CN112668133A - Automatic determination method for cross section of hot-rolled strip steel based on big data analysis - Google Patents
Automatic determination method for cross section of hot-rolled strip steel based on big data analysis Download PDFInfo
- Publication number
- CN112668133A CN112668133A CN201910937298.1A CN201910937298A CN112668133A CN 112668133 A CN112668133 A CN 112668133A CN 201910937298 A CN201910937298 A CN 201910937298A CN 112668133 A CN112668133 A CN 112668133A
- Authority
- CN
- China
- Prior art keywords
- seq
- strip steel
- cross
- section
- curve
- 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.)
- Pending
Links
Images
Abstract
The invention relates to a hot-rolled strip steel cross section automatic judgment method based on big data analysis, which comprises the following steps: step 1: periodically sampling the thickness of the cross sections of the hot-rolled strip steel at different positions; step 2: dividing all cross section curves of the single-coil strip steel into curve categories according to the head, middle front, middle rear and tail parts of the strip steel; and step 3: after classification of the head, the middle and the tail of all cross section curves of the strip steel is finished, centering the curves of all parts by using a central line, and constructing a two-dimensional array by each coordinate; and 4, step 4: and designing an automatic judging system module for the cross section of the hot-rolled strip steel, and automatically judging four curves of a head curve, a middle curve, a front curve, a middle curve, a rear curve and a tail curve. The method fully considers the detection flow and the real requirement of the production link of a steel mill on the cross section outline of the strip steel, has essential efficiency improvement compared with the traditional manual online judgment method, and provides an effective means for cross section quality evaluation and process improvement decision of the strip steel finished product.
Description
Technical Field
The invention relates to a judgment method, in particular to a hot-rolled strip steel cross section automatic judgment method based on big data analysis, and belongs to the technical field of big data analysis strip steel automatic judgment.
Background
The strip steel is a main product of a rolling production line and has wide application field. In the actual production process of hot rolling, the change of the thickness of the strip steel along the width direction is usually described by using the cross section shape, and the change of the cross section shape (the transverse thickness distribution) shows the characteristics that the middle part is gentle and the reduction is quicker as the section is closer to the edge part because of the factors of small metal transverse flow resistance in the edge part area of the strip steel, elastic flattening jump transition of the roll surface of a working roll, roll deflection and the like. The method adopts cross section quality data of a 1780 hot rolling production line, the abnormal cross section of the production line is mainly characterized in that local high points and edge reverse warping exist on the edge of strip steel, and the cross section is mostly asymmetrically rolled due to roller abrasion. Cross-sectional shape as one of the important factors for the control of the shape, shape defects generally cause the following problems:
1) influencing the self plate thickness and the transverse distribution of internal stress of the rolled strip steel;
2) causing the generation of high-order wave shape of hot-rolled thin specification products (below 2.75 mm);
3) meanwhile, the hot rolled product with the cross section defect as a cold rolled material affects the cold rolling production and causes 'ribbing' and local wave shape in severe cases.
The automatic determination of the cross section of the hot-rolled strip steel can provide an accurate and efficient determination method, and meanwhile, an operator can perform subsequent process adjustment according to the produced defective strip steel in time. At present, the cross section quality judgment based on curve identification is still in an artificial 24-hour real-time online monitoring and analyzing stage, and for a strip steel finished product, the method has the following defects:
1) for quality inspection personnel, the method does not only need rich judgment experience, but also needs a large amount of manpower and material resources for enterprises;
2) a lot of repetitive work is carried out, and the characteristics of lag and poor accuracy exist;
3) meanwhile, the quality inspection departments have different groups and groups with different experience standards, which causes different judgment results of the cross section quality of the finished strip steel products, so that the missed inspection rate is high.
The curve identification process participated by quality testing personnel through experience knowledge can be automatically and accurately completed through a computer, so that less humanization or even no humanization is realized, the labor efficiency is improved, and the research significance is profound.
At present, the research on the automatic determination method of the cross section quality of the hot rolled strip steel is less at home and abroad, and mainly aims to analyze the defects of the cross section of the hot rolled strip steel and evaluate various factors influencing the cross section shape through computer simulation, such as evaluation and control of the cross section shape of the hot rolled strip steel (steel, 2011, 46 th volume, 09 th period), but the method cannot be used as reference for the quality of steel in actual production, and needs manual judgment in the production process. In the patent, local high points of a strip steel cross section profile defect are researched, and are analyzed through curve fitting, but the method has high requirements on detection personnel, cannot be really applied to actual production, and cannot improve the production efficiency. In the patent "a method for detecting the shape of a steel plate by using a computer secondary data processing system", according to the type data of the steel plate manually input and the type, the plate convexity and the wedge value are obtained through computer calculation, so that field technicians can conveniently evaluate the type, but the method is not suitable for bottom layer workers, the interface is complex, and the operation is not easy, so that a new scheme is urgently needed to solve the technical problems.
Disclosure of Invention
The invention provides a hot-rolled strip steel cross section automatic judging method based on big data analysis aiming at the problems in the prior art, the technical scheme provides a cross section automatic judging method based on big data analysis, after cross section contour curves of different positions of strip steel are obtained, the curve types are divided according to four parts of a strip steel head, a middle front part, a middle back part and a tail part, then a two-dimensional array is constructed aiming at each part of the curve centering by a central line, an average value is made from the middle to the corresponding positions on two sides to form four new curves of the head, the middle front part, the middle back part and the tail part, and finally, the automatic judgment is realized according to the requirements of customized rules.
In order to achieve the purpose, the technical scheme of the invention is that the method for automatically judging the cross section of the hot-rolled strip steel based on big data analysis comprises the following steps:
step 1: periodically sampling the thickness of the cross sections of the hot-rolled strip steel at different positions;
step 2: dividing all cross section curves of the single-coil strip steel into curve categories according to the head, middle front, middle rear and tail parts of the strip steel;
and step 3: after classification of the head, the middle and the tail of all cross section curves of the strip steel is finished, centering each curve by a central line aiming at each part, constructing a two-dimensional array by each coordinate, taking an average value from the middle to the corresponding position of two sides, drawing a line on a main picture according to the average value, and thus obtaining four curves of the head, the middle front, the middle rear and the tail according to a distance equal division principle;
and 4, step 4: designing an automatic judging system module for the cross section of the hot rolled strip steel, setting corresponding upper and lower limit lines for the requirements of the cross section of the hot rolled strip steel, and judging whether four curves of a head curve, a middle curve, a front curve, a middle curve, a rear curve and a tail curve exceed the limit lines to realize automatic judgment.
As an improvement of the invention, in the step 1, sampling is performed once every 5mm, the sampling quantity of the strip steel with different specifications is different, and if the sampling quantity is less, the sampling quantity is more than a few strip steel, and the sampling quantity is more than a hundred strip steel, and all cross section curve profiles of the single-roll strip steel are drawn.
As a modification of the present invention, the step 2: dividing all cross section curves of the single-coil strip steel into curve categories according to the head, middle front, middle rear and tail parts of the strip steel;
the partitioning algorithm is as follows: seq is the number of samples:
(1)seq％4＝1
head: line ═ 1/4 (seq-1)
Before the process is carried out: line2 ═ 2-Line1 (seq-1)
After neutralization: line3 ═ 3 [ ((seq-1)/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line3
(2)seq％4＝2
Head: line ═ 2/4
Before the process is carried out: line2 ═ 2/2-Line 1
After neutralization: line3 ═ 3 [ ((seq-2)/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line3
(3)seq％4＝3
Head: line ═ s (seq +1)/4
Before the process is carried out: line2 ═ 2-Line1 (seq +1)
After neutralization: line3 ═ 3 [ (seq +1)/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line3
(4) And (3) the rest:
head: line ═ seq/4
Before the process is carried out: line 2-eq/2-Line 1
After neutralization: line3 ═ 3 ═ seq/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line 3.
Compared with the prior art, the method has the following advantages that 1) the cross section automatic judgment based on the big data analysis provided by the technical scheme mainly has the following advantages:
1) the scheme combines different curve characteristics, and avoids the inaccuracy of single curve judgment;
2) the scheme judges four sections of the cross section, namely a head section, a middle section, a front section, a middle section, a rear section and a tail section, through processing big data and integrating potential influence factors as much as possible, and has higher accuracy and strong pertinence;
3) according to the scheme, only a certain part (such as the head) of the strip steel is subjected to defect judgment in a segmented mode, the strip steel can be positioned more accurately, judgment is not influenced by other parts, and targeted post-treatment is carried out on part of defects.
The method fully considers the detection flow and the real requirement of the production link of a steel mill on the cross section outline of the strip steel, has essential efficiency improvement compared with the traditional manual online judgment method, and provides an effective means for cross section quality evaluation and process improvement decision of the strip steel finished product. The accuracy is high, the accuracy can be accurate to 0.01um, and the accuracy is improved by 100 times compared with the accuracy of manual judgment of 1 um.
Drawings
FIG. 1 is a typical defect curve of a cross section of a strip steel under different thickness specifications;
FIG. 2 is a schematic view of a cross section of a finished strip;
fig. 3 is a flow chart of a cross section quality automatic determination method.
The specific implementation mode is as follows:
for the purpose of enhancing an understanding of the present invention, the present embodiment will be described in detail below with reference to the accompanying drawings.
Example 1: referring to fig. 1 to 3, a method for automatically determining a cross section of a hot rolled strip based on big data analysis, the method comprising the steps of:
step 1: periodically sampling the thickness of the cross sections of the hot-rolled strip steel at different positions;
step 2: dividing all cross section curves of the single-coil strip steel into curve categories according to the head, middle front, middle rear and tail parts of the strip steel;
and step 3: after classification of the head, the middle and the tail of all cross section curves of the strip steel is finished, centering each curve by a central line according to each part, constructing a two-dimensional array by each coordinate, and averaging corresponding positions from the middle to two sides to form four new curves of the head, the middle, the front, the middle, the rear and the tail;
and 4, step 4: and designing an automatic judging system module for the cross section of the hot-rolled strip steel to finish automatic judgment. The method comprises the steps of setting standard local high points, edge reverse warping and upper and lower deviation tolerance limit lines of section fall according to three different dimensions formed by application codes, user codes and special requirements, measuring judgment standards of cross sections by using data, establishing a set of standard section judgment reference library, bringing four curves collected in real time and obtained according to the step 3 into the set of reference library, and finally automatically judging four curves of head, middle front, middle back and tail according to the fact that the local high points, the edge reverse warping and the section fall curves exceeding the upper and lower deviation tolerance limit lines are not identified.
In the step 1, sampling is carried out once every 5mm, the sampling quantity of the strip steels with different specifications is different, if the sampling quantity is less, the sampling quantity is more than a few strip steels, and the cross section curve profiles of all the strip steels in a single roll are drawn;
the step 2: dividing all cross section curves of the single-coil strip steel into curve categories according to the head, middle front, middle rear and tail parts of the strip steel;
the partitioning algorithm is as follows: seq is the number of samples:
(1)seq％4＝1
head: line ═ 1/4 (seq-1)
Before the process is carried out: line2 ═ 2-Line1 (seq-1)
After neutralization: line3 ═ 3 [ ((seq-1)/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line3
(2)seq％4＝2
Head: line ═ 2/4
Before the process is carried out: line2 ═ 2/2-Line 1
After neutralization: line3 ═ 3 [ ((seq-2)/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line3
(3)seq％4＝3
Head: line ═ s (seq +1)/4
Before the process is carried out: line2 ═ 2-Line1 (seq +1)
After neutralization: line3 ═ 3 [ (seq +1)/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line3
(4) And (3) the rest:
head: line ═ seq/4
Before the process is carried out: line 2-eq/2-Line 1
After neutralization: line3 ═ 3 ═ seq/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line 3.
The application example is as follows:
the typical defect curve of the cross section of the steel strip under different thickness specifications is shown in figure 1. Fig. 2 is a schematic diagram of the cross section of the finished strip. The specific implementation mode of the patent is described by taking a finished steel strip product with the steel coil number of 29516501700, the thickness specification of 2.75mm, the width specification of 1483mm and the steel grade designation SPHC of a certain steel enterprise as an example.
Step 1: the thickness of the cross section of the hot rolled strip steel at different positions is periodically sampled (every 5 mm), the number of the strip steel samples in different specifications is different, and if the number of the strip steel samples is less, the number of the strip steel samples is more than hundreds. The data is stored in an Oracle database, the ordinate of the curve is stored in a Blob data block, and the following steps are carried out:
step 2: dividing all cross section curves of the single-coil strip steel into curve categories according to the head, middle front, middle rear and tail parts of the strip steel;
the partitioning algorithm is as follows: (seq is the number of samples)
(1)seq％4＝1
Head: line ═ 1/4 (seq-1)
Before the process is carried out: line2 ═ 2-Line1 (seq-1)
After neutralization: line3 ═ 3 [ ((seq-1)/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line3
(2)seq％4＝2
Head: line ═ 2/4
Before the process is carried out: line2 ═ 2/2-Line 1
After neutralization: line3 ═ 3 [ ((seq-2)/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line3
(3)seq％4＝3
Head: line ═ s (seq +1)/4
Before the process is carried out: line2 ═ 2-Line1 (seq +1)
After neutralization: line3 ═ 3 [ (seq +1)/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line3
(4) And (3) the rest:
head: line ═ seq/4
Before the process is carried out: line 2-eq/2-Line 1
After neutralization: line3 ═ 3 ═ seq/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line3
And step 3: after classification of the head, the middle and the tail of all cross section curves of the strip steel is finished, centering each curve by a central line aiming at each part, constructing a two-dimensional array by each coordinate, and taking an average value from the middle to corresponding positions on two sides to form four new curves of the head, the middle, the front, the middle, the rear and the tail;
and 4, step 4: and designing an automatic judging system module for the cross section of the hot-rolled strip steel, and automatically judging four curves of a head curve, a middle curve, a front curve, a middle curve, a rear curve and a tail curve.
The specific method comprises the following steps:
(1) and (3) formulating a judgment rule meeting the requirements of a user according to different cross section defects: section drop, local high point: comparing the extreme value difference in the local range of the curve; and (3) edge reverse tilting: defining the range of the edge part, and comparing the local fall from the outer side to the inner side;
(2) and (4) performing cross section rule setting on the strip steel according to specification, flow direction, steel type and even user, and finishing automatic judgment of very-cut surfaces according to the set rule. And then automatically identifying the defect reasons of the unqualified products, recording a blocking code, and simultaneously notifying an operator by means of voice alarm. The automatic judgment method for the cross section of the hot-rolled strip steel based on big data analysis effectively and accurately meets the automatic judgment requirement for the cross section of the hot-rolled strip steel, and the overall precision reaches over 95 percent and is higher than the actual manual judgment accuracy of production; meanwhile, through comparison of cross section judgment data of one quarter, the automatic judgment reject ratio is 2.75 percent and is higher than the manual judgment reject ratio by 1.03 percent, the subsequent omission factor causing quality objection is greatly reduced while the automatic judgment precision is ensured, and the cross section quality benefit is improved.
It should be noted that the above-mentioned embodiments are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention, and all equivalent substitutions or substitutions made on the basis of the above-mentioned technical solutions belong to the scope of the present invention.
Claims (3)
1. A method for automatically judging the cross section of a hot-rolled strip steel based on big data analysis is characterized by comprising the following steps:
step 1: periodically sampling the thickness of the cross sections of the hot-rolled strip steel at different positions;
step 2: dividing all cross section curves of the single-coil strip steel into curve categories according to the head, middle front, middle rear and tail parts of the strip steel;
and step 3: after classification of the head, the middle and the tail of all cross section curves of the strip steel is finished, centering each curve by a central line aiming at each part, constructing a two-dimensional array by each coordinate, taking an average value from the middle to the corresponding position of two sides, drawing a line on a main picture according to the average value, and thus obtaining four curves of the head, the middle front, the middle rear and the tail according to a distance equal division principle;
and 4, step 4: designing an automatic judging system module for the cross section of the hot rolled strip steel, setting corresponding upper and lower limit lines for the requirements of the cross section of the hot rolled strip steel, and judging whether four curves of a head curve, a middle curve, a front curve, a middle curve, a rear curve and a tail curve exceed the limit lines to realize automatic judgment.
2. The method according to claim 1, wherein the sampling is performed every 5mm in the step 1.
3. The method for automatically determining the cross section of a hot-rolled strip based on big data analysis according to claim 2, wherein the step 2: dividing all cross section curves of the single-coil strip steel into curve categories according to the head, middle front, middle rear and tail parts of the strip steel;
the partitioning algorithm is as follows: seq is the number of samples:
(1)seq％4＝1
head: line ═ 1/4 (seq-1)
Before the process is carried out: line2 ═ 2-Line1 (seq-1)
After neutralization: line3 ═ 3 [ ((seq-1)/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line3
(2)seq％4＝2
Head: line ═ 2/4
Before the process is carried out: line2 ═ 2/2-Line 1
After neutralization: line3 ═ 3 [ ((seq-2)/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line3
(3)seq％4＝3
Head: line ═ s (seq +1)/4
Before the process is carried out: line2 ═ 2-Line1 (seq +1)
After neutralization: line3 ═ 3 [ (seq +1)/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line3
(4) And (3) the rest:
head: line ═ seq/4
Before the process is carried out: line 2-eq/2-Line 1
After neutralization: line3 ═ 3 ═ seq/4) -Line1-Line2
Tail: line4 ═ seq-Line1-Line2-Line 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910937298.1A CN112668133A (en) | 2019-09-29 | 2019-09-29 | Automatic determination method for cross section of hot-rolled strip steel based on big data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910937298.1A CN112668133A (en) | 2019-09-29 | 2019-09-29 | Automatic determination method for cross section of hot-rolled strip steel based on big data analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112668133A true CN112668133A (en) | 2021-04-16 |
Family
ID=75399628
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910937298.1A Pending CN112668133A (en) | 2019-09-29 | 2019-09-29 | Automatic determination method for cross section of hot-rolled strip steel based on big data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112668133A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114858120A (en) * | 2022-04-15 | 2022-08-05 | 首钢集团有限公司 | Method and device for evaluating outline warping of strip steel plate |
| CN117113156A (en) * | 2023-10-20 | 2023-11-24 | 浙江鸿昌铝业有限公司 | Saw cutting section quality analysis method for aluminum profile |
-
2019
- 2019-09-29 CN CN201910937298.1A patent/CN112668133A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114858120A (en) * | 2022-04-15 | 2022-08-05 | 首钢集团有限公司 | Method and device for evaluating outline warping of strip steel plate |
| CN114858120B (en) * | 2022-04-15 | 2023-12-12 | 首钢集团有限公司 | Evaluation method and device for strip steel plate profile warping |
| CN117113156A (en) * | 2023-10-20 | 2023-11-24 | 浙江鸿昌铝业有限公司 | Saw cutting section quality analysis method for aluminum profile |
| CN117113156B (en) * | 2023-10-20 | 2024-01-09 | 浙江鸿昌铝业有限公司 | Saw cutting section quality analysis method for aluminum profile |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107321801B (en) | Hot-rolled strip steel judgment method and hot-rolled online quality judgment system | |
| CN105486831B (en) | A kind of coil of strip quality detecting system | |
| CN104751288A (en) | Segment-based multi-dimensional online quality evaluation system and method for steel coils | |
| CN112668133A (en) | Automatic determination method for cross section of hot-rolled strip steel based on big data analysis | |
| CN111080054B (en) | Automatic evaluation method and system for width quality of hot rolled strip steel | |
| CN107179749A (en) | Hot dip zinc product whole process method of quality control | |
| CN110989510A (en) | Hot galvanizing product full-process quality control and grade automatic judgment system | |
| CN105203732A (en) | Spring steel slab component segregation quantitative analysis method | |
| CN109877650B (en) | Method for predicting service life of bar shearing tool | |
| CN114565314A (en) | Hot rolled steel coil end face quality control system and method based on digital twinning | |
| Ma et al. | Influence of profile indicators of hot-rolled strip on transverse thickness difference of cold-rolled silicon steel | |
| CN104942027A (en) | Method for detecting steel plate type through computer second-level data processing system | |
| CN107677677A (en) | A kind of continuous casting slab segregation degree quantification characterizing method | |
| CN112181956B (en) | Data mining method based on hot rolling industry big data | |
| CN112246878B (en) | Thickness judgment system and judgment method for pickling coil hot rolling process | |
| CN114331195A (en) | Process curve risk evaluation method for influencing overall length quality of hot-rolled strip steel | |
| CN116433194B (en) | Centralized area management informatization system for railway tunnel steel member processing | |
| CN103008363B (en) | Quantitative analysis method for hot-rolled piece cambers | |
| CN111366702A (en) | Online intelligent accurate slitting system and online intelligent accurate slitting method for non-oriented silicon steel | |
| CN106651945B (en) | A kind of silicon steel section thickness difference quality measurement patterned area analysis method | |
| CN105195523A (en) | Method for improving calculation precision of temperature of intermediate billet in hot rolling | |
| CN105728476B (en) | Made-up belt camber testing and analysis system in the middle of hot rolling | |
| CN117554186B (en) | Machine vision-based quality detection and analysis method for small-sized concrete prefabricated part | |
| CN117314187A (en) | Performance evaluation method and system for seamless steel tube | |
| CN110070543B (en) | Grading system for full-automatic continuous casting billet center segregation based on digital image processing |
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 |