CN111254243A - Method and system for intelligently determining iron notch blocking time in blast furnace tapping process - Google Patents
Method and system for intelligently determining iron notch blocking time in blast furnace tapping process Download PDFInfo
- Publication number
- CN111254243A CN111254243A CN202010165654.5A CN202010165654A CN111254243A CN 111254243 A CN111254243 A CN 111254243A CN 202010165654 A CN202010165654 A CN 202010165654A CN 111254243 A CN111254243 A CN 111254243A
- Authority
- CN
- China
- Prior art keywords
- operation mode
- projection
- blast furnace
- time
- regression model
- 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
Links
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 title claims abstract description 238
- 229910052742 iron Inorganic materials 0.000 title claims abstract description 119
- 238000000034 method Methods 0.000 title claims abstract description 101
- 230000000903 blocking effect Effects 0.000 title claims abstract description 74
- 230000008569 process Effects 0.000 title claims abstract description 59
- 238000010079 rubber tapping Methods 0.000 title claims abstract description 42
- 238000003723 Smelting Methods 0.000 claims abstract description 16
- 239000013598 vector Substances 0.000 claims description 38
- 230000006870 function Effects 0.000 claims description 35
- 239000011159 matrix material Substances 0.000 claims description 22
- 238000000513 principal component analysis Methods 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 5
- 230000001419 dependent effect Effects 0.000 claims description 5
- 238000013210 evaluation model Methods 0.000 claims description 5
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 claims 1
- 229910052710 silicon Inorganic materials 0.000 claims 1
- 239000010703 silicon Substances 0.000 claims 1
- 238000004519 manufacturing process Methods 0.000 description 19
- 238000001514 detection method Methods 0.000 description 13
- 239000002893 slag Substances 0.000 description 9
- 229910000831 Steel Inorganic materials 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 6
- 239000010959 steel Substances 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000005457 optimization Methods 0.000 description 5
- 238000005259 measurement Methods 0.000 description 4
- 230000001186 cumulative effect Effects 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 238000000691 measurement method Methods 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 239000000126 substance Substances 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000006073 displacement reaction Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000003034 coal gas Substances 0.000 description 1
- 238000009749 continuous casting Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000155 melt Substances 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 230000035699 permeability Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000004584 weight gain Effects 0.000 description 1
- 235000019786 weight gain Nutrition 0.000 description 1
Images
Classifications
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21B—MANUFACTURE OF IRON OR STEEL
- C21B7/00—Blast furnaces
- C21B7/12—Opening or sealing the tap holes
-
- 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
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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"
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Strategic Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- Chemical & Material Sciences (AREA)
- Human Resources & Organizations (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Manufacturing & Machinery (AREA)
- Game Theory and Decision Science (AREA)
- Materials Engineering (AREA)
- Metallurgy (AREA)
- Organic Chemistry (AREA)
- Manufacture Of Iron (AREA)
Abstract
The invention discloses a method and a system for intelligently determining the iron notch blocking time in the blast furnace tapping process, which are characterized in that a good operation mode library for representing the condition parameters, the operation parameters and the iron notch blocking time of the blast furnace smelting process is established, whether an optimal operation mode which corresponds to the current working condition and meets the preset similarity condition can be obtained by matching in the good operation mode library is judged, if yes, the iron notch blocking time of the blast furnace is obtained according to the optimal operation mode, otherwise, a projection pursuit regression model is established, and the iron notch blocking time is predicted based on the projection pursuit regression model.
Description
Technical Field
The invention mainly relates to the technical field of blast furnace taphole plugging time determination, in particular to an intelligent method and system for determining the taphole plugging time in the tapping process of a blast furnace.
Background
In the steel industry, the stability of blast furnace production is of great importance to the whole production process of steel enterprises. The process energy consumption and the production cost of blast furnace iron-making production account for more than 70% of the production of iron and steel enterprises, however, various multiphase substances in the blast furnace coexist and interact, and a plurality of physical and chemical phenomena occur simultaneously, which is considered as one of the most complex metallurgical reactors in the chemical field, and the pressure change in the blast furnace in the operation of a black box is difficult to be ascertained by the prior art means. The detection of the flow velocity of the molten iron at the tapping hole of the blast furnace can represent the pressure in the blast furnace, and can reflect the proportional relation between the produced metal and slag, so that abnormal working conditions can be found and eliminated in time, the air permeability of the blast furnace is improved, and the stable and smooth production of the blast furnace is ensured. Therefore, the detection of the flow velocity of the molten iron at the outlet of the blast furnace is particularly important for the energy conservation and emission reduction, quality improvement and efficiency improvement of the blast furnace production.
The detection object is molten iron with high temperature and high light, and the detection site has inevitable vibration and a large amount of dust and other strong interference factors which are not uniformly distributed, so that great challenge is brought to the detection. At present, the method for detecting the flow velocity of high-temperature molten iron is mainly a non-contact measurement method, and the non-contact measurement method comprises the following steps: image methods and numerical simulation methods.
The method comprises the steps of attaching a cross wire label to a tank body of a torpedo tank by an image method, carrying out rough positioning by a characteristic matching method through image processing of cross wires, realizing accurate positioning by applying angular point detection, obtaining the downward pressing moving distance of a spring of the torpedo tank car, calculating the mass of molten iron flowing into the torpedo tank car, and calculating the flow of real-time molten iron flow field workers to molten iron at a taphole. However, the measurement method has large time lag and inaccuracy, and is difficult to provide effective guiding significance for stable and efficient production of the blast furnace.
The numerical simulation method calculates the flow velocity value of molten iron at each stage of blast furnace tapping by establishing a mechanism model of molten iron outflow from a blast furnace tap hole and utilizing a numerical simulation method, but the method needs a good assumed environment and unknown parameter values, and cannot obtain an accurate flow velocity value.
The detection of the flow velocity of the high-temperature and high-speed molten iron is extremely challenging, related patents are few, and the defects of the existing patents are large.
The invention discloses a system and a method for measuring the flow velocity of high-temperature molten steel of a continuous casting crystallizer, and the system is characterized in that a bearing is fixed on a fixing system through a fixing shaft, a spring and a measuring rod are respectively arranged at the upper symmetrical position and the lower symmetrical position of the bearing, the spring is arranged on a T-shaped fixing system, a bearing sleeve is arranged on the bearing, the bearing sleeve and an angle displacement sensor are connected through a coupler, the angle displacement sensor is powered by a power supply, the real-time deflection angle of the measuring rod in the flowing molten steel is recorded, and the real-time deflection angle is transmitted to a data acquisition and analysis system through a data line, so that angle data are converted into the flow velocity value of the molten steel. However, the system needs to be re-registered according to different detection objects of the system, the system is preheated to 1200-1400 ℃ before use, the detection range is small, fluid with overlarge flow rate cannot be detected, the system cannot directly detect the next object after detection is finished, and the repeatability in use is limited.
The invention discloses a method for detecting the flow of blast furnace slag based on a fuzzy model, which is characterized in that a fuzzy inference model of the flow of the blast furnace slag is established, a fuzzy membership function about the height of the slag surface at the ith moment is set by combining the influence characteristics of the height of the slag surface at the ith moment on the flow of the blast furnace slag, a blast furnace slag flow calculation model is established by utilizing the fuzzy inference model and the fuzzy membership function, and the blast furnace slag flow calculation model is used for carrying out online detection on the total flow of the blast furnace slag in real time. However, the initial value in the patent is obtained from the experience knowledge of manual operation by a craft worker, the influence of human factors is large, the subjectivity is strong, and the design flow is an open loop, so that the accuracy of a long-term operation result cannot be ensured.
Disclosure of Invention
The method and the system for intelligently determining the time for blocking the iron notch in the blast furnace tapping process solve the technical problem that the time for blocking the iron notch of the blast furnace cannot be accurately judged in the prior art.
In order to solve the technical problem, the intelligent determination method for the iron notch blocking time in the blast furnace tapping process provided by the invention comprises the following steps:
establishing an excellent operation mode library representing condition parameters, operation parameters and iron notch plugging time of the blast furnace smelting process;
and judging whether an optimal operation mode which corresponds to the current working condition and meets the preset similarity condition can be obtained by matching in the excellent operation mode library, if so, obtaining the time of the iron notch blocking of the blast furnace according to the optimal operation mode, otherwise, establishing a projection pursuit regression model, and predicting the time of the iron notch blocking based on the projection pursuit regression model.
Further, whether an optimal operation mode which corresponds to the current working condition and meets the preset similarity condition can be obtained in the excellent operation mode library in a matching mode or not is judged, if yes, the blast furnace taphole plugging time is obtained according to the optimal operation mode, otherwise, a projection pursuit regression model is established, and the taphole plugging time is predicted based on the projection pursuit regression model, wherein the prediction comprises the following steps:
performing PCA analysis on the condition component and decision component matrix of the excellent operation mode in the excellent operation mode library to obtain the number of the principal elements and the attribute weight coefficient corresponding to the number of the principal elements;
constructing an operation mode reference object corresponding to the current working condition and an operation mode matching object corresponding to the excellent operation mode in the pivot element space;
removing good operation modes with the similarity of the current working condition smaller than a preset similarity lower limit value according to the Euclidean distance between the operation mode reference object and the operation mode matching object to obtain an initial similar operation mode set;
and on the basis of the initial similar operation mode set, judging whether an optimal operation mode which corresponds to the current working condition and meets the preset similarity condition can be obtained by matching in the initial similar operation mode set by using a Cauchy inequality, if so, obtaining the time of the iron blocking port of the blast furnace according to the initial similar operation mode set, otherwise, establishing a projection pursuit regression model, and predicting the time of the iron blocking port on the basis of the projection pursuit regression model.
Further, the concrete formula of the cauchy inequality is as follows:
wherein,
the object is referred to for the operation mode,
for matching objects to the operation pattern, ωjThe corresponding weight of the jth pivot of the operation mode matching object, l is the pivot number, α is the preset similarity threshold,
is composed of
And
the product of the Hadamard sum of (C),
is composed of
And
the product of the Hadamard sum of (C),
presentation pair
The norm is calculated,
presentation pair
And (5) calculating a norm.
Further, establishing a projection pursuit regression model, and predicting the iron notch plugging time based on the projection pursuit regression model comprises:
determining a projection index function of a projection pursuit regression model;
projecting a projection pursuit regression model input vector to a one-dimensional space according to a projection direction in the projection index function to obtain a projection value, wherein the projection pursuit regression model input vector is a data matrix extracted according to a condition parameter vector and an operation parameter vector of a candidate operation mode;
taking the projection value as an independent variable and the output vector of the projection pursuit regression model as a dependent variable, obtaining the Hermite polynomial projection pursuit regression model by orthogonal Hermite polynomial fitting, and taking the output vector of the projection pursuit regression model as the iron notch plugging time of the candidate operation mode;
optimizing projection direction parameters and polynomial parameters in a Hermite polynomial projection pursuit regression model by adopting a projection pursuit learning network based on a projection index function to obtain optimal projection directions and optimal polynomial parameters;
and acquiring the prediction time of the iron blocking port by using a Hermite polynomial projection pursuit regression model based on the optimal projection direction and the optimal polynomial parameters.
Further, based on the optimal projection direction and the optimal polynomial parameters, after the obtaining of the prediction time of the iron blocking hole by using the Hermite polynomial projection pursuit regression model, the method further comprises the following steps:
and calculating the error between the predicted time of the iron notch blocking and the actual time of the iron notch blocking, if the error is larger than the preset error, increasing the number of ridge functions, and obtaining a Hermite polynomial projection pursuit regression model by reusing orthogonal Hermite polynomial fitting, and repeating the steps until the error is smaller than the preset error.
Further, the determining of the projection index function of the projection pursuit regression model is specifically a projection index function of the projection pursuit regression model determined according to the actual value, the predicted value and the projection direction of the iron blocking port time, and the projection index function is specifically:
wherein n is the number of samples, akFor projection, finding the projection direction of regression model, s is projection index function value, yiThe actual tap hole plugging time corresponding to the ith sample,
and predicting time for the iron blocking opening corresponding to the ith sample.
Further, establishing a good operation mode library for representing condition parameters, operation parameters and iron notch plugging time of the blast furnace smelting process comprises the following steps:
establishing an operation mode according to the corresponding relation of the condition parameters, the operation parameters and the iron blocking time in the blast furnace smelting process;
obtaining an operation mode space according to a space formed by operation modes corresponding to all possible conditions in the actual blast furnace tapping process;
establishing a working condition evaluation model, and evaluating condition parameters of the same operation mode and different tap hole plugging time corresponding to the operation parameters according to the working condition evaluation model so as to obtain excellent operation modes corresponding to each group of operation modes;
based on the operation mode space, a good operation mode library is obtained according to a set of good operation modes under different condition parameters.
The invention provides an intelligent determination system for the iron notch blocking time in the blast furnace tapping process, which comprises the following steps:
the intelligent determination method comprises the following steps of a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the steps of the intelligent determination method for the iron notch blocking time in the blast furnace tapping process are realized when the processor executes the computer program.
Compared with the prior art, the invention has the advantages that:
according to the method and the system for intelligently determining the iron notch blocking time in the blast furnace tapping process, the condition parameters representing the smelting process of the blast furnace, the excellent operation mode library of the operation parameters and the iron notch blocking time are established, whether the excellent operation mode library can be matched to obtain the optimal operation mode which corresponds to the current working condition and meets the preset similarity condition or not is judged, if yes, the iron notch blocking time of the blast furnace is obtained according to the optimal operation mode, otherwise, a projection pursuit regression model is established, and the iron notch blocking time is predicted based on the projection pursuit regression model.
Drawings
FIG. 1 is a flowchart of a method for intelligently determining the time of a taphole plugging in a tapping process of a blast furnace according to a first embodiment of the invention;
FIG. 2 is a flowchart of a method for intelligently determining the taphole plugging time in the tapping process of a blast furnace according to a second embodiment of the invention;
FIG. 3 is a flow chart of PCA analysis according to a second embodiment of the present invention;
FIG. 4 is a block diagram of a hierarchical matching strategy of operation modes according to a second embodiment of the present invention;
fig. 5 is a schematic diagram of a projection pursuit learning network according to a second embodiment of the present invention;
FIG. 6 is a block diagram showing the structure of an intelligent determination system for the taphole plugging time in the tapping process of the blast furnace according to the embodiment of the present invention;
fig. 7 is a block diagram of a system for intelligently determining the taphole plugging time in the tapping process of the blast furnace according to a third embodiment of the present invention.
Description of reference numerals:
10, a memory; 20. a processor; u1: a data access unit; u2: an operation mode discovery unit; u3: an operation pattern matching unit; u4: a projection pursuit regression unit; u5: and an output unit.
Detailed Description
In order to facilitate an understanding of the invention, the invention will be described more fully and in detail below with reference to the accompanying drawings and preferred embodiments, but the scope of the invention is not limited to the specific embodiments below.
The embodiments of the invention will be described in detail below with reference to the drawings, but the invention can be implemented in many different ways as defined and covered by the claims.
Example one
Referring to fig. 1, a method for intelligently determining a taphole plugging time in a blast furnace tapping process according to an embodiment of the present invention includes:
step S101, establishing an excellent operation mode library representing condition parameters, operation parameters and iron notch plugging time of a blast furnace smelting process;
and S102, judging whether an optimal operation mode which corresponds to the current working condition and meets the preset similarity condition can be obtained in the excellent operation mode library in a matching mode, if so, obtaining the time of the blast furnace iron notch blocking according to the optimal operation mode, otherwise, establishing a projection pursuit regression model, and predicting the time of the iron notch blocking based on the projection pursuit regression model.
According to the method for intelligently determining the iron notch blocking time in the blast furnace tapping process, provided by the embodiment of the invention, the condition parameters representing the smelting process of the blast furnace, the excellent operation mode library of the operation parameters and the iron notch blocking time are established, whether the excellent operation mode library can be matched to obtain the optimal operation mode which corresponds to the current working condition and meets the preset similarity condition or not is judged, if yes, the iron notch blocking time of the blast furnace is obtained according to the optimal operation mode, otherwise, a projection pursuit regression model is established, and the iron notch blocking time is predicted based on the projection pursuit regression model.
Specifically, the embodiment first transmits a large amount of production data to the database server for storage; establishing an excellent operation mode library for representing working condition parameters, operation parameters and iron notch plugging time of the blast furnace smelting process; searching an optimal operation mode set suitable for the current working condition from the excellent operation mode library; if the similarity of all the operation modes in the excellent operation mode library and the current working condition does not meet the requirement of the preset similarity condition, establishing a projection pursuit regression model, and predicting the iron notch plugging time based on the projection pursuit regression model; and finally, outputting the optimal operation mode under the current working condition to obtain the optimal iron notch plugging time.
Example two
Referring to fig. 2, the method for intelligently determining the time for plugging the taphole in the tapping process of the blast furnace provided by the second embodiment of the invention comprises the following steps:
step S201, establishing an excellent operation mode library representing condition parameters, operation parameters and iron notch plugging time of the blast furnace smelting process.
Specifically, the present embodiment first uploads mass production data generated by various detection instruments and systems operating in a blast furnace site to a database server via an intermediary such as a control system and an internal network, and stores the mass production data for a long time. The data is effective reflection of production operation rules; the production data stored in the field devices is then read through the intranet and the associated data in the database is read remotely through the connection server.
In the large-scale blast furnace ironmaking production process, the relation among vectors formed by condition parameters, operation parameters and iron blocking time presents a 'many-to-one' mapping relation. Theoretically, when the vectors formed by the condition parameters and the operation parameters are the same, the same tapping time is required when the vectors are applied to the tapping process.
A vector composed of an operation mode unit is defined as an operation mode P, namely, a condition parameter which is a working condition reflecting the operation information of the smelting tapping process of the blast furnace, an operation parameter which is a relevant operation possibly performed in the tapping process, a group of condition parameters and the corresponding operation parameter form an operation mode
P＝[IT,UT]T(1)
Wherein, P ═ IT,UT]TRepresents a pair [ IT,UT]Find transpose, I ═ I1,i2…i11]TThe condition parameter vector respectively represents the hearth pressure, the molten iron flow velocity, the stream diameter, the weight gain rate of the molten iron tank, the average diameter of the iron notch, the gravity acceleration constant, the liquid level difference of the center line of the iron notch outlet, the depth of the iron notch, the pressure difference in the iron notch, the friction coefficient of the iron notch wall and the average density of the melt, and U is [ U ═ U [ [ U ]1,u2,u3]TThe operation parameter vector represents the operations of furnace top pressure reduction, wind reduction and iron penetration, the output of the mode is the iron blocking time, and O is equal to OT。
A full library of operating modes may be created by grouping all operating modes. The operation parameter vector U in the operation mode is obtained by making a decision from the condition parameter vector I through a certain relation, so that the condition parameter vector I is called as a mode condition component, the operation parameter vector is called as a mode decision component, a complete operation mode space is mapped based on the data, and P is setj(j ═ 1,2, …, k) characterizes any one of the operating modes of the blast furnace tapping process, and the space consisting of the operating modes corresponding to all possible conditions during the actual tapping of the blast furnace is called the operating mode space.
The excellent operation mode library module of the embodiment comprehensively considers indexes such as whether theoretical iron yield calculated according to batches is consistent with actual iron yield, whether a large amount of coal gas is sprayed out of an iron notch, manual experience judgment and the like, evaluates different iron notch plugging times corresponding to the same mode condition component and operation parameters by establishing a working condition evaluation model, and evaluates an excellent operation mode with the best result corresponding to each group of operation modes. After the operation mode library is input, data preprocessing is firstly carried out, then operation modes are classified, different excellent operation modes can be obtained according to the same mode evaluation method under different input conditions, and in an operation mode space, a set formed by the excellent operation modes under different mode conditions is called as an excellent operation mode library.
Step S202, PCA analysis is carried out on the condition component and decision component matrixes of the excellent operation modes in the excellent operation mode library, the number of the principal elements and the attribute weight coefficient corresponding to the number of the principal elements are obtained, and an operation mode reference object corresponding to the current working condition and an operation mode matching object corresponding to the excellent operation mode are constructed in the principal element space.
In this embodiment, PCA (Principal component Analysis) is performed on the condition component and the decision component matrix in the good operation mode library, a pivot attribute weight is set, and a reference object and a matching object of the operation mode are constructed in a pivot space. Referring to fig. 3, the PCA conversion of the present embodiment includes:
in step S301, a data sample matrix is extracted.
Taking a data matrix X consisting of n data samples of m dimensions as an example, the following details are provided:
assume that the present embodiment extracts a data sample matrix X ═ χij]n×mWherein n represents the number of data samples, and m is the number of variables; the data is normalized, so that the processed data matrix can still be decomposed into the sum of a plurality of vector outer products, namely: x is TPT＝t1P1 T+t2P2 T+…+tmPm TWherein T ∈ Rn×m，ti∈Rn(i-1, 2, …, m) is the principal element, pi∈RmIs a load vector.
In step S302, a covariance matrix is calculated.
Specifically, the specific formula of the covariance matrix for calculating the data sample matrix X in this embodiment is as follows:
step S303, orthogonal decomposition.
Specifically, the covariance matrix is subjected to orthogonal decomposition to obtain
Wherein D ═ diag (λ)1，λ2，…λm) Is a characteristic root matrix of the covariance matrix, and1≥λ2≥…≥λm，P＝[p1,p2,…,pm]is a load matrix.
And step S304, determining the number of the pivot elements.
Specifically, in the present embodiment, the eigenvalue variance cumulative contribution rate method is used to determine the number of principal elements, and the cumulative contribution rate of the first l principal elements to the total variance of the sample may be represented as:
and when the cumulative contribution rate of the principal elements exceeds a set threshold, considering that the corresponding principal elements are the number of the principal elements needing to be reserved.
In step S305, a pivot score matrix is calculated.
Specifically, the calculating the pivot score matrix in this embodiment specifically includes:
T＝XP (6)
principal component tiCan be expressed as:
ti＝Xpi,i＝1,2,….l (7)
step S306, determining a principal component weight coefficient.
Specifically, in PCA, the degrees of representation of data information amounts by different principal elements are different, and the eigenvalue corresponding to each principal element represents how much information amount, so that the contribution rate of the eigenvalue corresponding to each principal element can be used as the attribute weight coefficient of the corresponding principal element, and the weight coefficient of the ith (i ═ 1.2.…, l) principal element is:
and satisfy
Step S203, removing the excellent operation mode with the similarity of the current working condition being smaller than the lower limit value of the preset similarity according to the Euclidean distance between the operation mode reference object and the operation mode matching object to obtain an initial similar operation mode set, judging whether the optimal operation mode corresponding to the current working condition and meeting the preset similarity condition can be obtained in the initial similar operation mode set or not by using the Cauchy inequality on the basis of the initial similar operation mode set, if so, obtaining the blast furnace taphole plugging time according to the initial similar operation mode set, and if not, executing the step S204.
Reference object of the present embodiment
And between the matching objects
The degree of similarity of (c) can be calculated using the following formula,
in the formula, l is the number of principal elements; omegajRepresents the corresponding weight of the jth principal element, satisfies
Reference object
And matching the object
The more similar, skThe smaller the value, s when the reference object is identical to the matching objectk＝0。
In the embodiment, the Cauchy inequality is introduced into the similarity measurement criterion, and the norm comparison between the reference object and the matching object replaces the comprehensive measurement among the attribute components, so that each matching object can make a decision without performing complex comprehensive operation, and the efficiency of searching the optimal operation mode which is adaptive to the current working condition from the similar operation mode set is improved.
Order to
Then
And vector
The hadamard product of (a) can be expressed as:
in the form of a norm, it is expressed as:
on the other hand, from the Cauchy inequality:
the following results were obtained:
the present embodiment performs the norm before the operation pattern matching
Can be obtained by analyzing by a PCA method by utilizing a good operation mode library. Therefore, each time of similarity calculation only needs one simple addition, subtraction, multiplication and comparison operation, and complexity of matching calculation is greatly simplified.
The criterion for successful matching of the operation modes in this embodiment is that the similarity between the operation modes is less than or equal to the set threshold α, i.e. when s iskAnd when the operation mode is less than or equal to α, the operation mode is considered to be successfully matched.
If there are a plurality of satisfies skOperating mode ≦ α, then s is selectedkAnd the minimum operation mode is the optimal operation mode, and the iron blocking time in the operation mode is extracted as matching output. If s is not satisfiedkAnd α, executing step S204.
In this embodiment, the euclidean distance is used as a primary judgment criterion for the operation mode matching, the operation modes with small similarity are quickly removed, and a similar operation mode set required by the secondary matching is extracted, so that the secondary matching process is only performed in the similar operation mode set, thereby improving the subsequent mode matching speed, and in addition, the cauchy inequality is introduced into the secondary matching, so that the similarity calculation amount is further simplified, specifically referring to fig. 4. Therefore, the embodiment adopts a grading quick matching strategy, and can quickly and accurately obtain the optimal operation mode under the current working condition.
And step S204, determining a projection index function of the projection pursuit regression model.
In particular, assume that the present embodiment has n sets of data samples yi(i-1, 2, …, n) respectively represents the actual value of the tap hole plugging time of the ith sample,
respectively representing the predicted values of the time of the iron blocking opening of the ith sample,
and (4) representing the prediction error of the iron notch blockage time of the ith sample. The sum of the projection direction vector lengths of the iron blocking port time prediction model is 1, so that the projection index function can be expressed as:
wherein n is the number of samples, akFor projection, finding the projection direction of regression model, s is projection index function value, yiThe actual tap hole plugging time corresponding to the ith sample,
and predicting time for the iron blocking opening corresponding to the ith sample.
Step S205, projecting the input vector of the projection pursuit regression model into a one-dimensional space according to the projection direction in the projection index function to obtain a projection value, wherein the input vector of the projection pursuit regression model is a data matrix extracted according to the condition parameter vector and the operation parameter vector of the candidate operation mode.
Specifically, the embodiment first extracts the corresponding condition parameter vector and operation parameter vector in the candidate operation mode library set, and forms the data matrix { x }i1,xi2,…,xi14The (i is 1,2, m) is used as a projection pursuit regression input vector, and the iron notch blocking time y in the blast furnace iron tapping process is usedi＝[yi1](i-1, 2, …, m) as the projection pursuit regression output vector. Wherein m is the number of samples, and constructing an independent variable data matrix Xn×pAnd dependent variable data matrix Yn×lCalculating Xn×pThe projected value of (a) is calculated,
and step S206, obtaining a Hermite polynomial projection pursuit regression model by using the projection value as an independent variable and the projection pursuit regression model output vector as a dependent variable and using orthogonal Hermite polynomial fitting, wherein the projection pursuit regression model output vector is used as the taphole plugging time of the candidate operation mode.
Specifically, the present embodiment extracts the input vector xi＝[xi1，xi2，…，xi14]By projecting a direction vector a ═ a1，a2，…，a14]TProjecting to a one-dimensional space to obtain a projection value zi(i＝1,2,…,m)
Then in the projection value ziAs an independent variable, with yi1(i-1, 2, …, m) is a dependent variable, and the scatter point (z) is corrected by an orthogonal Hermite polynomiali,yi1) Performing curve fitting, the Hermite polynomial projection pursuit regression model can be expressed as:
wherein m is the number of ridge functions, r is the number of polynomial orders, cklIs a polynomial coefficient, hkl(. cndot.) is a Hermite polynomial and can be calculated by:
wherein,
is a standard Gaussian function, an
hr(z) is a Hermite polynomial, whose recursion form can be expressed as: h is0(z)＝1；h1(z)＝2z；hr(z)＝2(zHr-1(z)-(r-1)Hr-2(z))。
And step S207, optimizing projection direction parameters and polynomial parameters in the Hermite polynomial projection pursuit regression model by adopting a projection pursuit learning network based on the projection index function to obtain the optimal projection direction and optimal polynomial parameters.
In the process of optimizing the established model parameters, a proper optimization method is selected for the projection direction akSum polynomial coefficient hkOptimizing and solving the best akAnd hkThe value of (c).
Specifically, in the model parameter training process, a projection index function is taken as an optimization target, a projection direction and polynomial coefficients in the iron notch plugging time prediction model are optimized by using a Projection Pursuit Learning Network (PPLN), and the number of ridge functions, the model prediction precision, the model overall performance and the model parameter updating speed are ensured to be coordinated with one another. A schematic diagram of a projection pursuit learning network is shown in fig. 5.
The feed forward approximation for PPLN is:
in the formula, zkIs a projection weight, hkIs a Hermite polynomial, akjFor the projection direction, these three sets of parameters minimize the error loss function by training the network.
The algorithm implementation steps of PPLN corresponding to the k hidden layer neurons are as follows:
(1) for the projection weight zkHermite polynomial hkIn the projection direction akjAssigning an initial value;
(2) the gauss-newton optimization algorithm is used to estimate:
wherein Δ is calculated by the following formula:
wherein
Is provided with
Then:
(3) known as akEstimating h from the smoothed linear best-match scatter plotk，
Orthonormalizes Hermite polynomials enable more accurate derivative calculations and smoother interpolation.
(4) Repeating the steps (2) and (3) for several times of iteration;
(5) using the latest hkAnd akEstimate zik；
(6) Repeating the steps (2) to (5) until the error is reached
Below a given threshold, the best projection direction and projection value for each sample are obtained.
And S208, acquiring the prediction time of the iron blocking port by using a Hermite polynomial projection pursuit regression model based on the optimal projection direction and the optimal polynomial parameters.
In the embodiment, projection direction optimization is performed by using a Projection Pursuit Learning Network (PPLN), and a fitting error e under an optimal value is calculated1Whether y-y' meets the requirements. If e1If the requirements are met, optimizing is finished, and model parameters are output.
If e1If the requirement is not met, the number of the ridge functions is increased, a new round of Hermite polynomial fitting is started, and the projection pursuit regression model can be expressed as
Where a represents the number of times the ridge function is operated on. Each additional ridge function needs to optimize and solve model parameters such as projection direction, polynomial coefficients and the like of the additional ridge function, so that the updating speed of the model parameters is reduced. The updating speed of the model parameters and the model prediction precision are mutually opposite, and the updating speed of the model parameters can be reduced by improving the model prediction precision.
Calculating the fitting error e each time1If not, use e1Instead of y, the next ridge function fitting is started. And when the requirement is met, stopping increasing the number of the ridge functions and outputting the final result of the model.
In order to obtain an operation mode set similar to the current working condition, realize judgment of the working condition and guide stokehole blocking of a stokehole worker, an optimal operation mode under the current working condition is searched by adopting a weighted Euclidean distance measurement and Cauchy inequality similarity measurement twice operation mode matching strategy, if one or more groups of operation modes meet the threshold requirement, the group of operation mode most similar to the current working condition is taken as the optimal operation mode, and therefore the iron blocking time under the current working condition is obtained.
If the optimal operation mode cannot be obtained by utilizing the two-stage matching strategy, a projection pursuit multivariate regression model is established to predict the iron notch plugging time by utilizing a candidate operation mode set similar to the current working condition. And when model parameters such as projection direction, polynomial coefficient, ridge function number and the like in the iron notch blocking time prediction model are solved, a Projection Pursuit Learning Network (PPLN) is adopted for parameter optimization, a prediction result is finally output, and the iron notch blocking time predicted according to the current working condition is output.
According to the method for intelligently determining the iron notch blocking time in the blast furnace tapping process, provided by the embodiment of the invention, the condition parameters representing the smelting process of the blast furnace, the excellent operation mode library of the operation parameters and the iron notch blocking time are established, whether the excellent operation mode library can be matched to obtain the optimal operation mode which corresponds to the current working condition and meets the preset similarity condition or not is judged, if yes, the iron notch blocking time of the blast furnace is obtained according to the optimal operation mode, otherwise, a projection pursuit regression model is established, and the iron notch blocking time is predicted based on the projection pursuit regression model.
The invention aims to design an intelligent method for determining the time of iron blocking during the tapping process of a blast furnace, which comprises the steps of establishing an operation mode library for inputting working condition parameters, operation parameters and output iron blocking time, obtaining an optimal operation mode under the current working condition by using a grading fast matching strategy, and if the optimal operation mode corresponding to the current working condition does not exist in the mode library, intelligently predicting the time of the iron blocking by using a projection pursuit model and a projection pursuit learning network according to the input of the current working condition to form a complete method for intelligently determining the time of the iron blocking during the tapping process of the blast furnace.
Referring to fig. 6, the system for intelligently determining the iron notch plugging time in the blast furnace tapping process provided by the invention comprises a memory 10, a processor 20 and a computer program which is stored on the memory 10 and can run on the processor 20, wherein the processor 20 executes the steps of the method for intelligently determining the iron notch plugging time in the blast furnace tapping process provided by the invention.
EXAMPLE III
Referring to fig. 7, the system for intelligently determining the tap hole blockage time in the blast furnace tapping process according to the third embodiment of the present invention is composed of a data access unit U1, an operation pattern discovery unit U2, an operation pattern matching unit U3, a projection pursuit regression unit U4, and an output unit U5.
As shown in fig. 7, the specific work flow is as follows: firstly, a large amount of production data is transmitted to a database server for storage through a database data access unit U1; then establishing a good operation mode library for representing working condition parameters, operation parameters and iron notch plugging time of the blast furnace smelting process by using an operation mode discovery unit U2; then the operation mode matching unit U3 searches an optimal operation mode set suitable for the current working condition from the excellent operation mode library; if the similarity between all the operation modes in the mode library and the current working condition does not meet the threshold requirement, a projection pursuit regression model is established through a projection pursuit regression unit U4, and the current working condition parameters and the operation parameters are input to predict the iron notch plugging time through a projection pursuit learning network; and finally, the output unit U5 outputs the optimal operation mode under the current working condition to obtain the optimal iron notch plugging time.
Optionally, the database data access unit U1 of the present embodiment includes a data storage module, a local data reading module, and a remote database reading module. The data storage unit uploads mass production data generated by various detection instruments and devices operating in a blast furnace field to a database server through the intermediary of a control system, an internal network and the like, and the mass production data is stored for a long time. The data is effective reflection of production operation rules; the local data reading module reads the production data stored in the field device through the intranet, and the remote database reading module remotely reads the related data in the database through the connection server.
According to the intelligent determination system for the iron notch blocking time in the blast furnace tapping process, the condition parameters representing the smelting process of the blast furnace, the excellent operation mode library of the operation parameters and the iron notch blocking time are established, whether the excellent operation mode library can be matched to obtain the optimal operation mode which corresponds to the current working condition and meets the preset similarity condition or not is judged, if yes, the iron notch blocking time of the blast furnace is obtained according to the optimal operation mode, otherwise, a projection pursuit regression model is established, and the iron notch blocking time is predicted based on the projection pursuit regression model.
The working principle and the process of the intelligent determination system for the iron notch blocking time in the blast furnace tapping process can refer to the working principle and the process of the intelligent determination method for the iron notch blocking time in the blast furnace tapping process.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes will occur to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. An intelligent determination method for the iron notch blocking time in the blast furnace tapping process is characterized by comprising the following steps:
establishing an excellent operation mode library representing condition parameters, operation parameters and iron notch plugging time of the blast furnace smelting process;
and judging whether an optimal operation mode which corresponds to the current working condition and meets the preset similarity condition can be obtained by matching in the excellent operation mode library, if so, obtaining the time of the iron notch blocking of the blast furnace according to the optimal operation mode, otherwise, establishing a projection pursuit regression model, and predicting the time of the iron notch blocking based on the projection pursuit regression model.
2. The method for intelligently determining the iron notch blocking time in the blast furnace tapping process according to claim 1, wherein the step of judging whether an optimal operation mode which corresponds to the current working condition and meets a preset similarity condition can be obtained in the excellent operation mode library in a matching manner, if so, obtaining the iron notch blocking time of the blast furnace according to the optimal operation mode, otherwise, establishing a projection pursuit regression model, and predicting the iron notch blocking time based on the projection pursuit regression model comprises the following steps:
performing PCA analysis on the condition component and decision component matrix of the excellent operation mode in the excellent operation mode library to obtain the number of principal elements and the attribute weight coefficient corresponding to the number of the principal elements;
constructing an operation mode reference object corresponding to the current working condition and an operation mode matching object corresponding to the excellent operation mode in a pivot space;
removing good operation modes with the similarity of the current working condition smaller than a preset similarity lower limit value according to the Euclidean distance between the operation mode reference object and the operation mode matching object to obtain an initial similar operation mode set;
and on the basis of the initial similar operation mode set, judging whether an optimal operation mode which corresponds to the current working condition and meets a preset similarity condition can be obtained by matching in the initial similar operation mode set or not by using a Cauchy inequality, if so, obtaining the time of blocking the iron notch of the blast furnace according to the initial similar operation mode set, otherwise, establishing a projection pursuit regression model, and predicting the time of blocking the iron notch on the basis of the projection pursuit regression model.
3. The intelligent determination method for the iron notch blocking time in the blast furnace tapping process according to claim 2, characterized in that the specific formula of the Cauchy inequality is as follows:
wherein,
the object is referred to for the operation mode,
for matching objects to the operation pattern, ωjThe corresponding weight of the jth pivot of the operation mode matching object, l is the pivot number, α is the preset similarity threshold,
is composed of
And
the product of the Hadamard sum of (C),
is composed of
And
the product of the Hadamard sum of (C),
presentation pair
The norm is calculated,
presentation pair
And (5) calculating a norm.
4. The intelligent determination method for the iron notch plugging time in the blast furnace tapping process according to any one of claims 1-3, wherein establishing a projection pursuit regression model, and predicting the iron notch plugging time based on the projection pursuit regression model comprises:
determining a projection index function of a projection pursuit regression model;
projecting a projection pursuit regression model input vector to a one-dimensional space according to a projection direction in the projection index function to obtain a projection value, wherein the projection pursuit regression model input vector is a data matrix extracted according to a condition parameter vector and an operation parameter vector of a candidate operation mode;
obtaining a Hermite polynomial projection pursuit regression model by using the projection value as an independent variable and a projection pursuit regression model output vector as a dependent variable and utilizing orthogonal Hermite polynomial fitting, wherein the projection pursuit regression model output vector is the iron notch plugging time of the candidate operation mode;
optimizing projection direction parameters and polynomial parameters in the Hermite polynomial projection pursuit regression model by adopting a projection pursuit learning network based on the projection index function to obtain optimal projection directions and optimal polynomial parameters;
and acquiring the prediction time of the iron blocking port by utilizing the Hermite polynomial projection pursuit regression model based on the optimal projection direction and the optimal polynomial parameters.
5. The intelligent determination method for the iron-blocking time in the blast furnace tapping process according to claim 4, wherein the obtaining of the predicted time for the iron-blocking by the Hermite polynomial projection pursuit regression model based on the optimal projection direction and the optimal polynomial parameters further comprises:
and calculating the error between the predicted time of the iron blocking hole and the actual time of the iron blocking hole, if the error is larger than a preset error, increasing the number of ridge functions, and obtaining a Hermite polynomial projection pursuit regression model by reusing orthogonal Hermite polynomial fitting, and repeating the steps until the error is smaller than the preset error.
6. The intelligent determination method for the iron blocking time in the blast furnace tapping process according to claim 5, wherein the projection index function for determining the projection pursuit regression model is a projection index function for determining the projection pursuit regression model according to the actual value, the predicted value and the projection direction of the iron blocking time, and the projection index function is specifically:
wherein n is the number of samples, akFinding the projection direction of the regression model for the projection, s being the projection index function value, yiThe actual tap hole plugging time corresponding to the ith sample,
and predicting time for the iron blocking opening corresponding to the ith sample.
7. The method of claim 6, wherein establishing a library of good operating modes that characterize the condition parameters, operating parameters, and tap hole plugging times of the blast furnace smelting process comprises:
establishing an operation mode according to the corresponding relation of the condition parameters, the operation parameters and the iron blocking time in the blast furnace smelting process;
obtaining an operation mode space according to a space formed by operation modes corresponding to all possible conditions in the actual blast furnace tapping process;
establishing a working condition evaluation model, and evaluating condition parameters of the same operation mode and different tap hole plugging time corresponding to the operation parameters according to the working condition evaluation model so as to obtain excellent operation modes corresponding to each group of operation modes;
based on the operation mode space, a good operation mode library is obtained from a set of good operation modes under different condition parameters.
8. An intelligent forecasting system for the silicon content of blast furnace molten iron is characterized by comprising the following components:
memory, processor and computer program stored on the memory and executable on the processor, characterized in that the steps of the method according to any of the preceding claims 1 to 7 are implemented when the computer program is executed by the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010165654.5A CN111254243B (en) | 2020-03-11 | 2020-03-11 | Method and system for intelligently determining iron notch blocking time in blast furnace tapping process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010165654.5A CN111254243B (en) | 2020-03-11 | 2020-03-11 | Method and system for intelligently determining iron notch blocking time in blast furnace tapping process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111254243A true CN111254243A (en) | 2020-06-09 |
| CN111254243B CN111254243B (en) | 2020-11-24 |
Family
ID=70945801
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010165654.5A Active CN111254243B (en) | 2020-03-11 | 2020-03-11 | Method and system for intelligently determining iron notch blocking time in blast furnace tapping process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111254243B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112257590A (en) * | 2020-10-22 | 2021-01-22 | 中冶南方工程技术有限公司 | Automatic detection method and system for working state of blast furnace taphole and storage medium |
| CN112347893A (en) * | 2020-11-02 | 2021-02-09 | 深圳大学 | Model training method and device for video behavior recognition and computer equipment |
| CN113832276A (en) * | 2020-06-23 | 2021-12-24 | 宝山钢铁股份有限公司 | Automatic control method for intelligent iron blocking port of blast furnace tapping |
| CN114774605A (en) * | 2022-03-15 | 2022-07-22 | 中南大学 | Intelligent forecasting device for iron notch blocking time of blast furnace |
| CN115044725A (en) * | 2022-07-14 | 2022-09-13 | 广东韶钢松山股份有限公司 | Blast furnace taphole control method, device, equipment and storage medium |
| CN115595388A (en) * | 2022-10-14 | 2023-01-13 | 合肥视展光电科技有限公司（Cn） | Blast furnace tapping operation control method based on molten iron flowing state and electronic equipment |
| CN115652003A (en) * | 2022-09-06 | 2023-01-31 | 中南大学 | Blast furnace taphole plugging time online monitoring method and system based on two-stage classification |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH11222610A (en) * | 1998-02-06 | 1999-08-17 | Nkk Corp | Method for controlling furnace heat of blast furnace |
| CN1995401A (en) * | 2006-12-18 | 2007-07-11 | 冶金自动化研究设计院 | Intelligent diagnosis and determination support system for blast furnace |
| CN109815282A (en) * | 2018-12-03 | 2019-05-28 | 江苏省沙钢钢铁研究院有限公司 | Big data platform of iron-making system |
-
2020
- 2020-03-11 CN CN202010165654.5A patent/CN111254243B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH11222610A (en) * | 1998-02-06 | 1999-08-17 | Nkk Corp | Method for controlling furnace heat of blast furnace |
| CN1995401A (en) * | 2006-12-18 | 2007-07-11 | 冶金自动化研究设计院 | Intelligent diagnosis and determination support system for blast furnace |
| CN109815282A (en) * | 2018-12-03 | 2019-05-28 | 江苏省沙钢钢铁研究院有限公司 | Big data platform of iron-making system |
Non-Patent Citations (2)
| Title |
|---|
| 刘建华: ""基于操作模式的铜闪速熔炼过程工况迁移策略研究及应用"", 《中国博士学位论文全文数据库 工程科技I辑》 * |
| 顾东洋: ""基于数据驱动的高炉冶炼操作参数优化"", 《中国优秀硕士学位论文全文数据库 工程科技I辑》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113832276A (en) * | 2020-06-23 | 2021-12-24 | 宝山钢铁股份有限公司 | Automatic control method for intelligent iron blocking port of blast furnace tapping |
| CN113832276B (en) * | 2020-06-23 | 2022-12-16 | 宝山钢铁股份有限公司 | Automatic control method for intelligent iron blocking port of blast furnace tapping |
| CN112257590A (en) * | 2020-10-22 | 2021-01-22 | 中冶南方工程技术有限公司 | Automatic detection method and system for working state of blast furnace taphole and storage medium |
| CN112257590B (en) * | 2020-10-22 | 2023-08-01 | 中冶南方工程技术有限公司 | Automatic detection method, system and storage medium for working state of blast furnace tap hole |
| CN112347893A (en) * | 2020-11-02 | 2021-02-09 | 深圳大学 | Model training method and device for video behavior recognition and computer equipment |
| CN112347893B (en) * | 2020-11-02 | 2023-07-21 | 深圳大学 | Model training method and device for video behavior recognition and computer equipment |
| CN114774605A (en) * | 2022-03-15 | 2022-07-22 | 中南大学 | Intelligent forecasting device for iron notch blocking time of blast furnace |
| CN114774605B (en) * | 2022-03-15 | 2023-06-13 | 中南大学 | Intelligent forecasting device for blast furnace iron notch blocking time |
| CN115044725A (en) * | 2022-07-14 | 2022-09-13 | 广东韶钢松山股份有限公司 | Blast furnace taphole control method, device, equipment and storage medium |
| CN115652003A (en) * | 2022-09-06 | 2023-01-31 | 中南大学 | Blast furnace taphole plugging time online monitoring method and system based on two-stage classification |
| CN115652003B (en) * | 2022-09-06 | 2024-01-09 | 中南大学 | Blast furnace iron notch blocking time on-line monitoring method and system based on two-stage classification |
| CN115595388A (en) * | 2022-10-14 | 2023-01-13 | 合肥视展光电科技有限公司（Cn） | Blast furnace tapping operation control method based on molten iron flowing state and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111254243B (en) | 2020-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111254243B (en) | Method and system for intelligently determining iron notch blocking time in blast furnace tapping process | |
| CN110066895B (en) | Stacking-based blast furnace molten iron quality interval prediction method | |
| CN109935280B (en) | Blast furnace molten iron quality prediction system and method based on ensemble learning | |
| CN105807741B (en) | A kind of industrial process stream prediction technique | |
| CN116261690A (en) | Computer system and method for providing operating instructions for blast furnace thermal control | |
| CN113761787A (en) | Blast furnace molten iron silicon content online prediction method and system based on deep migration network | |
| AU2020102518A4 (en) | A method of gas sensor calibration based on linear optimization | |
| CN111006240B (en) | Biomass boiler furnace temperature and load prediction method | |
| CN106153550A (en) | Converter steel-smelting molten steel carbon content based on SVM online Real-time and Dynamic Detection method | |
| CN114678080A (en) | Converter end point phosphorus content prediction model, construction method and phosphorus content prediction method | |
| Cardoso et al. | Artificial neural networks for modelling and controlling the variables of a blast furnace | |
| Takalo-Mattila et al. | Explainable steel quality prediction system based on gradient boosting decision trees | |
| CN109992844A (en) | A kind of boiler flyash carbon content prediction technique based on ADQPSO-SVR model | |
| CN116085823A (en) | Boiler combustion control method and system | |
| Sun et al. | Design of an adaptive nonnegative garrote algorithm for multi-layer perceptron-based soft sensor | |
| CN114117919B (en) | Instant learning soft measurement modeling method based on sample collaborative representation | |
| CN114528764A (en) | Soft measurement modeling method and device based on integral optimization and instant learning | |
| CN107844679B (en) | Sectional type blast furnace molten iron silicon content prediction method and device | |
| CN117152102B (en) | Method and system for detecting working state of coke oven waste gas mound rod | |
| CN116861256A (en) | Furnace temperature prediction method, system, equipment and medium for solid waste incineration process | |
| CN114093433B (en) | Observer-based method and system for evaluating prediction precision of single-ton energy consumption in rectification process | |
| CN109858707A (en) | A kind of Ethylene Cracking Furnace Tubes intelligence coking diagnostic method based on ABC-ANFIS-CTF | |
| CN115512455A (en) | Heating furnace steel billet temperature prediction method integrating mechanism and model migration | |
| CN116324323A (en) | Method for generating learned prediction model for predicting energy efficiency of melting furnace, method for predicting energy efficiency of melting furnace, and computer program | |
| Shen et al. | Data-driven adaptive improved just-in-time learning for industrial process soft sensor |
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 | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 410000 Hunan province Changsha Lushan Road No. 932 Patentee after: CENTRAL SOUTH University Patentee after: Hefei Jinxing Intelligent Control Technology Co., Ltd Address before: 410000 Hunan province Changsha Lushan Road No. 932 Patentee before: CENTRAL SOUTH University Patentee before: Hefei Jinxing Electromechanical Technology Development Co., Ltd |
|
| CP01 | Change in the name or title of a patent holder |