CN115591947B - Distributed regulation and control method for strip quality in continuous rolling process - Google Patents
Distributed regulation and control method for strip quality in continuous rolling process Download PDFInfo
- Publication number
- CN115591947B CN115591947B CN202211609053.4A CN202211609053A CN115591947B CN 115591947 B CN115591947 B CN 115591947B CN 202211609053 A CN202211609053 A CN 202211609053A CN 115591947 B CN115591947 B CN 115591947B
- Authority
- CN
- China
- Prior art keywords
- rack
- data set
- quality
- process variables
- variables
- 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.)
- Active
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B21—MECHANICAL METAL-WORKING WITHOUT ESSENTIALLY REMOVING MATERIAL; PUNCHING METAL
- B21B—ROLLING OF METAL
- B21B37/00—Control devices or methods specially adapted for metal-rolling mills or the work produced thereby
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B21—MECHANICAL METAL-WORKING WITHOUT ESSENTIALLY REMOVING MATERIAL; PUNCHING METAL
- B21B—ROLLING OF METAL
- B21B1/00—Metal-rolling methods or mills for making semi-finished products of solid or profiled cross-section; Sequence of operations in milling trains; Layout of rolling-mill plant, e.g. grouping of stands; Succession of passes or of sectional pass alternations
- B21B1/22—Metal-rolling methods or mills for making semi-finished products of solid or profiled cross-section; Sequence of operations in milling trains; Layout of rolling-mill plant, e.g. grouping of stands; Succession of passes or of sectional pass alternations for rolling plates, strips, bands or sheets of indefinite length
- B21B1/24—Metal-rolling methods or mills for making semi-finished products of solid or profiled cross-section; Sequence of operations in milling trains; Layout of rolling-mill plant, e.g. grouping of stands; Succession of passes or of sectional pass alternations for rolling plates, strips, bands or sheets of indefinite length in a continuous or semi-continuous process
- B21B1/26—Metal-rolling methods or mills for making semi-finished products of solid or profiled cross-section; Sequence of operations in milling trains; Layout of rolling-mill plant, e.g. grouping of stands; Succession of passes or of sectional pass alternations for rolling plates, strips, bands or sheets of indefinite length in a continuous or semi-continuous process by hot-rolling, e.g. Steckel hot mill
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Control Of Metal Rolling (AREA)
Abstract
The invention discloses a distributed regulation and control method for the quality of a strip in a continuous rolling process, which comprises the following specific steps: acquiring historical data of a finish rolling process, including process variables and quality indexes; dividing the finish rolling flow into 7 subsystems, and obtaining an optimal data set of process variables and quality indexes of the subsystems according to the maximum correlation and minimum redundancy screening principle; mapping the optimal data set to a high-dimensional feature space by adopting a mixed kernel function method; and constructing a multi-system distributed coordination control model according to the optimal data set mapped into the high-dimensional feature space, introducing a group intelligent optimization algorithm, and realizing coordination control among the racks by adopting a cyclic correction strategy. The invention fully utilizes the prior rolling information, introduces a group intelligent optimization algorithm and a cyclic correction strategy, improves the coordination among the racks and meets the requirement of distributed regulation and control.
Description
Technical Field
The invention belongs to the field of rolling technology and control science and engineering, and particularly relates to a distributed regulation and control method for the quality of a strip in a continuous rolling process.
Background
The traditional control method for the hot continuous rolling process is distributed control based on a PID (proportion integration differentiation) controller, cannot systematically consider the coupling effect between production line parts, and is not beneficial to globally realizing the performance optimization of closed-loop control. The hot continuous rolling process is a typical multivariable, strong-coupling and nonlinear industrial production process, and the quality requirement of the market on hot rolled strips is higher and higher along with the high-end and intelligent transformation of the manufacturing industry in China. The plate shape and the plate thickness are important quality indexes which are particularly concerned by users, the plate shape and the plate thickness influence each other in actual production, the control effect of a hot continuous rolling production line is nearly limited due to the development of industrialization and informatization, the pursuit of quality is endless, and some key problems related to a rolling process are not completely solved. In terms of quality control, there have been many reports on the control of thickness and strip shape, and furthermore, the distributed control research for hot continuous rolling lines is also tedious. Therefore, timely introduction of advanced control strategies and presentation of novel distributed control systems have important guiding significance for improving the robustness of the whole production process and improving the quality of hot rolled strips.
Disclosure of Invention
In order to solve the problems, the invention provides a strip quality distributed control method in a continuous rolling process, which improves the performance of a hot continuous rolling strip quality control system and improves the quality of a hot continuous rolling strip steel product to a certain extent.
In order to achieve the aim, the invention provides a strip quality distributed regulation and control method in a continuous rolling process, which comprises the following steps:
acquiring historical data of a finish rolling process, wherein the historical data comprises: process variables and quality indicators;
dividing the finish rolling flow into 7 subsystems, dividing the process variables into different subsystems respectively, and obtaining an optimal data set of the process variables and quality indexes in each subsystem according to a maximum correlation and minimum redundancy screening principle;
mapping the preferred data set to a high-dimensional feature space by adopting a mixed kernel function method;
constructing a multi-system distributed coordination control model according to the preferred data set in the high-dimensional feature space;
and introducing a group intelligent optimization algorithm based on the multi-system distributed coordination control model, and realizing coordination control among the racks by adopting a cyclic correction strategy.
Preferably, the obtained historical data of the finish rolling process is as follows: rolling the plate strips of the same steel type, different steel grades, the same steel grade, different steel grades and different plate thicknesses to generate data; the process variables comprise rolling force, roll gap value, rolling speed, roll bending force and roll shifting amount; the quality indexes comprise thickness, width of the plate strip, convexity of the plate strip and flatness of the plate strip.
Preferably, the 7 subsystems dividing the finish rolling flow include: the device comprises a first rack, a second rack, a third rack, a fourth rack, a fifth rack, a sixth rack and a seventh rack.
Preferably, the method for dividing the process variables into different subsystems comprises the following steps:
wherein the content of the first and second substances,a data set representing process variables of the process,a data set consisting of process variables for a first rack process,a data set consisting of process variables for the second rack,a data set consisting of process variables for the third rack,a data set consisting of process variables for the fourth rack,a data set consisting of process variables for the fifth rack,a data set consisting of process variables for the sixth rack,a data set consisting of process variables for the seventh rack;
wherein the content of the first and second substances,a data set representing the quality indicator is presented,and the thickness and the convexity of the finished product corresponding to the quality index data set are represented.
Preferably, the method for obtaining the optimal data set of the process variables and the quality indexes in each subsystem according to the maximum correlation and minimum redundancy screening principle comprises the following steps:
obtaining the incidence relation between the process variable and the quality index in each subsystem according to the maximum correlation screening principle;
obtaining the redundancy relation between the process variables and the quality indexes in each subsystem according to the minimum redundancy screening principle;
respectively introducing a joint measurement coefficient to the process variable and the quality index in each subsystem based on the incidence relation and the redundancy relation to obtain an optimal data set of the process variable in each subsystemAnd a preferred data set of said quality indicators;
The formula corresponding to the maximum correlation principle is as follows:in the formula (I), wherein,represented as a set of input variables,a set of output variables is represented that are,andare respectively asAndthe elements (A) and (B) in (B),is defined asAndmaximum information coefficient therebetween;
the expression of the incidence relation between the process variables and the quality indexes is as follows:
in the formula (I), wherein,representing a set of process variables for each subsystem,a set of quality indicators is represented, and,andare respectively asAndthe elements (A) and (B) in (B),is defined asAndmaximum information coefficient therebetween;
the formula corresponding to the minimum redundancy screening principle is as follows:in the formula (I), wherein,in order to input the set of variables,andis composed ofThe elements (A) and (B) in (B),is composed ofAndmaximum information coefficient therebetween;
the expression of the redundancy relationship between the process variables and the quality indexes is as follows:
,in the formula (I), wherein,represents a set of process control variables for each subsystem,a set of quality indicators is represented, and,andis thatThe elements (A) and (B) in (B),andis thatThe elements (A) and (B) in (B),represents the maximum information coefficient between elements;
preferably, the expression of the mixing kernel function is:
in the formula (I), the compound is shown in the specification,representing process variablesAndthe mixing kernel function of (a) is,indicating quality indexAndthe mixing kernel function of (a) is,andis at a subsystemThe process variables of (1) are set,in order to be a quality index,、、are parameters of the corresponding kernel function, wherein,is a bandwidth parameter, the value of which is greater than 0, and。
preferably, the method for constructing the multi-system distributed coordination control model according to the preferred data set in the high-dimensional feature space comprises the following steps:
in the formula (I), the compound is shown in the specification,is composed ofAnda combined input-output matrix, wherein,is a matrix composed of high-dimensional features whose process parameters are mapped to a high-dimensional space through a mixed kernel function,is a matrix composed of high-dimensional features which are mapped to a high-dimensional space through a mixed kernel function,for future output matrices, i.e. quality index matrices,for future input matrices, i.e. process variable matrices,andrespectively, are the control matrices of the corresponding matrices,there is an unknown disturbance between the process variable and the quality index.
Preferably, a group intelligent optimization algorithm is introduced, a cyclic correction strategy is adopted, and the method for realizing coordination control among the racks comprises the following steps:
setting a comprehensive objective function related to the thickness and the plate shape and the dimension of an individual, wherein the dimension of the individual is set to be 14 dimensions, namely the outlet thickness and the convexity of 7 racks;
optimizing the outlet thickness and the convexity of each rack by taking the comprehensive objective function as a reference, and obtaining the optimal outlet thickness and the optimal convexity value of each rack after optimizing;
checking whether the outlet thickness and convexity of the end frame meet the comprehensive objective function, if so, terminating iteration, and otherwise, continuing the iteration until the error is within an allowable range;
after the iteration is ended, according to the optimal outlet thickness and the optimal convexity value of each rack, the rolling force, the roll bending force and the roll shifting amount applied by each rack are reversely deduced by combining the multi-system distributed coordination control model, and the specific strokes of the hydraulic cylinder and the roll bending cylinder are reversely deduced by combining the specific numerical values of the rolling force, the roll bending force and the roll shifting amount, so that the regulation and control of the process variables of the corresponding process are completed;
after the regulation and control of the process variables of the corresponding process are finished, the deviation vectors existing in the measured values and the target values of the thickness and the convexity of the outlet of the finished product are obtained through the measurement of a multifunctional instrument(ii) a According to a regulation matrixCalculating the adjustment amount required for eliminating the corresponding deviation vector(ii) a The production process is regulated and controlled in a circulating and reciprocating manner by taking time as a variable, so that the load distribution of each rack and the dynamic coordination among the racks are realized;
the expression of the comprehensive objective function is as follows:wherein, in the process,indicating the best output in the future, i.e. the best control effect set by the control system, ofThe corresponding optimal outlet thickness in the quality index is represented,the corresponding optimal plate shape in the quality index is shown.
Compared with the prior art, the invention has the following advantages and technical effects:
the invention provides a distributed regulation and control method for strip quality in a continuous rolling process, which screens data by using acquired field data and adopting the maximum correlation and minimum redundancy principle, and maps process variables and quality indexes in the data into a high-dimensional characteristic space by using a mixed kernel function method to lay a foundation for a multi-system distributed coordination control model. An objective function related to the thickness and the plate shape is established through a designed multi-system distributed coordination control model, and the thickness and the convexity of each rack are further optimized by using a group intelligent optimization algorithm with the objective function as a reference. And after the iteration of the algorithm is terminated, reversely deducing the rolling force, the roll bending force and the roll shifting amount of each rack through the thickness and the convexity of each rack, and then compensating and correcting the rolling force, the roll bending force and the roll shifting amount of each rack by adopting a cyclic correction strategy to realize the load distribution of each rack and the dynamic coordination among the racks.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the application and, together with the description, serve to explain the application and are not intended to limit the application. In the drawings:
FIG. 1 is a control schematic of an embodiment of the present invention;
FIG. 2 is a graphical representation of interaction information between a subsystem and a controller in accordance with an embodiment of the present invention;
FIG. 3 is a schematic diagram of an optimization algorithm for group intelligence optimization according to an embodiment of the present invention;
fig. 4 is a schematic configuration diagram of a distributed system process automation level according to an embodiment of the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer-executable instructions and that, although a logical order is illustrated in the flowcharts, in some cases, the steps illustrated or described may be performed in an order different than here.
As shown in fig. 1-4, the invention provides a distributed regulation and control method for the quality of a strip in a continuous rolling process, which comprises the following steps:
s1: acquiring historical data of a finish rolling process, wherein the historical data comprises: process variables and quality indicators;
s2: dividing the finish rolling flow into 7 subsystems, dividing the process variables and the quality indexes into different subsystems respectively, and obtaining an optimal data set of the process variables and the quality indexes in each subsystem according to a maximum correlation and minimum redundancy screening principle;
s3: mapping the preferred data set to a high-dimensional feature space by adopting a hybrid kernel function method;
s4: constructing a multi-system distributed coordination control model according to the preferred data set in the high-dimensional feature space;
s5: and introducing a group intelligent optimization algorithm based on the multi-system distributed coordination control model, and realizing coordination control among the racks by adopting a cyclic correction strategy.
The acquiring of the historical data of the finish rolling process comprises the following steps:
s1.1, production data of the metallurgical industry, especially the strip steel production industry, has the characteristics of multiple isomerism, large volume, multiple variables and multiple granularities, data acquisition generally manages and configures the data by taking a real-time database as a basic data platform depending on an actual production field, and corresponding data are acquired by adopting a standard I/O driving interface provided by the real-time data platform based on data acquisition protocols such as OPC, ODBC and the like;
s1.2, rolling the data generated by the plate strips of the same steel type, different steel grades, the same steel grade, different steel grades and different plate thicknesses; the process variables comprise rolling force, roll gap value, rolling speed, roll bending force and roll shifting quantity, and the quality indexes comprise thickness, width of the plate strip, convexity of the plate strip and flatness of the plate strip.
The step of dividing the finish rolling flow into 7 subsystems comprises the following steps:
s2.1, a first rack, a second rack, a third rack, a fourth rack, a fifth rack, a sixth rack and a seventh rack;
s2.2, dividing the process variables into different subsystems to be represented as:,. Wherein the content of the first and second substances,a data set representing process variable data of a process,a data set consisting of process variables for a first rack process,a data set consisting of process variables for the second rack,a data set consisting of process variables for the third rack,a data set consisting of process variables for the fourth rack,a data set consisting of process variables for the fifth rack,a data set consisting of process variables for the sixth rack,a data set consisting of process variables for the seventh rack;representing a quality indicator data set.And the thickness and the convexity of the finished product corresponding to the quality index data set are represented.
S2.3, obtaining the optimal data set of the process variables and the quality indexes in the subsystem according to the maximum correlation and minimum redundancy screening principle, wherein the optimal data set comprises the following steps: the maximum correlation principle is used for establishing a strong correlation relationship between process variables and quality indexes, and the minimum redundancy is used for avoiding mutual influence between the process variables to the maximum extent;
wherein, the maximum correlation corresponding formula is:in the formula (I), wherein,represented as a set of input variables,a set of output variables is represented that are,andare respectively asAndthe elements (A) and (B) in (B),is defined asAndthe largest information coefficient in between. Migrating the above formula to the present invention is represented as:
in the formula (I), the compound is shown in the specification,representing a set of process variables for each subsystem,a set of quality indicators is represented, and,andare respectively asAndthe elements in (1), similarly,is defined asAndthe largest information coefficient in between. The invention establishes the correlation between the process variables and the quality indexes in a two-way manner, and is the basis for subsequently establishing a multi-system distributed coordination control model;
the formula corresponding to the minimum redundancy is:in the formula (I), wherein,in order to input the set of variables,andis composed ofThe elements (A) and (B) in (B),is composed ofAndthe largest information coefficient in between. Migrating the above formula to the present invention is represented as:
in the formula (I), the compound is shown in the specification,representing sets of process control variables for each subsystem,A set of quality indicators is represented, and,andis thatThe elements (A) and (B) in (B),andis thatThe elements (A) and (B) in (B),represents the maximum information coefficient between elements;
introducing joint metric coefficientsLet us orderBy definition, the stronger the correlation, the weaker the redundancy, and the better the matching degree of the process parameters and the quality indexes.
The method adopting the mixed kernel function integrates the preferred dataAndrespectively mapping to high-dimensional feature spaces, including:
s3.1, the mixed kernel function is an integrated form of a radial basis kernel function and a Sigmoid kernel function, and is expressed as follows:,in order to be a function of the mixing kernel,in order to be a function of the radial basis kernel,is Sigmoid coreThe function of the function is that of the function,the weight is represented, the mixed kernel function has stronger generalization capability compared with the kernel function, and the established multi-system distributed control model is more stable based on the data mapped to the high-dimensional space by the mixed kernel function. Migration to the present invention is represented as:
in the formula (I), the compound is shown in the specification,representing process variablesAndthe mixing kernel function of (a) is,indicating quality indexAndthe mixing kernel function of (a) is,andto be in a subsystemThe process variables of (1) are set,to be in a subsystemThe quality index of (1).、、Are all necessary parameters of the corresponding kernel function. Wherein the content of the first and second substances,is a bandwidth parameter, the value of which is greater than 0, and。
s3.2 method of Mixed Kernel function optimization of data setAndrespectively mapping to high-dimensional feature space to respectively obtainAnd。
the method for constructing the multi-system distributed coordination control model according to the preferred data set mapped to the high-dimensional feature space comprises the following steps:
s4.1, constructing a multi-system distributed coordination control model
In the formula (I), the compound is shown in the specification,is composed ofAnda combined input-output matrix, wherein,is a matrix composed of high-dimensional features which are mapped to a high-dimensional space by a mixed kernel function for process parameters,is a matrix composed of high-dimensional features which are mapped to a high-dimensional space through a mixed kernel function,for future output matrices, i.e. quality index matrices,for future input matrices, i.e. process variable matrices,andrespectively, a control matrix corresponding to the matrix,unknown disturbances existing between process variables and quality indexes;
further, the air conditioner is provided with a fan,indicating the best output in the future, i.e. set by the control systemThe best control effect. In the formula (I), the compound is shown in the specification,represents the optimal set of outputs for each subsystem in the future,andthe closer the control effect, the better. ThereinThe corresponding optimal outlet thickness in the quality index is represented,and (4) representing the corresponding optimal plate shape in the quality index. In the usual case of the use of a magnetic tape,the related measurement indexes comprise convexity and flatness, and the convexity is mainly considered in the invention.
S4.2, in conclusion, the comprehensive objective function concerning the thickness and the plate shape is set as follows:
the final actuator for adjusting the thickness and the plate shape is on the actions of the hydraulic cylinder and the roll bending cylinder. The process variables particularly relate to rolling force, roll bending force and roll shifting amount.
The method for realizing coordination control among the racks by adopting the loop correction strategy comprises the following steps:
s5.1, introducing a group intelligent optimization algorithm for optimization, wherein the optimization comprises the following steps: the group intelligent optimization algorithm abstracts the concrete activities of foraging, hunting and the like of organisms by means of the concept of bionics, the dimensionality of each individual in a population is the optimized variable number, the vector parameters in each dimensionality are variables to be optimized, in the embodiment, the dimensionality of each individual is set to be 14 dimensionalities, and the dimensionalities are respectively the outlet thickness and the convexity of 7 racks. And optimizing the outlet thickness and the convexity of each rack by taking the comprehensive objective function as a reference, obtaining the optimal outlet thickness and the optimal convexity value of each rack after optimization, checking whether the thickness and the convexity of the last rack meet the comprehensive objective function, meeting the condition and terminating iteration, and continuing iteration until the thickness and the convexity reach the error allowable range.
After the iteration is ended, according to the optimal outlet thickness and the optimal convexity value of each rack, a rolling force, a roll bending force and a roll shifting amount applied by each rack are reversely deduced by combining a multi-system distributed coordination control model, and the specific strokes of a hydraulic cylinder and a roll bending cylinder are reversely deduced by combining the specific numerical values of the rolling force, the roll bending force and the roll shifting amount, so that the regulation and control of corresponding process variables are completed;
s5.2, the circulation optimization strategy is to obtain deviation vectors of measured values and target values of the outlet thickness and the convexity of the finished product through measurement of a multifunctional instrument after the regulation and control of the corresponding variables are completedAccording to a regulatory matrixThe adjustment quantity required for eliminating the deviation can be obtained by fast corresponding calculation. The cycle optimization strategy takes time as a variable to carry out the regulation and control on the production process in a cycle-to-cycle manner, and realizes the load distribution of each rack and the dynamic coordination among the racks.
The above description is only for the preferred embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (3)
1. A distributed regulation and control method for the quality of a strip in a continuous rolling process is characterized by comprising the following steps:
obtaining historical data of a finish rolling process, wherein the historical data comprises: process variables and quality indicators;
dividing the finish rolling flow into 7 subsystems, dividing the process variables into different subsystems, and obtaining an optimal data set of the process variables and quality indexes in each subsystem according to a maximum correlation and minimum redundancy screening principle;
mapping the preferred data set to a high-dimensional feature space by adopting a mixed kernel function method;
constructing a multi-system distributed coordination control model according to the preferred data set in the high-dimensional feature space;
introducing a group intelligent optimization algorithm based on the multi-system distributed coordination control model, and realizing coordination control among the racks by adopting a cyclic correction strategy;
a group intelligent optimization algorithm is introduced, a cyclic correction strategy is adopted, and the method for realizing coordination control among the racks comprises the following steps:
setting a comprehensive objective function related to the thickness and the plate shape and the dimension of an individual, wherein the dimension of the individual is set to be 14 dimensions, namely the outlet thickness and the convexity of 7 racks;
optimizing the outlet thickness and the convexity of each rack by taking the comprehensive objective function as a reference, and obtaining the optimal outlet thickness and the optimal convexity value of each rack after optimization;
checking whether the outlet thickness and convexity of the end frame meet the comprehensive objective function, if so, terminating iteration, and otherwise, continuing the iteration until the error is within an allowable range;
after iteration is ended, according to the optimal outlet thickness and the optimal convexity value of each rack, the rolling force, the roll bending force and the roll shifting quantity applied by each rack are reversely deduced by combining the multi-system distributed coordination control model, and the specific strokes of the hydraulic cylinder and the roll bending cylinder are reversely deduced by combining the specific values of the rolling force, the roll bending force and the roll shifting quantity, so that the regulation and control of the process variables of the corresponding process are completed;
after the regulation and control of the process variable of the corresponding process are finished, the process is startedThe measured value of the outlet thickness and convexity of the finished product and the deviation vector existing in the target value are obtained by measurement of an excessive function instrument(ii) a According to a regulation matrixCalculating the adjustment amount required for eliminating the corresponding deviation vector(ii) a The production process is regulated and controlled in a circulating and reciprocating manner by taking time as a variable, so that the load distribution of each rack and the dynamic coordination among the racks are realized;
the expression of the comprehensive objective function is as follows:wherein, in the step (A),indicating the best output in the future, i.e. the best control effect set by the control system, ofThe corresponding optimum outlet thickness in the quality indicator is indicated,representing the corresponding optimal plate shape in the quality index;
the 7 subsystems that divide the finish rolling flow include: the device comprises a first rack, a second rack, a third rack, a fourth rack, a fifth rack, a sixth rack and a seventh rack;
the method for dividing the process variables into different subsystems comprises the following steps:
wherein the content of the first and second substances,a data set representing process variable data of a process,a data set consisting of process variables for a first rack process,a data set composed of process variables for the second rack,a data set consisting of process variables for the third rack,a data set consisting of process variables for the fourth rack,a data set of process variables for the fifth rack,a data set consisting of process variables for the sixth rack,a data set consisting of process variables for the seventh rack;
wherein the content of the first and second substances,a data set representing the quality indicator is presented,the thickness and the convexity of the finished product corresponding to the quality index data set are represented;
the method for obtaining the optimal data set of the process variables and the quality indexes in each subsystem according to the maximum correlation and minimum redundancy screening principle comprises the following steps:
obtaining the incidence relation between the process variable and the quality index in each subsystem according to the maximum correlation screening principle;
obtaining the redundancy relation between the process variables and the quality indexes in each subsystem according to the minimum redundancy screening principle;
respectively introducing a joint measurement coefficient to the process variable and the quality index in each subsystem based on the incidence relation and the redundancy relation to obtain an optimal data set of the process variable in each subsystemAnd a preferred data set of said quality indicators;
The formula corresponding to the maximum correlation screening principle is as follows:in the formula (I), wherein,represented as a set of input variables,a set of output variables is represented that are,andare respectively asAndthe elements in (A) and (B) are selected,is defined asAndmaximum information coefficient in between;
the expression of the incidence relation between the process variables and the quality indexes is as follows:
in the formula (I), wherein,representing a set of process variables for each subsystem,a set of quality indicators is represented, and,andare respectively asAndthe elements in (A) and (B) are selected,is defined asAndmaximum information coefficient of (2);
the formula corresponding to the minimum redundancy screening principle is as follows:in the formula (I), wherein,in order to input the set of variables,andis composed ofThe elements (A) and (B) in (B),is composed ofAndmaximum information coefficient in between;
the expression of the redundancy relationship between the process variables and the quality indexes is as follows:,in the formula (I), wherein,represents a set of process control variables for each subsystem,a set of quality indicators is represented, and,andis thatThe elements (A) and (B) in (B),andis thatThe elements (A) and (B) in (B),represents the maximum information coefficient between elements;
the expression of the mixing kernel function is as follows:
in the formula (I), the compound is shown in the specification,representing process variablesAndthe mixing kernel function of (a) is,indicating quality indexAndthe mixing kernel function of (a) is,andis at a subsystemThe process variables of (1) are set to,in order to be a quality index,are parameters of the corresponding kernel function, wherein,is a bandwidth parameter, the value of which is greater than 0, and。
2. the distributed regulation and control method for the quality of the strip in the continuous rolling process according to claim 1, wherein the obtained historical data of the finish rolling process is as follows: rolling the plate strips of the same steel type, different steel grades, the same steel grade, different steel grades and different plate thicknesses to generate data; the process variables comprise rolling force, roll gap value, rolling speed, roll bending force and roll shifting amount; the quality indexes comprise thickness, width of the plate strip, convexity of the plate strip and flatness of the plate strip.
3. The method for the distributed regulation and control of the plate strip quality in the continuous rolling process according to claim 1, wherein the method for constructing the multi-system distributed coordination control model according to the preferred data set in the high-dimensional characteristic space comprises the following steps:in the formula (I), the compound is shown in the specification,is composed ofAnda combined input-output matrix, wherein,is a matrix composed of high-dimensional features whose process parameters are mapped to a high-dimensional space through a mixed kernel function,is a matrix composed of high-dimensional features which are mapped to a high-dimensional space through a mixed kernel function,for future output matrices, i.e. quality index matrices,for future input matrices, i.e. process variable matrices,andrespectively, are the control matrices of the corresponding matrices,there is an unknown disturbance between the process variable and the quality index.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211609053.4A CN115591947B (en) | 2022-12-15 | 2022-12-15 | Distributed regulation and control method for strip quality in continuous rolling process |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211609053.4A CN115591947B (en) | 2022-12-15 | 2022-12-15 | Distributed regulation and control method for strip quality in continuous rolling process |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115591947A CN115591947A (en) | 2023-01-13 |
CN115591947B true CN115591947B (en) | 2023-03-17 |
Family
ID=84853818
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211609053.4A Active CN115591947B (en) | 2022-12-15 | 2022-12-15 | Distributed regulation and control method for strip quality in continuous rolling process |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115591947B (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2004076086A2 (en) * | 2003-02-25 | 2004-09-10 | Siemens Aktiengesellschaft | Method for regulating the temperature of a metal strip, especially for rolling a metal hot strip in a finishing train |
CN101003063A (en) * | 2006-01-18 | 2007-07-25 | 宝山钢铁股份有限公司 | Method for comprehensive control of elongation coefficient and plate shape of steel band during even rolling process |
CN102069094A (en) * | 2010-11-16 | 2011-05-25 | 北京首钢自动化信息技术有限公司 | Data mining-based plate shape control key process parameter optimization system |
CN103143574A (en) * | 2011-08-04 | 2013-06-12 | 上海交通大学 | Distributed prediction control system of band steel hot continuous rolling process |
CN103722022A (en) * | 2013-12-29 | 2014-04-16 | 北京首钢自动化信息技术有限公司 | Friction coefficient model optimizing system and method in rolling process |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5193066A (en) * | 1989-03-14 | 1993-03-09 | Kabushiki Kaisha Kobe Seiko Sho | Equipment for adjusting the shape of a running band-like or plate-like metal material in the width direction |
CN102941232B (en) * | 2012-11-12 | 2014-12-10 | 东北大学 | Method for controlling hot continuous rolling finish rolling process |
CN104985003A (en) * | 2015-06-03 | 2015-10-21 | 燕山大学 | Rolling process slip diagnosis and prediction method based on data driving |
EP3461567A1 (en) * | 2017-10-02 | 2019-04-03 | Primetals Technologies Germany GmbH | Flatness control with optimiser |
CN111209967A (en) * | 2020-01-08 | 2020-05-29 | 太原科技大学 | Rolling process plate convexity prediction method based on support vector machine |
CN111250548B (en) * | 2020-01-18 | 2021-11-12 | 太原科技大学 | Board convexity prediction method based on kernel partial least square combined support vector machine |
CN111389915B (en) * | 2020-03-24 | 2020-12-29 | 燕山大学 | Micro-control device for simulating electric heating field quantity change of plate strip |
CN114048592A (en) * | 2021-10-25 | 2022-02-15 | 北京科技大学 | Finish rolling whole-flow distributed operation performance evaluation and non-optimal reason tracing method |
-
2022
- 2022-12-15 CN CN202211609053.4A patent/CN115591947B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2004076086A2 (en) * | 2003-02-25 | 2004-09-10 | Siemens Aktiengesellschaft | Method for regulating the temperature of a metal strip, especially for rolling a metal hot strip in a finishing train |
CN101003063A (en) * | 2006-01-18 | 2007-07-25 | 宝山钢铁股份有限公司 | Method for comprehensive control of elongation coefficient and plate shape of steel band during even rolling process |
CN102069094A (en) * | 2010-11-16 | 2011-05-25 | 北京首钢自动化信息技术有限公司 | Data mining-based plate shape control key process parameter optimization system |
CN103143574A (en) * | 2011-08-04 | 2013-06-12 | 上海交通大学 | Distributed prediction control system of band steel hot continuous rolling process |
CN103722022A (en) * | 2013-12-29 | 2014-04-16 | 北京首钢自动化信息技术有限公司 | Friction coefficient model optimizing system and method in rolling process |
Non-Patent Citations (1)
Title |
---|
谭树彬 ; 刘建昌 ; .Multi-Agent的连续轧制过程控制系统研究.2008,(02),全文. * |
Also Published As
Publication number | Publication date |
---|---|
CN115591947A (en) | 2023-01-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111522233B (en) | Parameter self-tuning MIMO different factor full-format model-free control method | |
CN103440368A (en) | Multi-model dynamic soft measuring modeling method | |
CN103809472A (en) | Online integration of model-based optimization and model-less control | |
CN111522229B (en) | Parameter self-tuning MIMO different factor offset format model-free control method | |
Song et al. | Transverse thickness profile control of electrical steel in 6-high cold rolling mills based on the GA-PSO hybrid algorithm | |
Chen et al. | Input-output dynamic model for optimal environmental pollution control | |
CN109856973A (en) | A kind of Technique for Controlling Greenhouse Environment and system based on fuzzy neural network | |
CN113075881A (en) | Parameter adjusting method and device of PID controller and electronic equipment | |
Vogt et al. | On-line adaptation of grid-based look-up tables using a fast linear regression technique | |
CN115591947B (en) | Distributed regulation and control method for strip quality in continuous rolling process | |
CN107085371A (en) | Crude(oil)unit economic model forecast control method based on data-driven | |
Pan et al. | Optimization of industrial boiler combustion control system based on genetic algorithm | |
CN110298385A (en) | One kind is based on * information and the online incipient fault detection method of increment SVDD | |
CN117094608B (en) | Titanium belt production control method and system combining application requirements | |
CN112180733B (en) | Fuzzy logic-based building energy consumption system prediction control parameter setting method | |
Liu et al. | Flatness Prediction of Cold Rolled Strip Based on EM-TELM | |
CN111444619A (en) | Online analysis method and equipment for injection mold cooling system | |
CN109814389A (en) | The tight format non-model control method of the different factor of the MIMO of parameter self-tuning | |
CN113641101A (en) | Multi-channel pool control parameter optimization algorithm based on numerical simulation | |
Wang et al. | Hybrid end-point static control model for 80 tons BOF steelmaking | |
Shaikh et al. | Computational Analysis of PID and PSO-PID Optimization for MIMO Process Control System | |
CN109388851B (en) | Spiral bevel gear shape cooperative manufacturing intelligent parameter driving decision and processing method | |
Chu et al. | Decision-making for structural parameters of injection mold gating system based on agent model and intelligent algorithm | |
CN110538881A (en) | hot continuous rolling thickness control method based on improved internal mold controller | |
CN113848839B (en) | Double-layer structure economic performance improvement strategy based on iterative learning control |
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 |