CN101901280B - Method for obtaining broadening by using computer program - Google Patents
Method for obtaining broadening by using computer program Download PDFInfo
- Publication number
- CN101901280B CN101901280B CN200910057338.XA CN200910057338A CN101901280B CN 101901280 B CN101901280 B CN 101901280B CN 200910057338 A CN200910057338 A CN 200910057338A CN 101901280 B CN101901280 B CN 101901280B
- Authority
- CN
- China
- Prior art keywords
- data
- width
- computing machine
- mean value
- spreading
- 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
Abstract
The invention discloses a method for obtaining broadening by using a computer program, which comprises the following steps: step one: collecting a plurality of actually measured data by a computer, wherein each group of data comprises length, width, thickness, width of a finished product and thickness of a finished product; step two: obtaining monotonous curves relative to data d1 and data d2 for each group of data, having a sole intersection for any non-coincident two curves, and obtaining two groups of data i.e. the data d1 and the data d2; step three: calculating mean value and standard variance for the data d1and the data d2 by the computer, taking a data except a positive and negative standard variance of the mean value as an abnormal data, and removing the abnormal data; and taking the mean value of the remained data d1 and data d2 as a finally optimal result of the data d1 and the data d2 for outputting. The invention can avoid blindness of only depending on experiences and a trial and error method to adjust related coefficients of the broadening, so that the finally obtained related coefficients of the broadening approximates the optimization, thereby improving products yield.
Description
Technical field
The present invention relates to a kind of method of utilizing computer program to calculate steel plate width, particularly relate to a kind of method of utilizing computer program to obtain and spread.
Background technology
In Medium and Heavy Plate Rolling field, in process computer (L2) rolling model, spread model in occupation of consequence very.The result of calculation of spreading model directly can affect the result of calculation of load distribution model, has influence on finished width and the rate of becoming a useful person.If spread model, calculate accurately, for the product of fixed length and Fixed width rolling, can improve the rate of becoming a useful person more than 1%, economic benefit is considerable.
At present, the conventional control method that spreads model is according to the finished width of actual measurement, the related coefficient of spreading of spreading in model is finely tuned, thereby reach product requirement.This method has following drawback:
1) technologist of adjustment model must be experienced, the method that adopts examination to gather, according to finished product thickness and width to coefficient rapid adjustment;
2) relevant to thickness because spread related coefficient, specifically spread related coefficient and may not be applicable to all finished product thicknesses;
3) very loaded down with trivial details to spreading related coefficient optimization, the coefficient being optimized is very difficult.
Summary of the invention
Technical matters to be solved by this invention is to provide a kind of method of utilizing computer program to obtain and spread, it can avoid by experience and method of trial and error, regulating the blindness of spreading related coefficient merely, make the related coefficient of spreading finally obtaining approach optimum, improve product lumber recovery.
In order to solve above technical matters, the invention provides a kind of method of utilizing computer program to obtain and spread; Comprise the following steps: a plurality of measured datas of computer acquisition, the length and width of every group of packet rubbing board base, thick, finished width and finished product thickness;
h
i=h
i-1+step
Obtain the monotonous curve of d1 data and d2 data;
Wherein d1 and d2 are for spreading related coefficient, and h0 and h1 are respectively inlet thickness and exit thickness, and b is exit width, and i is data sequence number; Every group of data all can obtain one about the monotonous curve of d1 data and d2 data, and two monotonous curves of d1 data and d2 data have a unique intersection point, and computing machine obtains d1 data and two groups of data of d2 data,
Step 3, computing machine, to d1 data, d2 data calculating mean value and standard variance, are considered as abnormal data by the data outside a positive and negative standard variance of mean value, remove these abnormal datas;
Computing machine using the mean value of remaining d1 data, d2 data as d1 data, the output of the final optimization pass result of d2 data.
Beneficial effect of the present invention is, can avoid by experience and method of trial and error, regulating the blindness of spreading related coefficient merely, makes the related coefficient of spreading finally obtaining approach optimum, improves product lumber recovery.
The present invention applies actual measurement width and rules and obtains actual measurement and spread related coefficient (d1 data and d2 data, can be directly used in production), and a model can cover all product thicknesses and width.Meanwhile, final optimization pass precision of the present invention is higher.
After step 3, also comprise: step 4: after obtaining the spreading of different finished product thicknesses and width, if increase new measured data, computing machine adds new measured data in legacy data, to recalculate and optimize d1 data and d2 data, finally obtains stablizing d1 data and d2 data.
Also comprise: if inlet thickness changes, computing machine repeating step one to step 3 recalculates, obtain new d1 data and d2 data.
Accompanying drawing explanation
Below in conjunction with the drawings and specific embodiments, the present invention is described in further detail.
Fig. 1 is method block diagram described in the embodiment of the present invention;
Fig. 2 is two intersections of complex curve schematic diagram described in the embodiment of the present invention;
Fig. 3 spreads Calculation of correlation factor method block diagram described in the embodiment of the present invention;
Fig. 4 spreads related coefficient optimization method block diagram described in the embodiment of the present invention.
Embodiment
In Medium and Heavy Plate Rolling field, in process computer rolling model, spread model in occupation of consequence very.The result of calculation of spreading model directly can have influence on finished width and the rate of becoming a useful person.
The present invention has designed a kind of computer program that utilizes and has automatically calculated the method for spreading related coefficient.According to two different rules, can obtain one group of related coefficient of spreading that meets rules; Then investigate the product specification of all productions, according to the specification of grouping, process the some groups of related coefficients of spreading that meet rules, obtain covering the related coefficient of spreading of all finished product thicknesses.Adopt in this way, avoided simple and regulated the blindness of spreading related coefficient by experience and method of trial and error, make the related coefficient of spreading finally obtaining approach optimum.
The technical problem to be solved in the present invention is according to actual measurement finished width, accurately obtains spreading related coefficient d1 data and d2 data, and then can accurately obtain spreading of any width and thickness finished product, and process flow diagram is shown in Fig. 1, and concrete steps are as follows:
Step 1, gather some measured datas, the length and width of every group of packet rubbing board base, thick, finished width and finished product thickness, i.e. every group of corresponding rules of data, the finished width that these data covers are all and finished product thickness.
Computing machine passes through formula:
h
i=h
i-1+step
Obtain the monotonous curve of d1 data and d2 data; Formula (1)
Wherein d1 and d2 are for spreading related coefficient, and h0 and h1 are respectively inlet thickness and exit thickness, and b is exit width, and i is data sequence number.Utilize formula (1) can calculate and infer spreading of different finished product thicknesses and width, for instructing every group of data of actual production all can obtain one about the monotonous curve (seeing Fig. 1) of d1 data and d2 data.Two curves that do not overlap arbitrarily have unique (d1 data, a d2 data) point (seeing Fig. 2).Can adopt dichotomy or Newton method iteratively faster to obtain the numerical solution of intersection point, solve flow process and see Fig. 3.
In step 3, some measured datas, data can be determined (d1 data, a d2 data) point between two, thereby obtain series of points, i.e. two groups of data of d1 data and d2 data.Respectively these two groups of data are carried out to data-optimized processing: first to d1 data calculating mean value and standard variance, the data outside a positive and negative standard variance of mean value are considered as to abnormal data, remove these data; To remaining d1 data calculating mean value and standard variance again, the data outside a positive and negative standard variance of mean value are considered as to abnormal data, remove these data; Final optimization pass result output using the mean value of remaining d1 data as d1 data.In like manner, can obtain the final optimization pass result output of d2 data, idiographic flow is shown in Fig. 4.
After step 4, newly-increased measured data, can add and in legacy data, recalculate and optimize d1 data and d2 data, finally obtain stablizing d1 data and d2 data.If rolling mill practice changes, the thickness of iron scale changes (being the variation that has influence on inlet thickness in essence), likely can have influence on ceofficient of spread, causes spreading calculating occurring deviation, at this moment need the step above repetition to recalculate, obtain new d1 data and d2 data.
The present invention program can be applicable to use computer program to improve the computational accuracy of spreading model, and the rate thereby raising product is become a useful person, can be widely used in Heavy Plate Production.
The rule of spreading in the present invention adopts exwponential function approximation to describe (seeing formula 1), also can adopt polynomial method to describe, and the optimized coefficients finally obtaining is the coefficient in polynomial expression, and method is identical with the present invention.
The present invention is not limited to embodiment discussed above.Above the description of embodiment is intended in order to describe and illustrate the technical scheme the present invention relates to.Apparent conversion based on the present invention enlightenment or substitute and also should be considered to fall into protection scope of the present invention.Above embodiment is used for disclosing best implementation method of the present invention, so that the ordinary skill person of entering of this area can be applied numerous embodiments of the present invention and multiple alternative reaches object of the present invention.
Claims (3)
1. utilize computer program to obtain a method of spreading, it is characterized in that, comprise the following steps:
Step 1, a plurality of measured datas of computer acquisition, the length and width of every group of packet rubbing board base, thick, finished width and finished product thickness;
Step 2, computing machine pass through formula:
h=h
i-1+step
Obtain the monotonous curve of d1 data and d2 data;
Wherein d1 and d2 are for spreading related coefficient, and h0 and h1 are respectively inlet thickness and exit thickness, and b is exit width, and i is data sequence number; Every group of data all can obtain one about the monotonous curve of d1 data and d2 data, and two monotonous curves of d1 data and d2 data have a unique intersection point, and computing machine obtains d1 data and two groups of data of d2 data,
Step 3, computing machine, to d1 data, d2 data calculating mean value and standard variance, are considered as abnormal data by the data outside a positive and negative standard variance of mean value, remove these abnormal datas;
Computing machine using the mean value of remaining d1 data, d2 data as d1 data, the output of the final optimization pass result of d2 data.
2. the method for utilizing computer program to obtain and spread as claimed in claim 1, is characterized in that,
After step 3, also comprise:
Step 4: after obtaining the spreading of different finished product thicknesses and width, if increase new measured data, computing machine adds new measured data to recalculate in legacy data and optimize d1 data and d2 data, finally obtains stablizing d1 data and d2 data.
3. the method for utilizing computer program to obtain and spread as claimed in claim 1, is characterized in that, also comprises:
If inlet thickness changes, computing machine repeating step one to step 3 recalculates, and obtains new d1 data and d2 data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910057338.XA CN101901280B (en) | 2009-06-01 | 2009-06-01 | Method for obtaining broadening by using computer program |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910057338.XA CN101901280B (en) | 2009-06-01 | 2009-06-01 | Method for obtaining broadening by using computer program |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101901280A CN101901280A (en) | 2010-12-01 |
| CN101901280B true CN101901280B (en) | 2014-03-05 |
Family
ID=43226814
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200910057338.XA Active CN101901280B (en) | 2009-06-01 | 2009-06-01 | Method for obtaining broadening by using computer program |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101901280B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102873105B (en) * | 2012-10-18 | 2014-08-27 | 广西柳州银海铝业股份有限公司 | Prediction method for rectangular flat material angular rolling width |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1833790A (en) * | 2006-02-21 | 2006-09-20 | 新疆八一钢铁股份有限公司 | Slot-less rolling technique of rolling section |
| CN1959682A (en) * | 2005-10-31 | 2007-05-09 | 宝山钢铁股份有限公司 | Method for setting up optimized width of rough rolling strip steel |
| CN101376139A (en) * | 2007-08-31 | 2009-03-04 | 宝山钢铁股份有限公司 | Control method for producing conical plate blank using side compression machine of fixed width plate blank |
| CN101403890A (en) * | 2008-11-08 | 2009-04-08 | 山西太钢不锈钢股份有限公司 | Method for improving model prediction precision by utilizing neuroid classification modeling method |
-
2009
- 2009-06-01 CN CN200910057338.XA patent/CN101901280B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1959682A (en) * | 2005-10-31 | 2007-05-09 | 宝山钢铁股份有限公司 | Method for setting up optimized width of rough rolling strip steel |
| CN1833790A (en) * | 2006-02-21 | 2006-09-20 | 新疆八一钢铁股份有限公司 | Slot-less rolling technique of rolling section |
| CN101376139A (en) * | 2007-08-31 | 2009-03-04 | 宝山钢铁股份有限公司 | Control method for producing conical plate blank using side compression machine of fixed width plate blank |
| CN101403890A (en) * | 2008-11-08 | 2009-04-08 | 山西太钢不锈钢股份有限公司 | Method for improving model prediction precision by utilizing neuroid classification modeling method |
Non-Patent Citations (4)
| Title |
|---|
| N.A.Fleck等.TOWARDS A NEW THEORY OF COLD ROLLING THIN FOIL.《Int.J.Mech.Sci.》.1987,第29卷(第7期), |
| TOWARDS A NEW THEORY OF COLD ROLLING THIN FOIL;N.A.Fleck等;《Int.J.Mech.Sci.》;19871231;第29卷(第7期);第507-524页 * |
| 在棒材轧制中方-椭圆、方-菱、圆-椭圆孔型系的宽展和延伸的计算方法;斋藤好弘 等;《冶金丛刊》;19911231;第29-39页 * |
| 斋藤好弘 等.在棒材轧制中方-椭圆、方-菱、圆-椭圆孔型系的宽展和延伸的计算方法.《冶金丛刊》.1991, |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101901280A (en) | 2010-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103092078B (en) | Whole-process optimizing method of multi-stage batch production process | |
| CN102728624B (en) | Method for setting load distribution of finish rolling band steel | |
| Kisi et al. | Prediction of lateral outflow over triangular labyrinth side weirs under subcritical conditions using soft computing approaches | |
| CN103177289B (en) | Modeling method for noise-uncertainty complicated nonlinear dynamic system | |
| CN105512404B (en) | Time-varying reliability Global sensitivity analysis method based on chaos polynomial expansion | |
| CN103745273A (en) | Semiconductor fabrication process multi-performance prediction method | |
| Tan et al. | The Impact of the subsidy policy on total factor productivity: an empirical analysis of China's cotton production | |
| CN102553941B (en) | Off-line self-learning method of plate-shaped regulating efficiency coefficient of cold rolling mill | |
| CN104668313B (en) | Roller gap adjustment method of seven-roller thick plate straightener | |
| Wang et al. | Research on the improvement effect of high tension on flatness deviation in cold strip rolling | |
| CN106540968A (en) | The compensation method of cold rolled sheet shape measured value and device | |
| CN106326525B (en) | Hot rolling deformation drag forecasting procedure based on Generalized Additive Models | |
| CN104451118A (en) | Forecasting method for segmental evolution of strip steel shape in continuous annealing process | |
| CN102375721A (en) | Matrix multiplying method, graphic processor and electronic equipment | |
| CN101901280B (en) | Method for obtaining broadening by using computer program | |
| CN104923571B (en) | A kind of hot-rolling plate belt steel rolling code dynamic setting method | |
| CN101131584B (en) | Multi-level material balance data correction method for petroleum refinery | |
| CN105855298A (en) | Method for calculating hot rolling force of high-precision boron-containing steel | |
| CN101745548A (en) | Finite element correcting method of deformation resistance of non-oriented electrical steel in CSP (cast steel plate) production line | |
| CN101927268A (en) | Method for controlling thickness of tandem cold-rolled striped steel | |
| CN101690949A (en) | Tertiary panel-shape closed-loop control method | |
| CN105373672A (en) | Real-time encryption and decryption method for quadrilateral unit | |
| CN105226644A (en) | Based on the conforming belt restraining equivalence method of active volume | |
| CN112700050B (en) | Method and system for predicting ultra-short-term 1 st point power of photovoltaic power station | |
| CN102814341A (en) | Novel decoupling control method for plate strip thickness and plate type |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |