large complicated industrial flow of the multi-process such as casting, hot rolling, cooling, cold rolling, annealing, more control levels, product have " primary steel
The title of material " is the important support of the national economic development.It is manually set by rule of thumb currently, China's flat steel product prepares still to be in
The stage being combined is controlled with single ingelligent optimizing procedure, the coordination for not forming full-range overall-in-one control schema and each level is excellent
Change, is faced with that product design size poor with internal performance control stability, high-end supply scarce capacity, labor productivity are low etc. to ask
It inscribes, the personalized customization urgently comprehensive propulsion under the conditions of enlargement, centralization, continuous production.Skill is prepared by intelligence
Art realizes multi-process, system-level, global product quality and preparation flow optimization, is the grand strategy side that flat steel product prepares development
To being to solve flat steel product preparation process complexity and inevitable choice the problems such as quality stability.How effective optimization is used
Control method prepares whole process to flat steel product and applies optimal control, for improving production efficiency, by resources advantage and technical force
Combine, the industry benefit for playing bigger has great importance.
Invention content
For the above problem present in existing flat steel product technology of preparing, now provide in a kind of collection flat steel product preparation flow
Production and technic index data build index storehouse, technic index and synthesis are established based on process rule and big data analysis technology
Decision relation table between production target, to optimize tune to each technological parameter of regular whole process for formulating concerned process steps
It is whole, build a closed-loop control system based on feedback compensation, the final method for realizing the optimization of whole process product quality.
Work is corresponded in product quality examination criteria and each process after the completion of the production and technic index respectively manufacture
The index for the parameter that skill changes.
The process rule and big data analysis method are according to passing production experience to including converter parameter, refining ginseng
The relationship between quality of technological parameter and generation product including number, plate embryo parameter, Cold-rolling Parameters, Hot Rolling Parameters etc. is modeled, structure
It builds technic index and generates the correspondence of index
The closed-loop control refers to completing to technological parameter and product quality relationship modeling, that is, after building decision table, subsequently
The index value of each generation technique is pushed back in production by the desired value to product quality, and these values are fed back into each process
Control section, to carry out the optimization of each process, and whole system is without external artificial control.
The present invention is as follows to realize above process：
Build flat steel product preparation process index storehouse.By big data analysis the relevant technologies, by flat steel product preparation process history
The comprehensive production index of each product object of data set and the technic index to be compensated are collected arrangement, build flat steel product
Prepare achievement data library.
Build decision table.According to this knowledge representation form of the information system of rough set theory, to collected index storehouse
The attribute of the creation data of middle collection is integrated, classify formation condition attribute set C and decision kind set D, wherein each
The codomain V of product attribute aa, attribute is all：V={ V1, V2,V3... ..., U is all product object set, and m is object and category
Mapping function of the property to attribute codomain：U×A→V.It is combined and is indicated according to above-mentioned knowledge, construct decision information system S=<U, C
∪ D, m>, i.e. decision table.
Discretization of decision table processing.Do not have for successive value or given price for the codomain of certain conditional attributes or decision attribute
The centrifugal pump of value, break point set is arranged within the scope of attribute codomain by by the sliding-model control based on comentropy in we, divide from
Section is dissipated, to obtain the centrifugal pump of more abstraction hierarchy.
Attribute reduction and technological compensa tion rule generate.A case in decision table just represents a basic decision
Rule, all such decision rules can be formed by the set of a decision rule.But basis decision rules therein do not have
Adaptability, the case where only mechanically having recorded each sample.The bigger rule of fitness are obtained in order to be extracted from decision table
Then, we remove decision table progress yojan under the premise of not losing original information from the conditional attribute of Decision Table Systems
To obtaining the unessential conditional attribute of decision, the decision of decision attribute is advised to analyze the conditional attribute in gained yojan
Then so that there is higher adaptability and representativeness by the decision rule in yojan treated decision table.
Uncertain process rule processing.The process decision rule with regularity and adaptability is obtained by step 3,4
Collection, however comprising with the same terms attribute but the inconsistent uncertainty rule of conclusion in current rule set, used here as setting
Reliability is identified current rule, for subsequent feedback compensation optimizing provides reference.
Rule query and feedback compensation.It will be used as conditional attribute value after the error amount discretization of the comprehensive production index at scene
Input, inquires matched rule, according to the process index value section of one or several rule inquired, provides anti-
Present offset.
Policy Updates.The new rule not included in meeting constantly regular library in industry manufactures production process occurs, this meeting
The rule set being previously achieved is caused to update therewith, we correct former using the more new algorithm of the management rule based on Apriori algorithm
There is decision table.
Beneficial effects of the present invention：
The present invention is that it is intelligent to be suitable for flat steel product whole process towards the technological parameter compensation optimizing method in preparation process
The quality optimization of preparation, Method And Principle is simple, high degree of automation, strong robustness, for flat steel product whole process intelligence production
Provide quality assurance.There is rule set extraction itself scalability and compatibility only to be updated when production environment changes
Rule set extraction and without again collect data handled, you can be continuing with the optimization system.
Description of the drawings
Fig. 1 is built and is run signal based on the process parameter optimizing method system of process rule and big data analysis technology
Figure；
Fig. 2 is the flow chart of the discretization method based on comentropy.
Specific implementation mode
The following describes the present invention in detail with reference to the accompanying drawings and specific embodiments.
As shown in Figure 1, flat steel product process rule collection structure specifically includes following steps：
1. building flat steel product preparation process achievement data library.According to the preparation process data set of historical accumulation, structure synthesis
Production target D (cc billet surface quality, Inner Quality of Billet) and feedback adjustment technic index C (degree of superheat, the negative slip time,
Level fluctuation, carbon content, Mn/s ratios etc.) between relation database table.Using i-th group of data as setting value, kth group data conduct
New desired value, the desired value difference △ Y=Y of the twok-YiIt regards the variation of desired output as, thus obtains comprehensive production
Index it is expected and technic index set point change amount △ C=Ck-CiBetween relationship maps：△Y→△C；
2. decision table is built., using technic index Ci and its variable quantity △ C as conditional attribute, comprehensive production index becomes for we
Change amount △ Y are as decision attribute, using achievement data library as data source, build the decision table based on process rule compensation optimizing；
3. discretization of decision table is handled.Before carrying out attribute reduction, need to the attribute that codomain is successive value be subjected to discretization
Processing, we will use the discretization method as shown in Figure 2 based on comentropy here：To arbitrary continuation value attribute, one is sought
Decision table is divided into two parts by a cut-point, the cut-point, calculates the entropy average value of the subset in two sections thus marked off,
Compare all candidate discrete points in median sequence, obtains the break point set for making entropy minimum.Two divided for above-mentioned discrete point
The secondary division of sub- attribute section recursive call, until meeting discretization determination requirement.Above-mentioned sliding-model control method is successively
To all conditions attribute Ci, △ C and decision attribute △ Y processing, the decision table after discretization is obtained.
4. attribute reduction and technological compensa tion rule generate.Since the characteristics of objects of conditional attribute description can not change, from can
The attribute most like with the classification of decision attribute is found in the conditional attribute of yojan to meet Decision Classfication most possibly, is passed through
A case in yojan treated decision table just represents a kind of case with identical rule characteristic, the decision obtained in this way
Rule just has higher generalization, applicability stronger.Then we carry out Value reduction processing, delete every number successively
According to each conditional attribute after, judge that decision conflict has occurred in new decision table, if changing the Indiscernible relation of decision table,
Then illustrate that this attribute value of the data cannot be deleted, if not changing, then indicates that the attribute value should be from this data
It deletes.The technological compensa tion rule after yojan is just generated after the completion of all yojan of the attribute value of all conditions；
5. uncertain process rule processing.For conditional attribute value having the same but there is different decision attributes
Value, i.e. rule are concentrated with a plurality of regular and inconsistent conclusion data, using the confidence level of computation rule to such uncertainty
Rule is handled.Here certain rule confidence level is to meet the decision attribute values D of this ruleiAnd meet precondition CiProduction
Product number/only meet precondition CiProduct number.
It is completed by the adaptable and representative process rule collection structure of step 1~5, one, when real-time flat steel product
It prepares production target data Y and it is expected that index error △ Y are incoming, Relational database will be passed through by possessing the closed-loop system of the rule set
Technology quick search transfers section to corresponding offset rule and technological parameter, and according to the codomain and discretization breakpoint of corresponding attribute
Reverse is that original continuous value is exported to corresponding process.If occurring the rule for not including in rule set in preparation process, we use
Association Rules Algorithm Updating based on Apriori algorithm：Such as IUA, FUP, dynamic update achievement data library and process rule collection.
Claims (1)
1. a kind of process parameter optimizing method based on process rule and big data analysis technology, which is characterized in that on realizing
The process present invention is stated to be as follows：
1.) flat steel product preparation process index storehouse is built；By big data analysis the relevant technologies, by flat steel product preparation process history number
It is collected arrangement according to the comprehensive production index and the technic index to be compensated of each product object of collection, builds flat steel product system
Standby achievement data library；
2.) decision table is built；According to this knowledge representation form of the information system of rough set theory, in collected index storehouse
The attribute of the creation data of collection is integrated, and classify formation condition attribute set C and decision kind set D, and wherein each is produced
The codomain V of product attribute aa, attribute is all：V={ V1, V2,V3... ..., U is all product object set, and m is object and attribute
To the mapping function of attribute codomain：U×A→V；It is combined and is indicated according to above-mentioned knowledge, construct decision information system S=<U, C ∪
D, m>, i.e. decision table；
3.) discretization of decision table is handled；Do not have for successive value or given price for the codomain of certain conditional attributes or decision attribute
The centrifugal pump of value, break point set is arranged within the scope of attribute codomain by by the sliding-model control based on comentropy in we, divide from
Section is dissipated, to obtain the centrifugal pump of more abstraction hierarchy；
4.) attribute reduction and technological compensa tion rule generate；A case in decision table just represents a basic decision
Rule, all such decision rules can be formed by the set of a decision rule；But basis decision rules therein do not have
Adaptability, the case where only mechanically having recorded each sample；The bigger rule of fitness are obtained in order to be extracted from decision table
Then, we remove decision table progress yojan under the premise of not losing original information from the conditional attribute of Decision Table Systems
To obtaining the unessential conditional attribute of decision, the decision of decision attribute is advised to analyze the conditional attribute in gained yojan
Then so that there is higher adaptability and representativeness by the decision rule in yojan treated decision table；
5.) uncertain process rule processing；The process decision rule with regularity and adaptability is obtained by step 3,4
Collection, however comprising with the same terms attribute but the inconsistent uncertainty rule of conclusion in current rule set, used here as setting
Reliability is identified current rule, for subsequent feedback compensation optimizing provides reference；
6.) rule query and feedback compensation；It will be used as conditional attribute value after the error amount discretization of the comprehensive production index at scene
Input, inquires matched rule, according to the process index value section of one or several rule inquired, provides anti-
Present offset；
7.) Policy Updates；The new rule not included in meeting constantly regular library in industry manufactures production process occurs, this can lead
The rule set being previously achieved is caused to update therewith, we correct original using the more new algorithm of the management rule based on Apriori algorithm
Decision table.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810499633.XA CN108664752A (en) | 2018-05-23 | 2018-05-23 | A kind of process parameter optimizing method based on process rule and big data analysis technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810499633.XA CN108664752A (en) | 2018-05-23 | 2018-05-23 | A kind of process parameter optimizing method based on process rule and big data analysis technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108664752A true CN108664752A (en) | 2018-10-16 |
Family
ID=63777503
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810499633.XA Pending CN108664752A (en) | 2018-05-23 | 2018-05-23 | A kind of process parameter optimizing method based on process rule and big data analysis technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108664752A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109919496A (en) * | 2019-03-12 | 2019-06-21 | 西北工业大学 | A kind of aircraft rigger skill design rule knowledge management method and system |
| CN110609523A (en) * | 2019-07-18 | 2019-12-24 | 浙江工业大学 | Cooperative control method for units in primary tea leaf making process |
| CN112580935A (en) * | 2020-08-20 | 2021-03-30 | 同济大学 | Industrial product production process traceability analysis method based on machine vision |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1652050A (en) * | 2005-02-24 | 2005-08-10 | 上海交通大学 | Plasticity forming technique regulation obtaining method based on simulation and fuzzy coarse central algorithm |
| WO2011023596A1 (en) * | 2009-08-28 | 2011-03-03 | Siemens Aktiengesellschaft | Method and system for monitoring wind turbine generator vibration |
| CN102879677A (en) * | 2012-09-24 | 2013-01-16 | 西北工业大学 | Intelligent fault diagnosis method based on rough Bayesian network classifier |
| CN107169628A (en) * | 2017-04-14 | 2017-09-15 | 华中科技大学 | A kind of distribution network reliability evaluation method based on big data mutual information attribute reduction |
-
2018
- 2018-05-23 CN CN201810499633.XA patent/CN108664752A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1652050A (en) * | 2005-02-24 | 2005-08-10 | 上海交通大学 | Plasticity forming technique regulation obtaining method based on simulation and fuzzy coarse central algorithm |
| WO2011023596A1 (en) * | 2009-08-28 | 2011-03-03 | Siemens Aktiengesellschaft | Method and system for monitoring wind turbine generator vibration |
| CN102879677A (en) * | 2012-09-24 | 2013-01-16 | 西北工业大学 | Intelligent fault diagnosis method based on rough Bayesian network classifier |
| CN107169628A (en) * | 2017-04-14 | 2017-09-15 | 华中科技大学 | A kind of distribution network reliability evaluation method based on big data mutual information attribute reduction |
Non-Patent Citations (1)
| Title |
|---|
| 杨秉宇: "基于数据的湿法冶金全流程优化控制补偿方法的研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109919496A (en) * | 2019-03-12 | 2019-06-21 | 西北工业大学 | A kind of aircraft rigger skill design rule knowledge management method and system |
| CN109919496B (en) * | 2019-03-12 | 2023-11-14 | 西北工业大学 | Method and system for managing design rule knowledge of aircraft assembly process |
| CN110609523A (en) * | 2019-07-18 | 2019-12-24 | 浙江工业大学 | Cooperative control method for units in primary tea leaf making process |
| CN112580935A (en) * | 2020-08-20 | 2021-03-30 | 同济大学 | Industrial product production process traceability analysis method based on machine vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106599519B (en) | The cut deal motherboard and slab collaborative design method and system of facing to manufacture order combination optimization | |
| CN108664752A (en) | A kind of process parameter optimizing method based on process rule and big data analysis technology | |
| JP2020500420A (en) | Yield prediction system and method for machine learning based semiconductor manufacturing | |
| CN110990784B (en) | Cigarette ventilation rate prediction method based on gradient lifting regression tree | |
| CN107818409A (en) | The optimization method of manufacture of cement parameter and regulation section based on fuzzy association rules | |
| CN1109310C (en) | Time-series signal predicting apparatus | |
| CN109787821B (en) | Intelligent prediction method for large-scale mobile client traffic consumption | |
| CN109271427A (en) | A kind of clustering method based on neighbour's density and manifold distance | |
| CN104077432A (en) | Process-adjustment choosing analysis method based on multidimensional correlation function | |
| Matuszny | Building decision trees based on production knowledge as support in decision-making process | |
| CN116611743A (en) | Building engineering construction quality evaluation method based on big data | |
| CN114117884B (en) | Intelligent ore blending system for sintering based on pre-iron big data | |
| CN111814979B (en) | Fuzzy set automatic dividing method based on dynamic programming | |
| CN103632014B (en) | A kind of technological parameter Forecasting Methodology of hot steel rolling production production procedure | |
| CN111125082B (en) | Cement process parameter data analysis method based on association rule mining | |
| CN115587579A (en) | Digital processing system, production system and production method | |
| CN111858530B (en) | Real-time correlation analysis method and system based on mass logs | |
| CN106095962A (en) | Weighted Fuzzy rule extraction | |
| CN113656372A (en) | Standard index database data mart architecture model and method | |
| CN102708241A (en) | Rapid particle swarm method for reservoir history matching | |
| CN111933226A (en) | Artificial intelligence material screening method based on classification model | |
| CN118092362B (en) | Method, device and equipment for analyzing abnormal reasons in sintering process | |
| CN111126726A (en) | Intelligent decision multi-target analysis method based on heterogeneous fusion data | |
| CN115292388B (en) | Automatic scheme mining system based on historical data | |
| CN102314535B (en) | Manufacturing method of flange forging based on tire mould database |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181016 |