CN114091800A - Intelligent design evaluation method for silicon steel product production scheme - Google Patents
Intelligent design evaluation method for silicon steel product production scheme Download PDFInfo
- Publication number
- CN114091800A CN114091800A CN202110778749.9A CN202110778749A CN114091800A CN 114091800 A CN114091800 A CN 114091800A CN 202110778749 A CN202110778749 A CN 202110778749A CN 114091800 A CN114091800 A CN 114091800A
- Authority
- CN
- China
- Prior art keywords
- data
- performance
- production
- product
- score
- 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
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 341
- 229910000976 Electrical steel Inorganic materials 0.000 title claims abstract description 115
- 238000013461 design Methods 0.000 title claims abstract description 74
- 238000011156 evaluation Methods 0.000 title claims abstract description 48
- 238000012827 research and development Methods 0.000 claims abstract description 109
- 229910000831 Steel Inorganic materials 0.000 claims description 141
- 239000010959 steel Substances 0.000 claims description 141
- 238000000034 method Methods 0.000 claims description 104
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 claims description 93
- 230000008569 process Effects 0.000 claims description 84
- 238000005265 energy consumption Methods 0.000 claims description 73
- 230000006698 induction Effects 0.000 claims description 69
- 239000000126 substance Substances 0.000 claims description 66
- 229910052742 iron Inorganic materials 0.000 claims description 46
- 238000004364 calculation method Methods 0.000 claims description 35
- 238000011031 large-scale manufacturing process Methods 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 15
- 238000002474 experimental method Methods 0.000 claims description 13
- 239000000203 mixture Substances 0.000 claims description 13
- 239000000463 material Substances 0.000 claims description 10
- 238000007689 inspection Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 7
- 238000010079 rubber tapping Methods 0.000 claims description 7
- 230000007547 defect Effects 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 6
- 238000011160 research Methods 0.000 claims description 6
- 238000002790 cross-validation Methods 0.000 claims description 4
- 230000008676 import Effects 0.000 claims description 4
- 230000008520 organization Effects 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 3
- 238000005261 decarburization Methods 0.000 description 17
- 238000000137 annealing Methods 0.000 description 15
- 239000000523 sample Substances 0.000 description 12
- 238000004422 calculation algorithm Methods 0.000 description 10
- 230000002159 abnormal effect Effects 0.000 description 9
- 238000005097 cold rolling Methods 0.000 description 7
- 238000012216 screening Methods 0.000 description 7
- 238000009628 steelmaking Methods 0.000 description 7
- 229910052799 carbon Inorganic materials 0.000 description 6
- 239000011248 coating agent Substances 0.000 description 6
- 238000000576 coating method Methods 0.000 description 6
- 229910052710 silicon Inorganic materials 0.000 description 6
- 238000005098 hot rolling Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- CPLXHLVBOLITMK-UHFFFAOYSA-N magnesium oxide Inorganic materials [Mg]=O CPLXHLVBOLITMK-UHFFFAOYSA-N 0.000 description 3
- 239000000395 magnesium oxide Substances 0.000 description 3
- 238000005096 rolling process Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000005674 electromagnetic induction Effects 0.000 description 2
- AXZKOIWUVFPNLO-UHFFFAOYSA-N magnesium;oxygen(2-) Chemical compound [O-2].[Mg+2] AXZKOIWUVFPNLO-UHFFFAOYSA-N 0.000 description 2
- 229910052751 metal Inorganic materials 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 230000010287 polarization Effects 0.000 description 2
- 238000004886 process control Methods 0.000 description 2
- 238000001953 recrystallisation Methods 0.000 description 2
- 238000000638 solvent extraction Methods 0.000 description 2
- 229910052717 sulfur Inorganic materials 0.000 description 2
- 238000012356 Product development Methods 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 239000002253 acid Substances 0.000 description 1
- 229910052782 aluminium Inorganic materials 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000009749 continuous casting Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 239000003112 inhibitor Substances 0.000 description 1
- 239000003041 laboratory chemical Substances 0.000 description 1
- 239000006101 laboratory sample Substances 0.000 description 1
- 239000000696 magnetic material Substances 0.000 description 1
- 229910052748 manganese Inorganic materials 0.000 description 1
- 238000007431 microscopic evaluation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005121 nitriding Methods 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 230000006911 nucleation Effects 0.000 description 1
- 238000010899 nucleation Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 229910052698 phosphorus Inorganic materials 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000006104 solid solution Substances 0.000 description 1
- 238000007711 solidification Methods 0.000 description 1
- 230000008023 solidification Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000005406 washing Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—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
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0206—Price or cost determination based on market factors
-
- 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
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Tourism & Hospitality (AREA)
- Finance (AREA)
- Data Mining & Analysis (AREA)
- Accounting & Taxation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Educational Administration (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Manufacturing & Machinery (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- General Factory Administration (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an intelligent design evaluation method for a silicon steel product production scheme, which comprises the following steps: 1. collecting data, and integrating user data, laboratory research and development data and mass production data; 2. constructing a user data theme, a laboratory research and development data theme and a mass production data theme; 3. inputting the requirement information of the type, specification and performance requirements of the silicon steel products, and generating a plurality of production schemes of the silicon steel products according to the requirement information, wherein the production schemes comprise a production scheme for optimizing the products and a production scheme for new products; 4. and 3, comprehensively evaluating the production scheme of each silicon steel product generated in the step 3 in a multi-dimensional manner to obtain a green design index of the production scheme of the silicon steel product, and recommending the production scheme according to the green design index. According to the invention, based on the product design requirement, experimental research and development data and mass production data are fused, a product design scheme model is generated and evaluated, the research and development period is shortened, and the research and development efficiency is improved.
Description
Technical Field
The invention relates to a production scheme of a steel product, in particular to an intelligent design evaluation method of a production scheme of a silicon steel product.
Background
At present, the research and development of steel products at home and abroad mainly depend on expert experience or laboratory trial and error type simulation experiments, the conventional steel product production scheme is provided from user requirements to be finally converted into steel products meeting the requirements, and the flow involved in the process is complex and various, for example: the method comprises the following steps of requirement analysis, failure mode analysis, research scheme formulation, laboratory chemical composition and process route parameter experiments, product trial production scheme output, industrial mass production trial production, process re-optimization, user authentication and the like. The scheme design mode in the prior art has the advantages of long development period, high product development cost and strong dependence on expert subjective experience and theoretical knowledge; meanwhile, in the process of outputting the on-site trial production scheme of the laboratory product, the laboratory product scheme is greatly uncertain due to the difference of the process conditions of laboratory equipment and the industrial production line of a steel mill, certain components or process parameters may need to be repeatedly demonstrated in the laboratory, and the development period is prolonged. In addition, key indexes such as yield, energy consumption and cost of each production process cannot be accurately evaluated in the existing mode.
Electrical steel, silicon steel, is a steel product with excellent soft magnetic performance, is mainly used for manufacturing iron cores of various motors and transformers, and is an important metal soft magnetic material in the industries of electric power and motors. The manufacturing process is complex, particularly for oriented silicon steel products, the whole process flow comprises the working procedures of molten iron pretreatment, RH refining, continuous casting, hot rolling, normalizing annealing, acid washing, cold rolling, decarburization annealing, MgO coating, high-temperature annealing, flattening stretching annealing, insulating layer coating and the like, the higher grade of the product also comprises the working procedures of nitriding treatment, laser nicking and the like, steel making components and inclusion elements are strictly controlled, the whole process parameters and influencing factors are numerous, and the product relates to metal solidification, rolling deformation, recrystallization nucleation and growth control, inhibitor solid solution precipitation control, secondary recrystallization, grain preferred orientation, steel plate decarburization, nitridation denitrification control, surface oxide layer control and the like, and is a very complex process. All the above influence factors need to be considered when developing new silicon steel products, and design and demonstration are carried out in the experimental scheme, if only depending on expert experience or laboratory trial and error experiments, the development cycle is very long, and meanwhile, the accuracy, the rationality and the manufacturability of the product design scheme cannot be ensured.
Disclosure of Invention
The invention aims to provide an intelligent design evaluation method for a silicon steel product production scheme, which is based on product design requirements, integrates laboratory research and development data and mass production data, generates and evaluates a product design scheme model, shortens the research and development period and improves the research and development efficiency.
The invention is realized by the following steps:
an intelligent design evaluation method for a silicon steel product production scheme comprises the following steps:
step 1: collecting data, and integrating user data, laboratory research and development data and mass production data;
step 2: constructing a user data theme, a laboratory research and development data theme and a mass production data theme;
and step 3: inputting the requirement information of the type, specification and performance requirements of the silicon steel products, and generating a plurality of production schemes of the silicon steel products according to the requirement information, wherein the production schemes comprise a production scheme for optimizing the products and a production scheme for new products;
and 4, step 4: and 3, comprehensively evaluating the production scheme of each silicon steel product generated in the step 3 in a multi-dimensional manner to obtain a green design index of the production scheme of the silicon steel product, and recommending the production scheme according to the green design index.
The user data comprises performance requirement indexes of users on various silicon steel products, user IDs (identity) and inquiry IDs and basic information of the users;
the laboratory research and development data comprise chemical components, research and development process parameters, various performance indexes and organization analysis results in the whole experimental process of various silicon steel products;
the large production data are production data of all units in the whole large production process of various silicon steel products, and comprise chemical components, steel tapping mark production material tracking information, production process parameter actual results, all performance data, inspection and test data, surface defect data, quality judgment data, outlet/inlet coil weight, yield, inlet coil weight apportionment amount, steel energy consumption per ton and steel cost per ton; the production process parameter actual performance comprises low-frequency data collected according to steel coils and high-frequency data collected according to distance or time intervals, and each steel coil has one piece of low-frequency data and a plurality of pieces of high-frequency data.
The step 2 comprises the following steps:
step 2.1: constructing a user demand data theme;
step 2.2: constructing a laboratory research and development data theme of the silicon steel product;
step 2.3: and constructing a mass production data theme of the silicon steel product.
The step 2.1 comprises the following steps: based on the user ID and the inquiry ID in the user data, linking the basic information of a certain user with the performance requirement index of the user on the silicon steel product to form a user requirement data theme;
the step 2.2 comprises the following steps:
step 2.2.1: configuring laboratory research and development data exception rules of chemical components, research and development process parameters and performance indexes of the silicon steel product;
step 2.2.2: preprocessing the laboratory research and development data based on the laboratory research and development data exception rule;
step 2.2.3: defining the inlet sample number and the outlet sample number of each experimental procedure based on the whole experimental procedure, wherein the outlet sample number of the previous experimental procedure is the inlet sample number of the next experimental procedure and is logic, and the whole experimental procedure data of the silicon steel product is connected in series to form a research and development data theme of a laboratory;
the step 2.3 comprises the following steps:
step 2.3.1: configuring large production data exception rules of chemical components, production process parameter actual performance and performance data of the silicon steel product;
step 2.3.2: preprocessing the mass production data based on the mass production data exception rule;
step 2.3.4: calculating the characteristic values of the high-frequency data and the performance data of the steel coil of the silicon steel product;
step 2.3.5: based on the whole process of large-scale production, the export coil number and the import coil number of each unit are defined, the export coil number of the former unit is equal to the entry coil number of the latter unit as logic, and the whole process data of large-scale production of the silicon steel products are connected in series to form the subject of the large-scale production data.
The step 3 comprises the following steps:
step 3.1: filtering a laboratory research and development data theme and a mass production data theme according to the demand information, and selecting two types of related data which meet the demand information; wherein, the performance index in the subject of laboratory research and development data is filtered after being multiplied by an adjustment factor;
step 3.2: respectively carrying out normalization and standardization data processing on each type of relevant data based on the performance requirements in the requirement information;
step 3.3: for each type of related data, respectively calculating a first performance similarity between the performance requirement and the normalized and standardized performance data;
step 3.4: setting a performance similarity threshold, if the first performance similarity is larger than or equal to the performance similarity threshold, marking the silicon steel product corresponding to the first performance similarity as an optimized product, and executing the step 3.5, if the first performance similarity is smaller than the performance similarity threshold, marking the silicon steel product corresponding to the first performance similarity as a new product, and executing the step 3.6;
step 3.5: generating a production scheme of an optimized product: dividing the production process parameter actual results by combining the production process parameter actual results corresponding to the first performance similarity and the control precision of each unit in large-scale production;
step 3.6: and establishing a performance index prediction model based on the performance requirements, and verifying the optimal production scheme model of the new product.
The step 3.6 comprises the following steps:
step 3.6.1: selecting historical data from the mass production data, wherein the type and the specification of the silicon steel product in the historical data are the same as those of the input silicon steel product, and the time range of the mass production data is the latest year;
step 3.6.2: establishing a plurality of performance index prediction models; the performance index prediction model comprises a plurality of performance index prediction models, wherein the input of each performance index prediction model is key chemical components and key process parameters selected by research personnel, the output of each performance index prediction model is a performance index, and the optimal performance prediction model in the performance index prediction models is determined through cross validation;
step 3.6.3: calculating the average value of the chemical components and the actual performance of the production process parameters in the large production data selected in the step 3.6.1, and taking the average value of the chemical components and the actual performance of the production process parameters as a scheme reference point;
step 3.6.4: and carrying out gridding search on the performance index prediction model from the scheme reference point.
In the step 4, the evaluation dimensions comprise performance similarity, yield, ton steel energy consumption and ton steel cost.
For each production scheme of the optimized product, the multi-dimensional comprehensive evaluation method comprises the following steps:
step 4.11: calculating a second performance similarity between the performance data of the production scheme of the optimized product and the actual performance data of mass production, if the second performance similarity is completely met, namely the performance data of the production scheme of the optimized product is completely matched with the actual performance data of mass production, marking the first performance score as 100, turning to step 4.13, and if the first performance score is not completely met, executing step 4.12;
step 4.12: respectively calculating the similarity of the magnetic induction and the iron loss of the production scheme of the optimized product and the actual magnetic induction and the iron loss of the large-scale production, and performing performance evaluation based on the weight to obtain a third performance similarity S3, wherein the evaluation formula is as follows: λ is S3 ═ λ11*SMagnetic induction 1+λ21*SIron loss 1The first performance score is denoted as S3;
wherein S isMagnetic induction 1In order to optimize the similarity between the magnetic induction of the production scheme performance data of the product and the magnetic induction of the mass production actual performance data, lambda11Is SMagnetic induction 1The weight of (c); sIron loss 1In order to optimize the similarity between the magnetic induction of the production scheme performance data of the product and the magnetic induction of the mass production actual performance data, lambda21Is SIron loss 1The weight of (c); lambda [ alpha ]11+λ21＝1；
Step 4.13: evaluating and optimizing the first material forming rate of each unit of the production scheme model of the product based on the weight to obtain a first material forming rate index;
the calculation formula of the first yield of each unit is as follows:
first yield η1100 (exit weight/entrance roll contribution weight);
wherein, the outlet weight is ═ Σ (outlet coil weight of each steel coil), and the inlet coil apportioned weight is ═ Σ (inlet coil apportioned weight of each steel coil);
the calculation formula of the first yield index is as follows: first yield index ═ Σ (λ)3*η1) Wherein λ is3The weight of the first yield corresponding to each unit, and ∑ λ3The first yield score is recorded as a first yield index when the yield score is 1;
step 4.14: evaluating the energy consumption of a production scheme model of an optimized product based on the energy consumption of steel coil production units, wherein when the energy consumption is the lowest, a first energy consumption score is marked as 100, and when the energy consumption is the highest, the first energy consumption score is marked as 0;
step 4.15: evaluating the ton steel cost of a production scheme model for optimizing a product based on the ton steel cost in steel coil production, wherein when the ton steel cost is the lowest, the first cost score is recorded as 100, and when the ton steel cost is the highest, the first cost score is recorded as 0;
step 4.16: calculating the green design index of the optimized product based on the weight, wherein the calculation formula is as follows:
green design index lambda of optimized product41First performance score + λ42First yield score + λ43First energy consumption score + λ44First cost score;
wherein，λ41Weight for the first performance score, λ42Weight, λ, for the first yield score43Weight, λ, for the first energy consumption score44A weight scored for the first cost, and41+λ42+λ43+λ44＝1。
for each new product production scheme, the multi-dimensional comprehensive evaluation method comprises the following steps:
step 4.21: and calculating a fourth performance similarity between the performance data of the production scheme of the new product and the actual performance data of the mass production, if the fourth performance similarity is completely met, namely the performance data of the production scheme of the new product is completely matched with the actual performance data of the mass production, marking the second performance score as 100, turning to the step 4.23, and if the fourth performance similarity is not completely met, executing the step 4.22.
Step 4.22: respectively calculating the similarity of the magnetic induction and the iron loss of the production scheme of the new product and the actual magnetic induction and the iron loss of the large production, and performing performance evaluation based on the weight to obtain a fifth performance similarity S5, wherein the evaluation formula is as follows: λ is S5 ═ λ12*SMagnetic induction 2+λ22*SIron loss 2(ii) a A second performance score is denoted as S5;
wherein S isMagnetic induction 2Is the similarity between the magnetic induction of the performance data of the production scheme of a new product and the magnetic induction of the actual performance data of mass production, lambda12Is SMagnetic induction 2The weight of (c); sIron loss 2Is the similarity between the magnetic induction of the performance data of the production scheme of a new product and the magnetic induction of the actual performance data of mass production, lambda22Is SIron loss 2The weight of (c); lambda [ alpha ]12+λ22＝1；
Step 4.23: evaluating a second yield of each unit of the production scheme of the new product based on the weight to obtain a second yield index;
the calculation formula of the second yield of each unit is as follows:
second yield η2100 (exit weight/entrance roll contribution weight);
wherein, the outlet weight is ═ Σ (outlet coil weight of each steel coil), and the inlet coil apportioned weight is ═ Σ (inlet coil apportioned weight of each steel coil);
the calculation formula of the second yield index is as follows: second yield index ═ Σ (λ)5*η2) And sigma lambda5The second yield score is recorded as a second yield index when the yield score is 1;
step 4.24: evaluating the energy consumption of the production scheme of the new product based on the energy consumption of the steel coil production unit, wherein when the energy consumption is the lowest, the second energy consumption score is 100, and when the energy consumption is the highest, the second energy consumption score is 0;
step 4.25: evaluating the cost of steel per ton of a production scheme of an optimized product based on the cost of steel per ton produced by a steel coil, wherein when the cost of steel per ton is lowest, the second cost score is recorded as 100, and when the cost of steel per ton is highest, the second cost score is recorded as 0;
step 4.26: calculating the green design index of the new product based on the weight, wherein the calculation formula is as follows:
green design index of new product lambda61Second performance score + λ62Second yield score + λ63Second energy consumption score + λ64Second cost score;
wherein λ is61Weight for the second performance score, λ62Weight, λ, for the second yield score63Weight, λ, for the second energy consumption score64A weight scored for the second cost, and61+λ62+λ63+λ64＝1；
step 4.27: calculating the deviation coefficient P ═ P between the new product production scheme and the most similar steel grade1+P2；
Wherein, P1For a first coefficient of deviation between the chemical composition in the production scheme of the new product and the chemical composition of the most similar steel grade, the calculation formula is:
wherein λ isiIs the weight of each chemical component,
is the design value, x, of each chemical component in the production scheme of a new productikIs the actual performance value of each chemical component of the most similar steel grade, NiRecording the number of the most similar steel grades, wherein i is the number of chemical components;
P2for a second deviation degree coefficient between the process parameters in the new product production scheme and the process parameters of the most similar steel grade, the calculation formula is as follows:
wherein, betaiAre the weights of the various process parameters,
is the design value, g, of each process parameter in the production scheme of a new productikActual performance values of the individual process parameters for the most similar steel grades, NiRecording the number of the most similar steel grades, wherein i is the number of process parameters;
step 4.28: correcting the green design index of the new product through the deviation degree coefficient P to obtain the corrected green design index of the new product, wherein the correction calculation formula is as follows:
the corrected green design index of the new product-P.
Compared with the prior art, the invention has the following beneficial effects:
1. because the method integrates the full-process data of laboratory research and development and mass production, the method can carry out scheme design aiming at key chemical components and process parameters of the full-process of silicon steel product production, and carry out multi-dimensional comprehensive evaluation on performance indexes and the like, can realize the trace back and adjustment of data of any steel coil, any process parameters, any performance indexes and the like in the whole process, and is favorable for assisting in optimizing the research and development design of products and new products.
2. The invention is based on the performance requirements of users and research and development, combines the research and development of laboratory laboratories and the mass data accumulation of mass production, can more specifically design a scheme model of a component product, greatly reduces the trial and error times and the dependence on expert experience, shortens the research and development period, obviously improves the research and development efficiency, and has good adaptability and high efficiency for the research and development of products such as silicon steel with various and complicated production procedures.
According to the invention, based on the product design requirement, experiment research and development data and mass production data are fused through a mass data technology, a product design scheme model is generated and multi-dimensional comprehensive evaluation is carried out, an optimal scheme is recommended to assist in optimizing the on-site rapid trial production of products and new products, the trial and error times are reduced, the research and development period is shortened, the research and development efficiency can be obviously improved, and the method is particularly suitable for the research and development of products with complex production procedures such as silicon steel.
Drawings
FIG. 1 is a flow chart of the intelligent design evaluation method for silicon steel product production schemes of the present invention.
Detailed Description
The invention is further described with reference to the following figures and specific examples.
Referring to fig. 1, an intelligent design evaluation method for a silicon steel product production scheme includes the following steps:
step 1: data are collected and user data, laboratory research and development data and mass production data are integrated.
The user data comprises performance requirement indexes (such as iron loss, magnetic induction and other electromagnetic performance) of the user on various silicon steel products, user ID, inquiry ID, basic information of the user and the like.
The laboratory research and development data comprise chemical components, research and development process parameters, various performance indexes (such as iron loss, electromagnetic performance of magnetic induction and the like), organization analysis results and other data in the whole experimental process of various silicon steel products. The whole experimental process comprises the working procedures of laboratory steel making, laboratory hot rolling, laboratory normalizing, laboratory cold rolling, laboratory continuous annealing, laboratory decarburization annealing, laboratory magnesium oxide coating, laboratory high-temperature annealing, laboratory coating and the like. Laboratory research and development data can be obtained from a research and development management system of silicon steel products, and each item of data in the laboratory research and development data is accumulated for at least more than one year so as to ensure the construction accuracy of a production scheme. The results of the tissue analysis are: the metallographic structure of the steel plate corresponding to each process in the manufacturing process of the oriented silicon steel product, such as a normalizing plate produced in the normalizing process, a decarburization plate corresponding to the decarburization annealing process, a high-temperature annealing plate corresponding to the high-temperature annealing process and the like, is subjected to metallographic microscopic analysis to obtain data such as a structure picture, a grain size and the like.
The large production data is the production data of each unit in the whole large production process of various silicon steel products, and comprises chemical components (such as elements C, Si, S and the like), steel tapping marks, production material tracking information (such as the unit number, the coil inlet number, the coil outlet number and the like corresponding to the production process of the silicon steel coils), production process parameter actual results (such as steel tapping temperature, normalizing maximum temperature and the like, low-frequency data collected according to the coils and high-frequency data corresponding to the coils collected according to distance or time intervals, wherein each coil only has one low-frequency data, each coil has a plurality of high-frequency data), various performance data (such as electromagnetic energy such as iron loss, magnetic induction and the like), inspection and test data and surface defect data, quality judgment data, outlet/inlet coil weight, yield, inlet coil weight apportionment, ton steel energy consumption, ton steel cost and the like. The mass production data can be obtained from a process control system and a manufacturing execution system of steel production, and each item of data in the mass production data is accumulated for at least more than one year so as to ensure the accuracy of constructing a scheme model.
Step 2: and constructing a user data theme, a laboratory research and development data theme and a mass production data theme.
Step 2.1: constructing a user requirement data theme, wherein the specific operation is as follows: and based on the user ID and the inquiry ID in the user data, linking the basic information of a certain user with the performance requirement index of the user on the silicon steel product to form a user requirement data theme.
Step 2.2: the method comprises the following specific operations of constructing a laboratory research and development data theme of a silicon steel product:
step 2.2.1: and configuring laboratory research and development data exception rules of chemical components, research and development process parameters and performance indexes of the silicon steel product. The reasonable data ranges of chemical components, research and development process parameters and performance indexes can be respectively determined according to specific silicon steel products.
Step 2.2.2: preprocessing the laboratory research and development data based on the laboratory research and development data exception rule, judging that the laboratory research and development data are abnormal if chemical components, research and development process parameters and performance indexes exceed reasonable data ranges, prompting research and development personnel to have exception data, facilitating the research and development personnel to perform relevant processing on the exception data, and otherwise judging that the laboratory research and development data are normal.
Step 2.2.3: based on the whole experimental process, according to the steelmaking date and the heat number, the slab cogging position of the silicon steel product and the hot rolling and cold rolling coil splitting information, the inlet sample number and the outlet sample number of each experimental process are defined, the outlet sample number of the previous experimental process is logic with the inlet sample number of the next experimental process, and the whole experimental process data of the silicon steel product are connected in series to form the research and development data theme of a laboratory, so that the whole experimental process can be traced from steelmaking to cold rolling.
Step 2.3: the method comprises the following specific operations of constructing a mass production data theme of the silicon steel product:
step 2.3.1: and configuring large production data exception rules of chemical components, production process parameter actual performance and performance data of the silicon steel product. The reasonable data ranges of the chemical components, the actual performance of the production process parameters and the performance data can be respectively determined according to specific silicon steel products.
Step 2.3.2: and preprocessing the mass production data based on a mass production data exception rule, if the chemical components, the production process parameter actual results and the performance data exceed the reasonable data range, judging that the mass production data is abnormal, prompting the research and development personnel that abnormal data exists, facilitating the research and development personnel to perform related processing on the abnormal data, otherwise, judging that the mass production data is normal.
Step 2.3.4: and calculating the characteristic values of the average value, the maximum value, the minimum value and the like of the high-frequency data and the performance data of the steel coil of the silicon steel product.
Step 2.3.5: based on the whole process of large-scale production, the export coil number and the import coil number of each unit are defined, the export coil number of the former unit is equal to the entry coil number of the latter unit as logic, and the whole process data of large-scale production of the silicon steel products are connected in series to form the subject of the large-scale production data.
The data of the whole large production process comprises production process parameter actual results, inspection and test data, surface defect data, quality judgment data, yield, ton steel energy consumption, ton steel cost and the like of each unit, and all the data of the whole large production process and the production process parameter actual results, the inspection and test results and the like of each unit can be traced for any steel coil from any process.
In the user demand data theme, the laboratory research and development data theme and the mass production data theme, the same data fields are in one-to-one correspondence, for example: the user demand data theme, the laboratory research and development data theme and the mass production data theme all comprise the same data field performance, the data field performance in the user demand data theme, the laboratory research and development data theme and the mass production data theme is correspondingly arranged, a reliable data base is provided for processing steps of filtering and selecting data in a subsequent production scheme and the like, and the data processing efficiency and precision are improved.
And step 3: inputting the requirement information of the type, specification, performance and the like of the silicon steel products, and generating a plurality of production schemes of the silicon steel products according to the requirement information, wherein the production schemes comprise a production scheme for optimizing the products and a production scheme for new products.
Step 3.1: the method comprises the steps of respectively filtering a laboratory research and development data theme and a mass production data theme according to demand information, and respectively selecting two types of related data which meet the demand information, namely the laboratory research and development related data and the mass production related data, particularly, when the laboratory research and development data theme is screened, because the detection standard of a mass production finished product is different from the detection standard of laboratory sample performance, the performance index of the laboratory research and development data theme needs to be multiplied by an adjustment factor (such as 1.2) and then participates in screening, and the adjustment factor is set by research and development personnel based on the comparison of the performance indexes of historical mass production data and research and development data under the same process setting. The demand information may be determined based on performance demand indicators of the user for the silicon steel product, or may be determined based on research and development requirements of research and development personnel for new silicon steel products.
Step 3.2: and (3) respectively processing the two types of relevant data screened in the step (3.1) based on the performance requirements in the requirement information, wherein the data processing comprises normalization and standardization so as to eliminate the influence of different performance indexes, different dimensions or abnormal values and improve the accuracy of the generation of the production scheme.
The normalized formula is: x ═ X-MIN)/(MAX-MIN);
the normalized formula is: x ═ X- μ)/σ.
Wherein X is the index of each performance requirement of the silicon steel product.
When the relevant data of laboratory research and development is processed in a normalization mode, MIN is the minimum value of the performance indexes of the relevant data of laboratory research and development, and MAX is the maximum value of the performance indexes of the relevant data of laboratory research and development; mu is the average value of the performance indexes of the relevant data of the laboratory research and development, and sigma is the standard deviation of the main performance indexes of the relevant data of the laboratory research and development.
When the large production related data are processed in a normalization mode, the MIN is the minimum value of the performance data of the large production related data, and the MAX is the maximum value of the performance data of the large production related data; mu is the average value of the performance indexes of the large production related data, and sigma is the standard deviation of the performance indexes of the large production related data.
Normalization and normalization are conventional means of data processing and will not be described in detail herein.
Step 3.3: for each type of relevant data, a first performance similarity between the performance requirement and the normalized and normalized performance data is calculated.
Preferably, the first performance similarity S1 may be calculated by using a euclidean distance similarity algorithm, where the calculation formula is: s1 ═ 1/(1+ DIST) × 100, where DIST is the distance between the performance demand indicator and the normalized, normalized performance data, where the performance data uses the average or maximum, minimum of multiple sampling points according to their characteristics. The euclidean distance similarity algorithm is a conventional technical means for data processing, and is not described herein again.
Step 3.4: and setting a performance similarity threshold based on the tolerance of the user to the performance deviation range, if the performance similarity threshold is 80%, if the first performance similarity is larger than or equal to the performance similarity threshold, marking the silicon steel product corresponding to the first performance similarity as an optimized product, and executing the step 3.5, if the first performance similarity is smaller than the performance similarity threshold, marking the silicon steel product corresponding to the first performance similarity as a new product, and executing the step 3.6.
Step 3.5: generating a production scheme of an optimized product: based on each type of related data, chemical components corresponding to the first performance similarity, production process parameter actual results and control precision of each unit in large-scale production are combined, research and development personnel select key chemical components (such as elements C, Si and the like) and key process parameters (such as normalized highest temperature and the like), the chemical components and the production process parameter actual results are divided one by one according to the control precision of each key chemical component and each key process parameter, the production process parameter actual results comprise characteristic values such as low-frequency process parameter actual results and average values, maximum values and minimum values of high-frequency data, all the chemical components and the division ranges of the production process parameter actual results are combined, and a production scheme for optimizing products is generated.
Step 3.6: and establishing a performance index prediction model based on the performance requirements, and verifying the optimal production scheme of the new product.
Step 3.6.1: and selecting historical data from the mass production data, wherein the type and the specification of the silicon steel product in the historical data are the same as those of the input silicon steel product, and the time range of the mass production data is the time range of the last year, namely the time range of the previous year from the current time point.
Step 3.6.2: establishing a plurality of performance index prediction models by using a multi-output xgboost regression algorithm and a neural network algorithm, wherein the number of the performance index prediction models is preferably 3-5; the input of each performance index prediction model is the selection of key chemical components and key process parameters for research personnel, the output is performance indexes (such as iron loss, magnetic induction and other electromagnetic performance), and the optimal performance index prediction model in the plurality of performance index prediction models is determined through cross validation of the minimum Mean Absolute Error (MAE).
Step 3.6.3: and (4) calculating the average values of the chemical components and the actual results of the production process parameters in the large production data selected in the step 3.6.1, and taking the average values of the chemical components and the actual results of the production process parameters as the plan datum points.
Step 3.6.4: and performing gridding search on the performance index prediction model from the scheme reference point, namely forming a plurality of schemes based on chemical components and production process parameter ranges through gridding search, and performing performance index prediction on the plurality of schemes based on the chemical components and the production process parameter ranges based on the performance index prediction model in 3.6.2 to form a plurality of production schemes. The control accuracy of the gridding search can be determined according to actual needs, and for example, the typical control accuracy of +/-0.0025 can be adopted.
And 4, step 4: and 3, comprehensively evaluating the production scheme of each silicon steel product generated in the step 3 in a multi-dimensional manner to obtain a green design index of the silicon steel product production scheme model, and recommending the production scheme according to the green design index, namely taking the production scheme with the highest green design index as a recommended scheme.
Preferably, the evaluation dimensions include performance similarity, yield, ton steel energy consumption and ton steel cost.
For each production scheme of the optimized product, the multi-dimensional comprehensive evaluation method comprises the following steps:
step 4.11: calculating a second performance similarity between the performance data of the production scheme of the optimized product and actual performance data of mass production S2, wherein the actual performance data of mass production corresponds to the design range of each chemical component and production process in the production scheme of the optimized product, the second performance similarity S2 is calculated by using the Euclidean distance similarity algorithm in the step 3.3, and the calculation formula is as follows: s2 ═ 1/(1+ DIST) × 100, where DIST is the distance between the performance data of the production schedule of the optimized product and the actual performance data of the large production, where the performance data uses the average or maximum, minimum of a plurality of sampling points according to their characteristics. The Euclidean distance similarity algorithm is a conventional technical means for data processing, and is not described herein again; if the performance data of the production scheme is completely met, namely the performance data of the production scheme is completely matched with the actual performance data of the mass production, the first performance score is recorded as 100, the step is switched to the step 4.13, and if the performance data of the production scheme is not completely met, the step 4.12 is executed.
Step 4.12: separately computationally optimizing production of productsSimilarity between magnetic induction and iron loss of the scheme and actual magnetic induction and iron loss of mass production is evaluated on the basis of weight, and third performance similarity S3 is obtained, wherein the evaluation formula is as follows: λ is S3 ═ λ11*SMagnetic induction 1+λ21*SIron loss 1The first performance score is denoted as S3.
Wherein S isMagnetic induction 1In order to optimize the similarity between the magnetic induction of the production scheme performance data of the product and the magnetic induction of the mass production actual performance data, SMagnetic induction 1The Euclidean distance similarity algorithm in the step 3.3 can be adopted for calculation, and details are not repeated here; lambda [ alpha ]11Is SMagnetic induction 1The weight of (c).
SIron loss 1In order to optimize the similarity between the magnetic induction of the production scheme performance data of the product and the magnetic induction of the mass production actual performance data, SIron loss 1The Euclidean distance similarity algorithm in the step 3.3 can be adopted for calculation, and details are not repeated here; lambda [ alpha ]21Is SIron loss 1Weight of (a), λ11+λ21＝1。
Step 4.13: and evaluating and optimizing the first material forming rate of each unit of the production scheme of the product based on the weight to obtain a first material forming rate index.
The calculation formula of the first yield of each unit is as follows:
first yield η1The outlet weight ∑ (exit coil weight per inlet coil contribution weight) and the inlet coil contribution weight ∑ (entry coil contribution weight) are 100.
The calculation formula of the first yield index is as follows: first yield index ═ Σ (λ)3*η1) Wherein λ is3The weight of the first yield corresponding to each unit, and ∑ λ31, the research and development personnel can properly adjust lambda according to the importance of the unit3The first yield score is recorded as a first yield index.
Step 4.14: the method is characterized in that the ton steel energy consumption of the production scheme of the optimized product is evaluated based on the unit energy consumption of steel coil production, and the evaluation formula is as follows: and N' ((N-Nmin)/(Nmax-Nmin)). 100, namely when the energy consumption N is the lowest, the first energy consumption score is marked as 100 points, when the energy consumption N is the highest, the first energy consumption score is marked as 0 point, and the rest are calculated according to a formula, wherein the first energy consumption score is between 0 and 100.
The method comprises the steps of obtaining a steel coil, obtaining N ═ Sigma (Ne)/N, Ne, N, Nmin and Nmax, wherein N is the unit energy consumption of the steel coil production, Ne is the energy consumption of each ton of steel coil, N is the number of the steel coils, Nmin is the unit energy consumption of the minimum steel coil production, and Nmax is the unit energy consumption of the minimum steel coil production.
Step 4.15: the ton steel cost of the production scheme model of the optimized product is estimated based on the ton steel cost in steel coil production, and the estimation formula is as follows: and C' ((C-Cmin)/(Cmax-Cmin)). 100, when the cost of steel per ton is lowest, the first cost score is recorded as 100, when the cost of steel per ton is highest, the first cost score is recorded as 0, and the rest of the cost scores are calculated according to a formula, and the first cost score is between 0 and 100.
Wherein, C is steel coil production ton steel cost, C ═ Σ (Ce)/n, Ce is steel coil production ton steel cost, n is steel coil number, Cmin is minimum steel coil production ton steel cost, and Cmax is maximum steel coil production ton steel cost.
Step 4.16: calculating the green design index of the optimized product based on the weight, wherein the calculation formula is as follows:
green design index lambda of optimized product41First performance score + λ42First yield score + λ43First energy consumption score + λ44First cost score.
Wherein λ is41Weight for the first performance score, λ42Weight, λ, for the first yield score43Weight, λ, for the first energy consumption score44A weight scored for the first cost, and41+λ42+λ43+λ44＝1。
for each new product production scheme, the multi-dimensional comprehensive evaluation method comprises the following steps:
step 4.21: calculating a fourth performance similarity between the performance data of the production scheme of the new product and the actual performance data of mass production S4, wherein the actual performance data of mass production corresponds to the design range of each chemical component and production process in the production scheme of the new product, and the fourth performance similarity S4 can be calculated by using the Euclidean distance similarity calculation method in the step 3.3, and is not repeated herein; if the product is completely satisfied, that is, the performance data of the production scheme of the new product is completely matched with the actual performance data of the mass production, the second performance score is marked as 100 (which may occur only when the production scheme of the new product is generated from the development data), the process goes to step 4.23, and if the product is not completely satisfied, the step 4.22 is executed.
Step 4.22: respectively calculating the similarity of the magnetic induction and the iron loss of the production scheme model of the new product and the actual magnetic induction and the iron loss of the large production, and performing performance evaluation based on the weight to obtain a fifth performance similarity S5, wherein the evaluation formula is as follows: λ is S5 ═ λ12*SMagnetic induction 2+λ22*SIron loss 2And the second performance score is denoted as S5.
Wherein S isMagnetic induction 2The similarity between the magnetic induction of the performance data of the production scheme of the new product and the magnetic induction of the actual performance data of the mass production, SMagnetic induction 2The Euclidean distance similarity algorithm in the step 3.3 can be adopted for calculation, and details are not repeated here; lambda [ alpha ]12Is SMagnetic induction 2The weight of (c).
SIron loss 2The similarity between the magnetic induction of the performance data of the production scheme of the new product and the magnetic induction of the actual performance data of the mass production, SIron loss 2The Euclidean distance similarity algorithm in the step 3.3 can be adopted for calculation, and details are not repeated here; lambda [ alpha ]22Is SIron loss 2Weight of (a), λ12+λ22＝1。
Step 3.23: and evaluating the second yield of each unit of the production scheme of the new product based on the weight to obtain a second yield index.
The calculation formula of the second yield of each unit is as follows:
second yield η2The outlet weight ∑ (exit coil weight per inlet coil contribution weight) and the inlet coil contribution weight ∑ (entry coil contribution weight) are 100.
The calculation formula of the second yield index is as follows: second yield index ═ Σ (λ)5*η2) Wherein λ is5Second yield eta for each unit2And sigma lambda of51, can be obtained by grindingThe generator properly adjusts lambda according to the importance of the generator set5And recording the second yield score as a second yield index.
Step 4.24: the energy consumption of the production scheme of the new product is evaluated based on the energy consumption of the steel coil production unit, and the evaluation formula is as follows: and N' ((N-Nmin)/(Nmax-Nmin)). 100, namely when the energy consumption N is the lowest, the second energy consumption score is marked as 100 points, when the energy consumption N is the highest, the second energy consumption score is marked as 0 point, and the rest are calculated according to a formula, wherein the second energy consumption score is between 0 and 100.
Wherein, N is the unit energy consumption of steel coil production, N ═ Σ (Ne)/N, Nmin is the minimum unit energy consumption of steel coil production, and Nmax is the minimum unit energy consumption of steel coil production.
Step 4.25: the ton steel cost of the production scheme model of the optimized product is estimated based on the ton steel cost in steel coil production, and the estimation formula is as follows: and C ═ 100 ((C-Cmin)/(Cmax-Cmin)). 100, namely when the cost per ton of steel is lowest, the second cost score is recorded as 100, when the cost per ton of steel is highest, the second cost score is recorded as 0, and the rest of the cost scores are calculated according to the formula, and the second cost score is between 0 and 100.
Wherein, C is the steel coil production ton steel cost, C ═ Σ (Ce)/n, Cmin is the minimum steel coil production ton steel cost, and Cmax is the minimum steel coil production ton steel cost.
Step 4.26: calculating the green design index of the new product based on the weight, wherein the calculation formula is as follows:
green design index of new product lambda61Second performance score + λ62Second yield score + λ63Second energy consumption score + λ64Second cost score.
Wherein λ is61Weight for the second performance score, λ62Weight, λ, for the second yield score63Weight, λ, for the second energy consumption score64A weight scored for the second cost, and61+λ62+λ63+λ64＝1。
step 4.27: calculating the chemical composition and the process parameter in the production scheme of the new product and the chemical composition and the production tool of the most similar steel grade (namely the steel grade corresponding to the steel tapping mark with the minimum sum of normalized Euclidean distance DIST)The deviation degree coefficient P between the actual results of the technological parameters is calculated by the following formula: p ═ P1+P2；
Wherein, P1For a first coefficient of deviation between the chemical composition in the production scheme of the new product and the chemical composition of the most similar steel grade, the calculation formula is:
wherein λ isiIs the weight of each chemical component,
is the design value, x, of each chemical component in the production scheme of a new productikIs the actual performance value of each chemical component of the most similar steel grade, NiThe number is recorded for the most similar steel grades, i is the number of chemical components.
P2For a second deviation degree coefficient between the process parameters in the new product production scheme and the process parameters of the most similar steel grade, the calculation formula is as follows:
wherein, betaiAre the weights of the various process parameters,
is the design value, g, of each process parameter in the production scheme of a new productikActual performance values of the individual process parameters for the most similar steel grades, NiThe number is recorded for the most similar steel grades, i is the number of process parameters.
Preferably, the chemical composition may include the content ranges of elements such as C, Si, Mn, Al, and P in the silicon steel product.
Step 4.28: correcting the green design index of the new product through the deviation degree coefficient P to obtain the corrected green design index of the new product, wherein the correction calculation formula is as follows:
the corrected green design index of the new product-P.
Example 1:
step 1: data are collected and user data, laboratory research and development data and mass production data are integrated.
The user data comprises performance requirement indexes (including iron loss, magnetic induction and other electromagnetic performances) of the user on various silicon steel products, user ID, inquiry ID, basic information of the user and the like.
The laboratory research and development data comprise chemical components, research and development process parameters, various performance indexes (such as iron loss, electromagnetic performance of magnetic induction and the like), organization analysis results and other data in the whole experimental process of various silicon steel products. The whole experimental process comprises the working procedures of laboratory steel making, laboratory hot rolling, laboratory normalizing, laboratory cold rolling, laboratory continuous annealing, laboratory decarburization annealing, laboratory magnesium oxide coating, laboratory high-temperature annealing, laboratory coating and the like. Laboratory research and development data can be obtained from a research and development management system of silicon steel products, and each item of data in the laboratory research and development data is accumulated for at least more than one year so as to ensure the construction accuracy of a production scheme.
The large production data is the production data of each unit in the whole large production process of various silicon steel products, and comprises chemical components (such as elements C, Si, S and the like), steel tapping marks, production material tracking information (such as the unit number, the coil inlet number, the coil outlet number and the like corresponding to the production process of the silicon steel coils), production process parameter actual results (such as steel tapping temperature, normalizing maximum temperature and the like, low-frequency data collected according to the coils and high-frequency data corresponding to the coils collected according to distance or time intervals, wherein each coil only has one low-frequency data, each coil has a plurality of high-frequency data), various performance data (such as electromagnetic energy such as iron loss, magnetic induction and the like), inspection and test data and surface defect data, quality judgment data, outlet/inlet coil weight, yield, inlet coil weight apportionment, ton steel energy consumption, ton steel cost and the like. The mass production data can be obtained from a process control system and a manufacturing execution system of steel production, and each item of data in the mass production data is accumulated for at least more than one year so as to ensure the accuracy of constructing a scheme model.
Step 2: and constructing a user data theme, a laboratory research and development data theme and a mass production data theme.
Step 2.1: constructing a user requirement data theme, wherein the specific operation is as follows: and based on the user ID and the inquiry ID in the user data, linking the basic information of a certain user with the performance requirement index of the user on the silicon steel product to form a user requirement data theme.
Step 2.2: the method comprises the following specific operations of constructing a laboratory research and development data theme of a silicon steel product:
step 2.2.1: and configuring laboratory research and development data exception rules of chemical components, research and development process parameters and performance indexes of the silicon steel product.
Step 2.2.2: preprocessing the laboratory research and development data based on the laboratory research and development data exception rule, judging that the laboratory research and development data are abnormal if chemical components, research and development process parameters and performance indexes exceed reasonable data ranges, prompting research and development personnel to have exception data, facilitating the research and development personnel to perform relevant processing on the exception data, and otherwise judging that the laboratory research and development data are normal.
Step 2.2.3: based on the whole experimental process, according to the steel-making date and the heat number, and the slab and the hot rolling and cold rolling coil dividing information of the silicon steel product, the inlet sample number and the outlet sample number of each experimental process are defined, the outlet sample number of the previous experimental process is logic as the inlet sample number of the next experimental process, and the whole experimental process data of the silicon steel product are connected in series to form the research and development data theme of a laboratory, so that the whole experimental process can be traced from steel-making to cold rolling conveniently.
Step 2.3: the method comprises the following specific operations of constructing a mass production data theme of the silicon steel product:
step 2.3.1: and configuring large production data exception rules of chemical components, production process parameter actual performance and performance data of the silicon steel product.
Step 2.3.2: and preprocessing the mass production data based on a mass production data exception rule, if the chemical components, the production process parameter actual results and the performance data exceed the reasonable data range, judging that the mass production data is abnormal, prompting the research and development personnel that abnormal data exists, facilitating the research and development personnel to perform related processing on the abnormal data, otherwise, judging that the mass production data is normal.
Step 2.3.4: and calculating the characteristic values of the average value, the maximum value, the minimum value and the like of the high-frequency data and the performance data of the steel coil of the silicon steel product.
Step 2.3.5: based on the whole process of large-scale production, the export coil number and the import coil number of each unit are defined, the export coil number of the former unit is equal to the entry coil number of the latter unit as logic, and the whole process data of large-scale production of the silicon steel products are connected in series to form the subject of the large-scale production data.
The data of the whole large production process comprises production process parameter actual results, inspection and test data, surface defect data, quality judgment data, yield, ton steel energy consumption, ton steel cost and the like of each unit, and all the data of the whole large production process and the production process parameter actual results, the inspection and test results and the like of each unit can be traced for any steel coil from any process.
And step 3: the type, specification and performance of silicon steel product demand information are input, and the user's demand information in this embodiment is: oriented silicon steel with thickness specification of 0.20mm and electromagnetic induction intensity B8(corresponding magnetic polarization strength when the magnetic field strength is 800A/m) is not less than 1.93T;
and respectively filtering the large production data theme and the research and development data theme based on the requirement information of the user, wherein the adjustment factor is 1.01, and corresponding data can be respectively obtained.
And carrying out standardization processing on the matched data, and calculating the similarity between the matched data and the magnetic performance requirement, wherein the mass production data is 1.93 correspondingly, the research and development data is 1.95, and the electromagnetic performance data adopts the mean value of each sampling point.
Setting the threshold value of the performance similarity as 50%, screening data which meet the performance similarity in the two types of data, marking the data as optimized products, and generating a production scheme of the optimized products.
The research personnel of the product select the key chemical component as the content of the element Si, the key process parameter as the decarburization temperature range, and the control precision is 0.05 and 5 respectively. The division is performed according to the control precision, and 5 division schemes can be obtained, as shown in table 1.
TABLE 1 partitioning scheme based on chemical composition Si content and decarburization temperature range
| Range of Si content | Range of decarburization temperature | Plan numbering |
| (3.20,3.25] | (840,845] | 1 |
| (3.20,3.25] | (835,840] | 2 |
| (3.20,3.25] | (830,835] | 3 |
| [3.15,3.25] | (835,840] | 4 |
| [3.15,3.25] | (830,835] | 5 |
And 4, step 4: carrying out multi-dimensional evaluation on the formed 5 schemes to obtain a green manufacturing index, and recommending the scheme:
performance scoring: production protocol 1 performance may be matched to the corresponding big production data performance data with a first performance score of 100 and the rest scored based on similarity.
Grading yield: the yield of the product concerns the normalizing, rolling and decarburization of the unit, wherein the decarburization weight is higher. The first yield is 0.3 normalization unit yield +0.3 rolling unit yield +0.4 decarburization unit yield.
And (3) scoring the first energy consumption: and comprehensively calculating the energy consumption of the ton steel according to the screening data theme.
First cost score: and comprehensively calculating the ton steel cost according to the screening data theme.
Green design index: the green design index is 0.5 × first performance score +0.3 × first yield score +0.1 × first energy consumption score +0.1 × first cost score, and the scoring results are shown in table 2.
TABLE 25 multidimensional evaluation of the schemes and their Green design indices
As can be seen from table 2, since the score of scenario 1 is the highest, scenario 1 is recommended as the preferred scenario.
Example 2:
step 1: data are collected and user data, laboratory research and development data and mass production data are integrated.
Step 2: and constructing a user data theme, a laboratory research and development data theme and a mass production data theme.
And step 3: inputting the type, specification and performance of the silicon steel product, wherein the user requirement information of the oriented silicon steel of the embodiment is that the thickness specification is 0.20mm and the electromagnetic induction intensity is B8(magnetic polarization strength corresponding to the magnetic field strength of 800A/m) performance of not less than 1.93T, and iron loss P1.7/50Not higher than 1.0W/kg.
And respectively filtering the mass production data theme and the research and development data theme based on the user demand information, wherein the magnetic induction and the iron loss adjustment factors are 1.01 and 0.98, and respectively acquiring corresponding data.
And carrying out standardization processing on the matched data, and calculating the similarity between the matched data and the magnetic performance requirement, wherein the mass production data correspond to magnetic induction of 1.93 and iron loss of 1.0, the research and development data correspond to magnetic induction of 1.95 and iron loss of 0.98, and the magnetic performance data adopt the mean value of each sampling point.
Setting the threshold value of the performance similarity as 50%, screening the data which meets the performance similarity in the two types of data, marking the data without the product which meets the threshold value of the similarity as a new product, and generating a production scheme of the new product.
And selecting the oriented silicon steel product data with the thickness specification of 0.02 in the last year in the large production data.
The research personnel select the key chemical component as element C, Si content, the key process parameter as decarburization temperature, and the control precision is 0.002, 0.05 and 5 respectively. Based on the data in the previous step, the average absolute error of the model trained by using the xgboost (the maximum depth of the tree is 5) is the minimum, and the model is the prediction model of the optimal performance index by using the xgboost (the maximum depth of the tree is 5), the neural network (two hidden layers), the training input (C content, Si content, decarburization temperature) and the output (magnetic induction and iron loss) and by using 10 cross validation for evaluation.
The reference points for obtaining the C content, Si content, and decarburization temperature based on mass production data were (0.056, 3.25, 845).
Performing a gridding search based on the fiducial points forms 5 scenarios, as shown in table 3.
TABLE 3 partitioning scheme based on chemical composition C and Si contents and decarburization temperature range
| Range of C content | Range of Si content | Range of decarburization temperature | Plan numbering |
| (0.056,0.058] | (3.25,330] | (845,850] | 1 |
| (0.056,0.058] | (3.25,330] | [840,845] | 2 |
| (0.056,0.058] | [3.20,3.25] | (845,850] | 3 |
| [0.054,0.056] | [3.20,3.25] | (845,850] | 4 |
| [0.054,0.056] | [3.20,3.25] | [840,845] | 5 |
And predicting magnetic induction and iron loss performance based on the optimal performance index prediction model.
And 4, step 4: multidimensional scheme evaluation: carrying out multi-dimensional evaluation on the formed 5 schemes to obtain a green manufacturing index, and recommending the scheme:
performance scoring: none of the above 5 schemes has been produced, that is, none of the schemes can match with the actual performance of mass production, and the calculation formula is: a second performance score of 0.4 magnetic induction similarity +0.6 core loss similarity;
second yield score: the unit concerned about the yield of the product comprises normalizing, decarbonizing and continuous annealing, wherein the decarbonizing weight is higher. The second yield is 0.3 × normalizing unit yield +0.4 × decarbonizing unit yield +0.3 × continuous annealing unit yield.
And (3) grading the second energy consumption: and comprehensively calculating the energy consumption of the ton steel according to the screening data theme.
Second cost score: and comprehensively calculating the ton steel cost according to the screening data theme.
Green design index: green design index 0.5 second performance score +0.3 second yield score +0.1 second energy consumption score +0.1 second cost score.
Calculating a deviation degree coefficient:
TABLE 45 multidimensional evaluation of schemes and Green design indices thereof
As can be seen from table 4, scenario 4 is recommended as the preferred scenario because scenario 4 has the highest score.
The present invention is not limited to the above embodiments, and any modifications, equivalent replacements, improvements, etc. within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. An intelligent design evaluation method for a silicon steel product production scheme is characterized by comprising the following steps: the method comprises the following steps:
step 1: collecting data, and integrating user data, laboratory research and development data and mass production data;
step 2: constructing a user data theme, a laboratory research and development data theme and a mass production data theme;
and step 3: inputting the requirement information of the type, specification and performance requirements of the silicon steel products, and generating a plurality of production schemes of the silicon steel products according to the requirement information, wherein the production schemes comprise a production scheme for optimizing the products and a production scheme for new products;
and 4, step 4: and 3, comprehensively evaluating the production scheme of each silicon steel product generated in the step 3 in a multi-dimensional manner to obtain a green design index of the production scheme of the silicon steel product, and recommending the production scheme according to the green design index.
2. The intelligent design evaluation method of the silicon steel product production scheme of claim 1, which is characterized in that: the user data comprises performance requirement indexes of users on various silicon steel products, user IDs (identity) and inquiry IDs and basic information of the users;
the laboratory research and development data comprise chemical components, research and development process parameters, various performance indexes and organization analysis results in the whole experimental process of various silicon steel products;
the large production data are production data of all units in the whole large production process of various silicon steel products, and comprise chemical components, steel tapping mark production material tracking information, production process parameter actual results, all performance data, inspection and test data, surface defect data, quality judgment data, outlet/inlet coil weight, yield, inlet coil weight apportionment amount, steel energy consumption per ton and steel cost per ton.
3. The intelligent design evaluation method of the silicon steel product production scheme of claim 1, which is characterized in that: the production process parameter actual performance comprises low-frequency data collected according to steel coils and high-frequency data collected according to distance or time intervals, and each steel coil has one piece of low-frequency data and a plurality of pieces of high-frequency data.
4. The intelligent design evaluation method of the silicon steel product production scheme of claim 1, which is characterized in that: the step 2 comprises the following steps:
step 2.1: constructing a user demand data theme;
step 2.2: constructing a laboratory research and development data theme of the silicon steel product;
step 2.3: and constructing a mass production data theme of the silicon steel product.
5. The intelligent design evaluation method of the silicon steel product production scheme of claim 4, which is characterized in that: the step 2.1 comprises the following steps: based on the user ID and the inquiry ID in the user data, linking the basic information of a certain user with the performance requirement index of the user on the silicon steel product to form a user requirement data theme;
the step 2.2 comprises the following steps:
step 2.2.1: configuring laboratory research and development data exception rules of chemical components, research and development process parameters and performance indexes of the silicon steel product;
step 2.2.2: preprocessing the laboratory research and development data based on the laboratory research and development data exception rule;
step 2.2.3: defining the inlet sample number and the outlet sample number of each experimental procedure based on the whole experimental procedure, wherein the outlet sample number of the previous experimental procedure is the inlet sample number of the next experimental procedure and is logic, and the whole experimental procedure data of the silicon steel product is connected in series to form a research and development data theme of a laboratory;
the step 2.3 comprises the following steps:
step 2.3.1: configuring large production data exception rules of chemical components, production process parameter actual performance and performance data of the silicon steel product;
step 2.3.2: preprocessing the mass production data based on the mass production data exception rule;
step 2.3.4: calculating the characteristic values of the high-frequency data and the performance data of the steel coil of the silicon steel product;
step 2.3.5: based on the whole process of large-scale production, the export coil number and the import coil number of each unit are defined, the export coil number of the former unit is equal to the entry coil number of the latter unit as logic, and the whole process data of large-scale production of the silicon steel products are connected in series to form the subject of the large-scale production data.
6. The intelligent design evaluation method of the silicon steel product production scheme of claim 1, which is characterized in that: the step 3 comprises the following steps:
step 3.1: filtering a laboratory research and development data theme and a mass production data theme according to the demand information, and selecting two types of related data which meet the demand information; wherein, the performance index in the subject of laboratory research and development data is filtered after being multiplied by an adjustment factor;
step 3.2: respectively carrying out normalization and standardization data processing on each type of relevant data based on the performance requirements in the requirement information;
step 3.3: for each type of related data, respectively calculating a first performance similarity between the performance requirement and the normalized and standardized performance data;
step 3.4: setting a performance similarity threshold, if the first performance similarity is larger than or equal to the performance similarity threshold, marking the silicon steel product corresponding to the first performance similarity as an optimized product, and executing the step 3.5, if the first performance similarity is smaller than the performance similarity threshold, marking the silicon steel product corresponding to the first performance similarity as a new product, and executing the step 3.6;
step 3.5: generating a production scheme of an optimized product: dividing the production process parameter actual results by combining the production process parameter actual results corresponding to the first performance similarity and the control precision of each unit in large-scale production;
step 3.6: and establishing a performance index prediction model based on the performance requirements, and verifying the optimal production scheme model of the new product.
7. The intelligent design evaluation method of the silicon steel product production scheme of claim 6, which is characterized in that: the step 3.6 comprises the following steps:
step 3.6.1: selecting historical data from the mass production data, wherein the type and the specification of the silicon steel product in the historical data are the same as those of the input silicon steel product, and the time range of the mass production data is the latest year;
step 3.6.2: establishing a plurality of performance index prediction models; the performance index prediction model comprises a plurality of performance index prediction models, wherein the input of each performance index prediction model is key chemical components and key process parameters selected by research personnel, the output of each performance index prediction model is a performance index, and the optimal performance prediction model in the performance index prediction models is determined through cross validation;
step 3.6.3: calculating the average value of the chemical components and the actual performance of the production process parameters in the large production data selected in the step 3.6.1, and taking the average value of the chemical components and the actual performance of the production process parameters as a scheme reference point;
step 3.6.4: and carrying out gridding search on the performance index prediction model from the scheme reference point.
8. The intelligent design evaluation method of the silicon steel product production scheme of claim 1, which is characterized in that: in the step 4, the evaluation dimensions comprise performance similarity, yield, ton steel energy consumption and ton steel cost.
9. The intelligent design evaluation method of the silicon steel product production scheme of claim 1, which is characterized in that: for each production scheme of the optimized product, the multi-dimensional comprehensive evaluation method comprises the following steps:
step 4.11: calculating a second performance similarity between the performance data of the production scheme of the optimized product and the actual performance data of mass production, if the second performance similarity is completely met, namely the performance data of the production scheme of the optimized product is completely matched with the actual performance data of mass production, marking the first performance score as 100, turning to step 4.13, and if the first performance score is not completely met, executing step 4.12;
step 4.12: respectively calculating the similarity of the magnetic induction and the iron loss of the production scheme of the optimized product and the actual magnetic induction and the iron loss of the large-scale production, and performing performance evaluation based on the weight to obtain a third performance similarity S3, wherein the evaluation formula is as follows: λ is S3 ═ λ11*SMagnetic induction 1+λ21*SIron loss 1The first performance score is denoted as S3;
wherein S isMagnetic induction 1In order to optimize the similarity between the magnetic induction of the production scheme performance data of the product and the magnetic induction of the mass production actual performance data, lambda11Is SMagnetic induction 1The weight of (c); sIron loss 1In order to optimize the similarity between the magnetic induction of the production scheme performance data of the product and the magnetic induction of the mass production actual performance data, lambda21Is SIron loss 1The weight of (c); lambda [ alpha ]11+λ21＝1；
Step 4.13: evaluating and optimizing the first material forming rate of each unit of the production scheme model of the product based on the weight to obtain a first material forming rate index;
the calculation formula of the first yield of each unit is as follows:
first yield η1100 (exit weight/entrance roll contribution weight);
wherein, the outlet weight is ═ Σ (outlet coil weight of each steel coil), and the inlet coil apportioned weight is ═ Σ (inlet coil apportioned weight of each steel coil);
the calculation formula of the first yield index is as follows: first yield index ═ Σ (λ)3*η1) Wherein λ is3The weight of the first yield corresponding to each unit, and ∑ λ3The first yield score is recorded as a first yield index when the yield score is 1;
step 4.14: evaluating the energy consumption of a production scheme model of an optimized product based on the energy consumption of steel coil production units, wherein when the energy consumption is the lowest, a first energy consumption score is marked as 100, and when the energy consumption is the highest, the first energy consumption score is marked as 0;
step 4.15: evaluating the ton steel cost of a production scheme model for optimizing a product based on the ton steel cost in steel coil production, wherein when the ton steel cost is the lowest, the first cost score is recorded as 100, and when the ton steel cost is the highest, the first cost score is recorded as 0;
step 4.16: calculating the green design index of the optimized product based on the weight, wherein the calculation formula is as follows:
green design index lambda of optimized product41First performance score + λ42First yield score + λ43First energy consumption score + λ44First cost score;
wherein λ is41Weight for the first performance score, λ42Weight, λ, for the first yield score43Weight, λ, for the first energy consumption score44A weight scored for the first cost, and41+λ42+λ43+λ44＝1。
10. the intelligent design evaluation method of the silicon steel product production scheme of claim 1, which is characterized in that: for each new product production scheme, the multi-dimensional comprehensive evaluation method comprises the following steps:
step 4.21: and calculating a fourth performance similarity between the performance data of the production scheme of the new product and the actual performance data of the mass production, if the fourth performance similarity is completely met, namely the performance data of the production scheme of the new product is completely matched with the actual performance data of the mass production, marking the second performance score as 100, turning to the step 4.23, and if the fourth performance similarity is not completely met, executing the step 4.22.
Step 4.22: respectively calculating the similarity of the magnetic induction and the iron loss of the production scheme of the new product and the actual magnetic induction and the iron loss of the large production, and performing performance evaluation based on the weight to obtain a fifth performance similarity S5, wherein the evaluation formula is as follows: λ is S5 ═ λ12*SMagnetic induction 2+λ22*SIron loss 2(ii) a A second performance score is denoted as S5;
wherein S isMagnetic induction 2Is the similarity between the magnetic induction of the performance data of the production scheme of a new product and the magnetic induction of the actual performance data of mass production, lambda12Is SMagnetic induction 2The weight of (c); sIron loss 2Is the similarity between the magnetic induction of the performance data of the production scheme of a new product and the magnetic induction of the actual performance data of mass production, lambda22Is SIron loss 2The weight of (c); lambda [ alpha ]12+λ22＝1；
Step 4.23: evaluating a second yield of each unit of the production scheme of the new product based on the weight to obtain a second yield index;
the calculation formula of the second yield of each unit is as follows:
second yield η2100 (exit weight/entrance roll contribution weight);
wherein, the outlet weight is ═ Σ (outlet coil weight of each steel coil), and the inlet coil apportioned weight is ═ Σ (inlet coil apportioned weight of each steel coil);
the calculation formula of the second yield index is as follows: second yield index ═ Σ (λ)5*η2) And sigma lambda5The second yield score is recorded as a second yield index when the yield score is 1;
step 4.24: evaluating the energy consumption of the production scheme of the new product based on the energy consumption of the steel coil production unit, wherein when the energy consumption is the lowest, the second energy consumption score is 100, and when the energy consumption is the highest, the second energy consumption score is 0;
step 4.25: evaluating the cost of steel per ton of a production scheme of an optimized product based on the cost of steel per ton produced by a steel coil, wherein when the cost of steel per ton is lowest, the second cost score is recorded as 100, and when the cost of steel per ton is highest, the second cost score is recorded as 0;
step 4.26: calculating the green design index of the new product based on the weight, wherein the calculation formula is as follows:
green design index of new product lambda61Second performance score + λ62Second yield score + λ63Second energy consumption score + λ64Second cost score;
wherein λ is61Weight for the second performance score, λ62Weight, λ, for the second yield score63Weight, λ, for the second energy consumption score64A weight scored for the second cost, and61+λ62+λ63+λ64＝1；
step 4.27: calculating the deviation coefficient P ═ P between the new product production scheme and the most similar steel grade1+P2；
Wherein, P1For a first coefficient of deviation between the chemical composition in the production scheme of the new product and the chemical composition of the most similar steel grade, the calculation formula is:
wherein λ isiFor each chemical compositionThe weight of the score is such that,
is the design value, x, of each chemical component in the production scheme of a new productikIs the actual performance value of each chemical component of the most similar steel grade, NiRecording the number of the most similar steel grades, wherein i is the number of chemical components;
P2for a second deviation degree coefficient between the process parameters in the new product production scheme and the process parameters of the most similar steel grade, the calculation formula is as follows:
wherein, betaiAre the weights of the various process parameters,
is the design value, g, of each process parameter in the production scheme of a new productikActual performance values of the individual process parameters for the most similar steel grades, NiRecording the number of the most similar steel grades, wherein i is the number of process parameters;
step 4.28: correcting the green design index of the new product through the deviation degree coefficient P to obtain the corrected green design index of the new product, wherein the correction calculation formula is as follows:
the corrected green design index of the new product-P.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110778749.9A CN114091800A (en) | 2021-07-09 | 2021-07-09 | Intelligent design evaluation method for silicon steel product production scheme |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110778749.9A CN114091800A (en) | 2021-07-09 | 2021-07-09 | Intelligent design evaluation method for silicon steel product production scheme |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114091800A true CN114091800A (en) | 2022-02-25 |
Family
ID=80296026
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110778749.9A Pending CN114091800A (en) | 2021-07-09 | 2021-07-09 | Intelligent design evaluation method for silicon steel product production scheme |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114091800A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115113596A (en) * | 2022-07-13 | 2022-09-27 | 浙江高裕家居科技股份有限公司 | Polyurethane material production parameter adjusting method and system based on quality monitoring |
| CN115840431A (en) * | 2023-02-27 | 2023-03-24 | 一夫科技股份有限公司 | Production control method and system for II-type anhydrous gypsum |
| CN115860592A (en) * | 2023-03-03 | 2023-03-28 | 一夫科技股份有限公司 | Quality detection and evaluation method and system for gypsum building blocks |
| CN116030900A (en) * | 2023-03-24 | 2023-04-28 | 安徽瑞邦数科科技服务有限公司 | Method, device, equipment and storage medium for controlling component content of chemical product |
| CN116501001A (en) * | 2023-06-27 | 2023-07-28 | 江苏宝孚克新能源科技有限公司 | Flexible aluminum alloy cable production process optimization control method and system |
| CN116501003A (en) * | 2023-06-29 | 2023-07-28 | 张家港广大特材股份有限公司 | Processing control method and system for smelting special steel |
| CN116702515A (en) * | 2023-08-03 | 2023-09-05 | 江苏甬金金属科技有限公司 | Nickel-plated steel strip preparation optimization method and system based on performance analysis |
-
2021
- 2021-07-09 CN CN202110778749.9A patent/CN114091800A/en active Pending
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115113596A (en) * | 2022-07-13 | 2022-09-27 | 浙江高裕家居科技股份有限公司 | Polyurethane material production parameter adjusting method and system based on quality monitoring |
| CN115113596B (en) * | 2022-07-13 | 2023-09-08 | 浙江高裕家居科技股份有限公司 | Polyurethane material production parameter adjusting method and system based on quality monitoring |
| CN115840431A (en) * | 2023-02-27 | 2023-03-24 | 一夫科技股份有限公司 | Production control method and system for II-type anhydrous gypsum |
| CN115840431B (en) * | 2023-02-27 | 2023-05-16 | 一夫科技股份有限公司 | Production control method and system for II-type anhydrous gypsum |
| CN115860592A (en) * | 2023-03-03 | 2023-03-28 | 一夫科技股份有限公司 | Quality detection and evaluation method and system for gypsum building blocks |
| CN116030900A (en) * | 2023-03-24 | 2023-04-28 | 安徽瑞邦数科科技服务有限公司 | Method, device, equipment and storage medium for controlling component content of chemical product |
| CN116501001A (en) * | 2023-06-27 | 2023-07-28 | 江苏宝孚克新能源科技有限公司 | Flexible aluminum alloy cable production process optimization control method and system |
| CN116501001B (en) * | 2023-06-27 | 2023-09-05 | 江苏宝孚克新能源科技有限公司 | Flexible aluminum alloy cable production process optimization control method and system |
| CN116501003A (en) * | 2023-06-29 | 2023-07-28 | 张家港广大特材股份有限公司 | Processing control method and system for smelting special steel |
| CN116501003B (en) * | 2023-06-29 | 2023-10-24 | 张家港广大特材股份有限公司 | Processing control method and system for smelting special steel |
| CN116702515A (en) * | 2023-08-03 | 2023-09-05 | 江苏甬金金属科技有限公司 | Nickel-plated steel strip preparation optimization method and system based on performance analysis |
| CN116702515B (en) * | 2023-08-03 | 2023-11-17 | 江苏甬金金属科技有限公司 | Nickel-plated steel strip preparation optimization method and system based on performance analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114091800A (en) | Intelligent design evaluation method for silicon steel product production scheme | |
| CN112036081B (en) | Method for determining addition amount of silicon-manganese alloy in converter tapping based on yield prediction | |
| CN107377634B (en) | A kind of hot-strip outlet Crown Prediction of Media method | |
| CN101391268B (en) | Reverse optimization method of steel plate rolling and cooling controlling-process temperature institution | |
| CN114897227A (en) | Multi-steel-grade mechanical property forecasting method based on improved random forest algorithm | |
| CN108469797B (en) | Neural network and evolutionary computation based ore grinding process modeling method | |
| CN113434994B (en) | Method for predicting cold rolling deformation resistance based on hot rolling process parameters of hot rolled original plate | |
| CN101320031A (en) | Austenitic stainless steel accurate steel strip performance prediction model and its cold rolling process planning | |
| CN108921232B (en) | Hot-rolled strip steel cooling historical data clustering and similarity measuring method | |
| CN106991242A (en) | A kind of control method of plate property optimization | |
| CN115169453A (en) | Hot continuous rolling width prediction method based on density clustering and depth residual error network | |
| CN116484745A (en) | Design method of hot-rolled low-alloy steel for physical metallurgy guidance industry big data mining | |
| CN115034437A (en) | Hot rolled plate convexity prediction method based on improved XGboost | |
| CN103632014B (en) | A kind of technological parameter Forecasting Methodology of hot steel rolling production production procedure | |
| CN112418522B (en) | Industrial heating furnace steel temperature prediction method based on three-branch integrated prediction model | |
| CN112598026B (en) | Continuous annealing production process fault diagnosis method based on unbalanced data deep learning | |
| CN116882776A (en) | Method and device for setting energy-saving strategy in steel production, electronic equipment and storage medium | |
| Dong et al. | Just-in-time learning-based soft sensor for mechanical properties of strip steel via multi-block weighted semisupervised models | |
| CN115456264A (en) | Method for predicting end point carbon content and end point temperature of small and medium-sized converters | |
| Wang et al. | A deep learning model for the accurate prediction of the microstructure performance of hot rolled steel | |
| CN117807424B (en) | Industrial big data driven wide and thick steel plate quality dynamic on-line identification method and device | |
| CN117265214B (en) | Optimized control method and system for special steel production | |
| CN115687899B (en) | Hybrid feature selection method based on high-dimensional spinning data | |
| CN116127345B (en) | Converter steelmaking process mode design method based on deep clustering generation countermeasure network | |
| Sehgal | Surface roughness optimization by respose surface methodology and particle swarm optimization |
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 |