CN113111092A - Silicon steel iron loss prediction method based on cold rolling full-process data - Google Patents

Silicon steel iron loss prediction method based on cold rolling full-process data Download PDF

Info

Publication number
CN113111092A
CN113111092A CN202110274559.3A CN202110274559A CN113111092A CN 113111092 A CN113111092 A CN 113111092A CN 202110274559 A CN202110274559 A CN 202110274559A CN 113111092 A CN113111092 A CN 113111092A
Authority
CN
China
Prior art keywords
data
iron loss
silicon steel
cold
data set
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110274559.3A
Other languages
Chinese (zh)
Other versions
CN113111092B (en
Inventor
王志军
贺立红
姚文达
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wisdri Engineering and Research Incorporation Ltd
Original Assignee
Wisdri Engineering and Research Incorporation Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Wisdri Engineering and Research Incorporation Ltd filed Critical Wisdri Engineering and Research Incorporation Ltd
Priority to CN202110274559.3A priority Critical patent/CN113111092B/en
Publication of CN113111092A publication Critical patent/CN113111092A/en
Application granted granted Critical
Publication of CN113111092B publication Critical patent/CN113111092B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24Querying
    • G06F16/245Query processing
    • G06F16/2458Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2462Approximate or statistical queries
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KGRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for recognising patterns
    • G06K9/62Methods or arrangements for pattern recognition using electronic means
    • G06K9/6217Design or setup of recognition systems and techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
    • G06K9/6218Clustering techniques
    • G06K9/622Non-hierarchical partitioning techniques
    • G06K9/6226Non-hierarchical partitioning techniques based on the modelling of probability density functions
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Computing arrangements based on biological models using neural network models
    • G06N3/04Architectures, e.g. interconnection topology
    • G06N3/0445Feedback networks, e.g. hopfield nets, associative networks
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Computing arrangements based on biological models using neural network models
    • G06N3/08Learning methods
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/04Forecasting or optimisation, e.g. linear programming, "travelling salesman problem" or "cutting stock problem"
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management, e.g. organising, planning, scheduling or allocating time, human or machine resources; Enterprise planning; Organisational models
    • G06Q10/063Operations research or analysis
    • G06Q10/0639Performance analysis
    • G06Q10/06395Quality analysis or management
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/04Manufacturing
    • YGENERAL 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
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02PCLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30Computing systems specially adapted for manufacturing

Abstract

The invention relates to a cold-rolled full-process data-based silicon steel iron loss prediction method, which is used for performing full-process material tracking on cold-rolled silicon steel, classifying the cold-rolled silicon steel based on a clustering algorithm, selecting a representative process as a training set of a neural network to predict the iron loss of the cold-rolled silicon steel, and exploring the influence of the processes of each cold-rolling unit on the iron loss of the cold-rolled silicon steel.

Description

Silicon steel iron loss prediction method based on cold rolling full-process data
Technical Field
The invention relates to the technical field of cold-rolled silicon steel, in particular to a silicon steel iron loss prediction method based on cold-rolled full-process data.
Background
The iron loss is the most important quality parameter of the cold-rolled silicon steel, and directly determines the service performance of the cold-rolled silicon steel. The current literature mainly considers steel-making components andthe temperature, hot rolling heating and cooling system have a large influence on the iron loss of silicon steel, for example, in the article of analysis and control method for poor iron loss of 50W1300 new steel, the analysis finds that the cause of poor iron loss of 50W1300 is mainly related to higher heating temperature of a hot rolling heating furnace, lower temperature of a contact part between a plate blank and a water beam of the heating furnace, uneven heating temperature of the heating furnace and the like. Also, for example, the influence of steelmaking process and component system on the iron loss of B50A1300 silicon steel discusses the RH oxygen blowing amount of refining, w [ Mn ]]2/w[Si]The ratio, the FeAl pre-deoxidation effect, the continuous casting molten steel calming time and the like on the iron loss of the B50A1300 silicon steel.
However, in actual production, each process of cold-rolled silicon steel production, including a normalizing pickling line, a rolling mill, a continuous annealing coating line and the like, has a great influence on the iron loss of the cold-rolled silicon steel, but at present, no mechanism model for the process of each cold-rolled line and the iron loss of the cold-rolled silicon steel is established, and even no relevant quantitative analysis data model is found.
Disclosure of Invention
The invention aims to provide a silicon steel iron loss prediction method based on cold rolling full-process data so as to find out the influence of cold rolling of each unit process on the iron loss of cold-rolled silicon steel.
The specific scheme is as follows:
a silicon steel iron loss prediction method based on cold rolling full-process data comprises the following steps:
s1, performing full-flow material tracking on cold-rolled silicon steel, wherein the full-flow material tracking at least comprises process data of a normalized pickling unit, a rolling mill and a continuous annealing coating unit, completing process data acquisition of each unit experienced by a steel coil and online iron loss data acquisition of the silicon steel, and obtaining length positions of each unit corresponding to each position of raw materials, process parameters and final iron loss values of the position through length mapping among the units, and each raw material position of each coil steel obtains a group of data d which comprises input data di (process data) and output data do (iron loss data);
s2, acquiring all group data D of each raw material roll to form a data set D; wherein the data set D comprises an input data set Di and an output data set Do;
carrying out cluster analysis on Di by using a DBSCAN clustering method to obtain N1 clusters;
performing cluster analysis on Do by using a DBSCAN clustering method to obtain N2 clusters;
s3, in step S2, the Di and Do clusters are completely and independently operated, and N1 × N2 is obtained as M cluster combinations, that is, each data D belongs to one of the M cluster combinations, and the current cnt group data is selected from the M clusters to form a new data set D; if the number of the data d included in the cluster is less than cnt, selecting all the data d;
s4, carrying out BP neural network training on the training data set, wherein input data of the BP neural network is an input data set Di, and output data of the BP neural network is an output data set Do; the BP neural network adopts a three-layer network, the number of neurons in a first layer is the number E of process data, the number of neurons in a hidden layer is E/2.5 and is an integer, and the number of neurons in an output layer is 1, wherein the number of neurons in the first layer is more than 6 and less than 15;
and S5, performing iron loss prediction on the new silicon steel coil by using the established neural network.
Further, the data selecting process in step S3 is as follows:
s31, calculating the Euclidean distance of each group of data relative to a zero point;
s32, arranging according to the ascending order of the distances;
and S33, and then selecting according to the distance at equal intervals.
Compared with the prior art, the silicon steel iron loss prediction method based on cold rolling full-process data provided by the invention has the following advantages:
1. an influence model of the iron loss of the cold-rolled silicon steel is established to find out the influence of the processes of each cold-rolling unit on the iron loss of the cold-rolled silicon steel.
2. The cold-rolled silicon steel iron loss is predicted based on cold-rolled full-process data, and the quality of the cold-rolled silicon steel can be effectively improved.
3. The influence of the processes of each cold rolling unit on the iron loss of the cold-rolled silicon steel is established by adopting a neural network, and the technical bottleneck of the mechanism model research process is overcome.
4. Based on clustering algorithm classification, and selecting a representative process as a training set of the neural network, the negative influence of the quality of a training sample on the training effect of the neural network is solved.
Drawings
Fig. 1 shows a flowchart of a silicon steel iron loss prediction method.
FIG. 2 is a diagram illustrating the data of the full-run material tracking process performed on cold-rolled silicon steel in one embodiment.
Detailed Description
To further illustrate the various embodiments, the invention provides the accompanying drawings. The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the embodiments. Those skilled in the art will appreciate still other possible embodiments and advantages of the present invention with reference to these figures. Elements in the figures are not drawn to scale and like reference numerals are generally used to indicate like elements.
The invention will now be further described with reference to the accompanying drawings and detailed description.
As shown in fig. 1, the present embodiment provides a method for predicting iron loss of silicon steel based on cold rolling full-process data, which includes the following steps:
s1: cold rolling full process material tracking
The method comprises the steps of carrying out full-flow material tracking on cold-rolled silicon steel, at least comprising process data of a normalizing pickling unit, a rolling mill and a continuous annealing coating unit, completing process data acquisition of each unit experienced by a steel coil, online iron loss data acquisition of the silicon steel, and obtaining length positions of each unit corresponding to each position of raw materials, process parameters and a final iron loss value of the position through length mapping between the units, so that each raw material position of each coil of steel can obtain a group of data d, and each raw material position of each coil of steel comprises input data di (process data) and output data do (iron loss data). Specific data for one embodiment is shown in fig. 2.
S2: DBSCAN cluster analysis
For each roll of stock, all their sets of data D are acquired, constituting a data set D. Where data set D includes an input data set Di (process data set) and an output data set Do (core loss data set).
And (5) carrying out cluster analysis on the Di by using a DBSCAN clustering method to obtain N1 clusters.
And (4) performing cluster analysis on the Do by using a DBSCAN clustering method to obtain N2 clusters.
DBSCAN (sensitivity-Based Spatial Clustering of Applications with Noise) is a relatively representative Density-Based Clustering algorithm. Given a set of points in a space, the algorithm can group nearby points (points with many neighboring points) and mark out-of-office points (points closest to it are also quite far away) that are located in low density areas.
DBSCAN requires two parameters: epsilon (eps) and the minimum number of points needed to form a high density region (minPts), starting with an arbitrary unvisited point, then explore the epsilon-neighborhood of this point, and if there are enough points in the epsilon-neighborhood, create a new cluster, otherwise this point is labeled as a murmurmur. This point may then be found in the epsilon neighborhood of other points, which may have enough points to be added to the cluster.
The DBSCAN has the following advantages compared with the K-means algorithm:
(1) DBSCAN does not require a pre-declared cluster number. Moreover, the DBSCAN can find any shape of cluster, even one cluster, which surrounds but is not connected to another cluster, and in addition, due to the minPts parameter, single-link effect (different clusters are connected by a point or a very young line and are regarded as one cluster) can be effectively avoided.
(2) DBSCAN can distinguish noise (local point).
(3) DBSCAN requires only two parameters and is almost insensitive to the order of the points within the database (the points of the edge between two clusters have a chance to be sorted into different clusters affected by the order, and the order of the other clusters will be affected by the order of the points).
(4) DBSCAN can be designed as a database structure that can be accessed in conjunction with acceleration ranges.
(5) Appropriate parameters may also be selected to obtain the best classification. If a point is in a dense region of a cluster, the points in its epsilon-neighborhood also belong to the cluster, and when these new points are added to the cluster, the points in its epsilon-neighborhood will also be added to the cluster if it(s) are also in the dense region. This process is repeated until no more points can be added, such that a densely connected cluster is completely found, and then an unaccessed point is explored to find a new cluster or noise.
The DBSCAN has the advantages that the DBSCAN can achieve a better effect in the silicon steel iron loss prediction process based on cold rolling full-flow data.
S3: training sample selection
The above two clustering runs are completely independent. This results in N1 × N2 being M cluster combinations, i.e., each data d belongs to one of the M cluster combinations. And selecting the existing cnt group data in the M kinds of clusters respectively to form a new data set D. And if the number of the data d included in the cluster is less than the cnt, selecting all the data d.
The data selection process is as follows:
s31: and calculating the Euclidean distance of each group of data relative to the zero point.
S32: arranged in ascending order of distance.
S33: and then picked at equal intervals according to the distance.
The main reason for selecting the training samples is that the iron loss produced by the same process parameters fluctuates, if the data are put into a neural network training set, the training results have larger deviation, and therefore, after clustering is adopted, the representative samples are selected, so that the problem can be solved.
S4: BP neural network based training
And carrying out BP neural network training on the training data set. Obviously, the input data of the BP neural network is an input data set Di, and the output data is an output data set Do. The BP neural network adopts a three-layer network, the number of neurons in the first layer is the number E of process data, the number of neurons in the hidden layer is E/2.5, and integers are taken to ensure that the number is more than 6 and less than 15. The number of neurons in the output layer is 1.
S5: prediction of iron loss in silicon steel
The iron loss prediction can be carried out on the new silicon steel coil by using the established neural network. The process parameters can be reversely controlled based on the neural network, so that the iron loss is lower or the iron loss consistency in the full-length direction of the strip steel is better.
While the invention has been particularly shown and described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (2)

1. A silicon steel iron loss prediction method based on cold rolling full-process data is characterized by comprising the following steps:
s1, performing full-flow material tracking on cold-rolled silicon steel, wherein the full-flow material tracking at least comprises process data of a normalized pickling unit, a rolling mill and a continuous annealing coating unit, completing process data acquisition of each unit experienced by a steel coil and online iron loss data acquisition of the silicon steel, and obtaining length positions of each unit corresponding to each position of raw materials, process parameters and final iron loss values of the position through length mapping among the units, and each raw material position of each coil steel obtains a group of data d which comprises input data di (process data) and output data do (iron loss data);
s2, acquiring all group data D of each raw material roll to form a data set D; wherein the data set D comprises an input data set Di and an output data set Do;
carrying out cluster analysis on Di by using a DBSCAN clustering method to obtain N1 clusters;
performing cluster analysis on Do by using a DBSCAN clustering method to obtain N2 clusters;
s3, in step S2, the Di and Do clusters are completely and independently operated, and N1 × N2 is obtained as M cluster combinations, that is, each data D belongs to one of the M cluster combinations, and the current cnt group data is selected from the M clusters to form a new data set D; if the number of the data d included in the cluster is less than cnt, selecting all the data d;
s4, carrying out BP neural network training on the training data set, wherein input data of the BP neural network is an input data set Di, and output data of the BP neural network is an output data set Do; the BP neural network adopts a three-layer network, the number of neurons in a first layer is the number E of process data, the number of neurons in a hidden layer is E/2.5 and is an integer, and the number of neurons in an output layer is 1, wherein the number of neurons in the first layer is more than 6 and less than 15;
and S5, performing iron loss prediction on the new silicon steel coil by using the established neural network.
2. The silicon steel iron loss prediction method of claim 1, wherein the data selection process in step S3 is as follows:
s31, calculating the Euclidean distance of each group of data relative to a zero point;
s32, arranging according to the ascending order of the distances;
and S33, and then selecting according to the distance at equal intervals.
CN202110274559.3A 2021-03-15 2021-03-15 Silicon steel iron loss prediction method based on cold rolling full-process data Active CN113111092B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110274559.3A CN113111092B (en) 2021-03-15 2021-03-15 Silicon steel iron loss prediction method based on cold rolling full-process data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110274559.3A CN113111092B (en) 2021-03-15 2021-03-15 Silicon steel iron loss prediction method based on cold rolling full-process data

Publications (2)

Publication Number Publication Date
CN113111092A true CN113111092A (en) 2021-07-13
CN113111092B CN113111092B (en) 2022-06-24

Family

ID=76711438

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110274559.3A Active CN113111092B (en) 2021-03-15 2021-03-15 Silicon steel iron loss prediction method based on cold rolling full-process data

Country Status (1)

Country Link
CN (1) CN113111092B (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040088139A1 (en) * 2002-08-16 2004-05-06 Blazek Kenneth E. Method of predicting motor performance based on CRML material properties
JP2010033536A (en) * 2007-12-20 2010-02-12 Nippon Steel Corp Method, device for predicting product material value, method, program for determining handling condition, and computer readable recording medium
CN104998913A (en) * 2015-06-29 2015-10-28 东北大学 Prediction method of motor power in cold rolling process
CN108021938A (en) * 2017-11-29 2018-05-11 中冶南方工程技术有限公司 A kind of Cold-strip Steel Surface defect online detection method and detecting system
US20190010572A1 (en) * 2015-12-24 2019-01-10 Posco Method for manufacturing grain-oriented electrical steel sheet
CN109272176A (en) * 2018-12-10 2019-01-25 贵州电网有限责任公司 Calculation method is predicted to platform area line loss per unit using K-means clustering algorithm
US20200089826A1 (en) * 2018-09-14 2020-03-19 Northwestern University Integrated process-structure-property modeling frameworks and methods for design optimization and/or performance prediction of material systems and applications of same
CN111652355A (en) * 2020-06-02 2020-09-11 中南大学 Method and device for predicting silicon content of blast furnace molten iron based on LSTM and DNN
CN112037209A (en) * 2020-09-02 2020-12-04 上海大学 Steel plate roller wear loss prediction method and system

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040088139A1 (en) * 2002-08-16 2004-05-06 Blazek Kenneth E. Method of predicting motor performance based on CRML material properties
JP2010033536A (en) * 2007-12-20 2010-02-12 Nippon Steel Corp Method, device for predicting product material value, method, program for determining handling condition, and computer readable recording medium
CN104998913A (en) * 2015-06-29 2015-10-28 东北大学 Prediction method of motor power in cold rolling process
US20190010572A1 (en) * 2015-12-24 2019-01-10 Posco Method for manufacturing grain-oriented electrical steel sheet
CN108021938A (en) * 2017-11-29 2018-05-11 中冶南方工程技术有限公司 A kind of Cold-strip Steel Surface defect online detection method and detecting system
US20200089826A1 (en) * 2018-09-14 2020-03-19 Northwestern University Integrated process-structure-property modeling frameworks and methods for design optimization and/or performance prediction of material systems and applications of same
CN109272176A (en) * 2018-12-10 2019-01-25 贵州电网有限责任公司 Calculation method is predicted to platform area line loss per unit using K-means clustering algorithm
CN111652355A (en) * 2020-06-02 2020-09-11 中南大学 Method and device for predicting silicon content of blast furnace molten iron based on LSTM and DNN
CN112037209A (en) * 2020-09-02 2020-12-04 上海大学 Steel plate roller wear loss prediction method and system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
王昕等: "神经网络在配网电能损耗预测中的应用研究", 《软件》 *

Also Published As

Publication number Publication date
CN113111092B (en) 2022-06-24

Similar Documents

Publication Publication Date Title
Kao et al. Optimization of electrochemical polishing of stainless steel by grey relational analysis
US6546310B1 (en) Process and device for controlling a metallurgical plant
CN112036081B (en) Method for determining addition amount of silicon-manganese alloy in converter tapping based on yield prediction
CN114329940A (en) Continuous casting billet quality prediction method based on extreme learning machine
CN113111092B (en) Silicon steel iron loss prediction method based on cold rolling full-process data
Sha et al. Refinement of coarse grained austenite in Nb–V–Ti microalloyed steel during roughing rolling
CN111798023A (en) Method for predicting comprehensive coke ratio in steelmaking sintering production
Chen et al. Recent Advances on Drawing Technology of Ultra-Fine Steel Tire Cord and Steel Saw Wire
Wu et al. Optimal design of hot rolling process for C-Mn steel by combining industrial data-driven model and multi-objective optimization algorithm
CN102277477A (en) Establishment method for prediction model of tensile strength of cold rolled strip
Jansto The integration of process and product metallurgy in niobium bearing steels
CN112149272A (en) Cold-rolled steel strip mechanical property prediction model based on multiple linear regression analysis
He et al. Smart-shifting strategy of work rolls for downstream stands in hot rolling
Chen et al. KDPAG expert system applied to materials design and manufacture
EP3916651A1 (en) Design assistance method for metal material, prediction model generation method, metal material manufacturing method, and design assistance device
Wang et al. An improved locally weighted regression for a converter re-vanadium prediction modeling
D’Arcangelo et al. The impact of the European carbon market on firm productivity: Evidence from Italian manufacturing firms
CN112966870A (en) Cold-rolled silicon steel quality defect prediction method based on neural network
CN112845617B (en) Plate type control method and device for hot-rolled strip steel
Dong et al. Just-in-Time Learning-Based Soft Sensor for Mechanical Properties of Strip Steel via Multi-Block Weighted Semisupervised Models
Duarte et al. Heat-loss cycle prediction in steelmaking plants through artificial neural network
Pettersson et al. A genetic algorithm evolving charging programs in the ironmaking blast furnace
CN113255102B (en) Method and device for predicting carbon content and temperature of molten steel at converter end point
Toktassynova et al. Application of grey system theory to phosphorite sinter process: From modeling to control
CN108364095B (en) Molten steel quality diagnosis method in steelmaking production process based on data mining

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant