CN114309781B - Circle shear force forecasting method based on combination of big data and neural network - Google Patents

Circle shear force forecasting method based on combination of big data and neural network Download PDF

Info

Publication number
CN114309781B
CN114309781B CN202011056260.2A CN202011056260A CN114309781B CN 114309781 B CN114309781 B CN 114309781B CN 202011056260 A CN202011056260 A CN 202011056260A CN 114309781 B CN114309781 B CN 114309781B
Authority
CN
China
Prior art keywords
strip steel
coefficient
force
neural network
shearing
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
Application number
CN202011056260.2A
Other languages
Chinese (zh)
Other versions
CN114309781A (en
Inventor
王辉
付顺鸣
张光星
张亚林
夏惊秋
赵昀
高玉强
范群
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Baoshan Iron and Steel Co Ltd
Original Assignee
Baoshan Iron and Steel Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Baoshan Iron and Steel Co Ltd filed Critical Baoshan Iron and Steel Co Ltd
Priority to CN202011056260.2A priority Critical patent/CN114309781B/en
Publication of CN114309781A publication Critical patent/CN114309781A/en
Application granted granted Critical
Publication of CN114309781B publication Critical patent/CN114309781B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Investigating Strength Of Materials By Application Of Mechanical Stress (AREA)

Abstract

A circle shear force forecasting method based on the combination of big data and a neural network belongs to the field of control. Collecting circle shear cutterThe method comprises the steps of (1) establishing a disc shear force forecast database according to sequence related data; calculation of group i pure shear force P 1i Bending force P of sheared part of ith group of strip steel 2i And bending force P of the sheared part of the ith group of strip steel 2i Coefficient Z of (a) 1i The method comprises the steps of carrying out a first treatment on the surface of the Establishing a coefficient Z 1i A neural network model for forecasting parameters in a database with the shearing force of the circle shear; establishing a coefficient Z 1 Is a computational model of (a); and the forecasting of the shearing force P of the disc shear is realized. The method adopts a method of combining big data with a neural network by mining a coefficient Z 1 The method is internally connected with parameters such as strip steel edge wire width, strip steel thickness, strip steel strength, blade gap amount, blade overlapping amount, tension and the like, a network model is built among the parameters, accurate forecasting and shearing capability assessment of the disc shear force are realized, and expansion of the shearing product strength and specification is realized within the shearing capability range of the disc shear.

Description

Circle shear force forecasting method based on combination of big data and neural network
Technical Field
The invention belongs to the field of shear force control of disc shears, and particularly relates to a shear force forecasting method for a disc shear.
Background
With the rapid development of the steel industry to the high-speed and continuous direction, the disc shearing edge becomes one of the important links in the production of plate and strip products.
The shearing force parameter of the disc shear is the basis of shearing quality analysis and equipment capacity evaluation, the shearing force setting directly relates to the shearing quality of the edge part of the strip steel, and the unreasonable shearing force setting can cause burr defects on the edge part of the strip steel, so that the quality of a finished product is seriously affected; meanwhile, the shearing capacity of the equipment is directly determined by the shearing force, and the shearing force is a key point of whether the strength and specification of a sheared product can be expanded. Therefore, the accurate prediction of the shearing force of the circular shears plays an important role in controlling the quality of the edge part of the strip steel and the expansion of equipment capacity.
At present, for the calculation problem of the shearing force of the disc shear, the oblique shearing theory is mainly adopted, namely the shearing force is decomposed into two parts, namely the pure shearing force P 1 The principle of determination of pure shearing force is similar to that of oblique blade shears, as shown in fig. 1; secondly, bending force P of the sheared part of the strip steel 2 The shearing is generated along with the complex bending of the strip steel, and is more obvious particularly when the width of the sheared edge wire is narrower.
The pure shearing force P can be accurately calculated on the premise of knowing the thickness, the strength and the elongation of the strip steel 1 . But bending force P for sheared portion of strip 2 In terms of the coefficient Z therein 1 The coefficient Z is related to the ratio of the width to the thickness of the sheared strip 1 Can only rely on empirical tables or curves, as shown in figure 2.
Because the specifications of the field strip steel products are complex and changeable, the width of the cut edge wire is different, and the coefficient Z is the same 1 The accurate prediction of the shearing force of the disc shears cannot be realized only by relying on empirical values.
Disclosure of Invention
The invention aims to provide a disc shear force forecasting method based on combination of big data and a neural network. The method fully utilizes mass actual production data on site, adopts a method of combining big data with a neural network, and uses the mining coefficient Z 1 Internal connection with parameters such as strip steel edge wire width, strip steel thickness, strip steel strength, blade gap amount, blade overlapping amount, tension and the like is established, a network model is established among the parameters, accurate prediction of the shearing force of the disc shear is realized, and theoretical basis and technical support are provided for accurate setting of the shearing force of the disc shear; meanwhile, the shearing capacity of the disc shears can be evaluated, and the expansion of the strength and the specification of sheared products is realized within the shearing capacity range of the disc shears.
The technical scheme of the invention is as follows: the disc shear force forecasting method based on the combination of big data and a neural network is characterized by comprising the following steps of:
1) Constructing an automatic production data acquisition system, acquiring data related to a shearing process of the disc shear, and inputting the diameter D of the blade of the upper machine and the expected shearing mileage L of the blade;
2) Establishing a disc shear force forecast database; the shearing force forecasting database of the disc shear adds up m groups of data, and any group of serial numbers are marked as i, i is more than or equal to 1 and less than or equal to m;
3) Calculation of group i pure shear force P 1i ；
4) Calculating bending force P of sheared part of ith group of strip steel 2i ；
5) Calculating the bending force P of the sheared part of the ith group of strip steel 2i Coefficient Z of (a) 1i ；
6) Establishing a coefficient Z 1i A neural network model for forecasting parameters in a database with the shearing force of the circle shear;
7) Establishing a coefficient Z 1 Is a computational model of (a);
8) And the forecasting of the shearing force P of the disc shear is realized.
Specifically, the ith group of parameters in the disc shear force forecast database at least comprises: strip steel edge width d i Thickness h of strip steel i Strength sigma of strip steel bi Elongation delta of strip steel material i Gap amount xi between upper and lower blades i Overlap amount theta of upper and lower blades i Number of cut mileage L i Total shear force detection value P si ；
Wherein, the width d of the strip steel edge wire i Is half of the difference between the width of the strip steel at the inlet and the width of the strip steel at the outlet of the circular shear.
Specifically, the i-th group of pure shear forces P 1i The calculation is performed by the following way:
(1) Calculating unit shearing work gamma in i-th group of pure shearing force i ：
γ i ＝K 1 σ bi K 2 δ i ＝σ bi δ i ；
Wherein K is 1 And K is equal to 2 Is the conversion coefficientTaking K 1 K 2 ＝1；
(2) Calculation of group i pure shear force P 1i ：
Figure BDA0002710962540000021
Wherein α is the blade shearing angle, and is determined by the following formula:
Figure BDA0002710962540000022
wherein ε 0 For relative cut-in rate, ε is taken here 0 ＝1.25δ i 。
Specifically, the bending force P of the sheared part of the ith group of strip steel is calculated 2i The method comprises the following steps:
(1) Firstly, establishing a functional relation between a shear force detection value and a calculated value:
P si ＝λ(P 1i +P 2i )
where lambda is the blade passivation influence coefficient, considered as maximum 1.2,
Figure BDA0002710962540000031
(2) Calculating bending force P of sheared part of strip steel 2i ：
Figure BDA0002710962540000032
Specifically, the bending force P of the sheared part of the ith group of strip steel is calculated 2i Coefficient Z of (a) 1i Comprising the following steps:
from the following components
Figure BDA0002710962540000033
The method can obtain: />
Figure BDA0002710962540000034
Specifically, the establishment coefficient Z 1i The neural network model for predicting parameters in the database with the disc shear force comprises the following steps:
(1) Setting an input layer of the neural network model, wherein the parameters of the input layer at least comprise: strip steel edge width d i Thickness h of strip steel i Strength sigma of strip steel bi Gap amount xi between upper and lower blades i Overlap amount θ of upper and lower blades i ；
(2) Determining structural parameters of the neural network model, and formulating reasonable hidden layer and learning efficiency parameters;
(3) Setting a neural network model output layer as a coefficient Z 1i ；
(4) Formation coefficient Z 1i And a neural network model for predicting parameters in the database with the shearing force of the circle shear.
Further, the structural parameters of the neural network model include: the relative width quadratic term coefficient a of the sheared strip steel; the coefficient b of the primary term of the relative width of the sheared strip steel; a relative strip steel strength influence coefficient c; the relative gap amount influence coefficient d'; a relative overlap amount influence coefficient e; a constant term f;
the hidden layer is expressed as:
Figure BDA0002710962540000035
the learning efficiency parameters are as follows: taking n groups of hidden layer calculation results as one learning, and taking the ratio of the average value of the first n groups of hidden layer calculation results to the average value of the last n groups of hidden layer calculation results as a learning efficiency parameter.
Specifically, the establishment coefficient Z 1 Can be represented by the following formula:
Figure BDA0002710962540000036
further, at the coefficient Z 1 In the calculation model of (2), all parameter variables are the parameters of the strip steel to be sheared corresponding to the variables.
Further, the prediction of the disc shear force P is expressed by the following formula:
Figure BDA0002710962540000041
compared with the prior art, the invention has the advantages that:
1. by adopting a method of combining big data with a neural network, a network model is built by internal connection of excavation coefficient Z1 and parameters such as strip steel edge wire width, strip steel thickness, strip steel strength, blade gap amount, blade overlap amount, tension and the like, and a disc shear force forecasting method based on the combination of big data with the neural network is provided.
2. The method has the advantages that mass actual production data on site is fully utilized, meanwhile, the influence of blade passivation on shearing force is considered, the accurate prediction of the shearing force of the circular shears is realized, theoretical basis and technical support are provided for the accurate setting of the shearing force of the circular shears on site, the shearing quality of the edge of strip steel can be further improved, meanwhile, the shearing capacity of the circular shears can be evaluated, and the expansion of the strength and specification of sheared products in the shearing capacity range of the circular shears is realized.
Drawings
FIG. 1 is a schematic diagram of the pressure of a circle shear when shearing a strip;
FIG. 2 is a schematic diagram of the relative width of a sheared strip versus the coefficient Z1;
FIG. 3 is a schematic flow diagram of the overall method of the present invention;
fig. 4 is a schematic diagram of a neural network model of the present invention.
Detailed Description
The invention is further described below with reference to the drawings and examples.
The invention relates to a disc shear force forecasting method based on big data and a neural network, the general flow chart of which is shown in figure 3, and the technical scheme mainly comprises the following steps:
1. constructing an automatic production data acquisition system, acquiring data related to a shearing process of the disc shear, and inputting the diameter D of the blade of the upper machine and the expected shearing mileage L of the blade;
2. screening and processing related parameters according to a large amount of historical data, and establishing a disc shear force forecast database; in total, m groups of data are counted, any group of serial numbers are marked as i, i is more than or equal to 1 and less than or equal to m, and the i group of parameters comprise: strip steel edge width d i Thickness h of strip steel i Strength sigma of strip steel bi Elongation delta of strip steel material i Gap amount xi between upper and lower blades i Overlap amount theta of upper and lower blades i Number of cut mileage L i Total shear force detection value P si Equal parameters, wherein the width d of the strip steel edge wire i Is half of the difference between the width of the strip steel at the inlet and the width of the strip steel at the outlet of the circular shear.
3. Calculation of group i pure shear force P 1i ：
(1) Calculating unit shearing work gamma in i-th group of pure shearing force i ：
γ i ＝K 1 σ bi K 2 δ i ＝σ bi δ i
Wherein K is 1 And K is equal to 2 For conversion coefficient, take K 1 K 2 ＝1。
(2) Calculation of group i pure shear force P 1i ：
Figure BDA0002710962540000051
Wherein α is the blade shearing angle, and is determined by the following formula:
Figure BDA0002710962540000052
wherein ε 0 For relative cut-in rate, ε is taken here 0 ＝1.25δ i 。
4. Calculating bending force P of sheared part of ith group of strip steel 2i ：
(1) Firstly, establishing a functional relation between a shear force detection value and a calculated value:
P si ＝λ(P 1i +P 2i )
where lambda is the blade passivation influence coefficient, considered as maximum 1.2,
Figure BDA0002710962540000053
(2) Calculating bending force P of sheared part of strip steel 2i ：
Figure BDA0002710962540000054
5. Calculating the bending force P of the sheared part of the ith group of strip steel 2i Coefficient Z of (a) 1i ：
From the following components
Figure BDA0002710962540000055
The method can obtain:
Figure BDA0002710962540000056
6. establishing a coefficient Z 1i Neural network model (as shown in fig. 4) with parameters in the circle shear force forecast database:
(1) Setting an input layer of a neural network model, wherein parameters of the input layer comprise: strip steel edge width d i Thickness h of strip steel i Strength sigma of strip steel bi Gap amount xi between upper and lower blades i Overlap amount theta of upper and lower blades i And the like.
(2) And determining structural parameters of the neural network model, and formulating reasonable hidden layer and learning efficiency parameters.
The structural parameters include: the relative width quadratic term coefficient a of the sheared strip steel; the coefficient b of the primary term of the relative width of the sheared strip steel; a relative strip steel strength influence coefficient c; the relative gap amount influence coefficient d'; a relative overlap amount influence coefficient e; a constant term f.
The hidden layer is as follows:
Figure BDA0002710962540000061
the learning efficiency parameter: taking n groups of hidden layer calculation results as one learning, and taking the ratio of the average value of the first n groups of hidden layer calculation results to the average value of the last n groups of hidden layer calculation results as a learning efficiency parameter.
(3) Setting a neural network model output layer as a coefficient Z 1i 。
(4) Formation coefficient Z 1i And a neural network model for predicting parameters in the database with the shearing force of the circle shear.
7. Establishing a coefficient Z 1 Can be represented by the following formula:
Figure BDA0002710962540000062
it should be noted that, all parameter variables in the above formula are parameters of the strip steel to be sheared corresponding to the variables, and are not historical data.
8. The prediction of the shearing force P of the disc shear is realized:
Figure BDA0002710962540000063
examples
In order to further explain the application process of the related technology of the invention, a certain production line circle shear unit is taken as an example, and the application process of a circle shear force forecasting method based on the combination of big data and a neural network is described in detail:
1. and (3) constructing an automatic production data acquisition system, acquiring data related to the shearing process of the disc shear, inputting the diameter D=300 mm of the cutter blade of the upper machine, and the expected shearing mileage L=6000 km of the cutter blade.
2. According to a large amount of historical data, screening and processing related parameters, establishing a disc shear force forecast database, and recording 300 groups of data in total, wherein any group of serial numbers are i, i is more than or equal to 1 and less than or equal to 300, and the i group of parameters comprise: strip steel edge width d i Thickness h of strip steel i Strength sigma of strip steel bi Elongation delta of strip steel material i Clearance between upper and lower bladesQuantity xi i Overlap amount theta of upper and lower blades i Number of cut mileage L i Total shear force detection value P si Equal parameters, for example, the first 10 sets, table 1 shows the values of the relevant parameters of the first 10 sets.
Table 1 top 10 sets of related parameter tables
Figure BDA0002710962540000071
/>
3. Calculation of group i pure shear force P 1i The previous 10 groups are shown in Table 2, for example.
Table 2 top 10 sets of calculated values for pure shear force
Figure BDA0002710962540000072
4. Calculating bending force P of sheared part of ith group of strip steel 2i The previous 10 groups are shown in Table 3, for example.
TABLE 3 bending forces of the front 10 cut-out portions
Figure BDA0002710962540000073
5. Calculating bending force P of sheared part of ith group of strip steel 2i Medium coefficient Z 1i The previous 10 groups are shown in Table 4, for example.
TABLE 4 first 10 sets of coefficients Z 1i
Figure BDA0002710962540000074
6. Establishing a coefficient Z 1i Neural network model with parameters in the circle shear force forecast database:
(1) Setting an input layer of a neural network model, wherein parameters of the input layer comprise: strip steel edge width d i Thickness h of strip steel i Strength sigma of strip steel bi Gap amount xi between upper and lower blades i Overlap amount theta of upper and lower blades i And the like.
(2) And determining structural parameters of the neural network model, and formulating reasonable hidden layer and learning efficiency parameters.
(3) Setting a neural network model output layer as a coefficient Z 1i 。
(4) Formation coefficient Z 1i And a neural network model for predicting parameters in the database with the shearing force of the circle shear.
The neural network model of the invention is shown in fig. 4, and the main body of the neural network model is as follows:
the structural parameters include: the relative width quadratic term coefficient a of the sheared strip steel; the coefficient b of the primary term of the relative width of the sheared strip steel; a relative strip steel strength influence coefficient c; the relative gap amount influence coefficient d'; a relative overlap amount influence coefficient e; the constant term f is shown in table 5.
Table 5 list of structural parameters
Figure BDA0002710962540000081
Hidden layer:
Figure BDA0002710962540000082
learning efficiency parameters: taking 10 groups of hidden layer calculation results as one learning, and taking the ratio of the average value of the first 10 groups of hidden layer calculation results to the average value of the last 10 groups of hidden layer calculation results as a learning efficiency parameter.
Subsequently, a coefficient Z is established 1 Is shown in the following formula:
Figure BDA0002710962540000083
/>
finally, the prediction of the shearing force P of the disc shear is realized:
Figure BDA0002710962540000084
according to the technical scheme, a method of combining big data with a neural network is adopted, and through the internal connection of the excavation coefficient Z1 with parameters such as strip steel edge wire width, strip steel thickness, strip steel strength, blade gap amount, blade overlap amount, tension and the like, a network model is built, so that accurate prediction of the shearing force of the disc shear is realized, and theoretical basis and technical support are provided for accurate setting of the shearing force of the disc shear; meanwhile, the shearing capacity of the disc shears can be evaluated, and the expansion of the strength and the specification of sheared products is realized within the shearing capacity range of the disc shears.
According to the production experience and theoretical analysis of the disc shear unit, the scheme of the invention is feasible, can be further popularized to other similar units in China, is used for forecasting the shearing force of the disc shear, and has relatively wide popularization and application prospects.
The invention can be widely applied to the field of forecasting or evaluating the shearing capacity of the disc shears.

Claims (10)

1. A circle shear force forecasting method based on the combination of big data and a neural network is characterized in that:
1) Constructing an automatic production data acquisition system, acquiring data related to a shearing process of the disc shear, and inputting the diameter D of the blade of the upper machine and the expected shearing mileage L of the blade;
2) Establishing a disc shear force forecast database; the shearing force forecasting database of the disc shear adds up m groups of data, and any group of serial numbers are marked as i, i is more than or equal to 1 and less than or equal to m;
3) Calculation of group i pure shear force P 1i ；
4) Calculating bending force P of sheared part of ith group of strip steel 2i ；
5) Calculating the bending force P of the sheared part of the ith group of strip steel 2i Coefficient Z of (a) 1i ；
6) Establishing a coefficient Z 1i A neural network model for forecasting parameters in a database with the shearing force of the circle shear;
7) Establishing a coefficient Z 1 Is a computational model of (a);
8) And the forecasting of the shearing force P of the disc shear is realized.
2. According to claim 1The method is characterized in that the ith group of parameters in the disc shear force forecasting database at least comprises the following steps of: strip steel edge width d i Thickness h of strip steel i Strength sigma of strip steel bi Elongation delta of strip steel material i Gap amount xi between upper and lower blades i Overlap amount theta of upper and lower blades i Number of cut mileage L i Total shear force detection value P si ；
Wherein, the width d of the strip steel edge wire i Is half of the difference between the width of the strip steel at the inlet and the width of the strip steel at the outlet of the circular shear.
3. The method for predicting shearing force of circle shear based on big data and neural network as recited in claim 1, wherein said i-th set of pure shearing forces P 1i The calculation is performed by the following way:
(1) Calculating unit shearing work gamma in i-th group of pure shearing force i ：
γ i ＝K 1 σ bi K 2 δ i ＝σ bi δ i ；
Wherein K is 1 And K is equal to 2 For conversion coefficient, take K 1 K 2 ＝1；
(2) Calculation of group i pure shear force P 1i ：
Figure FDA0004086412830000011
Wherein α is the blade shearing angle, and is determined by the following formula:
Figure FDA0004086412830000012
wherein ε 0 For relative cut-in rate, ε is taken here 0 ＝1.25δ i ；
σ bi The strength of the strip steel; delta i The elongation percentage of the strip steel material; h is a i The thickness of the strip steel is the thickness of the strip steel; θ i Is the amount of blade overlap.
4. The method for predicting shearing force of circular shear based on big data and neural network as set forth in claim 1, wherein said calculating the bending force P of the sheared portion of the ith band steel 2i The method comprises the following steps:
(1) Firstly, establishing a functional relation between a shear force detection value and a calculated value:
P si ＝λ(P 1i +P 2i )
where lambda is the blade passivation influence coefficient, considered as maximum 1.2,
Figure FDA0004086412830000021
(2) Calculating bending force P of sheared part of strip steel 2i ：
Figure FDA0004086412830000022
Wherein L is i Is the number of cut-out mileage; p (P) si Is the total shear force detection value.
5. The method for predicting shearing force of circular shear based on big data and neural network as set forth in claim 1, wherein said calculating the bending force P of the sheared part of the ith band steel 2i Coefficient Z of (a) 1i Comprising the following steps:
from the following components
Figure FDA0004086412830000023
The method can obtain: />
Figure FDA0004086412830000024
Wherein alpha is the blade shearing angle; delta i Is the elongation of the strip steel material.
6. Pressing the buttonThe method for forecasting the shearing force of the circle shear based on the combination of big data and a neural network as claimed in claim 1, wherein the establishment coefficient Z 1i The neural network model for predicting parameters in the database with the disc shear force comprises the following steps:
(1) Setting an input layer of the neural network model, wherein the parameters of the input layer at least comprise: strip steel edge width d i Thickness h of strip steel i Strength sigma of strip steel bi Gap amount xi between upper and lower blades i Overlap amount θ of upper and lower blades i ；
(2) Determining structural parameters of the neural network model, and formulating reasonable hidden layer and learning efficiency parameters;
(3) Setting a neural network model output layer as a coefficient Z 1i ；
(4) Formation coefficient Z 1i And a neural network model for predicting parameters in the database with the shearing force of the circle shear.
7. The method for forecasting the shearing force of the circle shear based on the combination of big data and the neural network according to claim 6, wherein the structural parameters of the neural network model comprise: the relative width quadratic term coefficient a of the sheared strip steel; the coefficient b of the primary term of the relative width of the sheared strip steel; a relative strip steel strength influence coefficient c; the relative gap amount influence coefficient d'; a relative overlap amount influence coefficient e; a constant term f;
the hidden layer is expressed as:
Figure FDA0004086412830000025
wherein h is the thickness of the strip steel; d is the width of the strip steel edge wire; d' is a relative gap amount influence coefficient; sigma (sigma) b The strength of the strip steel; ζ is the blade clearance; θ is the amount of blade overlap; sigma is the strength of the strip steel;
the learning efficiency parameters are as follows: taking n groups of hidden layer calculation results as one learning, and taking the ratio of the average value of the first n groups of hidden layer calculation results to the average value of the last n groups of hidden layer calculation results as a learning efficiency parameter.
8. The method for forecasting the shearing force of the circle shear based on the combination of big data and the neural network as set forth in claim 1, wherein the establishment coefficient Z 1 Can be represented by the following formula:
Figure FDA0004086412830000031
wherein h is the thickness of the strip steel; d is the width of the strip steel edge wire; sigma (sigma) b The strength of the strip steel; ζ is the blade clearance; θ is the amount of blade overlap; sigma is the strength of the strip steel; a is the quadratic term coefficient of the relative width of the sheared strip steel; b is a coefficient of a primary term of the relative width of the sheared strip steel; c is the relative strip steel strength influence coefficient; d' is a relative gap amount influence coefficient; e is the relative overlap amount influence coefficient; f is a constant term.
9. The method for predicting shearing force of circle shear based on big data and neural network as recited in claim 8, wherein said coefficient Z 1 In the calculation model of (2), all parameter variables are the parameters of the strip steel to be sheared corresponding to the variables.
10. The method for forecasting the shearing force of the circle shear based on the combination of big data and the neural network according to claim 1, wherein the forecasting of the shearing force P of the circle shear is realized by the following formula:
Figure FDA0004086412830000032
wherein lambda is the blade passivation influence coefficient; h is the thickness of the strip steel; z is Z 1 Coefficients in bending force of the sheared part of the strip steel; delta is the elongation of the strip steel material; alpha is the blade shearing angle; sigma (sigma) b Is the strength of the strip steel.
CN202011056260.2A 2020-09-30 2020-09-30 Circle shear force forecasting method based on combination of big data and neural network Active CN114309781B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011056260.2A CN114309781B (en) 2020-09-30 2020-09-30 Circle shear force forecasting method based on combination of big data and neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011056260.2A CN114309781B (en) 2020-09-30 2020-09-30 Circle shear force forecasting method based on combination of big data and neural network

Publications (2)

Publication Number Publication Date
CN114309781A CN114309781A (en) 2022-04-12
CN114309781B true CN114309781B (en) 2023-05-09

Family

ID=81011348

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011056260.2A Active CN114309781B (en) 2020-09-30 2020-09-30 Circle shear force forecasting method based on combination of big data and neural network

Country Status (1)

Country Link
CN (1) CN114309781B (en)

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0710452B2 (en) * 1988-11-30 1995-02-08 日本鋼管株式会社 How to install a round blade shearing device
WO2011026190A1 (en) * 2009-09-03 2011-03-10 Outotec Oyj A method and apparatus for testing a shearing device
CN102825328B (en) * 2011-06-17 2014-12-24 宝山钢铁股份有限公司 Shearing control method of passive disc shear
CN106552963B (en) * 2015-09-30 2018-11-06 宝山钢铁股份有限公司 A kind of strip trimming cut burr control method suitable for rewinding unit circle shear
CN105728823B (en) * 2016-04-29 2017-08-29 燕山大学 Measure and lap comprehensive optimization method strip shear history intermediate gap
CN110576214A (en) * 2019-09-17 2019-12-17 攀钢集团西昌钢钒有限公司 trimming quality control method
CN111400870A (en) * 2020-02-25 2020-07-10 唐山钢铁集团有限责任公司 Heavy coil production line tension group model calculation method based on BP neural network

Also Published As

Publication number Publication date
CN114309781A (en) 2022-04-12

Similar Documents

Publication Publication Date Title
CN113333474B (en) Strip steel hot-rolled strip shape control method and system based on digital twinning
CN103761423B (en) A kind of hot rolled plate microstructure and property prediction method based on PSO-ELM
CN102513351B (en) Rolling method and device for strip steel tandem cold rolling
CN101607264A (en) A kind of periodic longitudinal variable-thickness strip, longitudinal variable-thickness sheet material and preparation method thereof
CN103586286B (en) Rolling schedule comprehensive optimization method for cold continuous rolling unit taking scratch prevention as objective
CN102825328B (en) Shearing control method of passive disc shear
CN102189237A (en) System integration technology for dynamic soft reduction of continuously-cast billet
CN107179749A (en) Hot dip zinc product whole process method of quality control
CN105363794B (en) A kind of finish rolling energy-saving control method based on mechanical properties forecast and required power model
CN103123483B (en) Rolling deformation resistance prediction system of new steel based on clustering analysis
CN1270838C (en) Automatic controlling technical parameter optimization method of metal plate rolling
CN106345823A (en) On-line real-time mechanical property prediction method based on hot rolled steel coil production processes
CN102722104A (en) Method for controlling acid-washing and edge-shearing quality of high-strength steel
CN106126941A (en) A kind of cold mill complex modeling power method
CN106424969A (en) Method for precisely predicting dynamic cutting force of plunge milling by taking cutter eccentricity into consideration
CN110287456A (en) Bulk lots volume rolled surface defect analysis method based on data mining
CN114309781B (en) Circle shear force forecasting method based on combination of big data and neural network
CN104307886B (en) A kind of large dead time AGC control method of composite rolling mill sandwich rolling
CN102688896A (en) Mechanism model-based virtual crown gauge setting method for four-roll cold continuous rolling machine set
CN101685302B (en) Method for calculating cold rolling material by using computer program
CN113118869B (en) Method and system for determining use mileage of cutting edge of circle shear and industrial control equipment
CN114386679A (en) Ball mill energy consumption state monitoring method based on ensemble learning
CN111318579B (en) Roughing camber control method based on data driving
CN109604709B (en) Method and device for judging abrasion state of trimming shears of continuous hot-dip galvanizing production line
CN114309780B (en) Disc shear force feedback control method based on big data theory

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