CN109759452B - Method for constructing steel rail specification adjustment quantity model - Google Patents
Method for constructing steel rail specification adjustment quantity model Download PDFInfo
- Publication number
- CN109759452B CN109759452B CN201910244235.8A CN201910244235A CN109759452B CN 109759452 B CN109759452 B CN 109759452B CN 201910244235 A CN201910244235 A CN 201910244235A CN 109759452 B CN109759452 B CN 109759452B
- Authority
- CN
- China
- Prior art keywords
- rolling
- reduction
- increment
- steel rail
- adjustment
- 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
Landscapes
- Metal Rolling (AREA)
- Control Of Metal Rolling (AREA)
Abstract
The invention relates to the field of steel rolling production, and discloses a method for constructing a steel rail specification adjustment model, which is used for improving the control precision of steel rail specification in the steel rail rolling process. Firstly, calculating the specification adjustment amount in the steel rail rolling process, and then calculating the rolling reduction increment of a roller in the steel rail rolling process by taking the difference value between the same parameters of the rolling processes of two steel rails before and after the non-roller-changing period as an independent variable; determining the type of the increment of the reduction amount which corresponds to the specification adjustment amount and meets the requirement of the correlation through correlation analysis; then, aiming at the specification adjustment quantity and the corresponding reduction increment category, a least square multiple regression algorithm is used for establishing a specification adjustment quantity-reduction increment model; and finally, carrying out steel rail rolling adjustment by using the established specification adjustment quantity-rolling reduction increment model. The invention is suitable for controlling the specification of the steel rail.
Description
Technical Field
The invention relates to the field of steel rolling production, in particular to a method for constructing a steel rail specification adjustment quantity model.
Background
At present, the method is still insufficient in the aspect of precision control of the final specification of the high-precision steel rail, a manual adjustment method is adopted, the adjustment process and the quality are different from person to person, the precision of the control size of the steel rail is also different from person to person, the adjustment difference on duty is large, the quality difference between shifts is large, and the improvement of the size precision of the steel rail is greatly influenced.
The steel rail adjusting model used in steel rolling production has the problems of multiple related factors, large cross influence among parameters, parameter accuracy and the like, so that the analysis results of equipment parameters and specification parameters have the phenomenon different from the actual deformation rule and the actual adjusting method. How to deal with such phenomena is the key to the success or failure of the model.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the method for constructing the steel rail specification adjustment model is used for improving the control precision of the steel rail specification in the steel rail rolling process.
In order to solve the problems, the invention adopts the technical scheme that: the method for constructing the steel rail specification adjustment quantity model comprises the following steps of:
A. calculating the rolling reduction increment of the roller in the rolling process of the steel rail by taking the difference value between the same parameters of the rolling processes of the two steel rails before and after the non-roller-changing period as an independent variable;
B. calculating the specification adjustment quantity of the front steel rail and the rear steel rail during the non-roll-changing period;
C. performing correlation analysis between the specification adjustment quantity and the rolling reduction increment, and determining the rolling reduction increment category corresponding to the specification adjustment quantity and meeting the correlation requirement;
D. aiming at the specification adjustment quantity and the corresponding reduction increment category, a steel rail specification adjustment quantity model, namely a specification adjustment quantity-reduction increment model, is established by using a least square multiple regression algorithm;
E. and D, utilizing the specification adjustment quantity-rolling reduction increment model established in the step D to perform steel rail rolling adjustment.
Further, the specification adjustment amount comprises an rail height adjustment amount, and the specification adjustment amount-rolling reduction increment model established in the step D comprises an rail height adjustment amount-rolling reduction increment model.
Further, the rail height adjustment amount-rolling reduction increment model is as follows:
△RH＝(C+M1*△Xil+M2*△Xir+M3*△Zjud+M4*△Fnl)/a+o
wherein △ RH is the rail height adjustment amount, C, M1、M2、M3、M4Respectively, the coefficient of regression, 1/a is the coefficient of natural cooling thermal contraction, o is the regression error, △ Xil is the increment of the left roll reduction of the ith pass of the universal roughing mill, △ Xir is the increment of the right roll reduction of the ith pass of the universal roughing mill, △ Zjud is the horizontal roll reduction increment for pass j of the edger, △ Fnl is the increment of the left roll reduction of the nth pass of the universal finishing mill.
The invention has the beneficial effects that: under the condition of production stress of the rolling mill, the phenomena of mechanical clearance, roller bounce and the like caused by roller assembly are caused, so that the direct modeling by using an actual value can bring about larger common interference.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
In the rolling process parameters of the universal rolling mill, the roll gap parameters are calibrated in a stressed state, and roll gap compensation of TCS is added, so that the roll gap precision is high theoretically, the original parameters of big data can be made, and the processing and analysis of the data show that the horizontal correction correlation is strong, the vertical roll correlation is weak, and even the parameter with the strongest relation with the deformation principle has no strong correlation. Through further analysis and investigation, the applicant finds that the zero deviation of the roll gap of the vertical roll before and after roll change is large and can reach 2mm generally, and the roll gap change in the roll change process covers the adjustment increment of the rolling reduction in the normal adjustment process, so that the data analysis cannot find the correlation. Under the condition of production stress of the rolling mill, the phenomena of mechanical clearance, roller bounce and the like caused by roller assembly exist, so that the direct modeling by using an actual value can bring larger common interference, the method can eliminate the common interference (including bearing clearance, roller bounce and the like) by adopting a data processing mode of difference increment, and the established rolling process model has higher precision.
In summary, the present invention provides a method for constructing a rail gauge adjustment model, the flow of which is shown in fig. 1, and the method specifically includes the following steps:
1. and calculating the rolling reduction increment of the roller in the rolling process of the steel rail by taking the difference value between the same parameters of the rolling processes of the two steel rails before and after the non-roller-changing period as an independent variable. For example, the pressing amount of the Y roller of the X stand during the first steel rolling is XY1+ SC+SDThe rolling reduction of the X stand and the Y roll in the second steel rolling is XY2+ TC+TDThe difference is △ XY (XY2-XY1) + (T)C-SC)+(TD-SD) Wherein T isC、SCIs a common-mode component in two steel reduction signals, TD、SDThe difference mode component in the two steel rolling reduction signals is the T mode component because the common mode component changes a little in the rolling stabilization section and the initial adjustment sectionC-SCWhen the reduction increment is 0, the reduction increment is only influenced by the fluctuation of a differential model signal, and compared with the method which directly uses the reduction modeling, the processing method greatly reduces the interference influence.
2. And calculating a difference value between certain specifications of the front steel rail and the rear steel rail during the non-roll changing period, namely specification adjustment quantity, wherein the specification adjustment quantity comprises specification adjustment quantities such as rail height, waist thickness, head width, bottom width and the like.
3. Performing correlation analysis between the specification adjustment quantity and the rolling reduction increment, and determining the rolling reduction increment category corresponding to the specification adjustment quantity and meeting the correlation requirement;
4. and aiming at the specification adjustment quantity and the corresponding rolling reduction increment category, a steel rail specification adjustment quantity model, namely a specification adjustment quantity-rolling reduction increment model, is established by using a least square multiple regression algorithm.
Due to some differential mode signals (T) during rollingD-SD) The interference will be more obvious, so when the algorithm is actually used, a group of steel rail data of adjacent production should be selected as much as possible for regression, and grouping regression should be performed if necessary.
Taking the incremental rolling reduction and the high specification adjustment of the finished rail as an example, the method is realized as follows:
setting the natural cooling thermal contraction coefficient as 1/a, the finished rail height as RH, and the horizontal rolling reduction of the i-th pass roughing mill frame as Xiud, left roll pressing amount Xil, the right roller pressing amount is Xir, the horizontal rolling pressing amount of the edge rolling machine frame of the jth pass is Zjud, the horizontal rolling reduction of the finishing stand of the nth pass is Fnud, left roll pressing amount is Fnl，
(1) The rail height adjustment △ RH and the rolling reduction increments △ Xud, △ Xl, △ Xr, △ Zud, △ Fud and △ Fl for each pass were calculated.
(2) Outliers outside t standard deviations are rejected, where t depends on the statistical distribution of the values of the parameters, and may be 3 if the distribution is more concentrated.
(3) Respectively carrying out correlation analysis between the increment of the rolling reduction and the gauge adjustment of the rail height, wherein the correlation analysis shows that △ RH and △ Xil、△Xir、△Zjud、△Fnl has strong correlation and the interference of multiple linear correlation is small.
(4) Establishing an adjustment model by using a least square multiple regression algorithm:
△RH＝(C+M1*△Xil+M2*△Xir+M3*△Zjud+M4*△Fnl)/a+o (1-1)
c, M in formula (1-1)1、M2、M3、M4As regression coefficient, o is regression error, △ Xil is the increase in the left-hand reduction of the roughing stand of the i-th pass, △ Xir is the right roll reduction increment of the universal roughing mill of the ith pass, △ Zjud is the horizontal roll reduction increment for the edger of pass j, △ Fnl is the left roll reduction increment of the finishing mill of the nth pass. The regression coefficient solving process is as follows:
substituting the rail height adjustment amount based on the actual data and the incremental data of the related rolling reduction amount into (1-1) forms the related variable C, M1、M2、M3、M4The system of multiple linear equations is such that o is in the formula (1-1)2Minimum, i.e. (△ RH- (C + M)1*△Xil+M2*△Xir+M3*△Zjud+M4*△Fnl)/a)2Should be minimal, its first derivative should be 0, thus let o2For each coefficient, the first order partial derivative is 0, and then each regression coefficient value can be obtained.
Similarly, the above method can also be used to establish a model of the adjustment between each reduction and other specifications.
And finally, in the process of rolling the steel rail, the specification of the steel rail can be controlled by using the specification adjustment quantity-rolling reduction increment model established in the step 4.
Claims (3)
1. The method for constructing the steel rail specification adjustment quantity model is characterized by comprising the following steps of:
A. calculating the rolling reduction increment of the roller in the rolling process of the steel rail by adopting the difference value between the same rolling reduction parameters of the two steel rails in the rolling process before and after the non-roller-changing period as an independent variable;
B. calculating the specification adjustment quantity of the front steel rail and the rear steel rail during the non-roll-changing period;
C. performing correlation analysis between the specification adjustment quantity and the rolling reduction increment, and determining the rolling reduction increment category corresponding to the specification adjustment quantity and meeting the correlation requirement;
D. aiming at the specification adjustment quantity and the corresponding reduction increment category, a least square multiple regression algorithm is used for establishing a specification adjustment quantity-reduction increment model;
E. and D, utilizing the specification adjustment quantity-rolling reduction increment model established in the step D to perform steel rail rolling adjustment.
2. The method of constructing a rail gauge adjustment model according to claim 1, wherein the gauge adjustment comprises a rail height adjustment and the gauge adjustment-reduction incremental model created in step D comprises a rail height adjustment-reduction incremental model.
3. The method for constructing a rail gauge adjustment model according to claim 2, wherein the rail height adjustment-rolling reduction increment model is:
△RH= (C+M1*△Xil+ M2*△Xir+ M3*△Zjud+ M4*△Fnl)/a +o ，
wherein △ RH is the rail height adjustment amount, C, M1、M2、M3、M4Respectively, the coefficient of regression, 1/a is the coefficient of natural cooling thermal contraction, o is the regression error, △ Xil is the increase in the rolling reduction of the left roll in the i-th pass of the roughing mill, △ Xir is the increase of the right roll reduction of the i-th pass of the roughing mill, △ Zjud is the water of the j pass of the edge rolling millFlat roll reduction increment, △ Fnl is the left roll reduction increment of the nth pass of the finishing mill.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910244235.8A CN109759452B (en) | 2019-03-28 | 2019-03-28 | Method for constructing steel rail specification adjustment quantity model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910244235.8A CN109759452B (en) | 2019-03-28 | 2019-03-28 | Method for constructing steel rail specification adjustment quantity model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109759452A CN109759452A (en) | 2019-05-17 |
| CN109759452B true CN109759452B (en) | 2020-06-09 |
Family
ID=66459920
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910244235.8A Active CN109759452B (en) | 2019-03-28 | 2019-03-28 | Method for constructing steel rail specification adjustment quantity model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109759452B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| PL442342A1 (en) * | 2022-09-21 | 2024-03-25 | Firma Codogni Spółka Jawna | Method of rolling balls |
| PL442341A1 (en) * | 2022-09-21 | 2024-03-25 | Firma Codogni Spółka Jawna | Method of rolling balls |
| PL442343A1 (en) * | 2022-09-21 | 2024-03-25 | Firma Codogni Spółka Jawna | Method of rolling balls |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101075124A (en) * | 2007-06-15 | 2007-11-21 | 武汉钢铁(集团)公司 | Stanermo wind-cooling linear fuzzy controlling method and system |
| CN102049420A (en) * | 2009-11-05 | 2011-05-11 | 刘斌 | Decision tree-based method for extracting key characteristic variables of finish rolling temperature control 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 |
| CN107999545A (en) * | 2017-11-30 | 2018-05-08 | 中冶南方工程技术有限公司 | Cold-rolling mill second flow method for controlling thickness and system based on System Discrimination and parameter adaptive |
| CN108580560A (en) * | 2018-06-14 | 2018-09-28 | 攀钢集团攀枝花钢钒有限公司 | The method for building rail profile specification adjustment optimizing model |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100348060B1 (en) * | 1997-10-31 | 2002-10-25 | 주식회사 포스코 | Method for controlling thickness of front end part of steel plate in plate rolling process |
-
2019
- 2019-03-28 CN CN201910244235.8A patent/CN109759452B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101075124A (en) * | 2007-06-15 | 2007-11-21 | 武汉钢铁(集团)公司 | Stanermo wind-cooling linear fuzzy controlling method and system |
| CN102049420A (en) * | 2009-11-05 | 2011-05-11 | 刘斌 | Decision tree-based method for extracting key characteristic variables of finish rolling temperature control 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 |
| CN107999545A (en) * | 2017-11-30 | 2018-05-08 | 中冶南方工程技术有限公司 | Cold-rolling mill second flow method for controlling thickness and system based on System Discrimination and parameter adaptive |
| CN108580560A (en) * | 2018-06-14 | 2018-09-28 | 攀钢集团攀枝花钢钒有限公司 | The method for building rail profile specification adjustment optimizing model |
Non-Patent Citations (1)
| Title |
|---|
| 万能轧制线高速钢轨轧制参数优化模型研究;陶功明等;《中国机械工程》;20100531;第21卷(第10期);1200-1202,1207 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109759452A (en) | 2019-05-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109759452B (en) | Method for constructing steel rail specification adjustment quantity model | |
| CN103934278B (en) | A kind of hot fine rolling band steel method for controlling thickness | |
| CN101829687B (en) | Strip steel finishing mill roll gap control method for eliminating specification change influence | |
| CN103752623B (en) | Improve the autocontrol method of camber of intermediate slab in rough rolling | |
| EP3278889A1 (en) | Rolling method for boards with different longitudinal thicknesses | |
| CN109821909B (en) | Method for controlling thickness deviation of two sides of wide and thick plate | |
| CN103506404A (en) | Control method for roll gap in precision rolling process of strip steel | |
| CN103831304B (en) | A kind of hot continuous rolling intermediate blank target width computational methods and system | |
| CN110851994B (en) | Online shape target curve adjusting method for cold-rolled strip | |
| CN107442577A (en) | A kind of fine-rolling strip steel sharing of load establishing method | |
| CN112916624A (en) | Method for obtaining regulation and control efficiency coefficient of plate-shaped execution mechanism of UCM rolling mill | |
| CN111842507A (en) | Method for realizing deviation control of center line of plate blank | |
| CN102601124B (en) | Method for controlling bottom width full-length fluctuation of steel rail | |
| CN116020885A (en) | Prediction method for hot continuous rolling finish rolling force of composite plate | |
| CN117696635A (en) | Control method and system for reducing sizing mill rolling | |
| CN114634294B (en) | Automatic substrate glass thickness adjusting method and system | |
| CN115007656B (en) | Setting method of normalized plate-shaped target curve | |
| CN115106384B (en) | Thick plate rolling roll gap correction method based on random forest | |
| CN112547809B (en) | Method for improving setting precision of roll gap of rolling mill | |
| CN115582439A (en) | Self-adaptive rough rolling adjustment method and system | |
| CN112329198A (en) | Data-driven method for optimizing length of wide and thick plate in scale-up mode | |
| CN106984650B (en) | The method for controlling thickness of aluminum and Aluminum Alloy Plate | |
| CN114393033B (en) | Rolling method for rolling 0.24mm ultrathin material by acid continuous rolling unit | |
| CN101745542A (en) | Nonlinear speed compensation method for thickness control of irreversible aluminum cold rolling machine | |
| CN111046508B (en) | Plate-shaped load distribution weight calculation method based on shortest path |
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 |